University of Southern California Dissertations and Theses

Investigating the modulatory mechanisms of ILC2s as a therapeutic strategy in the context of health and disease

(USC Thesis Other)

Investigating the modulatory mechanisms of ILC2s as a therapeutic strategy in the context of health and disease

PDF

Download a page range

Download transcript

Copy asset link

Request this asset

Transcript (if available)

Content

Investigating the Modulatory Mechanisms of ILC2s as a Therapeutic Strategy
in the Context of Health and Disease

by

Emily Howard

A Dissertation Presented to the
FACULTY OF THE USC GRADUATE SCHOOL
UNIVERSITY OF SOUTHERN CALIFORNIA
In Partial Fulfillment of the
Requirements for the Degree
DOCTOR OF PHILOSOPHY
(INFECTIOUS DISEASES, IMMUNOLOGY AND PATHOGENESIS)

December 2022

Copyright 2022 Emily Howard

ii

Acknowledgements

Thank you to my mentor Dr. Akbari for always supporting me. Thank you for the way that you
have encouraged me to take ownership of my projects, as it has helped make me confident to share
my ideas and data interpretation in rooms with scientists far senior to me. Thank you for the ways
you have challenged me to grow into the scientist I am today.

Thank you to the Akbari lab postdocs for their incredible mentorship, patience, and guidance. Dr.
Benjamin Hurrell, Dr. Lauriane Galle-Treger, and Dr. Georges Helou, thank you dearly for your
friendship, for the laughter, and for the endless editing help. Thank you for always pushing me to
be a better scientist and for the sure knowledge that you were always in my corner. I will forever
be grateful that our lab resembled a family more than a workplace.

Thank you to my family, mom and dad, Becca and Andrew. Thank you for continuing to travel
across the country to Los Angeles so I could show you the city I love. Thank you for your endless
support and for never caring what degree I have so long as I was happy.

Thank you to our dear friends, Audrey and Philip, who helped make Los Angeles our home. I
cannot express my gratitude for how you’ve walked this journey with me. Thank you to my church
family who made sure that I never felt alone in all my failures, insecurities, and celebrations.

Finally, thank you to my husband, Ian, for more than I could ever express here. Thank you for your
unwavering support, incredible love, and for never allowing me to doubt that I would make it to
this point. Thank you for the ways that our marriage and nightly sit-down dinners were a safe
space, away from the pressures of the lab. Thank you for the ways you showed me that my worth
was so much more than my papers or academic accomplishments.

iii
Table of Contents
Acknowledgments………………………………………………………………………………. ii
List of Figures…………………………………………………………………………………… iv
Abstract…………………………………………………………………………………………... vi
Chapter 1: Introduction…………………………………………………………………………… 1
Chapter 2: IL-10 production by ILC2s requires Blimp-1 and cMaf, modulates cellular
metabolism, and ameliorates airway hyperreactivity…………………………………………….29
Abstract……………………………………………………………………………………... 29
Introduction………………………………………………………………………………… 29
Results……………………………………………………………………………………… 31
Discussion………………………………………………………………………………….. 38
References………………………………………………………………………………….. 41
Materials and Methods……………………………………………………………………... 46
Figures……………………………………………………………………………………… 50
Supplemental Figures………………………………………………………………………. 62

Chapter 3: PD-1 Blockade on Tumor Microenvironment-Resident ILC2s Promotes TNF-α
Production and Restricts Progression of Metastatic Melanoma…………………………………66
Abstract…………………………………………………………………………………….. 66
Introduction………………………………………………………………………………… 66
Results……………………………………………………………………………………… 68
Discussion………………………………………………………………………………….. 73
References………………………………………………………………………………….. 76
Materials and Methods…………………………………………………………………….. 80
Figures……………………………………………………………………………………… 84
Supplemental Figures……………………………………………………………………… 91

Chapter 4: Orai inhibition modulates pulmonary ILC2 metabolism, effector function and
alleviates airway hyperreactivity……………………………………………………...…………93
Abstract…………………………………………………………………………………….. 93
Introduction………………………………………………………………………………… 93
Results……………………………………………………………………………………… 95
Discussion………………………………………………………………………………… 101
References………………………………………………………………………………… 105
Materials and Methods……………………………………………………………………. 109
Figures…………………………………………………………………………………… 114

Chapter 5: Future Perspectives……………………………………………………………….. 124
Part 1: The regulatory role of ILC210s in disease context………………………………. 124
Part 2: Protective role of ILC2s in B16 melanoma……………………………………….126
Part 3: Inhibiting calcium flux leading to attenuation of allergic asthma………………….128

iv
List of Figures

Chapter 2
Figure 1. A subset of IL-10
+
ILC2s is naturally present in an allergic inflammation
murine model .................................................................................................................................50

Figure 2. IL-4 induces ILC210 generation and downregulates effector function ..........................52

Figure 3. cMaf and Blimp-1 are necessary for IL-10 generation in ILC210s……………………54

Figure 4. ILC210s suppress activated ILC2 effector function in vitro…………………………….56

Figure 5. ILC210s downregulate airway hyperreactivity and lung inflammation .........................58

Figure 6. ILC210 generation is mediated by a metabolic shift from the fatty acid oxidation
pathway to the glycolytic pathway ................................................................................................60

Supplemental Figure 1. ILC2 gating strategy, related to Figure 1………………………………62

Supplemental Figure 2. ILC2 cytokine and transcription factor expression, related to
Figure 3 ..........................................................................................................................................63

Supplemental Figure 3. ILC2 viability and proliferation, related to Figure 4………………….64

Supplemental Figure 4. ILC2 proliferation, related to Figure 5………………………………..64

Supplemental Figure 5. ILC2 viability, related to Figure 6……………………………………..65

Chapter 3
Figure 1. B16 melanoma induces PD-1 expression on pulmonary ILC2s………………………..84

Figure 2. PD-1 expression on pulmonary ILC2s promotes B16 tumor growth
and drives melanoma-induced fatality ...........................................................................................85

Figure 3. PD-1 deficiency on IL-33 stimulated ILC2s enhances TNF- α
expression and phosphorylation of canonical NF κB pathway ......................................................86

v
Figure 4. Blocking PD-1 on ILC2s increases TNF- α production and
enhances cytotoxic properties ........................................................................................................88

Figure 5. Blocking ILC2 PD-1 expression increases TNF- α production and
inhibits tumor progression in vivo .................................................................................................89

Figure 6. Blocking PD-1 engagement on human blood ILC2 cells increases TNF- α secretion….90

Supplemental Figure 1. PD-1 expression on pulmonary ILC2s promotes B16 tumor
growth and drives melanoma-induced fatality ...............................................................................91

Supplemental Figure 2. PD-1 deficiency on IL-33 stimulated ILC2s enhances TNF-a
expression and phosphorylation of canonical NF κB pathway ......................................................92

Chapter 4
Figure 1. Murine pulmonary ILC2s express calcium release activated calcium channels
Orai1 and Orai2............................................................................................................................114

Figure 2. Pulmonary ILC2 effector function is dependent on Orai1 and Orai2…………………115

Figure 3. Orai inhibition alters pro-inflammatory transcriptomic profile in pulmonary
murine ILC2s ...............................................................................................................................116

Figure 4. Orai channels significantly affect metabolic function in pulmonary murine ILC2s…..118

Figure 5. Orai channels significantly downregulate mitochondrial function, affecting
ILC2 effector function and cytokine production .........................................................................119

Figure 6. Inhibition of Orai channels in pulmonary ILC2s significantly
downregulates development of airway inflammation ..................................................................121

Figure 7. Inhibition of Orai channels ameliorates human ILC2-mediated AHR………………..122

vi
Abstract

ILC2s are the dominant innate lymphoid cell population in the lungs at steady state and
their release of type-2 cytokines upon activation play a crucial role in the maintenance and
homeostasis of body health. First described in 2001 as a population of non-B and non-T cells that
produce cytokines IL-5 and IL-13, the improper activation or suppression of ILC2s has been
implicated as a driving force behind the development and exacerbation of a variety of diseases
throughout the human body. There are currently no treatments specifically targeting these cells,
though they are increasingly identified as being responsible for the underlying pathological
mechanisms behind diverse diseases. Understanding their biology and behavior therefore is
essential for the development of ILC2-based therapeutic strategies.
Allergic asthma is an inflammatory disease of the upper airway that results from the
inappropriate immune response to environmental allergens. In the context of this disease, ILC2s
are the first responders to the subsequent secreted epithelial alarmins. ILC2s’ responding cytokine
secretion is a central driver in responding eosinophil infiltration, increased airway hyperreactivity
and associated lung tissue injury. In this study, we identified a subset of ILC2s (ILC210s) that
actively produce and secrete IL-10, an anti-inflammatory cytokine with the ability to ameliorate
allergic lung inflammation. We further identified key molecular and transcriptional requirements
may be required for the induction of IL-10, with the potential for targeted modulation.
Transcription factors Blimp-1 and cMaf are crucial for the induction of IL-10 production.
Furthermore, glycolytic metabolism appears to be essential for the development of ILC210s, in
contrast to the traditionally utilized fatty acid oxidation pathway that is important for the
conventional pro-inflammatory phenotype. We address specific transcriptional and metabolic

vii
requirements for the modulation of pathogenic ILC2s with the intention of targeted conversion to
ILC210s with the ability to regulate airway hyperreactivity.
In contrast to the uncontrolled ILC2 activation found in allergic asthma, we also studied
the improper suppression of proinflammatory effector function in the context of metastatic
melanoma. The role of ILC2s in cancer is highly debated and appears to be dependent on cancer
location and type. We found that blocking PD-1 on pulmonary ILC2s inhibited the growth of
metastatic melanoma through the enhancement of cytokine secretion TNF- α, intensifying the pro-
inflammatory phenotype of ILC2s and subsequent anti-tumor activity.
Finally, we further addressed the urgent need of asthma patients by taking advantage of the
recent breakthroughs in the field of Ca
2+
signaling to benefit asthma therapy. We identified Orai1
and Orai2, long-sought pore component of calcium release activated calcium (CRAC) channels.
We found that blocking the channels with the drug 5D inhibited the development of airway
inflammation in both murine and humanized models. Moreover, inhibition of these channels
severely affected metabolism and mitochondrial function, leading to the decrease in cytokine
production and proliferation in ILC2s. This information will have a broad impact on basic
understanding of the function of Ca
2+
signaling in allergic ILC2s, and the development of novel
therapy to treat severe forms of allergic and atopic inflammatory diseases
We believe that the results obtained from this study will provide insights into an important
and understudied role of ILC2 modulation in diseases. Importantly, this dissertation identifies
novel targets and knowledge related to the metabolism, development, and modulation of
pulmonary ILC2 cells, offering new avenues for the development of therapeutic strategies in the
treatment of ILC2-associated diseases.

1

Chapter 1: Introduction

The discovery of Group 2 Innate Lymphoid Cells (ILC2s) nearly two decades ago
dramatically introduced a novel immune cell that was quickly discovered to be crucial in the
functioning of the immune system
1
. First described in the early 2000’s, it’s become apparent that
ILC2s play a previously unrecognized but essential part in a variety of bodily diseases
1
. Innate
lymphoid cells are composed of three subclasses, broadly mimicking the cytokine production and
effector function of their T helper cell counterparts
2
. Unlike T helper cells however, ILC2s are
primarily tissue resident cells that lack lineage and antigen specific markers, and instead are
activated through epithelial cell alarmins, including interleukin (IL)-33, IL-25 and thymic stromal
lymphopoietin (TSLP)
3
. ILC2s are largely found deep in mucosal tissues, including the gut, lung,
skin and brain, and are increasingly being recognized for their unique role in the homeostasis of
the broader immune system
4
.
ILC2s, with their location throughout the body and their rapid activation by surrounding
cytokines, are uniquely designed to be first responders to bodily damage or infiltrating microbes.
They contribute broadly to the body’s type 2 immune response and are principally known for their
IL-5 and IL-13 cytokine secretion following activation
5
. Their secreted cytokines are responsible
for the subsequent response of a variety of immune cells, including eosinophils, macrophages, T
cells, and dendritic cells, and are a key contributor to the health of the immune system
1
. As with
any part of the immune system, however, inappropriate activation or an unregulated response in
these cells disrupts the homeostasis of the immune system and drastically affects the surrounding
cells and tissues, often resulting in a chronic disease or disorder. Recently, a wealth of growing
research has detailed the immense protective or pathologic contribution of ILC2s to a host of
diseases, including diabetes, allergic asthma, atopic dermatitis, and cancer
6–8
. In the following

2
chapters, we utilize various disease models to further explore the specific role of ILC2s and their
modulation as a potential for therapeutic interventions.
Figure 1. An overview of currently known pulmonary ILC2 regulators, mediators, surface
receptors and responding effector function. (Adapted from Reference 3).

Biology of ILC2s
Ontogeny
In contrast to adaptive immune cells, ILC2 cells are primarily tissue-resident immune cells
that migrate during early development
9
. All ILC cells are derived from common lymphoid
progenitors (CLP) found primarily in the bone marrow and fetal liver
10
. During development,

3
transcription factors TOX, TCF-1 and NFIL3 are essential for commitment to the ILC lineage and
are the key drivers for the early ILC progenitor stages α-lymphoid precursor ( αLP) and early ILC
progenitor (EILP)s
11
. Seillet et al. and others found that commitment to this stage does not allow
for further development into T and B cells but the cells at this point have the capacity to develop
all ILC subsets, including NK cells
11,12
. Downstream of αLP and EILPs, the cells utilize
transcription factors ID2 and PLZF to differentiate further into subsets CHILP and ILCP
respectively
11,13
, losing the ability to differentiate into NK cells. Differentiation into ILC1, ILC2
and ILC3 from this stage is dependent on the subclasses’ specific classical transcription factors.
ILC1s depend on T-bet expression and are known primarily for their IFN- γ cytokine secretion.
ILC2s rely on GATA3, BCL11B, and ROR α expression for their IL-5 and IL-13 secretion, while
ILC3 depend on ROR γ τ and ROR α expression to secrete IL-17 and IL-22 (Figure 2)
14,15
.

4
Figure 2. An overview of BM-dependent ILC and thymus-dependent T-cell development. Current
models for both ILC and T-cell development suggest that they are of distinct lineages and that the
site for maturation do not overlap beyond the CLP stage. ILCs have been proposed to develop in
the BM whereas committed T-cell progenitors undergo intense development processes that are
heavily influenced by the thymic niche. (Figure from reference 11).

The mechanisms of differentiation from the progenitors into the separate subclasses of
ILCs remains largely unresolved. However, the unique ILC2 progenitor population, termed ILC2p,
derives directly from the ILCp stage and has been characterized by the expression of ILC2-
dependent transcription factors SCA1, ID2 and GATA3
16,17
. Specifically, ILC2s have been shown
to require ID2 expression and retain high expression of the transcription factor into cellular
maturity
9
. Furthermore, these specific transcription factors and others have been shown to be
crucial in the development of ILC2s, including BCL11B, ROR- α, GFI1, and ETS1, providing
some hints towards the full pathway of ILC2 maturity and development
18
. Temporally, ILC2s are
shown to distribute into specific tissues at three specific time-points: before birth in the fetus, post-
birth before weaning, and adult-derived ILC2s
9
. Through parabiosis experiments, ILC2s and
ILC2ps have been shown to be primarily tissue-resident, though circulating cells do contribute to
disease progression
19
. Recent studies have revealed incredible heterogeneity in the chromatin of
tissue-resident ILC2s based on their location and microenvironment
20
. Whether this heterogeneity
arises pre-maturity from a variety of possible ILC2 progenitor cells, or post-maturity in that the
chromatin is altered after development based on location is unknown and requires further study.
While bone marrow and fetal liver have been the traditional locations for the rise in
progenitor cells, recently Ferreria et al. demonstrated that the thymus also has the potential to

5
differentiate ILC2 cells based on Notch and IL-7R signaling
14
. Here, progenitor cells that have the
potential to differentiate into T cells or ILC2s in the thymus utilizing transcription factor BCL11b
to commit to the T cell lineage via suppression of ILC2 ID2
14
. Contrastingly, ILC2s also depend
on BCL11B for commitment, further complicating ontogeny of ILC2s. Regardless, Ferreria et al.
showed that ROR α expression is the critical regulator of ILC2 development in the thymus, a
previously unrecognized location of ILC2 development
14
.

Transcription Factors
As outlined above there are a host of transcription factors and proteins responsible for the
development and effector function of ILC2 cells. Specifically, BCL11B, ROR α, GFI1, and ETS1,
ID2 and GATA3 are crucial for ILC2 maturity and development
18
. GATA3 remains the
transcription factor traditionally associated with type-2 inflammation from ILC2 cells
21
. However
countless transcription factors expressed by ILC2s in the body control cell homeostasis and
inflammation.
IL-33/ST2 signaling, one of the primary alarmins known to stimulate ILC2 cells, has an
extensive signaling pathway with the ability to enhance or inhibit a variety of transcription factors
downstream activation. For example, it is well known that the NF- κB pathway lies downstream of
the IL-33 receptor and indeed, Mindt et al. found that the c-Rel subunit of NF- κB is crucial for an
inflammatory response by ILC2s
22
. Furthermore, signaling pathways p38α/ β and ERK1/2 MAPK,
activated by IL-33 signaling are associated with significantly upregulated ILC2 effector function
and cytokine production, suggesting they are also critical for proinflammatory ILC2 activity
23
.
Recently it’s been shown that transcription factor PPAR γ regulates ST2/IL-33 receptor expression
on ILC2s, thereby controlling the potential for ILC2 pathogenicity via transcription factors

6
upstream of activation signaling
24
. The published literature points toward IL-33 signaling and the
downstream transcription factors as being a crucial as an immune target in the event of unregulated
proinflammation. Moreover, Van der ploeg et al. recently described how the IL-33 signaling
pathway activates transcription factor Stat5 and subsequently causes the upregulation of
transcription factors IRF4 and BATF
25
. These proteins induce ILC2 CD45RO expression, a
variation of the receptor CD45
25
. These CD45RO+ ILC2s are corticosteroid resistant, thereby
opposing any potential for traditional asthma treatment
25
.
The binding motif for BATF is enriched in specifically ILC2s, as opposed to ILC1/3s,
suggesting its importance in ILC2 function
26
. And indeed, the transcription factor appears to be
involved in a variety of important ILC2 functions, including migration and pulmonary wound
healing via ILC2s
27,28
. It is clear the signaling pathways and proteins responsible for the
proinflammatory effector function of ILC2s are complicated and require much further study.
Recently, an anti-inflammatory role of ILC2s via IL-10 production has been detailed.
While the signaling pathways and specific inducers are still being resolved, the literature provides
some hints towards potential transcription factors responsible. Zhang et al. identified that IL-33
and IL-10 induces the expression of transcription factor Blimp-1 in ILC2s; these Blimp-1
expressing cells consequently lose the ability to produce and secrete their proinflammatory
cytokines
29
. And indeed, Blimp-1 is crucial for the IL-10 dependent regulatory pathway in T cells
during disease
30
. The transcription factor cMaf has also been implicated in IL-10 production in a
variety of immune cells, while the combination of Blimp-1 and cMaf induces IL-10 in CD4+ T
cells
31,32
. An exhaustive list of transcription factors and their importance in ILC2 function is near
impossible, but it is clear from the growing wealth of literature that targeting these proteins in the
enhancement or regulation of ILC2 cells has a promising therapeutic potential in disease.

7

Cytokines
As previously mentioned, ILCs are subdivided into three main classes
2
. ILC1s traditionally
secrete IFN- γ and TNF- α, ILC2s their type 2 cytokines, and ILC3 secrete IL-17
1
. However,
plasticity among the three subclasses is retained post-differentiation and is often the result of
extracellular cytokine stimulation. IL-12, for example, reduces ILC2 effector function and
stimulates the cells to transition to ILC1s as classified by their identifying transcription factors
33
.
Golebski et al. identified that IL-1β, IL-23 and TGF- β drive human ILC2s to produce and secrete
IL-17 in nasal polyps, while upregulating the transcription factor ROR γ τ that is used to identify
ILC3s
34
. In addition, ILC2 Th2 cytokine production is inhibited by traditional Th1 cytokines,
including IFN- β, IFN- γ and IL-27
35
.

Cytokine function Cytokine class Cytokine
Pro-inflammatory Th2 cytokines IL-4, IL-5, IL-6,
IL-13, IL-9, amphiregulin
Non-Th2 cytokines GM-CSF, TNF- α
Anti-inflammatory Th2 cytokines IL-10, TGF- β, amphiregulin
Table 1. Identified cytokines secreted by Gata3
+
ILC2s in disease setting.

In addition to the traditional effector function cytokines IL-5 and IL-13, ILC2s activated
by IL-33, IL-25 and TSLP are conventionally known to secrete IL-6, IL-9, GM-CSF,
amphiregulin, and in humans, IL-4 (Table 1)
3
. The majority of these cytokines fit the type-2 milieu

8
expected of activated pulmonary ILC2s. However further investigation into how ILC2s respond in
specific microenvironments have uncovered a flexibility in cytokine production that has the
potential for therapeutic modulation. And indeed, even amongst the classical epithelial alarmins
Camelo et al. found a distinct difference in cytokine response, with IL-33/IL-2 producing the most
abundant levels of IL-5 and IL-13 from the ILC2 cells
36
.
While there is plasticity amongst the classes of ILCs, there is also flexibility in the
cytokines produced specifically by ILC2s. Recently, studies have drawn attention to the potential
regulatory role of ILC2s. And indeed, ILC2s expressing their identifiable transcription factor
GATA3 have been shown to secrete TGF- β and Amphiregulin during intestinal damage to promote
repair
37
. Further studies have demonstrated ILC2s produce the anti-inflammatory cytokine IL-10
in response to a host of cytokine stimuli, including IL-33, IL-2 and retinoic acid
38–40
. IL-10 is a
well-documented anti-inflammatory cytokine traditionally attributed to T regulatory cells.
However, Cao et al. demonstrated these IL-10 producing ILC2s’ protective role in vivo against the
murine renal damage, suggesting a potential avenue for regulatory modulation through cytokine
stimulation and response
41
. The IL-10 cytokine superfamily includes IL-10, IL-19, IL-20, IL-22,
IL-24, IL-28a, IL-28b, and IL-29
42
. IL-19 and IL-24 have been shown in different immune cells
to also have anti-inflammatory properties that are further enhanced by IL-10 stimulation
42
.
Whether the IL-10
+
ILC2s secrete additional IL-10 superfamily cytokines is unknown.
The versatility of cytokines allows for the exciting potential to approach cell-based
therapeutics focused on ILC2s as it relates to disease. Understanding the response of ILC2s to
different microenvironmental stimuli opens the possibility for modulation towards regulatory roles
or pathogenic activation in different disease settings.

9
Co-stimulatory molecules
The use of co-stimulatory molecules as a means for cellular modulation has been well
documented in T cell activation and regulation
43
. Specifically in ILC2s, these co-stimulatory or
co-inhibitory molecules are receptors that bind to ligands on surrounding immune cells and have
already been proven important for regulation in a variety of diseases (Table 2).
Costimulatory
molecule family
Name of molecule Role in ILC2s References
Ig superfamily CD28 Inflammatory Roan et al.
44

CTLA-4 Inhibitory Morita et al.
45

ICOS Inflammatory Maazi et al.
46

PD-1 Inhibitory Helou et al.
47

TNF superfamily CD27 Unknown in ILC2s You et al.
48

CD30 Inhibitory Liu et al.
49

GITR Inflammatory Galle et al.
50

OX40L Inflammatory Halim et al.
51

Other CD80/86 Inflammatory Hepworth et al.
52

NKP30 Inhibitory Salimi et al.
53

DR3 Inflammatory Shafiei-Jahani et al.
54

Btn2a2 Inhibitory Frech et al.
55

Table 2. Identified co-stimulatory molecules on ILC2s or ILC210s.
We and others have demonstrated, for example, the expression of inducible T-cell co-
stimulator (ICOS) on pulmonary ILC2 cells that is critical for ILC2 activation and effector function
in the development of murine models of asthma
46,56
. Inhibition of ICOS significantly

10
downregulated ILC2 effector function and ameliorated the murine lung inflammation, suggesting
this as a potential therapeutic target in the context of ILC2-dependent asthma
46
. Similarly, co-
stimulatory molecule TNFRSF18 (GITR) has been shown to enhance ILC2 cytokine production
and the development of murine lung inflammation
57,58
.
In contrast, PD-1 is a co-inhibitory molecule expressed on ILC2s at steady state and after
activation that negatively regulates ILC2 effector function through inhibition of proliferation and
cytokine production. In the context of allergic asthma, our group has demonstrated that PD-1 plays
a critical regulatory role in the development of airway hyperreactivity
47
. Use of the FDA-approved
PD-1 agonist reduces ILC2-dependent AHR and inflammation
47
. In the context of cancer however,
PD-1 antagonists have been well documented, as previously described, in the activation and
enhancement of immune cell activation in an anti-tumor manner
59
. Thus, PD-1 is one clear
example of how modification of a co-stimulatory molecule in the appropriate context leads to the
desired ILC2 manipulation.

Metabolism
One previously underrecognized mechanism that has the potential for affecting effector
functions of immune cells is cellular metabolism. Immune activation is highly dependent on
energetic fuel, offering a new platform by which to manipulate immune cell function (Figure 3).
Activated Th2 cells, for example, favor the glycolytic pathway to support their immune functions,
while anti-inflammatory T regulatory cells (Tregs) rely on oxidative phosphorylation
(OXPHOS)
60,61
. Altering the metabolic pathways available to the immune cells therefore offers a
potential mechanism by which we control effector function.

11
Figure 3. Modified schematic representation of metabolic pathways in cellular mitochondria.
(Ramirez AK, Lynes MD, Shamsi F, et al. Integrating Extracellular Flux Measurements and Genome-Scale Modeling Reveals
Differences between Brown and White Adipocytes. Cell Rep. 2017;21(11):3040-3048. doi:10.1016/j.celrep.2017.11.065)

Previous reports suggest that ILC2s are metabolically distinct from other lymphocytes,
including T helper and T regulatory subsets, in that they preferentially use lipid-fueled fatty acid
oxidation (FAO) to support proliferation and effector functions
57
. A recent report showed that
intestinal ILC2s express a genetic signature enriched in fatty acid metabolism genes, as they have
been shown to uptake extracellular fatty acids from the local microenvironment
62
. Furthermore,
inhibition of FAO – but not of glycolysis – in vivo reduced ILC2 accumulation and effector
cytokine secretion in response to helminth infection
62
. We were recently the first to show that
impaired FAO due to a lack of autophagy in lung ILC2s negatively affected pro-inflammatory
effector cytokine secretion upon IL-33 stimulation, suggesting that such an observation is not
restricted to intestinal ILC2s
57
. Importantly, we further found that lipid accumulation in ILC2s
lacking autophagy was associated with an increased glycolytic signature, suggesting that ILC2s
induce a context-dependent metabolic reprogramming from FAO towards glycolysis in the context
of airway inflammation. Interestingly, Surace et al. recently reported that in human ILC2s IL-13

12
cytokine secretion is supported by glycolysis in activated ILC2s, while proliferation is dependent
on OXPHOS
63
. They suggest the cellular functions are therefore metabolically uncoupled in
disease settings
63
. Further investigation into whether cytokine secretion may be uncoupled from
proliferation is needed.
Microenvironment plays a crucial role in the metabolic pathways utilized by an immune
cell. ILC2s, as demonstrated by their switch to glycolysis when FAO was compromised, have a
measure of flexibility and adaptability in their fuel source, though their immune function was
critically affected
57
. While not explored in this manuscript, Harmon et al. identified that
competition with surrounding tumor cells for glycolytic metabolites is a primary reason immune
cells do not efficiently activate anti-tumor defenses during cancer
64
. Xia et al. suggests that
metabolic reprogramming of tumor and immune cells has the potential to regulate and enhance
anti-tumor responses by the body through the further activation of the immune system
65
.
Alteration of ILC2 metabolism can be controlled by intracellular and extracellular
conditions and have a dramatic effect on the cells’ effector function. STAT3 deficiency, for
example, has the ability to block methionine metabolism that appears to be critical for ILC2
inflammatory response in lung inflammation
21
. Moreover, infection with Mycobacterium
tuberculosis skews the Th2 ILC2 inflammation toward a protective Th1 phenotype through the
control of glycolysis
66
. Lastly, blockade of the cholesterol biosynthesis pathway has a profound
inhibitory effect on IL-10 production in regulatory B and Th1 cells
67,68
. It is clear further
investigation into metabolism as a means for therapeutic modulation is required.

Migration and tissue specificity

13
As evidenced by metabolism, ILC2 biology is highly tissue dependent, as they are greatly
influenced by their microenvironment. The majority of ILC2 cells are considered to be tissue-
resident, the perinatal period serving as a crucial period for distribution into organs
9
. While there
is growing evidence of the importance of circulating ILC2s later in disease progression, a large
portion of the mature ILC2s found throughout the adult body are the result of local ILC2 cells
expanding or decreasing dependent on the surrounding signaling
11
. Recently, Ricardo-Gonzalez et
al. demonstrated that ILC2s generated prenatal or in the early fetal period are replaced by de novo
ILC2 production during the adult life. But the group demonstrates that even this replacement rate
is tissue-specific
19
. Lungs, fat and the gut had slower replacement of the original ILC2 population,
whereas bone marrow and skin are more rapid
19
.
Utilizing single-cell RNA sequencing, Schneider et al. showed that murine ILC2s as early
as 14 days postnatal demonstrated a unique activation and tissue-specific imprinting
9
. The ILC2
cells throughout the body upregulate specific receptors and cytokines, likely in an effort to be
readily available for the tissues’ specific activation. For example, ILC2s have shown an
upregulation of IL-25R expression in the gut, IL-18R in the skin and ST2/IL-33R in the lung and
adipose tissue
19
. Additionally, there are tissue-specific markers that help to identify the origin of
ILC2 cells. Neuopilin-1 (Nrp-1) for example is a specific marker of pulmonary ILC2s, while aryl
hydrocarbon receptor is most highly expressed by gut ILC2s
69,70
. Utilizing tissue-specific markers,
Ricardo-Gonzalez et al. was able to identify the bodily origin of circulating blood ILC2s after
infection with N. brasiliensis, demonstrating how these markers can be used for ILC2 trafficking
71
.
Recently there has been a growing interest in the migration and trafficking of ILC2 cells.
Though largely tissue-resident, populations of ILC2 cells have been found to migrate throughout
the body dependent on tissue-specific alarmins in the event of disease or damage. Huang et al.

14
shows that ILC2s in the small intestine respond to IL-25 stimulation to upregulate their
sphingosine 1-phosphate (S1P) receptors
72
. This receptor allows them to then traffic to the lung
72
.
Moreover, parabiosis mouse experiments indicated that inflamed pulmonary ILC2s during airway
hyperreactivity migrate to the liver, upregulating Th2 inflammation and downregulating Th1
response
73
. Delineating the details of migration in ILC2 biology is essential for understanding how
ILC2 inflammation is linked throughout the body.

The role of ILC2s in health and disease
ILC2s in health and homeostasis
ILC2s play a crucial role in health and homeostasis throughout the body. Their presence as
the first responders ensures a level of protection independent of adaptive immunity. For example,
Sonnenberg et al. found that IL-22 secreted by ILC2s is essential for the homeostasis of the
intestinal tissue
74
. Absent the ILC2s, systemic inflammation and dissemination of commensal
bacteria occurred, resulting in disease
74
. ILC2s have also been highly characterized for their
protective role in type 2 diabetes and are known to be crucial in preventing the development of
metabolic diseases
75
. In adipose tissue ILC2 cells interact closely with the surrounding
microenvironment to regulate the development of type 2 diabetes
75
.
In the lung, ILC2s have been shown to be important in wound healing and tissue repair,
dependent on their secretion of amphiregulin
38
. Further, ILC2-derived IL-5 is critical for the
maintenance of steady state eosinophil levels in the lungs. Eosinophils subsequently protects
against acute lung injury, suggesting ILC2 IL-5 secretion plays a crucial role in lung barrier
protection and homeostasis
76
. They are also known to protect pulmonary endothelial cells during
sepsis
77
. IL-33 that is secreted by the endothelial cells during septic inflammation activates and

15
expands pulmonary ILC2s, resulting in ILC2-derived IL-9. IL-9 protects the endothelial layer from
undergoing pryoptosis, a form of cell-death
77
. ILC2s also modulate CD4 T cell activity to ensure
lung health and repair. Through cell-to-cell contact, ILC2s enhance Th2 cell activity and have the
additional potential of direct T cell activation through antigen presentation
78
. Growing research
has indicated various methods by which ILC2s regulate health throughout the body. Left
uncontrolled, however, their pro-inflammatory response tips the balance into disease and disorder.

Airway inflammation
Allergic asthma continues to rise at an alarming rate and has reached epidemic proportions
over the last several decades. The highly heterogenous pulmonary disease is characterized by the
development of airway hyperreactivity (AHR) and bronchoconstriction
3
. Patients with asthma
have demonstrated higher numbers of activated ILC2s in both the blood and bronchioalveolar
lavage (BAL) fluid as compared to the healthy control patients
8,79
. This, along with the correlation
between elevated BAL eosinophils and ILC2 number in patients with disease, was the first
suggestion that ILC2s played a critical role in the development and persistence of allergic asthma
7
.
Allergic airway hyperreactivity results from allergens entering the lungs and interacting
with the pulmonary epithelial layer
80
. The epithelial cells subsequently produce alarmins that, in
addition to activating dendritic cells to present the antigen in the lymph nodes, immediately
activates the ILC2 cells that culminate at the epithelial layer
81
. The ILC2s are rapidly stimulated
via the alarmins to produce cytokines, primarily IL-5, IL-13, and in humans, IL-4
81
. IL-5 has been
well documented to be crucial for the maturation, recruitment, and activation of eosinophils
82
. IL-
13 secretion is responsible for the resulting smooth muscle contraction and the generation of
fibrogenic processes in the lungs
83
.

16
In murine studies of AHR, reports utilizing mice that lack all adaptive immunity, named
Rag2
-/-
mice, readily establish the development of acute and chronic airway inflammation
46,84,85
.
These results clearly suggest the innate immune system independent of adaptive immunity is
competent to mount an immune response in the development of murine allergic asthma. Research
over the last two decades has provided an explanation to this phenomenon through the exploration
of ILC2 cells.
According to the American Lung Association, allergic diseases and asthma affect 10-15%
of people in the United States
86
. The World Health Organization projects this trend to double over
the next decade, escalating the already staggering economic burden, which according to the
Centers for Disease Control, amounts to $56B/year
87
. Despite significant advances in our
understanding of the underlying pathogenesis of allergic asthma, most of the FDA-approved
pharmacopoeia is aimed at alleviating symptoms rather than addressing the underlying
pathological mechanisms
88
. Additionally, the heterogeneity of the disease results in many patients
experiencing complications with their treatment courses
89
. Given that more than half of the U.S.
population is estimated to be sensitized to at least one allergen, novel, effective and mechanism-
based therapeutic strategies are urgently needed.
Recently, utilizing cell signaling and cellular effector functions as a therapeutic means
against allergic asthma has gained interest
89
. Currently however there are no therapeutic strategies
aimed at addressing the pathogenic role of ILC2s in allergic asthma.

Intestinal Inflammation
ILC2s in the intestine are largely beneficial to the host body. Pathogenically, ILC2s have
been implicated in the development of oxazolone colitis
90
. Mice in this study had significantly

17
elevated IL-25 in their colon, enhancing the secretion of ILC2-derived IL-13
90
. This led to reduced
colon length and chronic inflammation that was ameliorated upon ablation of IL-25
90
.
Though unregulated ILC2s may contribute to forms of intestinal inflammation, the
overwhelming majority of literature suggests ILC2s found in the gut promotes overall intestinal
health. ILC1 and ILC3 cells have been implicated in the development of Crohn’s Disease. IL-10-
producing ILC2s (ILCregs or ILC210s) in this environment limit ILC1/3 activation, leading to a
resolution of the intestinal inflammation
37
. And indeed, as described above, amphiregulin secreted
from intestinal ILC2s play a critical role in the intestinal integrity barrier, as well as enhances T
regulatory cell activation
90
. While pathogenic properties of ILC2s in the intestine may arise, their
function in the gut highlights the importance of ILC2s in health and homeostasis.

Adipose tissue
Similarly, ILC2s have a protective role in adipose tissue in the fight against metabolic
diseases. ILC2s are present in the visceral adipose tissue (VAT) where they are primary
contributors of Th2 cytokines IL-5 and IL-13 at steady state
75
. This secretion is essential for the
maintenance of glucose homeostasis, as well as for alternatively activated or M2-like adipose
tissue macrophages and T regulatory cells
91
. Halim et al. demonstrated that OX40L signaling on
adipose ILC2 cells is required to maintain a healthy population of T regulatory cells, cells that
subsequently are essential in the prevention of insulin resistance
51,92
. And indeed, the number of
ILC2s resident in the adipose tissue is statistically decreased in adults with obesity, subsequently
causing a decline in the levels of eosinophils and M2 macrophages, resulting in insulin resistance
after high-fat diet feeding in mice
93
. Adoptively transferring activated ILC2s in these models
suppressed weigh t gain and reversed insulin resistance
94
.

18
Two independent groups have shown the involvement of ILC2s in the white adipose tissue
browning, as well
95,96
. In one study, IL-33 activated ILC2s produced methionine-enkephalin
peptide. This peptide upregulates uncoupling protein 1 (UPC1) to uncouple mitochondrial
respiration and allow energy to escape as heat, thereby promoting browning
96
. The other suggested
that IL-33 activated ILC2 cells produced IL-5 to recruit eosinophils
95
. The eosinophil cytokines
then activated proliferation of the adipocyte precursors, leading to the browning of the fat tissue
95
.

Cancer
While ILC2s are clearly pathogenic in the establishment of allergic asthma, their role in
cancer is highly contested
97
. The supposed discrepancy seen in the published research on whether
ILC2s act in a pro-tumor or anti-tumor manner is representative of the heterogeneity and
complexity of cancer. What is clear from the research however is that the role of ILC2s in cancer
is largely dependent on microenvironment and cancer type
97
.
ILC2s, for example, are shown by one group to contribute to tumor growth in colorectal
cancer (CRC). CRC tumors secrete high levels of IL-25, activating ILC2s and consequently
leading to the enhancement of immunosuppressive myeloid cells
98
. Blocking IL-25 in this context
alleviated tumor burden and allowed for an anti-tumorigenic microenvironment in the colon
98
.
Contradictorily, a separate group demonstrated that IL-9 secreted from ILC2 cells in the intestine
led to the activation of anti-tumor CD8+ T cells, inhibiting CRC progression and thereby acting in
an anti-tumor manner
99
.
Similar contrasting results have arisen in murine melanoma models. Strikingly, one recent
study that has gained interest found that IL-33 administration can inhibit pulmonary metastasis in
mice
100
. The group identified natural killer (NK) cells as the principal cell recruited to the tumor

19
site by the IL-33 addition
100
. Furthermore, Jacquelot et al. identified a key anti-tumor role for
ILC2s with IL-33 addition in melanoma through their secretion of GM-CSF and subsequent
recruitment of eosinophils, thereby promoting an anti-tumor immune response
59
. In another study,
Schuijus et al. also found that IL-33 activation of ILC2s in melanoma models recruited
eosinophils, but in direct contrast to the previous studies, the group found that the higher numbers
of eosinophils in the tumor microenvironment suppressed anti-tumor natural killer cell Th1
immune response, leading to the progression of tumor burden
101
.

Understanding the cellular mechanisms by which ILC2 pro-inflammatory or anti-
inflammatory immune response functions is critical to unleash their potential in the course of
diseases. Whatever the specific disease, uncovering the role of ILC2s in the pathology of the
disease, as well as the mechanisms by which homeostasis may be restored is paramount to the
health of patients. Approaches that manipulate cellular metabolism, cytokine stimulation, and co-
stimulatory molecules can have profound effects on cell function and disease development. My
dissertation explores the way in which ILC2s respond to these specific factors with the goal of
harnessing these intrinsic mechanisms to help create therapies for the prevention and treatment of
various ILC2-involved diseases. Each chapter in this dissertation focuses on different aspects of
manipulating ILC2s in various disease models, and therefore each has its own research question
and specific therapeutic application. In Chapter 2, we investigate the anti-inflammatory and
regulatory role of IL-10 producing ILC2s in the context of allergic asthma, focusing on
metabolism, transcription factors and cytokines as a means of modulation. In Chapter 3, we
concentrate on enhancing pro-inflammatory ILC2 effector function and harnessing their cytokine

20
flexibility through checkpoint inhibitors in the context of metastatic melanoma. Finally in Chapter
4, we focus on how the internal mechanisms of metabolism rely on the external signaling of Ca
2+

and their profound alteration of effector function. Together, scientific knowledge gained from
these chapters has the potential to educate on the modulation of ILC2s directly in a variety of
disease contexts. Findings from the studies in the following chapters could lead to identification
of therapeutic targets that can be pharmacologically manipulated to specifically modulate ILC2s
in either a regulatory or pathologic capacity for the treatment of various diseases, including asthma
and cancer.

21
References
1. Vivier, E. et al. Innate Lymphoid Cells: 10 Years On. Cell 174, 1054–1066 (2018).

2. Spits, H. et al. Innate lymphoid cells — a proposal for uniform nomenclature. Nat Rev Immunol
13, 145–149 (2013).

3. Hurrell, B. P., Shafiei Jahani, P. & Akbari, O. Social Networking of Group Two Innate
Lymphoid Cells in Allergy and Asthma. Front Immunol 9, 2694 (2018).

4. Panda, S. K. & Colonna, M. Innate Lymphoid Cells in Mucosal Immunity. Frontiers in
Immunology 10, (2019).

5. Price, A. E. et al. Systemically dispersed innate IL-13–expressing cells in type 2 immunity. Proc
Natl Acad Sci U S A 107, 11489–11494 (2010).

6. Mattiola, I. & Diefenbach, A. Enabling anti-tumor immunity by unleashing ILC2. Cell Res 30,
461–462 (2020).

7. Smith, S. G. et al. Increased numbers of activated group 2 innate lymphoid cells in the airways
of patients with severe asthma and persistent airway eosinophilia. Journal of Allergy and Clinical
Immunology 137, 75-86.e8 (2016).

8. Bartemes, K. R., Kephart, G. M., Fox, S. J. & Kita, H. Enhanced innate type 2 immune response
in peripheral blood from patients with asthma. Journal of Allergy and Clinical Immunology 134,
671-678.e4 (2014).

9. Schneider, C. et al. Tissue-Resident Group 2 Innate Lymphoid Cells Differentiate by Layered
Ontogeny and In Situ Perinatal Priming. Immunity 50, 1425-1438.e5 (2019).

10. Kondo, M., Weissman, I. L. & Akashi, K. Identification of Clonogenic Common Lymphoid
Progenitors in Mouse Bone Marrow. Cell 91, 661–672 (1997).

11. Shin, S. B. & McNagny, K. M. ILC-You in the Thymus: A Fresh Look at Innate Lymphoid
Cell Development. Frontiers in Immunology 12, (2021).

12. Seillet, C. et al. Deciphering the Innate Lymphoid Cell Transcriptional Program. Cell Reports
17, 436–447 (2016).

13. Constantinides, M. G., McDonald, B. D., Verhoef, P. A. & Bendelac, A. A committed
precursor to innate lymphoid cells. Nature 508, 397–401 (2014).

14. Ferreira, A. C. F. et al. RORα is a critical checkpoint for T cell and ILC2 commitment in the
embryonic thymus. Nat Immunol 22, 166–178 (2021).

22
15. Zhou, W. & Sonnenberg, G. F. Activation and Suppression of Group 3 Innate Lymphoid Cells
in the Gut. Trends in Immunology 41, 721–733 (2020).

16. Walker, J. A. et al. Polychromic Reporter Mice Reveal Unappreciated Innate Lymphoid Cell
Progenitor Heterogeneity and Elusive ILC3 Progenitors in Bone Marrow. Immunity 51, 104-
118.e7 (2019).

17. Koga, S. et al. Peripheral PDGFRα+gp38+ mesenchymal cells support the differentiation of
fetal liver-derived ILC2. J Exp Med 215, 1609–1626 (2018).

18. Michieletto, M. F. & Henao-Mejia, J. Ontogeny and heterogeneity of innate lymphoid cells
and the noncoding genome. Immunological Reviews 300, 152–166 (2021).

19. Ricardo-Gonzalez, R. R., Molofsky, A. B. & Locksley, R. M. ILC2s – development,
divergence, dispersal. Current Opinion in Immunology 75, 102168 (2022).

20. Spits, H. & Mjösberg, J. Heterogeneity of type 2 innate lymphoid cells. Nat Rev Immunol 1–
12 (2022) doi:10.1038/s41577-022-00704-5.

21. Fu, L. et al. A mitochondrial STAT3-methionine metabolism axis promotes ILC2-driven
allergic lung inflammation. Journal of Allergy and Clinical Immunology 149, 2091–2104 (2022).

22. Mindt, B. C. et al. The NF-κB Transcription Factor c-Rel Modulates Group 2 Innate Lymphoid
Cell Effector Functions and Drives Allergic Airway Inflammation. Frontiers in Immunology 12,
(2021).

23. Petrova, T., Pesic, J., Pardali, K., Gaestel, M. & Arthur, J. S. C. p38 MAPK signalling regulates
cytokine production in IL-33 stimulated Type 2 Innate Lymphoid cells. Sci Rep 10, 3479 (2020).

24. Xiao, Q. et al. PPARγ enhances ILC2 function during allergic airway inflammation via
transcription regulation of ST2. Mucosal Immunol 14, 468–478 (2021).

25. van der Ploeg, E. K. et al. Steroid-resistant human inflammatory ILC2s are marked by
CD45RO and elevated in type 2 respiratory diseases. Science Immunology 6, eabd3489 (2021).

26. Liu, Q., Kim, M. H., Friesen, L. & Kim, C. H. BATF regulates innate lymphoid cell
hematopoiesis and homeostasis. Sci Immunol 5, eaaz8154 (2020).

27. Miller, M. M. et al. BATF acts as an essential regulator of IL-25–responsive migratory ILC2
cell fate and function. Science Immunology 5, eaay3994 (2020).

28. Wu, X. et al. BATF promotes group 2 innate lymphoid cell–mediated lung tissue protection
during acute respiratory virus infection. Science Immunology 7, eabc9934 (2022).

23
29. Zhang, J., He, K., Ford, B. R., Hettinga, A. & Poholek, A. C. Blimp-1 expressing lung ILC2s
are a distinct cell population in allergic asthma. The Journal of Immunology 206, 94.15-94.15
(2021).

30. De Trez, C. et al. A Critical Blimp-1-Dependent IL-10 Regulatory Pathway in T Cells Protects
From a Lethal Pro-inflammatory Cytokine Storm During Acute Experimental Trypanosoma brucei
Infection. Front Immunol 11, 1085 (2020).

31. Liu, M. et al. Transcription factor c-Maf is essential for IL-10 gene expression in B cells. Scand
J Immunol 88, e12701 (2018).

32. Ahlers, J. et al. A Notch/STAT3-driven Blimp-1/c-Maf-dependent molecular switch induces
IL-10 expression in human CD4+ T cells and is defective in Crohn´s disease patients. Mucosal
Immunol 15, 480–490 (2022).

33. Almeida, F. F. & Belz, G. T. Innate lymphoid cells: models of plasticity for immune
homeostasis and rapid responsiveness in protection. Mucosal Immunol 9, 1103–1112 (2016).

34. Golebski, K. et al. IL-1β, IL-23, and TGF-β drive plasticity of human ILC2s towards IL-17-
producing ILCs in nasal inflammation. Nature Communications 10, 2162 (2019).

35. Steinmann, S. et al. Hepatic ILC2 activity is regulated by liver inflammation-induced cytokines
and effector CD4+ T cells. Sci Rep 10, 1071 (2020).

36. Camelo, A. et al. IL-33, IL-25, and TSLP induce a distinct phenotypic and activation profile
in human type 2 innate lymphoid cells. Blood Advances 1, 577–589 (2017).

37. Wang, S. et al. Regulatory Innate Lymphoid Cells Control Innate Intestinal Inflammation. Cell
171, 201-216.e18 (2017).

38. Monticelli, L. A. et al. Innate lymphoid cells promote lung-tissue homeostasis after infection
with influenza virus. Nat Immunol 12, 1045–1054 (2011).

39. Seehus, C. R. et al. Alternative activation generates IL-10 producing type 2 innate lymphoid
cells. Nat Commun 8, 1900 (2017).

40. Bando, J. K. et al. ILC2s are the predominant source of intestinal ILC-derived IL-10. J Exp
Med 217, e20191520 (2020).

41. Cao, Q. et al. Regulatory innate lymphoid cells suppress innate immunity and reduce renal
ischemia/reperfusion injury. Kidney International 97, 130–142 (2020).

42. Ricciardi, A. & Nutman, T. B. IL-10 and Its Related Superfamily Members IL-19 and IL-24
Provide Parallel/Redundant Immune-Modulation in Loa loa Infection. J Infect Dis 223, 297–305
(2021).

24
43. Chen, L. & Flies, D. B. Molecular mechanisms of T cell co-stimulation and co-inhibition. Nat
Rev Immunol 13, 227–242 (2013).

44. Roan, F. et al. CD4+ group 1 innate lymphoid cells form a functionally distinct ILC subset that
is increased in systemic sclerosis. J Immunol 196, 2051–2062 (2016).

45. Morita, H. et al. Induction of human regulatory innate lymphoid cells from group 2 innate
lymphoid cells by retinoic acid. Journal of Allergy and Clinical Immunology 143, 2190-2201.e9
(2019).

46. Maazi, H. et al. ICOS:ICOS-ligand interaction is required for type 2 innate lymphoid cell
function, homeostasis, and induction of airway hyperreactivity. Immunity 42, 538–551 (2015).

47. Helou, D. G. et al. PD-1 pathway regulates ILC2 metabolism and PD-1 agonist treatment
ameliorates airway hyperreactivity. Nat Commun 11, 3998 (2020).

48. You, Y. et al. ILC2 Proliferated by IL-33 Stimulation Alleviates Acute Colitis in Rag1-/-
Mouse through Promoting M2 Macrophage Polarization. Journal of Immunology Research 2020,
e5018975 (2020).

49. Liu, S. et al. Optimal identification of human conventional and nonconventional (CRTH2–
IL7Rα–) ILC2s using additional surface markers. Journal of Allergy and Clinical Immunology
146, 390–405 (2020).

50. Galle-Treger, L. et al. Costimulation of type-2 innate lymphoid cells by GITR promotes
effector function and ameliorates type 2 diabetes. Nat Commun 10, 713 (2019).

51. Halim, T. Y. F. et al. Tissue-Restricted Adaptive Type 2 Immunity Is Orchestrated by
Expression of the Costimulatory Molecule OX40L on Group 2 Innate Lymphoid Cells. Immunity
48, 1195-1207.e6 (2018).

52. Hepworth, M. R. et al. Innate lymphoid cells regulate CD4+ T-cell responses to intestinal
commensal bacteria. Nature 498, 113–117 (2013).

53. Salimi, M. et al. Group 2 innate lymphoid cells express functional NKp30 receptor inducing
type 2 cytokine production. J Immunol 196, 45–54 (2016).

54. Shafiei-Jahani, P. et al. DR3 stimulation of adipose resident ILC2s ameliorates type 2 diabetes
mellitus. Nat Commun 11, 4718 (2020).

55. Frech, M. et al. Btn2a2 Regulates ILC2–T Cell Cross Talk in Type 2 Immune Responses.
Frontiers in Immunology 13, (2022).

56. Kamachi, F., Isshiki, T., Harada, N., Akiba, H. & Miyake, S. ICOS promotes group 2 innate
lymphoid cell activation in lungs. Biochem Biophys Res Commun 463, 739–745 (2015).

25
57. Galle-Treger, L. et al. Autophagy is critical for group 2 innate lymphoid cell metabolic
homeostasis and effector function. Journal of Allergy and Clinical Immunology 145, 502-517.e5
(2020).

58. Nagashima, H. et al. GITR cosignal in ILC2s controls allergic lung inflammation. Journal of
Allergy and Clinical Immunology 141, 1939-1943.e8 (2018).

59. Jacquelot, N. et al. Blockade of the co-inhibitory molecule PD-1 unleashes ILC2-dependent
antitumor immunity in melanoma. Nat Immunol 22, 851–864 (2021).

60. Berod, L. et al. De novo fatty acid synthesis controls the fate between regulatory T and T helper
17 cells. Nat Med 20, 1327–1333 (2014).

61. Michalek, R. D. et al. Cutting edge: distinct glycolytic and lipid oxidative metabolic programs
are essential for effector and regulatory CD4+ T cell subsets. J Immunol 186, 3299–3303 (2011).

62. Wilhelm, C. et al. Critical role of fatty acid metabolism in ILC2-mediated barrier protection
during malnutrition and helminth infection. J Exp Med 213, 1409–1418 (2016).

63. Surace, L. et al. Dichotomous metabolic networks govern human ILC2 proliferation and
function. Nat Immunol 22, 1367–1374 (2021).

64. Harmon, C., O’Farrelly, C. & Robinson, M. W. The Immune Consequences of Lactate in the
Tumor Microenvironment. Adv Exp Med Biol 1259, 113–124 (2020).

65. Xia, L. et al. The cancer metabolic reprogramming and immune response. Molecular Cancer
20, 28 (2021).

66. Corral, D. et al. ILC precursors differentiate into metabolically distinct ILC1-like cells during
Mycobacterium tuberculosis infection. Cell Rep 39, 110715 (2022).

67. Bibby, J. A. et al. Cholesterol metabolism drives regulatory B cell IL-10 through provision of
geranylgeranyl pyrophosphate. Nat Commun 11, 3412 (2020).

68. Perucha, E. et al. The cholesterol biosynthesis pathway regulates IL-10 expression in human
Th1 cells. Nat Commun 10, 498 (2019).

69. Zhang, J. et al. Neuropilin-1 mediates lung tissue-specific control of ILC2 function in type 2
immunity. Nat Immunol 23, 237–250 (2022).

70. Li, S. et al. Aryl Hydrocarbon Receptor Signaling Cell Intrinsically Inhibits Intestinal Group
2 Innate Lymphoid Cell Function. Immunity 49, 915-928.e5 (2018).

71. Ricardo-Gonzalez, R. R. et al. Tissue-specific pathways extrude activated ILC2s to
disseminate type 2 immunity. Journal of Experimental Medicine 217, e20191172 (2020).

26
72. Huang, Y. et al. S1P-dependent interorgan trafficking of group 2 innate lymphoid cells
supports host defense. Science 359, 114–119 (2018).

73. Mathä, L. et al. Migration of Lung Resident Group 2 Innate Lymphoid Cells Link Allergic
Lung Inflammation and Liver Immunity. Frontiers in Immunology 12, (2021).

74. Sonnenberg, G. F. et al. Innate lymphoid cells promote anatomical containment of lymphoid-
resident commensal bacteria. Science 336, 1321–1325 (2012).

75. Painter, J. D. & Akbari, O. Type 2 Innate Lymphoid Cells: Protectors in Type 2 Diabetes.
Front Immunol 12, 727008 (2021).

76. Krishack, P. A. et al. IL-33-mediated Eosinophilia Protects against Acute Lung Injury. Am J
Respir Cell Mol Biol 64, 569–578 (2021).

77. Lai, D. et al. Group 2 innate lymphoid cells protect lung endothelial cells from pyroptosis in
sepsis. Cell Death Dis 9, 1–12 (2018).

78. Mirchandani, A. S. et al. Type 2 Innate Lymphoid Cells Drive CD4+ Th2 Cell Responses. The
Journal of Immunology 192, 2442–2448 (2014).

79. Christianson, C. A. et al. The Persistence of Asthma requires Multiple Feedback Circuits
Involving ILC2 and IL33. J Allergy Clin Immunol 136, 59-68.e14 (2015).

80. Doherty, T. A. & Broide, D. H. Airway innate lymphoid cells in the induction and regulation
of allergy. Allergology International 68, 9–16 (2019).

81. Helfrich, S., Mindt, B. C., Fritz, J. H. & Duerr, C. U. Group 2 Innate Lymphoid Cells in
Respiratory Allergic Inflammation. Front Immunol 10, 930 (2019).

82. Pelaia, C. et al. Interleukin-5 in the Pathophysiology of Severe Asthma. Front Physiol 10, 1514
(2019).

83. Passalacqua, G. et al. IL-13 and idiopathic pulmonary fibrosis: Possible links and new
therapeutic strategies. Pulm Pharmacol Ther 45, 95–100 (2017).

84. Howard, E. et al. IL-10 production by ILC2s requires Blimp-1 and cMaf, modulates cellular
metabolism, and ameliorates airway hyperreactivity. J Allergy Clin Immunol 147, 1281-1295.e5
(2021).

85. Rigas, D. et al. Type 2 innate lymphoid cell suppression by regulatory T cells attenuates airway
hyperreactivity and requires inducible T-cell costimulator-inducible T-cell costimulator ligand
interaction. J Allergy Clin Immunol 139, 1468-1477.e2 (2017).

27
86.Sicherer, S. H. & Leung, D. Y. M. Advances in allergic skin disease, anaphylaxis, and
hypersensitivity reactions to foods, drugs, and insects in 2014. Journal of Allergy and Clinical
Immunology 135, 357–367 (2015).

87. Moorman, J. E. et al. National surveillance of asthma: United States, 2001-2010. Vital Health
Stat 3 1–58 (2012).

88. Sampson, H. A. Food allergy--accurately identifying clinical reactivity. Allergy 60 Suppl 79,
19–24 (2005).

89. Athari, S. S. Targeting cell signaling in allergic asthma. Sig Transduct Target Ther 4, 1–19
(2019).

90.Ochel, A., Tiegs, G. & Neumann, K. Type 2 Innate Lymphoid Cells in Liver and Gut: From
Current Knowledge to Future Perspectives. Int J Mol Sci 20, 1896 (2019).

91. Wu, D. et al. Eosinophils Sustain Adipose Alternatively Activated Macrophages Associated
with Glucose Homeostasis. Science 332, 243–247 (2011).

92. Feuerer, M. et al. Lean, but not obese, fat is enriched for a unique population of regulatory T
cells that affect metabolic parameters. Nat Med 15, 930–939 (2009).

93. Bénézech, C. & Jackson-Jones, L. H. ILC2 Orchestration of Local Immune Function in
Adipose Tissue. Frontiers in Immunology 10, (2019).

94. Hams, E., Locksley, R. M., McKenzie, A. N. J. & Fallon, P. G. Cutting Edge: IL-25 Elicits
Innate Lymphoid Type 2 and Type II NKT Cells That Regulate Obesity in Mice. The Journal of
Immunology 191, 5349–5353 (2013).

95. Lee, M.-W. et al. Activated Type 2 Innate Lymphoid Cells Regulate Beige Fat Biogenesis.
Cell 160, 74–87 (2015).

96. Brestoff, J. R. et al. Group 2 innate lymphoid cells promote beiging of white adipose tissue
and limit obesity. Nature 519, 242–246 (2015).

97. Zebboudj, A., Amisaki, M. & Balachandran, V. P. ILC2s—Bipartisan politicians in cancer.
Science Immunology 7, eabq2791 (2022).

98. Jou, E. et al. An innate IL-25–ILC2–MDSC axis creates a cancer-permissive
microenvironment for Apc mutation–driven intestinal tumorigenesis. Science Immunology 7,
eabn0175 (2022).

99. Wan, J. et al. ILC2-derived IL-9 inhibits colorectal cancer progression by activating CD8+ T
cells. Cancer Letters 502, 34–43 (2021).

28
100. Qi, L. et al. Interleukin-33 activates and recruits natural killer cells to inhibit pulmonary
metastatic cancer development. International Journal of Cancer 146, 1421–1434 (2020).

101. ßSchuijs, M. J. et al. ILC2-driven innate immune checkpoint mechanism antagonizes NK cell
anti-metastatic function in the lung. Nat Immunol 21, 998–1009 (2020).

29
Chapter 2:
IL-10 production by ILC2s requires Blimp-1 and cMaf, modulates
cellular metabolism and ameliorates airway hyperreactivity

DOI: 10.1016/j.jaci.2020.08.024

Abstract
Background: ILC2s are the dominant innate lymphoid cell population in the lungs at
steady state and their release of type 2 cytokines is a central driver in responding eosinophil
infiltration and increased airway hyperreactivity (AHR). Our laboratory has identified a unique
subset of ILC2s in the lungs that actively produce IL-10 (ILC210).
Objective: We characterized the effector functions of ILC210s in the development and pathology
of allergic asthma.
Methods: IL-4 stimulated ILC210s were isolated to evaluate cytokine secretion, transcription
factor signaling, metabolic dependence, and effector functions in vitro. ILC210s were also
adoptively transferred into Rag2
-/-
γc
-/-
mice, which were then challenged with IL-33 and assessed
for airway hyperreactivity and lung inflammation.
Results: We have identified that transcription factors cMaf and Blimp-1 regulate IL-10 expression
in ILC210s. Strikingly, our results demonstrate that ILC210s can utilize both autocrine and paracrine
signaling to suppress pro-inflammatory ILC2 effector functions in vitro. Further, this subset
dampens AHR and significantly reduces lung inflammation in vivo. Interestingly, ILC210s
demonstrated a metabolic dependency on the glycolytic pathway for IL-10 production, shifting
from the fatty acid oxidation pathway conventionally utilized for pro-inflammatory effector
functions.
Conclusion: These findings provide an important and previously unrecognized role of ILC210s in
diseases associated with ILC2s such as allergic lung inflammation and asthma. They also provide
new insights into the metabolism dependency of pro-inflammatory and anti-inflammatory ILC2
phenotypes.

Introduction

30
Group 2 Innate Lymphoid Cells (ILC2s) are non-B and non-T innate immune cells that are
rapidly activated by epithelium-derived cytokines, interleukin (IL)-33, thymic stromal
lymphopoietin (TSLP), and IL-25(1-3). In response to activation, ILC2s produce type 2 cytokines,
namely IL-5 and IL-13, and are important drivers of eosinophil infiltration and the development
of airway inflammation(3,4). ILC2s are the dominant innate lymphoid cell population in the lungs
at steady state and when challenged with IL-33 or IL-25, strategically cluster at the airway
epithelium and alveolar space(5,6).
While classically defined by their effector functions, recent studies have drawn attention
to the immunoregulatory potential of ILC2s(7-9). ILC2s have been shown to produce TGF- β and
Amphiregulin to promote intestinal homeostasis and repair(10). Additionally, several groups have
demonstrated a subset of ILC2s that produce IL-10, an anti-inflammatory cytokine shown to
downregulate airway hyperreactivity (AHR) and dampen the secretion of various inflammatory
cytokines(7-9). Importantly, ILC2s have substantial plasticity in inflamed tissue. In response to
IL-1β and IL-12 they produce IFN- γ, while in response to IL-6 and TGF- β or Notch signaling they
produce IL17α(11-14). Despite the presence of IL-10 producing ILC2s in many peripheral tissues,
the transcriptional factors and conditions that regulate polarization of ILC2s to ILC210s remain
unresolved.
Delineating the intracellular signaling pathways and cellular mechanisms that drive
immune effector functions is essential to the understanding and modulation of immune cells.
Activated (a)ILC2s have been shown to generate fuel predominantly through the fatty acid
oxidation pathway (FAO), but our group and others have reported activated (a)ILC2s can
metabolically adapt to environmental stimuli (15,16). Moreover, metabolic reprogramming has
increasingly been reported to have a determinate effect of an immune cells’ homeostasis and
effector functions. T-helper (Th) cells, for example, favor the glycolytic pathway for immune
function, while their regulatory counterpart, T-regulatory (Treg) cells, preferentially utilize the
mitochondrial oxidative phosphorylation(17,18). Further investigation into whether there is a
similar metabolic preference shift driving IL-10 producing ILC2s has yet to be explored.
Transcription factor cMaf has been reported to regulate IL-10 production in a variety of
immune cells, including Th1, Th2, Treg, B-regulatory and Th17 cells(19-23). Recently Blimp-1
has been implicated as having a crucial regulatory role in IL-10 production, as well(24-28). In a
number of these studies, IL-4 stimulation has been demonstrated to induce cMaf and Blimp-1

31
intranuclearly, driving the production of IL-10 expression. While traditionally considered a type
2, pro-inflammatory cytokine, IL-4 has recently been implicated as necessary for immune
regulation in a host of diseases. Immune suppression demonstrated by Treg cells for example, is
dependent on the presence of IL-4(29). Similarly, adoptive transfer of IL-4-induced M2
macrophages secreting IL-10 prevents Type 1 diabetes in NOD mice(30). Furthermore, Zhu et al
reported that IL-4 stimulation in dendritic cells results in a shift from pro-inflammatory cell profile
to anti-inflammatory profiles(31). Importantly, Gata3 expression controlling effector functions in
pulmonary ILC2s is dependent on the expression of the IL-4 receptor(32). Th2 cells are major
sources of IL-4 early during lung inflammation and recent reports demonstrate that their crosstalk
is necessary for pulmonary ILC2 accumulation(33). However, the relationship between ILC2 IL-
10 production and IL-4 secretion by Th2 cells is unknown.
In this study, we show that a natural subpopulation of IL-10 producing ILC2s are present
in murine models of asthma described by their copious Th2 cytokines. Further, we identify that
IL-4 stimulation induces IL-10 production and secretion in ILC2 cells (ILC210) via transcription
factors cMaf and Blimp-1 signaling. We demonstrate the immunoregulatory role of ILC210s on
surrounding ILC2 effector function, as well as their striking ability to dampen AHR and lung
eosinophil infiltration in murine models of asthma. The IL-10 production and the generation of
ILC210s are dependent on a metabolic shift from the traditionally preferred fatty acid oxidation
pathway (FAO) for cytokine production to the glycolytic pathway. Together, our findings give
further insight into the fundamental biology of ILC2s and identify therapeutic targets that can be
pharmacologically manipulated to specifically modulate ILC2s for the treatment of a broad array
of inflammatory diseases. Specifically, our studies contribute to significant progress in our
understanding of the underlying immunomodulatory mechanisms responsible for the manipulation
of ILC2s.

Results
IL-10
+
ILC2s are naturally present in allergic lung inflammation murine model.
We first assessed whether IL-10 producing ILC2s were present in the lungs of mice after
induction of a conventionally-utilized allergic murine model of asthma. To do this, we sensitized
and challenged a cohort of IL-10
GFP
mice with OVA-alum, according to our previously established
protocol(39). We sensitized a cohort of IL-10
GFP
mice by an intraperitoneal injection of OVA-
alum on days 0 and 7 (Figure 1A). We then challenged the mice on days 14, 16, and 18 with OVA

32
or PBS intranasally. On day 19, we measured the number (Figure 1C) and frequency (Figure
E1A, Figure 1B) of IL-10 producing ILC2s present in the lungs, as well as the number of T
regulatory and T helper cells (Figure E1B). After OVA challenge, there was an increase in the
number of ILC2s, Th2 cells, and T regulatory cells the murine lung. Importantly, however, the
number of IL-10-producing pulmonary ILC2s (ILC210s) increased significantly, from a mean of
10,000 cells per lung in the control group, to 70,000 per lung in the OVA-treated group, suggesting
a significant induction of ILC210 generation during allergic inflammation. Additionally, the
number of ILC210s in the lung after OVA challenge is remarkably similar to the number of T
regulatory cells.
As Th2 cells producing IL-4 and IL-13 are the dominant cells driving the OVA-alum
allergic airway model, we investigated the direct effect of Th2 cells and their IL-4 secretion on the
production of IL-10 in culture(40). To test this, we first isolated CD4
+
CD25
-
cells from WT
spleens (Figure 1D) and stimulated with anti-CD3 and anti-CD28 to ensure Th2 differentiation,
according to a previously established protocol(36). Concurrently, we activated ILC2s in a different
cohort of WT mice with three days of intranasal recombinant mouse (rm)IL-33. We then FACS-
sorted aILC2s from the lungs of the rmIL-33 challenged mice, and Th2 cells from the stimulated,
differentiated spleen cells. We co-cultured the two populations with anti-IL-4 or isotype together
for 48 hours, collected the supernatants, and measured cytokine levels by ELISA. We also
collected the cells in order to confirm the identity the corresponding cell populations by FACS
(Figure E1C). Importantly, we first established that the cultured Th2 cells secreted on average 95
pg/ml of IL-4 and this secretion is completely undetectable with the addition of anti-IL-4 (Figure
1E). While some IL-13 was secreted, the most abundant cytokine was IL-4 (data not shown). We
then measured the level of IL-10 in the supernatant by ELISA. In line with our previous findings,
supernatant from wells with ILC2s and Th2 cells demonstrated a dramatic increase in the level of
IL-10, and this significantly reduced with the addition of anti-IL-4 (Figure 1F), suggesting IL-4
from T cells is necessary for the production of IL-10.
These results indirectly suggest IL-10 is produced by ILC2s after IL-4 stimulation. To
investigate further the direct effect of Th2 and IL-4 secretion on surrounding ILC2 effector
function, we measured the frequency of IL-5, IL-13- producing ILC2s after co-culture by
intracellular flow cytometry. Surprisingly, co-culture with Th2 cells dose-dependently reduced the
frequency of IL-5
+
IL-13+ ILC2s in culture (Figure 1G). To confirm that this observed effect is

33
indeed dependent on IL-4 signaling, we treated cells with anti-IL-4 monoclonal antibody (mAb)
or isotype. Consistent with our previous findings, absence of IL-4 signaling restored ILC2 effector
function (Figure 1H), Together, our findings show that T-cell derived IL-4 drives IL-10
production in pulmonary ILC2s in a model of OVA-alum allergic airway inflammation.

IL-4 induces ILC2
10
generation and downregulates effector function.
In order to study the ILC210 effector function further, we explored whether we could
develop an ex vivo method of ILC210 generation. IL-10
GFP
mice were given three days of rmIL-33
intranasally to activate lung ILC2s (Figure 2A). The IL-10-GFP negative ILC2s were then isolated
and cultured for 48 hours with rmIL-2, rmIL-7, with and without IL-33 (as indicated by
experiment) and IL-4 (10ng/ml) (Figure 2A). After 48 hours with rmIL-4, an average of 50% of
the cultured ILC2s became ILC210s (Figure 2B). Importantly, as our selection criteria for aILC2
dictates that we begin experiments with IL-10
-
aILC2s and then stimulate with rmIL-4, the results
reported here focus on the newly generated ILC210s.
We investigated whether IL-4 had any further effect on aILC2 effector function. We
activated and FACS-sorted aILC2s, as previously demonstrated, and cultured these cells with IL-
2, IL-7, with and without IL-33 and IL-4 for 48 hours. After 48 hours, cytokine secretion levels
were measured in the supernatant by ELISA (Figure 2C) and by flow cytometry with intracellular
staining (Figure 2D). Interestingly, while rmIL-33 induced a small percentage of ILC210s, the co-
stimulation of rmIL-33 and rmIL-4 gave the highest percentage of ILC210s. In line with our
previous findings, rmIL-4 stimulation specifically downregulated IL-5
+
IL-13
+
ILC2s, compared
to continuous rmIL-33 stimulation which drives copious cytokine secretion (Figure 2D).
As previously mentioned, Th2 cells secrete both IL-4 and IL-13 in the pathogenesis of
allergic airway inflammation(40). To test whether rmIL-13 had a similar effect on ILC210
generation and effector function, we cultured aILC2s as previously described with rmIL-13 to
measure the frequency of IL-5
+
IL-13+ ILC2s (Figure 2E) and ILC210s (Figure 2F). Strikingly
rmIL-13 did not have the same effect as rmIL-4 and did not induce ILC210 production. Together,
these results suggest rmIL-4 induces ILC210 generation and decreases aILC2 effector function in
a manner separate from rmIL-33 or rmIL-13.

cMaf and Blimp-1 are necessary for IL-10 generation in ILC2
10
s.

34
We next investigated the molecular mechanism by which IL-10 is produced in ILC210s.
The transcription factor cMaf binding to the MARE motif of IL-10 has been identified in the
signaling pathway responsible for IL-4 stimulated IL-10 production in B cells, macrophages, and
Th17 cells(22,23,41). In Th1 cells, Blimp-1 (Prdm1) controls IL-10 production by IL-4 stimulation
but is assisted by cMaf(24). Blimp-1 regulation has already been associated with both Stat4 and
Stat1/3 pathways, as well(24,25) .
As a proof of concept, we demonstrated the necessity for the IL-4 receptor on ILC210 IL-
10 production by adding anti-IL-4 or isotype to cultured aILC2s with rmIL-4 for 48 hours. We
quantified ILC210 generation by flow cytometry (Figure 3A). Not surprisingly, addition of anti-
IL-4 inhibited IL-10 production, suggesting the IL-4 receptor is required for IL-4 stimulated IL-
10 production. A step further down the signaling pathway, the IL-4 receptor, when activated,
utilizes the transcription factor Stat6 to carry out downstream functions(42). To determine if there
was an effect of Stat6 knock-down on IL-10 production, we cultured aILC2s with a Stat6
morpholino (MO), an anti-sense oligomer, for 24 hours before addition of rmIL-4. After an
additional 48 hours with the morpholinos and rmIL-4, cells were collected for intranuclear protein
expression staining by flow cytometry (Figure 3B) and cytokine levels in the supernatant were
measured by ELISA (Figure 3C). And indeed, addition of rmIL-4 induced Stat6 protein
expression levels in cells cultured with control morpholino, while the Stat6 morpholino
significantly reduced protein expression (Figure 3B). Additionally, cells with little to no Stat6
protein expression secreted significantly less IL-10 than those cultured with the control (Figure
3C), suggesting Stat6 is crucial for IL-10 production in ILC210s.
As previously mentioned, cMaf and Blimp-1 are transcription factors implicated in the IL-
4 stimulated IL-10 production in various immune cells. To determine if these are also responsible
for the IL-10 production in ILC210s downstream IL-4R and Stat6 activation, we first investigated
whether the transcription factors were detectable in ILC210s and aILC2s by intranuclear staining
(Figure 3D-E). Both transcription factors stained positive above isotype staining by intracellular
flow cytometry in ILC210s and aILC2s, while no protein was detected in naïve ILC2s (Figure
E3A). cMaf and Blimp-1 are further upregulated by IL-4 as compared to aILC2 controls. We next
replicated the experiments performed previously with the Stat6 morpholino, but instead with the
addition of the appropriate cMaf and Blimp-1 morpholinos. We show that addition of the
appropriate morpholino, but not the control morpholino, knocked down protein expression after

35
24 hours by intranuclear staining (Figure 3F-G). After culture with the relative morpholino for 24
hours, we added rmIL-4 for 48 hours, and subsequently measured ILC210 generation by flow
cytometry, as well as IL-5 and IL-13 expression level (Figure 3H, Figure E3B). Remarkably,
addition of either transcription factor morpholino significantly reduced the frequency of ILC210s,
suggesting both transcription factors are vital for IL-10 production after IL-4 stimulation.
Together, these results suggest that the IL-4 receptor and downstream Stat6 transcription factor
are necessary for IL-10 production in ILC210s via upregulation of both cMaf and Blimp-1
transcription factors.

ILC2
10
s suppress activated ILC2 effector function in vitro.
As IL-10 is considered both a Th2 and an anti-inflammatory cytokine, further exploration
into the effect ILC210s have on their surrounding microenvironment was required. As a proof of
concept, we first investigated whether aILC2s express the IL-10 receptor and have the capability
to respond to the cytokine. aILC2s do indeed express the IL-10 receptor (Figure 4A). Our group
has previously demonstrated a decrease in ILC2 cytokine release following contact with Tregs, an
immune cell characterized by its release of IL-10(38). Due to these results, we next tested if aILC2s
cultured with rmIL-10 also exhibit a decrease in effector function. aILC2s were cultured with
media, rmIL-4, or rmIL-10 for 48 hours and effector function was measured intracellularly by flow
cytometry (Figure 4B). As expected, aILC2s effector function, measured by the frequency of IL-
5
+
IL-13
+
ILC2s, is downregulated after co-culture with rmIL-10 to similar extent as rmIL-4
addition.
Cultured ILC210s secrete lower levels of IL-5 and IL-13 as compared to aILC2s (Figure
E3A-B). While our previous results suggest IL-4 downregulates ILC210 IL-5 and IL-13
production, we wondered if the secreted IL-10 possessed an autocrine function in the further
downregulation of pro-inflammatory effector function. To test this, we cultured sorted ILC210s
with rmIL-4 for 48 hours to stimulate continued IL-10 production, though with the addition of
anti-IL-10R mAb or isotype. The antibody, while having no effect on the rmIL-4, neutralizes any
IL-10 and therefore inhibits any downstream effect. We show here that after the addition of anti-
IL-10R, there is partial restoration of ILC210 pro-inflammatory effector function detectable by
flow cytometry, suggesting that the secreted IL-10 serves an additional autocrine function in
downregulating inflammatory cytokines (Figure 4C).

36
To uncover if ILC210 IL-10 production directly has an effect on aILC2s, we first cultured
ILC210s and control aILC2s for 48 hours (Figure 4D). After 48 hours, we collected and applied
the supernatant to fresh aILC2s for an additional 48 hours. An aliquot of the supernatant was
assessed for IL-10 levels by ELISA (Figure 4E). We then collected the cells, and measured IL-5
+

and IL-13
+
aILC2 levels by intracellular flow cytometry (Figure 4F). Remarkably, we found that
aILC2s cultured with ILC210 supernatant showed a significant decrease in both IL-5 and IL-13
production (Figure 4F), while viability and proliferation via Ki67 expression remained unaltered
as compared to the cells in the control group (Figure E3C). Additionally, while cell proliferation
remained unchanged between the conditions, the Gata3 expression level, the transcription factor
vital to cytokine production and aILC2 development, was mildly but significantly decreased in
cells cultured with supernatant from ILC210s (Figure 4G). To investigate whether the decreased
effector function was indeed a consequence of the IL-10 from the ILC210 culture and not a
different, unidentified protein, we repeated the experiment with the addition of anti-IL-10R mAb.
Cells cultured with ILC210s and anti-IL-10R mAb did indeed partially, though significantly, restore
aILC2 effector function compared to control cells (Figure 4H). Together, these findings suggest
ILC210s have the ability to directly decrease aILC2 effector function through the secretion of IL-
10. Moreover, the secreted IL-10 serves the additional autocrine function of even further
downregulating ILC210s’ IL-5 and IL-13 production.

ILC2
10
s downregulate airway hyperreactivity and lung inflammation.
In order to further explore the biological relevance of ILC210s in vivo in the context of
allergic asthma, we sorted both ILC210s and aILC2s from WT and IL-10
GFP
mice. We then
adoptively transferred only aILC2s or a 1:1 mix of ILC210s and ILC2s (Figure 5A) by tail injection
to Rag
-/-
γc
-/-
mice, a strain that lacks all B, T, and innate lymphoid cells. 24 hours after transfer,
we began three consecutive days of intranasal challenge with rmIL-33 in PBS. To test the
dependency on IL-10 signaling we included a group of Rag
-/-
γc
-/-
mice with the same 1:1 mix
ILC210: aILC2 with the addition of anti-IL-10R mAb i.p. injections at the time of intranasal
challenge.
After three days of rmIL-33 intranasal challenge, lung functions were assessed, and BAL
fluid was collected on day 4. ILC210s significantly downregulated ILC2-dependent airway
hyperreactivity compared to control mice with ILC2 alone (Figure 5B). Additionally, AHR

37
reduction was partially abrogated by i.p. injections of anti-IL-10R, suggesting ILC210s biologically
reduce inflammation (Figure 5B). With the presence of ILC210s, the decrease in AHR induction
corresponded with a decrease in eosinophils infiltrating the BAL with presence of ILC210s (Figure
5C). To explore whether the decrease in AHR was due to an increase in ILC2s trafficking to the
lung after transfer, we harvested the lung tissue and counted the number of ILC2s and their
proliferation in the respective groups. Despite the reduced AHR induction, there were more ILC2s
in the lungs of groups with a 1:1 mix of ILC210s: aILC2s compared to ILC2 alone, potentially due
to the decreased ILC2 proliferation in these mice (Figure 5D, Figure E4).
We next confirmed our findings in a more physiologically relevant setting utilizing
Alternaria alternata (A. alternata), an allergen known to indirectly stimulate ILC2s (Figure 5E).
In line with our previous findings, mice adoptively transferred a 1:1 mix ILC210: aILC2 exhibited
significantly damped AHR, as well as a decrease in BAL eosinophils (Figure 5F-G). This was
restored in mice given injections of anti-IL-10R throughout the allergen challenge. Importantly,
mice adoptively transferred only aILC2s but with injections of anti-IL-10R did not show the same
AHR restoration, suggesting the dampened ILC2 response was due to the IL-10 secreted by the
transferred ILC210s and not another cell in the mouse. Cell analysis of the lungs demonstrated
similar numbers of ILC2s in each group, though effector function of the ILC2s in the 1:1 mix
group was significantly downregulated (Figure 5I). Together, our findings provide strong
evidence that ILC210s play a crucial and previously underrecognized regulatory role in disease
pathogenesis.

ILC2
10
generation is mediated by a metabolic shift from the fatty acid oxidation pathway to
the glycolytic pathway.
Emerging evidence investigating T-effector cells, macrophages, T-regulatory cells, and
neutrophils support the idea that metabolic mechanisms possess the ability to alter immune
responses(15,43,44). Our group and others have previously established that traditional aILC2s
favor the fatty acid oxidation (FAO) pathway as a mechanism to generate energy(15,16). We
therefore investigated if this unique subtype, ILC210s, prefers the same pathway or if metabolic
changes were associated with the shift in inflammatory properties.
To further explore, we performed RNA sequencing (RNAseq) on aILC2s and ILC210s after
FACS-sorting (Figure 6A). Surprisingly, 306 genes were significantly downregulated in ILC210s,

38
while only 56 genes were significantly upregulated (Fold change > 2, P<0.05, Figure 6B). A
myriad of the genes that was significantly upregulated were related in some capacity to the
glycolytic metabolic pathway (Figure 6C), suggesting a possible metabolic shift in metabolism.
To test whether ILC210s do indeed preferentially utilize glycolysis over the conventional FAO
pathway, we measured glucose uptake and lipid droplet quantification in ILC210s compared to
aILC2s (Figure 6D-E). ILC210s demonstrated a significant increase in glucose uptake and
decreased intracellular lipid droplet accumulation as compared to aILC2s. Additionally, we
quantified levels of L-lactate, the end product of glycolysis, in the supernatant of ILC210s and
aILC2s (Figure 6F). There was a significant increase in the level of L-lactate produced by ILC210s,
suggesting an increase in the use of the glycolytic pathway in ILC210s. To test whether ILC210
generation also relies on glycolysis, we cultured aILC2s with rmIL-4 and various metabolic
pathway inhibitors for 48 hours before measuring ILC210 generation by flow cytometry (Figure
6G). DEUP, orlistat, and etomoxir are well-known inhibitors of the FAO pathway. Blocking the
FAO pathway did not inhibit ILC210 generation, but addition of 2-DG, a glycolysis inhibitor, did
significantly reduce IL-10 production (Figure 6G, Figure E5). We then repeated the ILC210
generation experiment but cultured the aILC2s with rmIL-4 in conventional cRPMI (with 10mM
glucose) and in RPMI supplemented with 10mM galactose for 48 hours. Cells grown in media
with galactose, as opposed to glucose, have shown enhanced oxidative phosphorylation and a
decrease in the rate of glycolysis(45,46). In line with our previous findings, aILC2s cultured with
rmIL-4 in galactose media did not produce any ILC210s detectable by flow cytometry (Figure 6H).

Discussion
Overall this study introduces a novel, IL-4-inducible IL-10-producing population of
ILC210s in the lung are naturally present in a high Th2 model of murine asthma. We show that the
IL-10 production in these cells corresponds with transcription factors cMaf and Blimp-1
upregulation, and that knock-down of either protein inhibits the generation of IL-10. Moreover,
these cells have the ability, through both autocrine and paracrine signaling, to downregulate
immune cell effector function and ameliorate AHR in vivo. In addition to the transcription factors
responsible for IL-10 production, this ILC2 pro-inflammatory to regulatory shift is modulated by
a metabolic preference toward the glycolytic pathway.

39
In patients with high Th2 asthma, characterized by an overproduction of T cell activation
and subsequent abundance of Th2 cytokine secretion, it is essential for the lungs to control the
dysregulated immune response of various immune cells, including ILC2s(47). Based on our
results, we propose that the production of IL-4, predominantly secreted by Th2 cells in allergic
models of asthma, inhibits activation and effector function of murine ILC2s. Moreover, the IL-4
promotes a shift in ILC2s from pro-inflammatory to anti-inflammatory in order to act as a potent
immunomodulatory mechanism in the development of AHR. Previously, several groups
demonstrated the induction of IL-10 expression in ILC2s through distinct cytokine stimuli, such
as IL-33 and IL-2, using various disease models(8,48). Seehus et al. reported that Retinoic Acid
along with IL-2 could also induce IL-10 producing cells in vitro(8). However, these cells
maintained high levels of IL-5 and IL-13 expression. Indeed, Bando et al. described IL-4
stimulation induced ILC210s in the gastrointestinal tract, though the transcription factors necessary
for the IL-10 production had not been explored(7). IL-4 can also maintain human ILC2
proliferation in vitro and maintain type 2 identity(14).
Our group, to our knowledge, is the first to report ILC210s and their role in dampening
AHR development. However, the role of ILC210s and their immunoregulatory function has been
previously reported in the kidneys and intestinal tissue. Interestingly, Morita et al. demonstrated
an induction of ILC210s in the nasal tissue of human patients with CRSwNP(49). Cao et al. showed
in both that this population was actively present in both human and mouse kidneys, and their
protective role in vivo against the development of murine renal damage(50). Bando et al. in also
showed that ILC2s are a major source of IL-10 in the gut(7). Although it is not clear if IL-4
signaling and receptor is important in the induction of ILC210 development in these models, several
lines of research suggest that this population is capable of immune regulation. There is emerging
evidence of their importance in the dampening of pro-inflammatory cells and the maintenance of
peripheral immunological tolerance.
Our group and others have previously established the important immunoregulatory
function of secreted IL-10 from T regulatory cells in the lung during asthmatic challenge(38,51).
Additionally, there is accumulating evidence that IL-4 stimulation induces IL-10 secretion in a
variety of immune cells, including B cells, macrophages, Th1, Th2 and Th17 cells(22-24,29,30).
These studies have identified transcription factors cMaf and Blimp-1 as the active master
regulators of IL-10 production, though no such investigation has been done to determine if this is

40
similar in ILC210s. Here we demonstrate that both cMaf and Blimp-1 are essential for IL-10
production in ILC210s. Knock-down of either protein resulted in a significant decrease of ILC210
generation.
In agreement with previous reports, the secreted IL-10 from ILC210s significantly
suppressed immune cell effector function in vitro and in vivo utilizing both autocrine and paracrine
signaling methods. Strikingly, in addition to discovery of the master IL-10-controlling
transcription factors in ILC210s, we also discovered ILC210 generation is dependent on glucose
availability and functional completion of the glycolytic pathway. Our group and others have
previously shown the FAO pathway as the preferred metabolic pathway for aILC2s in proliferation
and cytokine secretion(15,16). We show ILC210s’ IL-10 production, unlike aILC2s’ pro-
inflammatory cytokine secretion, has no dependency on the FAO pathway, as inhibition of the
FAO pathway resulted in no significant reduction in the ILC210 production in response to IL-4.
In summary, in this report we reveal a previously unrecognized immunoregulatory role for
IL-4 stimulated ILC210s in AHR development. Our results suggest a protective role of ILC210s in
the development of IL-33-induced asthma. The identification of important controlling
transcription factors and the metabolic reprogramming responsible for the ILC210 generation
allows for a potentially inducible, anti-inflammatory role for a cell that is uniquely positioned in
the lung to act as the first responder to an allergen response. Similar to the therapeutic diagnostic
role of T effector to Treg ratio in asthma, we propose a parallel importance to the aILC2 to ILC210
ratio for determining disease outcome(47). We believe the knowledge gained by this report
increases the therapeutic potential of anti- cytokine therapies. We also believe these data could
help explain the lack of efficacy on eosinophil reduction using anti-IL-4 or IL-4R targeting
therapeutics in the field of allergic asthma and lung inflammation.

41
References
1.Hurrell BP, Shafiei Jahani P, Akbari O. Social Networking of Group Two Innate Lymphoid Cells
in Allergy and Asthma. Front Immunol 2018;9:2694.

2. Maazi H, Akbari O. Type two innate lymphoid cells: the Janus cells in health and disease.
Immunol Rev 2017 07;278(1):192-206.

3. Barlow JL, Peel S, Fox J, Panova V, Hardman CS, Camelo A, et al. IL-33 is more potent than
IL-25 in provoking IL-13-producing nuocytes (type 2 innate lymphoid cells) and airway
contraction. J Allergy Clin Immunol 2013 Oct;132(4):933-941.

4. Chang Y, Kim HY, Albacker LA, Baumgarth N, McKenzie ANJ, Smith DE, et al. Innate
lymphoid cells mediate influenza-induced airway hyper-reactivity independently of adaptive
immunity. Nat Immunol 2011 May 29,;12(7):631-638.

5. Mindt BC, Fritz JH, Duerr CU. Group 2 Innate Lymphoid Cells in Pulmonary Immunity and
Tissue Homeostasis. Front Immunol 2018;9:840.

6. Li Y, Chen S, Chi Y, Yang Y, Chen X, Wang H, et al. Kinetics of the accumulation of group 2
innate lymphoid cells in IL-33-induced and IL-25-induced murine models of asthma: a potential
role for the chemokine CXCL16. Cell Mol Immunol 2019 01;16(1):75-86.

7. Bando JK, Gilfillan S, Di Luccia B, Fachi JL, Sécca C, Cella M, et al. ILC2s are the predominant
source of intestinal ILC-derived IL-10. J Exp Med 2020 Feb 03,;217(2).

8. Seehus CR, Kadavallore A, Torre Bdl, Yeckes AR, Wang Y, Tang J, et al. Alternative activation
generates IL-10 producing type 2 innate lymphoid cells. Nat Commun 2017 12 01,;8(1):1900.

9. Miyamoto C, Kojo S, Yamashita M, Moro K, Lacaud G, Shiroguchi K, et al. Runx/Cbfβ
complexes protect group 2 innate lymphoid cells from exhausted-like hyporesponsiveness during
allergic airway inflammation. Nature Communications 2019;10.

10. Monticelli LA, Osborne LC, Noti M, Tran SV, Zaiss DMW, Artis D. IL-33 promotes an innate
immune pathway of intestinal tissue protection dependent on amphiregulin–EGFR interactions.
Proc Natl Acad Sci U S A 2015 -8-25;112(34):10762-10767.

11. Zhang K, Xu X, Pasha MA, Siebel CW, Costello A, Haczku A, et al. Cutting Edge: Notch
Signaling Promotes the Plasticity of Group-2 Innate Lymphoid Cells. J Immunol 2017 03
01,;198(5):1798-1803.

12. Golebski K, Ros XR, Nagasawa M, van Tol S, Heesters BA, Aglmous H, et al. IL-1β, IL-23,
and TGF-β drive plasticity of human ILC2s towards IL-17-producing ILCs in nasal inflammation.
Nat Commun 2019 05 14,;10(1):2162.

42
13. Ohne Y, Silver JS, Thompson-Snipes L, Collet MA, Blanck JP, Cantarel BL, et al. IL-1 is a
critical regulator of group 2 innate lymphoid cell function and plasticity. Nat Immunol 2016
06;17(6):646-655.

14. Sm B, Jh B, M N, J G, Mm S, K G, et al. IL-1β, IL-4 and IL-12 control the fate of group 2
innate lymphoid cells in human airway inflammation in the lungs. Nat Immunol 2016
/04/25;17(6):636-645.

15. Berod L, Friedrich C, Nandan A, Freitag J, Hagemann S, Harmrolfs K, et al. De novo fatty
acid synthesis controls the fate between regulatory T and T helper 17 cells. Nature Medicine 2014
-11;20(11):1327-1333.

16. Michalek RD, Gerriets VA, Jacobs SR, Macintyre AN, MacIver NJ, Mason EF, et al. Cutting
edge: distinct glycolytic and lipid oxidative metabolic programs are essential for effector and
regulatory CD4+ T cell subsets. J Immunol 2011 Mar 15,;186(6):3299-3303.

17. Xu M, Pokrovskii M, Ding Y, Yi R, Au C, Harrison OJ, et al. c-MAF-dependent regulatory T
cells mediate immunological tolerance to a gut pathobiont. Nature 2018 02 15,;554(7692):373-
377.

18. Xie Q, McGreal R, Harris R, Gao CY, Liu W, Reneker LW, et al. Regulation of c-Maf and
αA-Crystallin in Ocular Lens by Fibroblast Growth Factor Signaling. J Biol Chem 2016 Feb
19,;291(8):3947-3958.

19. Gabryšová L, Alvarez-Martinez M, Luisier R, Cox LS, Sodenkamp J, Hosking C, et al. c-Maf
controls immune responses by regulating disease-specific gene networks and repressing IL-2 in
CD4 + T cells. Nature Immunology 2018 -05;19(5):497-507.

20. Xu J, Yang Y, Qiu G, Lal G, Wu Z, Levy DE, et al. c-Maf Regulates IL-10 Expression during
Th17 Polarization. Journal of immunology (Baltimore, Md.: 1950) 2009;182(10):6226-6236.

21. Liu M, Zhao X, Ma Y, Zhou Y, Deng M, Ma Y. Transcription factor c-Maf is essential for IL-
10 gene expression in B cells. Scand J Immunol 2018;88(3):e12701.

22. Neumann C, Heinrich F, Neumann K, Junghans V, Mashreghi M, Ahlers J, et al. Role of
Blimp-1 in programing Th effector cells into IL-10 producers. J Exp Med 2014;211(9):1807-1819.

23. Martins GA, Cimmino L, Shapiro-Shelef M, Szabolcs M, Herron A, Magnusdottir E, et al.
Transcriptional repressor Blimp-1 regulates T cell homeostasis and function. Nat Immunol 2006
May;7(5):457-465.

24. Sun M, He C, Chen L, Yang W, Wu W, Chen F, et al. RORγt Represses IL-10 Production in
Th17 Cells To Maintain Their Pathogenicity in Inducing Intestinal Inflammation. J Immunol 2019
01 01,;202(1):79-92.

43
25. Garg G, Muschaweckh A, Moreno H, Vasanthakumar A, Floess S, Lepennetier G, et al. Blimp1
Prevents Methylation of Foxp3 and Loss of Regulatory T Cell Identity at Sites of Inflammation.
Cell Rep 2019 02 12,;26(7):1854-1868.e5.

26. Wang Y, Tsai D, Ko Y, Yang T, Lin I-, Hung K, et al. Blimp-1 Contributes to the Development
and Function of Regulatory B Cells. Front Immunol 2019;10:1909.

27.Yang W, Hwang Y, Chen Y, Liu C, Shen C, Hong W, et al. Interleukin-4 Supports the
Suppressive Immune Responses Elicited by Regulatory T Cells. Frontiers in Immunology 2017;8.

28. Parsa R, Andresen P, Gillett A, Mia S, Zhang X, Mayans S, et al. Adoptive transfer of
immunomodulatory M2 macrophages prevents type 1 diabetes in NOD mice. Diabetes
2012;61(11):2881-2892.

29. Zhu B, Buttrick T, Bassil R, Zhu C, Olah M, Wu C, et al. Interleukin 4 and retinoic acid
synergistically induce regulatory dendritic cells expressing Aldh1a2. Journal of immunology
(Baltimore, Md.: 1950) 2013;191(6):3139-3151.

30. Zhu J. GATA3 Regulates the Development and Functions of Innate Lymphoid Cell Subsets
at Multiple Stages. Frontiers in immunology 2017;8:1571.

31. Symowski C, Voehringer D. Th2 cell-derived IL-4/IL-13 promote ILC2 accumulation in the
lung by ILC2-intrinsic STAT6 signaling in mice. European Journal of Immunology
2019;49(9):1421-1432.

32.Galle-Treger L, Sankaranarayanan I, Hurrell BP, Howard E, Lo R, Maazi H, et al.
Costimulation of type-2 innate lymphoid cells by GITR promotes effector function and ameliorates
type 2 diabetes. Nat Commun 2019 02 12,;10(1):713.

33. Galle-Treger L, Lewis BP, Lewis G, Howard E, Shafiei Jahani P, Banie H, et al. Autophagy
is critical for group 2 innate lymphoid cell metabolic homeostasis and effector function. Journal
of Allergy and Clinical Immunology 2020 /02/01;145(2):502-517.e5.

34. Lewis G, Wang B, Shafiei Jahani P, Hurrell BP, Banie H, Aleman Muench GR, et al. Dietary
Fiber-Induced Microbial Short Chain Fatty Acids Suppress ILC2-Dependent Airway
Inflammation. Frontiers in immunology 2019;10:2051.

35. Akbari O, Stock P, Meyer EH, Freeman GJ, Sharpe AH, Umetsu DT, et al. ICOS/ICOSL
Interaction Is Required for CD4+ Invariant NKT Cell Function and Homeostatic Survival. J
Immunol 2008 -4-15;180(8):5448-5456.

36. Hurrell BP, Galle-Treger L, Jahani PS, Howard E, Helou DG, Banie H, et al. TNFR2 Signaling
Enhances ILC2 Survival, Function, and Induction of Airway Hyperreactivity. Cell Reports 2019
Dec 24,;29(13):4509-4524.e5.

44
37. Rigas D, Lewis G, Aron JL, Wang B, Banie H, Sankaranarayanan I, et al. Type 2 innate
lymphoid cell suppression by regulatory T cells attenuates airway hyperreactivity and requires
inducible T-cell costimulator-inducible T-cell costimulator ligand interaction. J Allergy Clin
Immunol 2017;139(5):1468-1477.e2.

38. Oh J, Seroogy CM, Meyer EH, Akbari O, Berry G, Fathman CG, et al. CD4 T-helper cells
engineered to produce IL-10 prevent allergen-induced airway hyperreactivity and inflammation. J
Allergy Clin Immunol 2002 Sep;110(3):460-468.

39. Oeser K, Maxeiner J, Symowski C, Stassen M, Voehringer D. T cells are the critical source
of IL-4/IL-13 in a mouse model of allergic asthma. Allergy 2015 Nov;70(11):1440-1449.

40. Cao S, Liu J, Song L, Ma X. The Protooncogene c-Maf Is an Essential Transcription Factor
for IL-10 Gene Expression in Macrophages. Journal of immunology (Baltimore, Md.: 1950)
2005;174(6):3484-3492.

41. Junttila IS. Tuning the Cytokine Responses: An Update on Interleukin (IL)-4 and IL-13
Receptor Complexes. Front Immunol 2018;9.

42. Wilhelm C, Harrison OJ, Schmitt V, Pelletier M, Spencer SP, Urban JF, et al. Critical role of
fatty acid metabolism in ILC2-mediated barrier protection during malnutrition and helminth
infection. J Exp Med 2016;213(8):1409-1418.

43. Buck MD, Sowell RT, Kaech SM, Pearce EL. Metabolic Instruction of Immunity. Cell
2017;169(4):570-586.

44. Angajala A, Lim S, Phillips JB, Kim J, Yates C, You Z, et al. Diverse Roles of Mitochondria
in Immune Responses: Novel Insights Into Immuno-Metabolism. Frontiers in Immunology
2018;9.

45. Kase ET, Nikolić N, Bakke SS, Bogen KK, Aas V, Thoresen GH, et al. Remodeling of
Oxidative Energy Metabolism by Galactose Improves Glucose Handling and Metabolic Switching
in Human Skeletal Muscle Cells. PLoS One 2013 -4-1;8(4).

46. Aguer C, Gambarotta D, Mailloux RJ, Moffat C, Dent R, McPherson R, et al. Galactose
Enhances Oxidative Metabolism and Reveals Mitochondrial Dysfunction in Human Primary
Muscle Cells. PLoS One 2011 -12-15;6(12).

47. Till SJ, Francis JN, Nouri-Aria K, Durham SR. Mechanisms of immunotherapy. The Journal
of Allergy and Clinical Immunology 2004;113(6):1025-1034.

48. Cameron GJM, Cautivo KM, Loering S, Jiang SH, Deshpande AV, Foster PS, et al. Group 2
Innate Lymphoid Cells Are Redundant in Experimental Renal Ischemia-Reperfusion Injury. Front
Immunol 2019;10:826.

45
49. Morita H, Kubo T, Rückert B, Ravindran A, Soyka MB, Rinaldi AO, et al. Induction of human
regulatory innate lymphoid cells from group 2 innate lymphoid cells by retinoic acid. J Allergy
Clin Immunol 2019 Jun;143(6):2190-2201.e9.

50. Cao Q, Wang R, Wang Y, Niu Z, Chen T, Wang C, et al. Regulatory innate lymphoid cells
suppress innate immunity and reduce renal ischemia/reperfusion injury. Kidney International 2020
/01/01;97(1):130-142.

51. Aron JL, Akbari O. Regulatory T cells and type 2 innate lymphoid cell-dependent asthma.
Allergy 2017;72(8):1148-1155.

46
Materials and Methods
Mouse experiments
Experimental protocols were approved by the USC institutional Animal Care and Use Committee
(IACUC) and conducted in accordance with the USC Department of Animal Resources’
guidelines. 5-10 week old age and sex matched mice were used in the studies. C57BL/6J,
BALB/cByJ, Vert-X/IL-10
GFP
(B6(Cg)-Il10tm1.1Karp/J), and Rag2
-/-
γc
-/-
(C;129S4-
Rag2tm1.1Flv Il2rgtm1.1Flv/J) mice were bred in our animal facility at the Keck School of
Medicine, University of Southern Calißfornia (USC).

In vivo experiments and tissue preparation
For induction of OVA-induced allergic airway inflammation models, mice were sensitized with
100µg OVA complexed with aluminum hydroxide (alum, 2.25mg) on days 0 and 7 by
intraperitoneal (i.p.) injection. On days 14, 16, and 18, mice were intranasally (i.n.) challenged
with OVA (100µg in 50 µL) or PBS after administration of light anesthesia. On day 19, mice were
euthanized, lungs were collected and processed for the indicated readout as described
previously(16,34). For IL-33 in vivo treatment, when indicated, mice were challenged on 3
consecutive days with carrier-free rmIL-33 (0.5µg/mouse in 50µL, BioLegend) or PBS. On day 4,
lungs were collected and processed as previously described(35). Briefly, the lungs were perfused
with PBS and digested in Collagenase IV (MP Biomedicals, LLC) for 1 hour at 37°C. The lungs
were then stained with antibodies to identify aILC2s. aILC2s were gated as lineage (CD3e, CD5,
CD45R, Gr-1, CD11c, CD11b, Ter119, TCR γ δ, TCR β and FC εRI α) negative, CD45
+
, ST2
+
,
CD127
+
cells. Where indicated, ILC210s from IL-10
GFP
mice were identified by their positive GFP
fluorescence. For Th2 differentiation, spleens from BALB/cByJ mice were harvested and
processed as previously described(36).

Flow Cytometry
The following murine antibodies were used: biotinylated anti-mouse lineage CD3e (145-2C11),
CD5 (53-7.3), TCR β (H57-597), CD45R (RA3-6B2), Gr-1 (RB6-8C5), CD11c (N418), CD11b
(M1/70), Ter119 (TER-119), Fc εRI α (MAR-1), TCR γ δ (eBioGL3), Streptavidin-FITC,
Streptavidin-APC, PE-Cy7 anti-mouse CD127 (A7R34), APCCy7 anti-mouse CD45 (30-F11),
PECy7 anti-mouse CD45 (30-F11), APCCy7 anti-mouse CD11c (N418), FITC anti-mouse CD19

47
(6D5), APC anti-mouse Gr-1 (RB6-8C5), PerCPCy5.5 anti-mouse CD3 (17A2), APC anti-mouse
CD4 (GK1.5), PE anti-mouse CD25 (3C7), PE anti-mouse CD210 (1B1.3a) were purchased from
BioLegend. PE anti-mouse SiglecF (E50-2440) was purchased from BD Biosciences. PerCP-
eFluor710 anti-mouse ST2 (RMST2-2), eFluor450 anti-mouse CD11b (M1/70) were purchased
from Thermofisher. Intranuclear staining was performed using the Foxp3 Transcription Factor
Staining Kit (Thermofisher) per the manufacturer’s instructions. APC anti-mouse Ki67 (SolA15,
Thermofisher), APC anti-mouse Blimp-1 (5E7, BioLegend), PE anti-mouse cMaf (sym0F1,
Thermofisher), PE anti-mouse Gata3 (TWAJ, Thermofisher), PE anti-mouse Phospho-Stat6
(Tyr641, Thermofisher) were used. Intracellular staining was performed using the BD Biosciences
Cytofix/Cytoperm kit. When indicated, cells were stimulated in vitro for 4 hours with 50µg/mL
PMA, 500µg/mL ionomycin (both Sigma) and 1µg/mL Golgi plug (BD Biosciences) before
cytokine assessment. APC anti-mouse IL-13 (85BRD, Thermofisher), PE anti-mouse IL-5
(TRFK5, BioLegend) were used. Live/dead fixable violet cell stain kit was used to exclude dead
cells (Thermofisher) and CountBright absolute counting beads (Thermofisher) to calculate
absolute cell numbers when indicated. Stained cells were analyzed on FACSCanto II and/or
FACSARIA III systems and the data were analyzed with FlowJo version 10 software.

Murine ILC2 and in vitro culture
Murine ILC2s were FACS-sorted to a purity of >95% on a FACSARIA III system. Isolated ILC2s
were cultured at 37°C (5x10
4
/mL) for 24-48 hours as indicated with rmIL-2 (10ng/mL), rmIL-7
(10ng/mL), +/- rmIL-13 (10ng/ml), +/- rmIL-10 (10ng/ml), +/- rmIL-33 (10ng/ml), +/- rmIL-4
(10ng/ml) purchased from BioLegend in complete RPMI (cRPMI). For cRPMI, RPMI (Gibco)
was supplemented with 10% heat-inactivated FBS (Omega Scientific), 100 units/mL penicillin
and 100mg/mL streptomycin (GenClone). For Th2 differentiation, naïve T cells from the spleen
were identified and sorted as CD4
+
CD25
-
. Cells were cultured in cRPMI with anti-CD3, anti-
CD28 and rmIL-4 (20ng/ml) (BioLegend) for 4 days. In experiments involving cytokine
neutralizers, cells were cultured for 48 hours with anti-mouse IL-10R (10µg/ml, 1B1.3A), anti-
mouse IL-4 (10 µg/ml, 11B11) or appropriate control purchased from BioXCell. In experiments
involving drug treatments, etomoxir (20 μM; Sigma, St Louis, Mo), DEUP (5 μM, Sigma), orlistat
(100 μM, Sigma), and 2-DG (5 μM, Sigma) were added to the cells in cRPMI and cultured for 48
hours, as indicated. For glucose uptake measurements, cells were incubated in media containing

48
1000 μg/mL 2-NBDG (Thermo Fisher Scientific) for 20 minutes at 37°C after surface antibody
staining. For lipid droplet quantification, cells were incubated in media containing 1000 ng/mL
Bodipy (Thermo Fisher Scientific) at 37°C for 30 minutes. In experiments culturing cells outside
of the conventional cRPMI described, aILC2s were sorted and cultured for 48 hours with IL-7
(10ng/ml), IL-2 (10ng/ml), IL-33 (10ng/ml), with or without rmIL-4 (10ng/ml) in RPMI 1640
without glucose or glutamine (BI Biological Industries), but supplemented with 10% heat-
inactivated FBS (Omega Scientific), 100 units/mL penicillin and 100mg/mL streptomycin
(GenClone), and 10mM galactose (Sigma). For measurement of lactate levels in the populations,
aILC2s and ILC210s were isolated and cultured for 24 hours as described above. After 24 hours,
supernatants were analyzed for L-lactate levels using the Glycolysis Cell-Based Assay Kit
(Cayman Chemicals). For knock-down studies, 2.5 µM Stat6 in-vivo morpholinos 5′-
ACTCCCTGTGAGGAAAGTAGGATTA-3′, Prdm1 5′-ACATCTGAGATA
AGCCTCTCTCATG-3′, cMaf 5’-GTTCATTGCCAGTTCTGAAGCCATC-3’ or random control
oligo (Gene Tools, LLC) were added as free uptake oligos 24 h before rmIL-4 (10ng/ml) addition.

Cytokine measurements
The amounts of cytokines in culture supernatants were measured by ELISA. Murine IL-5, IL-13,
IL-6, GM-CSF and IL-10 ELISA kits were used according to the manufacturer’s instructions
(Thermofisher).

Adoptive Transfer
Experiments were performed as described previously(37) . Briefly, aILC2 from IL-33 treated
C57BL/6J mice were isolated as above and cultured for 48 hours. Concurrently, ILC210s were
generated and isolated from IL-10
GFP
mice as described above. 2.5 x 10
4
of the appropriate
population of cells were transferred in PBS to Rag
-/-
γc
-/-
mice by tail intravenous (i.v.) for a total
of 5.0 x 10
4
cells per mouse. A cohort was also given i.p. injections of anti-mouse IL-10R or
isotype (300 μg, BioXCell). 24 hours later, 0.5 μg rmIL-33 i.n. in 50 μL was given once a day for
3 days. AHR was then measured on day 4. Alternatively, 100 μg A. Alternaria i.n. in 50 μl was
given once a day for 4 days. AHR was then measured on day 5.

Measure of Airway Hyperreactivity

49
Experiments were performed as described previously(35,38). Lung function was evaluated by
direct measurement of lung resistance using the FinePointe RC system (Buxco Research Systems,
Wilmington, NC) under general anesthesia. AHR was measured by exposure to an aerosol
containing increasing doses of Methacholine (Sigma), following a baseline measurement after the
delivery of a saline aerosol.

BAL collection
BAL fluid was collected as previously described(37). The trachea was cannulated, the lungs
lavaged three times with 1.0 ml PBS and the collected fluid pooled. Eosinophils were gated as
CD45
+
CD11c
-
SiglecF
+
single cells.

RNA sequencing and data analysis
Transcriptomic analysis was performed as described previously(34,37). Briefly, RNA was
extracted from cultured cells using the MicroRNeasy kit, and cDNA was generated for library
preparation using 10ng of RNA. Samples were then amplified and sequenced on a NextSeq 500
system (Illumina) and data processed on Partek® Genomics Suite® software, version 7.0
Copyright ©; Partek Inc.

Statistical analysis
Experiments were repeated at least three times (n=4-8 each) and data are shown as the
representative of >2 independent experiments. A student t-test for unpaired data was used for
comparisons between each group using Prism Software (GraphPad Software Inc.). Mean, Standard
Deviation and degree of significance was indicated as: *p<0.05, **p<0.01, ***p<0.001.

50
Figures
Figure 1
A subset of IL-10
+
ILC2s is naturally present in an allergic inflammation murine model.
(A) A cohort of Vert-X/IL-10
GFP
mice was given an intraperitoneal injection of Ova-alum on days
0 and 7. The mice were then intranasally treated with Ova (100µg) or PBS on days 14, 16, and 18.
On day 19 lung ILC2s were isolated and analyzed by flow cytometry according to (B, Figure
E1A).

51
(C) Total number of IL-10
+
ILC2s in the lungs.
(D) Spleen CD4
+
cells were isolated from a cohort of WT control mice and stimulated with anti-
CD3 and anti-CD28 for 72h to ensure Th2 differentiation. An additional cohort of WT control
mice was intranasally treated with rmIL-33 (0.5µg) on days 1-3. On day 4 lung ILC2s were sorted
by flow cytometry. Th2 cells and ILC2 cells were co-cultured with anti-IL-4 or isotype. The levels
of IL-4 (E) and IL-10 (F) were measured by ELISA.
(G) Th2 and ILC2 cells were co-cultured in varying ratios, as described in (D). Percent of IL-5
+
IL-13
+
ILC2s were isolated and analyzed by flow cytometry.
(H) Percent of IL-5
+
IL-13
+
ILC2s were measured by flow cytometry when co-cultured with Th2
cells and +/- anti-IL-4.
Error bars are the mean ± SEM. Data are representative of 3 individual experiments with n=5.
Student’s t-test, *p < 0.05, **p < 0.01, ***p < 0.001. Mouse image provided with permission from
Servier Medical Art.

52
Figure 2
IL-4 induces ILC2
10
generation and downregulates effector function.
(A) A cohort of Vert-X/IL-10
GFP
mice was intranasally treated with rmIL-33 (0.5µg) or PBS on
days 1-3. On day 4 lung ILC2s were sorted by flow cytometry according to (Figure E1A) and
cultured in complete RPMI for 48h with rmIL-7, rmIL-2 (10ng/ml) and +/- rmIL-4 (10ng/ml).

53
(B) Representative flow cytometry plots of IL-10
+
ILC2s (ILC210s) from ILC2s cultured with and
without IL-4 as described in (A) and corresponding quantification presented as the percentage of
ILC210s.
(C) ILC2s were stimulated and sorted as described in 2A. Cells were then cultured in previously
described conditions with the cytokines indicated for 48h. Supernatant was collected and secreted
cytokines were measured by ELISA.
(D) Representative flow cytometry plots of IL-5
+
IL-13
+
ILC2s from ILC2s cultured with rmIL-4
(10ng/ml) or rmIL-33 (10ng/ml) and corresponding quantitation presented as the percentage of IL-
5
+
IL-13
+
ILC2s.
(E) Representative flow cytometry plots of IL-5
+
IL-13
+
ILC2s from ILC2s cultured with rmIL-4
or rmIL-13 at various concentrations and corresponding quantification presented as the percentage
of IL-5
+
IL-13
+
ILC2s.
(F) Representative flow cytometry plots of ILC210s

from ILC2s cultured with rmIL-4 (10ng/ml)
or rmIL-13 (10ng/ml) and corresponding quantification presented as the percentage of ILC210s.
Error bars are the mean ± SEM. Data are representative of 3 individual experiments with n=5.
Student’s t-test, *p < 0.05, **p < 0.01, ***p < 0.001. Mouse image provided with permission from
Servier Medical Art.

54

55
Figure 3
cMaf and Blimp-1 are necessary for IL-10 generation in ILC2
10
s.
(A) Sorted ILC2s were cultured +/- IL-4 and anti-IL-4 or isotype for 48h. Cells were collected and
analyzed for percentage ILC210s by flow cytometry. Corresponding quantification presented as the
percentage of ILC210s.
(B) Representative flow cytometry of pStat6 transcription factor expression in ILC2s cultured with
Stat6 morpholino or vehicle for 24h.
(C) IL-10 in supernatant of ILC2s cultured with and without morpholino (2µM) and IL-4
(10ng/ml) for 48h is measured by ELISA.
(D) Representative flow cytometry plots of cMaf transcription factor expression in ILC2s
stimulated +/- IL-4 (ng/ml) for 48h. Corresponding quantification is presented as the cMaf mean
fluorescence intensity (MFI).
(E) Representative flow cytometry plots of Blimp-1 transcription factor expression in ILC2s
stimulated +/- IL-4 (ng/ml) for 48h. Corresponding quantification is presented as the Blimp1 MFI.
(F) Representative flow cytometry of cMaf transcription factor expression in ILC2s with and
without knock-down by cMaf morpholino.
(G) Representative flow cytometry of Blimp1 transcription factor in ILC2s with and without
knock-down by Prdm1 morpholino (gene name).
(H) Sorted ILC2s were cultured for 48h +/- IL-4 (10ng/ml) and appropriate morpholino (2.5µM).
Cells were collected and the percentage of ILC210s was measured by flow cytometry.
Error bars are the mean ± SEM. Data are representative of 3 individual experiments with n=5.
Student’s t-test, *p < 0.05, **p < 0.01, ***p < 0.001.

56

Figure 4
ILC2
10
s suppress activated ILC2 effector function in vitro.

57
(A) Representative flow cytometry plot of IL-10R expression on lung aILC2s and corresponding
quantification.
(B) Sorted aILC2s were cultured 48h with rmIL-4 or rmIL-10 (10ng/ml). Cells were collected and
percentage of IL-5
+
IL-13
+
ILC2s was quantified by flow cytometry.
(C) Sorted ILC210s were cultured an additional 48h with IL-4 (10ng/ml) and anti-IL-10R or isotype
(10µg/ml). Cells were collected and percentage of IL-5
+
IL-13
+
ILC210s was quantified by flow
cytometry.
(D) aILC2s and ILC210s were generated, sorted into two populations, and cultured in media with
appropriate cytokines for 72h. Supernatant was collected and secreted IL-10 from the separate
populations was quantified by ELISA (E). The supernatant was diluted 2x with fresh media and
new survival cytokines (IL-2, IL-7). Concurrently a cohort of WT mice was intranasally treated
with rmIL-33 (0.5µg) on days 1-3. On day 4 lung aILC2s were sorted by flow cytometry. The
aILC2s were then cultured in diluted ILC210 or ILC2 supernatant or fresh media containing rmIL-
10 (5ng/ml) for 48h.
(E) Concentration of IL-10 in ILC210s and aILC2s supernatant measured by ELISA after 72h.
(F) Representative flow cytometry plots of IL-5
+
IL-13
+
ILC2s after 48h culture in control (aILC2)
supernatant, ILC210 supernatant, or rmIL-10 (5ng/ml) and corresponding quantification.
(G) Representative flow cytometry histogram plot of Gata3 expression in ILC2s from (F) with
corresponding quantification.
(H) Sorted aILC2s were cultured as described in (D) in aILC2 or ILC210 supernatant with the
addition of anti-IL-10R or isotype (10µg/ml). Representative flow cytometry plots and
corresponding quantification presented as IL-5
+
IL-13
+
ILC2s.
Error bars are the mean ± SEM. Data are representative of 3 individual experiments with n=5.
Student’s t-test, *p < 0.05, **p < 0.01, ***p < 0.001.

58

Figure 5
ILC2
10
s downregulate airway hyperreactivity and lung inflammation.

59
(A) ILC210s and aILC2s were generated and sorted into separate populations as previously
described. 5x10
4
1:1 mix of aILC2:ILC210s or a control 5x10
4
aILC2s were intravenously injected
to Rag
-/-
γc
-/-
host mice. 24 hours after transfer, mice were challenged intranasally with 0.5µg rmIL-
33 for three consecutive days. Mice were also given injections of anti-IL-10R or isotype for three
consecutive days. On day 4, AHR (B) was assessed. Additionally total number of eosinophils in
the BAL (C), number of lung ILC2s (D) is presented as mean numbers +/- SEM.
(E) ILC210s and aILC2s were generated and sorted into separate populations as previously
described. 5x10
4
1:1 mix of aILC2:ILC210s or a control 5x10
4
aILC2s were intravenously injected
to Rag
-/-
γc
-/-
host mice. 24 hours after transfer, mice were challenged intranasally with 100µg
Alternaria alternata for four consecutive days. Mice were also given injections of anti-IL-10R or
isotype for four consecutive days. On day 5, AHR (F) was assessed. Additionally total number of
eosinophils in the BAL (G), number of lung ILC2s (H) is presented as mean numbers +/- SEM.
Intracellular IL-5, IL-13 expression levels (I) of the lung ILC2s from (H) were measured by flow
cytometry and presented as %IL-5
+
IL-13
+
ILC2s.
Error bars are the mean ± SEM. Data are representative of 3 individual experiments with n=5.
Student’s t-test, *p < 0.05, **p < 0.01, ***p < 0.001. Mouse image provided with permission from
Servier Medical Art.

60

Figure 6
ILC2
10
generation is mediated by a metabolic shift from the fatty acid oxidation pathway to
the glycolytic pathway.
(A) ILC210s and activated ILC2s generated and sorted for RNA sequencing, n=3.

61
(B) Volcano plot comparison of whole transcriptome gene expression of sorted ILC210s and aILC2
control, n=3. Differentially expressed genes (defined as statistically significant adjusted p-
value<0.05) with changes of at least 2.0 fold-change (FC) are shown in yellow (upregulated genes)
or blue (downregulated genes).
(C) Heat plot of selected differentially expressed genes involved in the glycolytic pathway
(p<0.05; n=3).
(D) ILC210s and aILC2s were generated as previously described. Glucose uptake was measured
by 2-NBDG staining and quantified as 2-NBDG MFI.
(E) ILC210s and aILC2s were generated as previously described. Lipid droplet quantification was
measured by Bodipy staining and quantified as Bodipy MFI.
(F) ILC210s and aILC2s were generated as previously described. Enzymatic quantification of
lactate accumulation in the supernatants of cultured cells.
(G) Representative flow cytometry plots of IL-10
+
ILC2s from ILC2s cultured with IL-4 (10ng/ml)
and various metabolic inhibitors. Corresponding quantification is presented as the percentage of
ILC210s.
(H) Representative flow cytometry plots of IL-10
+
ILC2s from ILC2s cultured with IL-4 (10ng/ml)
in RPMI supplemented with control glucose (10mM) or OXPHOS-inducing galactose (10mM).
Corresponding quantification is presented as the percentage of ILC210s.
Error bars are the mean ± SEM. Data are representative of 3 individual experiments with n=5.
Student’s t-test, *p < 0.05, **p < 0.01, ***p < 0.001.

62
Supplemental Figures

Supplemental Figure E1
ILC2 gating strategy, related to Figure 1.
(A) Gating strategy used for Live, CD45
+
Lineage
-
CD127
+
ST2
+
ILC2s from murine lungs.
Biotinylated-lineage markers were stained by Streptavidin-APC in Figure 1A. All additional
pulmonary aILC2s were identified by Streptavidin-FITC.
(B) Total number of lung Th2, T regulatory and ILC2s from OVA-challenged mice in (1A). Th2
cells were identified as CD45
+
CD3
+
CD4
+
CD19
-
CD44
+
FoxP3
-
Gata3
+
. T regulatory cells were
identified as CD4
+
CD3
+
CD45
+
CD25
+
CD44
+
FoxP3
+
.
(C) Cells from co-cultures described in (D) were collected and populations were confirmed by
flow cytometry. Th0 cells were identified as CD4
+
CD3
+
Gata3
-/low
and Th2 cells were identified
as CD4
+
CD3
+
Gata3
+
. All cells were FoxP3
-
.

63

Supplemental Figure E2
ILC2 cytokine and transcription factor expression, related to Figure 3.
(A) Cells from (3H) were stained intracellularly for IL-5 and IL-13 expression levels following
culture with morpholino.
(B) Lung ILC2s from naïve mice were sorted and stained for cMaf and Blimp-1 expression levels.
Representative flow cytometry histograms are shown.

64

Supplemental Figure E3
ILC2 viability and proliferation, related to Figure 4.
(A) ILC210s and aILC2s were generated and sorted as previously described. Cells were cultured
separately in cRPMI for 48h. Supernatant was collected and IL-5 and IL-13 (B) levels were
measured by ELISA.
(C) Representative flow cytometry histogram plots of Ki-67
+
ILC2s from (4F). Corresponding
quantification is presented as the percentage of live and Ki-67
+
ILC2s.

Supplemental Figure E4
ILC2 proliferation, related to Figure 5.
Representative flow cytometry plots of Ki67 expression of lung ILC2s in (D). Corresponding
quantification is presented as the Ki67 mean fluorescence intensity (MFI).

65

Supplemental Figure E5
ILC2 viability, related to Figure 6.
Representative flow cytometry histogram plots of live ILC2s from (6G). Corresponding
quantification is presented as the percentage of live ILC2s.

66
Chapter 3:
PD-1 blockade on tumor microenvironment-resident ILC2s promotes
TNF-α production and restricts progression of metastatic melanoma

DOI: 10.3389/fimmu.2021.733136

Abstract
While pulmonary ILC2s represent one of the major tissue-resident innate lymphoid cell
populations at steady state and are key drivers of cytokine secretion in their occupational niche,
their role in pulmonary cancer progression remains unclear. As the programmed cell death protein-
1 (PD-1) plays a major role in cancer immunotherapy and immunoregulatory properties, here we
investigate the specific effect of PD-1 inhibition on ILC2s during pulmonary B16 melanoma
cancer metastasis. We demonstrate that PD-1 inhibition on ILC2s suppresses B16 tumor growth.
Further, PD-1 inhibition upregulates pulmonary ILC2-derived TNF- α

production, a cytotoxic
cytokine that directly induces cell death in B16 cells, independent of adaptive immunity. Together,
these results highlight the importance of ILC2s and their anti-tumor role in pulmonary B16 cancer
progression during PD-1 inhibitory immunotherapy.

Introduction
Melanoma is a highly aggressive form of cancer that spreads from primary skin sites to
various organs throughout the human body[1; 2]. While effective treatment is attainable if
diagnosed at the early stages, 20% of melanomas diagnosed in advanced stages resist
treatment[3]
,
[4]. Pulmonary metastasis is the most common progression in the late stages of
cancer diagnostics[5]. According to the World Health Organization, over 106,000 melanomas
were diagnosed, and 7,000 new fatalities occurred in the United States in 2020.
While the rates of melanoma diagnosis continue to be alarming, molecularly targeted
approaches, including inhibition of immune checkpoints such as programmed death-1 receptor
(PD-1) have significantly improved patient outcomes over the last decade. Nivolumab, a Food and
Drug Administration-approved antibody directed at PD-1 inhibition enhanced median overall
survival in patients with metastatic melanoma[6]. PD-1 binds to ligands PD-L1 or PD-L2 to act as
an inhibitory immune checkpoint, negatively regulating immune cell activation and effector
function[5; 7; 8]. Under healthy conditions, PD-1 and its ligands ensure immune cell function

67
homeostasis by restricting unchecked pro-inflammation in the body. In a tumor microenvironment,
however, cancer cells exploit the negative regulators to elude healthy antitumor immunity[9].
Group 2 Innate Lymphoid cells (ILC2s) are tissue resident non-B and non-T innate immune
cells that reside in mucosal tissues, including the lung, gut and skin[10; 11]. Though primarily
activated by epithelial alarmin cytokines, including interleukin (IL-) 33, IL-25 and TSLP, to
produce their effector function cytokines IL-5 and IL-13, our group and others have well
documented the immense plasticity in both ILC2 stimuli and responding cytokine production and
secretion[12-16]. Due to their unique occupational niche in the mucosal tissue, ILC2s are primed
to be first responders in the initiation of infiltrating cancer tumor cells. Recently, our group and
others reported that ILC2s facultatively express PD-1 and PD-L1 when activated with alarmin
cytokine IL-33 [8; 17; 18]. And indeed consistent with T, NK and B cells, engagement of PD-1 on
ILC2s dramatically downregulates proliferation, viability and effector function[19-21].
Conversely, inhibition of PD-1 engagement results in increases in total ILC2 number, increased
production of cytokines, and overall cell function[8; 17; 18]. While their role in Th-2 inflammatory
diseases has been well established, ILC2s’ contribution in cancer has long been controversial and
requires further investigation.
Tumor Necrosis Factor alpha (TNF- α) is a cytokine that plays a vital role in melanoma
tumor control in both mouse and lung cancer. Traditionally attributed to macrophages, several
groups have shown that TNF- α therapy can be used clinically in combination with established
therapies to enhance cancer treatment[22-24]. For example, TNF- α therapy can be used to target
tumor vasculature, ensuring the melanoma cells are more susceptible to anti-tumor anti-PD-1
therapy[22; 23]. Moreover, T-cell derived TNF- α has been demonstrated to kill tumor cells and
loss of TNF- α increases melanoma tumor cell invasion in the lung[25]. It was later detailed that
TNF- α kills tumor cells through enhancement of tumor oxidative stress in melanoma cells[26].
In this study, we show that B16 melanoma significantly enhances PD-1 expression on
pulmonary ILC2s, effectively promoting tumor progression and limiting anti-tumor immune
response. Blocking PD-1 on ILC2s however inhibits tumor growth progression and surprisingly,
induces the production ILC2-derived TNF- α. Through targeted adoptive transfer and in vitro
studies, we identify that lack of PD-1 on ILC2s directly affects B16 tumor cell apoptosis, mediated
through the cytotoxic properties of ILC2-derived TNF- α. Together, our findings give further
insight into the anti-tumoral immune biology of pulmonary ILC2s and their direct role in

68
conventional immune checkpoint therapies. Specifically, our studies can make significant progress
and advance our understanding of the underlying therapeutic value of ILC2s relating to clinical
cancer treatments.

Results
B16 melanoma induces PD-1 expression on pulmonary ILC2s.
We first assessed the effect of B16 melanoma tumors on ILC2 PD-1 expression. To do this,
we intravenously (i.v.) injected B16 melanoma cells or PBS control into a cohort of PD-1
-/-
or WT
mice on day 0 (Figure 1A). On day 14, mice were euthanized and total pulmonary tumor colonies
on the surface of the lungs was assessed and quantified (Figure 1B). As shown in Figure 1B, WT
mice inoculated with B16 cells displayed significantly more tumor colonies per field when
compared to mice deficient in PD-1, suggesting PD-1 dramatically promotes B16 melanoma
metastasis. The lungs of the mice were then digested and ILC2 PD-1 expression of WT mice with
and without B16 melanoma was assessed (Figure 1C). The percentage of PD-1
+
ILC2s
significantly increased after B16 inoculation, from a mean of 15% PD-1
+
ILC2s to 60% PD-1
+

ILC2s, demonstrating B16 melanoma induces PD-1 expression on pulmonary ILC2s and may
subsequently alter anti-tumor effector function (Figure 1D). PD-1
-/-
mice also displayed a
significant increase in the percentage and number of pulmonary ILC2s when compared to WT
mice (Figure 1E). Together, our results suggest B16 melanoma upregulates ILC2 PD-1
expression, driving the progression of tumor growth.

PD-1 expression on pulmonary ILC2s promotes B16 tumor growth and drives melanoma-
induced fatality.
Several groups have reported an anti-tumor role for ILC2s in the suppression of cancers
through a CD8+ T cell dependent mechanism[7; 30]. To explore whether ILC2s in our context
relied upon adaptive immunity for potential anti-tumor properties and cytokine production, we
utilized Rag2
-/-
PD-1
-/-
and control

Rag2
-/-
mice. These mice lack all B and T cells, allowing us to
explore more specifically the behavior of ILC2s in this microenvironment. We first gave a cohort
of Rag2
-/-
and Rag2
-/-
PD-1
-/-
mice i.v. injection of B16 melanoma cells on day 0 (Figure 2A).
Since NK cells are known to have a significant anti-tumor effect on various cancers, including
pulmonary melanoma, we additionally injected the mice with either anti-asialo GM1 or its isotype,
an antibody used to deplete NK cells in the mouse[31]. Repeated injections of anti-asialo GM1

69
significantly reduced the number of spleen NK cells in both Rag2
-/-
and Rag2
-/-
PD-1
-/-
mice
(Supplementary 1A). In a first cohort, we found that Rag2
-/-
PD-1
-/-
mice survived significantly
longer as compared to the Rag2
-/-
control mice (Figure 2B). Additionally, tumor metastasis
drastically decreased in Rag2
-/-
PD-1
-/-
mice as compared to the Rag2
-/-
control (Figure 2C). As a
proof of concept, we demonstrated that mice injected with the isotype as opposed to the NK
depleting antibody do indeed have less tumor colonies on the surface of their lungs, consistent
with published results (Figure 2C)[31]. However PD-1 expression on ILC2s greatly enhances
tumor progression, as evidenced by the increase in tumors in Rag2
-/-
control mice depleted of NK
cells. Importantly, we also found that the number of total ILC2 cells were significantly increased
in Rag2
-/-
PD-1
-/-
mice, further suggesting ILC2s play an important anti-tumor role independent of
surrounding B and T cells (Figure 2D). In order to directly test whether ILC2s had a direct effect
on the inhibition of tumor progression, we injected B16 cells into a cohort of Rag2
-/-
γc
-/-
mice, a
strain that lacks all B, T, and innate lymphoid cells (Figure 2E). Concurrently we adoptively
transferred IL-33-activated pulmonary ILC2s FACS-sorted from either PD-1
-/-
or WT mice every
six days, using our previously established protocol[16]. On day 14, lungs were removed and tumor
burden on the surface was assessed (Figure 2F). Strikingly, mice that did not receive either ILC2
phenotype demonstrated a significant increase in tumor metastasis as compared to the mice
adoptively transferred either WT or PD-1
-/-
ILC2s, suggesting that ILC2s do indeed play an
essential role in an anti-tumor immune response independent of adaptive immunity. Most
importantly, however, the mice adoptively transferred with PD-1
-/-
pulmonary ILC2s had the least
number of B16 tumor colonies at day 14, indicating that PD-1 inhibits important aspects of anti-
tumor ILC2 response in vivo. Interestingly, total number of pulmonary ILC2s was not significantly
different between the adoptively transferred groups (Supplementary 1B), indicating the decrease
in tumor burden was independent of a difference in ILC2 number. Taken together, our results
suggest that PD-1

deficiency in ILC2s play a specific role in anti-tumor immunotherapy in the
inhibition of tumor progression and improved survival.

PD-1 deficiency on IL-33 stimulated ILC2s enhances TNF- α expression and phosphorylation
of canonical NF κB pathway.
Several groups have previously reported IL-33 to be present in the melanoma tumor
microenvironment, though it is unclear whether the IL-33 is secreted from the melanoma cells

70
themselves or the lung epithelial layer due to injury[32; 33]. To further explain the relationship of
PD-1 expression with the loss of important ILC2 anti-tumor properties in a more controlled setting,
we intranasally (i.n.) challenged a cohort of WT mice with either rmIL-33 or PBS for three
consecutive days (Figure 3A). After three challenges, a variety of intracellular cytokines important
for the arrest of tumor growth was assessed. Surprisingly, we found a significant increase ILC2
TNF- α production, a potent anti-tumor cytokine known to severely inhibit melanoma growth
amongst others (Figure 3B). We repeated the same experiment on a cohort of WT and PD-1
-/-

mice and remarkably found that PD-1 deficiency combined with IL-33 stimulation further
increased the percentage of TNF- α
+
ILC2s (Figure 3C). In addition to TNF- α, we previously
showed that Th-2 cytokines, including IL-5 and IL-13 were increased in ILC2s deficient in PD-
1[8]. Interestingly, GM-CSF was also upregulated, both at an RNA sequencing level[8] and at the
protein level (Supplementary 2A).
TNF- α

production is controlled primarily by the canonical NF κB pathway[25]. To
elucidate the mechanism by which ILC2s produce TNF- α, we reanalyzed our group’s previously
published RNA sequencing data of sorted, IL-33-activated ILC2s from PD-1
-/-
and WT mice with
the new focus of genes involved in the canonical NF κB pathway[8]. Samples were characterized
by group according to RNA quality and gene expression (Supplementary 2B) and differential
gene expression between the two groups was plotted on a volcano plot (Fold change > 1.45,
P<0.05, Figure 3D). The results remained consistent with our previous findings in that ILC2s
deficient in PD-1 expression had significantly upregulated Tnf expression (Fold change = 1.5,
P<0.001) (Figure 3D).
Further transcriptomic analysis by Ingenuity Pathway Analysis (IPA) suggests the NF κB
complex and related genes upstream of TNF- α production are increased (Figure 3E). And indeed,
heatmap analysis focusing on genes highlighted in Figure 3E are upregulated in activated PD-1
-/-

ILC2s (Figure 3F). As the results from this analysis are from IL-33-activated ILC2s, we
questioned whether NF κB upregulation was present in the ILC2s of our B16 melanoma model.
We therefore injected B16 cells into PD-1
-/-
and WT mice as described above (Figure 3G) and
measured phosphorylation levels of p52 and p65, non-canonical and canonical members of the
NF κB pathway respectively, in pulmonary ILC2s on day 14 (Figure 3H-I). p52 phosphorylation
levels were unchanged in our model (Figure 3H). In line with the findings from the IL-33-
activated RNA sequencing data, phosphorylation levels of p65 were significantly inhibited in WT

71
ILC2s isolated from the TME as compared to PD-1
-/-
ILC2s (Figure 3I). Altogether, our data
suggest that the B16 melanoma tumor microenvironment induces phosphorylation of the NF κB
pathway in PD-1 deficient pulmonary ILC2s, potentially stimulating ILC2 TNF- α

production.

Blocking PD-1 on tumor microenvironment ILC2s increases TNF- α

production and
enhances cytotoxic properties ex vivo.
PD-1 blocking antibodies are currently used therapeutically to upregulate anti-tumor
immune responses in human patients[9]. To investigate the in vitro effect of PD-1 blocking
antibody on TME ILC2s, we injected a cohort of WT mice with B16 melanoma on day 0 (Figure
4A). After 14 days, the lungs were digested and ILC2s were FACS-sorted as previously
demonstrated[10]. ILC2s were then cultured with recombinant mouse (rm)IL-2, IL-7, IL-33 for 48
hours, with the addition of the mouse PD-1 blocking antibody or the isotype. After 48 hours, cells
cultured with the PD-1 blocking antibody demonstrated a significant upregulation of TNF- α

production in the ILC2s by intracellular flow cytometry, from 22% to 45% (Figure 4B).
Excitingly, a relatively high percentage of ILC2s cultured with either the antibody or the isotype
produced TNF- α, suggesting the B16 melanoma context from which they originated primed the
cells for TNF- α production. Moreover, blocking PD-1 significantly increased the percentage of
TNF- α
+
ILC2s, further suggesting a novel mechanism by which PD-1 blocking antibody may lead
to tumor inhibition by ILC2s in vivo.
Since TNF- α is a well-established inhibitor of tumor growth in a variety of cancers, we
wondered if the ILC2-derived TNF- α had any direct effect on B16 melanoma cells. To test this,
we first cultured B16 melanoma cells in the bottom well of a 0.4µm Transwell plate with either a
mouse neutralizing TNF- α antibody or the isotype. We next FACS-sorted activated PD-1
-/-
or WT
ILC2s to culture on top of the insert. After 48 hours, we collected the B16 cells from the bottom
well and used Dapi and Annexin V to assess early apoptotic and dead cells (Figure 4C). Cells
cultured with PD-1
-/-
ILC2s displayed a significant decrease in viability compared to both those
cultured with WT ILC2s and those cultured alone. Importantly, cells cultured with anti-TNF- α
neutralizing antibody saw a significant restoration in viability, suggesting that TNF- α production
from ILC2 has a direct effect on melanoma apoptosis. Together these results suggest a novel anti-
tumor mechanism of action by which PD1 deficiency on ILC2s affects melanoma cells.

72

Blocking ILC2 PD-1 expression increases TNF- α production and inhibits tumor progression
in vivo.
To determine if ILC2 TNF- α production was taking place in our in vivo B16 melanoma
models, we i.v. injected B16 melanoma to a cohort of WT and PD-1
-/-
mice as described above
(Figure 5A). And indeed, pulmonary ILC2s from PD-1
-/-
statistically increased the percentage of
TNF- α production 3-fold on day 14 by intracellular flow cytometry (Figure 5B). Of note, this was
again demonstrated in our previous in vivo model utilizing mice deficient in adaptive immunity
(Figure 5C-D). As previously mentioned, PD-1 blocking antibodies are vital immune checkpoint
inhibitors used for therapeutic means in today’s diagnostic world. To determine if the phenotypic
difference exhibited in Rag2
-/-
and Rag2
-/-
PD-1
-/-
mice is clinically translatable, we injected a
cohort of Rag2
-/-
mice with B16 melanoma cells (Figure 5E). In addition to the NK-depleting anti-
asialo GM1 injections, we included repeated i.p. injections of mouse PD-1 blocking antibody or
its isotype every four days. After 14 days, the lungs were removed for tumor colony assessment
and quantification (Figure 5F). Mice that received the PD-1 blocking antibody demonstrated a
significant decrease in the number of tumor colonies on the lung surface. Of note, the number of
pulmonary ILC2 cells between the two groups was unchanged (Figure 5G). However, there
remained a significant increase in the percentage of TNF- α

producing ILC2s in mice that received
the PD-1 blocking antibody (Figure 5H). Together these results suggest that the ILC2s treated
with PD-1 blocking antibody have significant clinical relevance for anti-tumorigenic function.

Blocking PD-1 engagement on human blood ILC2 cells increases TNF- α secretion.
To better understand the clinical relevance of PD-1 inhibition in TNF- α production, we
sought to investigate its role in human blood ILC2s. Human ILC2s were identified and sorted as
CD45+ lineage− CD127+ CRTH2+ (Figure 6A, Figure 6B). The cells were then cultured in
cRPMI with recombinant human (rh)IL-2, IL-7, with and without IL-33, and with either human
PD-1 blocking antibody or isotype for 72 hours as previously described. PD-1 expression in human
ILC2s cultured with IL-33 was significantly induced after 72 hours (Figure 6C). Additionally,
TNF- α secretion was significantly increased in supernatant of human ILC2s cultured with the PD-
1 blocking antibody as compared to the isotype (Figure 6D). These data, together with the previous

73
findings, provide insight into an exciting new mechanism of action by which human ILC2s can be
utilized for anti-tumor therapies.

Discussion
Overall this study introduces a novel anti-tumor role of pulmonary ILC2s in the context of
metastatic melanoma through the production of TNF- α. We demonstrate that B16 melanoma
significantly induces PD-1 expression on pulmonary ILC2s, thereby dampening pro-inflammatory
properties vital to the inhibition of tumor progression. Blocking or deficiency of PD-1 on ILC2s
slows tumor progression independent of adaptive immunity. Additionally, inhibition of PD-1
significantly enhances human and mouse ILC2 TNF- α production, a cytokine with the potential
to directly induce apoptosis in B16 melanoma cancer cells.
Many groups have reported the importance of ILC2 involvement in both anti-tumor and
pro-tumoral immunity in a variety of mouse and human cancers. For example, ILC2s’ rapid
secretion of inflammatory cytokines IL-13 and IL-4 in human cells have been reported to promote
tumor growth by recruitment of monocytic myeloid-derived suppressor cells, immune cells
important for the inhibition of anti-tumor immunity[34]. And indeed, in prostate cancer, Trabanelli
et al found that PGD2 secreted from tumor cells led to increased numbers of ILC2s in mouse and
human prostate cancer, resulting in an immunosuppressive pathway primed for tumor
progression[35]. In contrast, Moral et al published that increased number of ILC2s in human
pancreatic ductal adenocarcinoma led to improved survival, relying on the cells’ ability to recruit
and activate CD8+ T cells[30]. In similar duality, Schuijs et al reports that IL-33 treatment
enhances ILC2 IL-5 immune response in the lung, thus activating eosinophils that subsequently
suppress vital, anti-tumor Th-1 immunity, and lending to increased lung cancer metastasis and
mortality[36]. Lucarini et al, on the other hand, agrees that IL-33 activates eosinophils, but
demonstrates their cytotoxic capabilities against lung metastatic tumors and successive growth
delay[37]. It is therefore evident that additional studies investigating the individual contribution of
ILC2s to tumor progression are required.
The plasticity of ILC2 cytokine secretion has been well established. ILC2s are not
traditionally credited with TNF- α secretion. We show that both mouse and human ILC2s
stimulated in vitro with IL-33 do indeed produce TNF- α, and this is increased with PD-1
deficiency or inhibition. ILC2s deficient in PD-1 also exhibit an increase in Th-2 cytokines, such

74
as IL-5 and IL-13 as previously published by our group[8]. Furthermore, we also see an increase
in GM-CSF at both the RNA sequencing and protein level[8]. These cytokines may contribute
synergistically with TNF- α in the observations reported here in vivo. However, our results suggest
ex vivo neutralization of TNF- α alone is sufficient to abrogate the anti-tumor effect of PD-1
deficient ILC2s. Furthermore, in the B16 melanoma model, the percentage of TNF- α-producing
ILC2s significantly increases, suggesting the melanoma tumor microenvironment is primed to
allow for ILC2 TNF- α secretion. TNF- α is secreted after activation of the NFκB pathway, but
exactly how PD-1 deficiency enhances production upstream remains unclear. Metryka et al details
regulation of PI3K/AKT and IKKs are important for NFκB activation and subsequent TNF- α
secretion, but whether these are involved in our model remains to be investigated[38]. Moreover,
whether TME IL-33 alone or in combination with additional unidentified stimuli induces the high
levels of ILC2 TNF- α in our melanoma model requires further exploration. Golebski et al. reported
increased ILC2 derived TNF- α production after stimulation with IL-1B, IL-23 and TGF-B[13].
Maggi et al also demonstrated that human ILC2s produce TNF- α after various stimulations[39].
Based on our combined results of PD-1 deficient/inhibitory mouse models and cytotoxicity assay,
we propose a novel, direct role of ILC2-derived TNF- α in the apoptosis of B16 melanoma cells
that is enhanced by anti-PD-1 therapy in the significant delay of melanoma tumor growth.
One limitation of our study is that the NK depleting antibody used, anti-asialo GM1, may
recognize a subset of basophils in vivo as reported before [40]. However, we believe it is unlikely
that basophils play a role in our melanoma model.
In summary, this report details a previously unrecognized immunotherapeutic role for
blocking PD-1 on ILC2s in B16 melanoma metastasis. The identification of potent, anti-tumor
cytokine, TNF- α, dramatically induced in pulmonary ILC2s is vitally important for a cell that is
uniquely positioned to act as the first responder during tumor infiltration. Based on our findings
together, we propose an essential role of both human and mouse ILC2s in PD-1 immunotherapies
dependent on their dramatic ability to produce TNF- α. We believe the knowledge gained by this
report increases the therapeutic potential of anti-PD-1 therapies.

75
References
[1] E. Atallah and L. Flaherty, “Treatment of metastatic malignant melanoma,” p. 9.

[2] K. V. Liang, S. O. Sanderson, G. S. Nowakowski, and A. S. Arora, “Metastatic Malignant
Melanoma of the Gastrointestinal Tract,” Mayo Clinic Proceedings, vol. 81, no. 4, pp. 511–516,
Apr. 2006, doi: 10.4065/81.4.511.

[3] A. H. Shain and B. C. Bastian, “From melanocytes to melanomas,” Nat Rev Cancer, vol. 16,
no. 6, pp. 345–358, Jun. 2016, doi: 10.1038/nrc.2016.37.

[4] K. Khorsandi, R. Hosseinzadeh, and E. Chamani, “Molecular interaction and cellular studies
on combination photodynamic therapy with rutoside for melanoma A375 cancer cells: an in vitro
study,” Cancer Cell International, vol. 20, no. 1, p. 525, Oct. 2020, doi: 10.1186/s12935-020-
01616-x.

[5] J. Hwang et al., “Astragalus membranaceus polysaccharides potentiate the growth-inhibitory
activity of immune checkpoint inhibitors against pulmonary metastatic melanoma in mice,”
International Journal of Biological Macromolecules, vol. 182, pp. 1292–1300, Jul. 2021, doi:
10.1016/j.ijbiomac.2021.05.073.

[6] L. Liu, J. Xu, J. Yang, C. Feng, and Y. Miao, “Metastatic melanoma imaging using a novel Tc-
99m-labeled lactam-cyclized alpha-MSH peptide,” Bioorganic & Medicinal Chemistry Letters,
vol. 27, no. 22, pp. 4952–4955, Nov. 2017, doi: 10.1016/j.bmcl.2017.10.022.

[7] F. R. Mariotti, L. Quatrini, E. Munari, P. Vacca, and L. Moretta, “Innate Lymphoid Cells:
Expression of PD-1 and Other Checkpoints in Normal and Pathological Conditions,” Front
Immunol, vol. 10, p. 910, 2019, doi: 10.3389/fimmu.2019.00910.

[8] D. G. Helou et al., “PD-1 pathway regulates ILC2 metabolism and PD-1 agonist treatment
ameliorates airway hyperreactivity,” Nat Commun, vol. 11, no. 1, p. 3998, Dec. 2020, doi:
10.1038/s41467-020-17813-1.

[9] S. Kleffel et al., “Melanoma Cell-Intrinsic PD-1 Receptor Functions Promote Tumor Growth,”
Cell, vol. 162, no. 6, pp. 1242–1256, Sep. 2015, doi: 10.1016/j.cell.2015.08.052.

[10] B. P. Hurrell et al., “Distinct Roles of LFA-1 and ICAM-1 on ILC2s Control Lung Infiltration,
Effector Functions, and Development of Airway Hyperreactivity,” Front. Immunol., vol. 11, 2020,
doi: 10.3389/fimmu.2020.542818.

[11] H. Maazi et al., “ICOS:ICOS-ligand interaction is required for type 2 innate lymphoid cell
function, homeostasis, and induction of airway hyperreactivity,” Immunity, vol. 42, no. 3, pp. 538–
551, Mar. 2015, doi: 10.1016/j.immuni.2015.02.007.

76
[12] K. Zhang et al., “Cutting Edge: Notch Signaling Promotes the Plasticity of Group-2 Innate
Lymphoid Cells,” J Immunol, vol. 198, no. 5, pp. 1798–1803, Mar. 2017, doi:
10.4049/jimmunol.1601421.

[13] K. Golebski et al., “IL-1β, IL-23, and TGF-β drive plasticity of human ILC2s towards IL-17-
producing ILCs in nasal inflammation,” Nature Communications, vol. 10, no. 1, Art. no. 1, May
2019, doi: 10.1038/s41467-019-09883-7.

[14] Y. Ohne et al., “IL-1 is a critical regulator of group 2 innate lymphoid cell function and
plasticity,” Nat Immunol, vol. 17, no. 6, pp. 646–655, Jun. 2016, doi: 10.1038/ni.3447.

[15] S. M. Bal et al., “IL-1β, IL-4 and IL-12 control the fate of group 2 innate lymphoid cells in
human airway inflammation in the lungs,” Nat Immunol, vol. 17, no. 6, pp. 636–645, Jun. 2016,
doi: 10.1038/ni.3444.

[16] E. Howard et al., “IL-10 production by ILC2s requires Blimp-1 and cMaf, modulates cellular
metabolism, and ameliorates airway hyperreactivity,” J Allergy Clin Immunol, vol. 147, no. 4, pp.
1281-1295.e5, Apr. 2021, doi: 10.1016/j.jaci.2020.08.024.

[17] C. Schwartz et al., “ILC2s regulate adaptive Th2 cell functions via PD-L1 checkpoint
control,” J Exp Med, vol. 214, no. 9, pp. 2507–2521, Sep. 2017, doi: 10.1084/jem.20170051.

[18] S. Taylor et al., “PD-1 regulates KLRG1+ group 2 innate lymphoid cells,” Journal of
Experimental Medicine, vol. 214, no. 6, pp. 1663–1678, May 2017, doi: 10.1084/jem.20161653.

[19] M.-L. Thibult et al., “PD-1 is a novel regulator of human B-cell activation,” Int Immunol, vol.
25, no. 2, pp. 129–137, Feb. 2013, doi: 10.1093/intimm/dxs098.

[20] J. L. Riley, “PD-1 signaling in primary T cells,” Immunol Rev, vol. 229, no. 1, pp. 114–125,
May 2009, doi: 10.1111/j.1600-065X.2009.00767.x.

[21] L. Quatrini, F. R. Mariotti, E. Munari, N. Tumino, P. Vacca, and L. Moretta, “The Immune
Checkpoint PD-1 in Natural Killer Cells: Expression, Function and Targeting in Tumour
Immunotherapy,” Cancers (Basel), vol. 12, no. 11, Nov. 2020, doi: 10.3390/cancers12113285.

[22] A. R. Elia et al., “Targeting Tumor Vasculature with TNF Leads Effector T Cells to the Tumor
and Enhances Therapeutic Efficacy of Immune Checkpoint Blockers in Combination with
Adoptive Cell Therapy,” Clin Cancer Res, vol. 24, no. 9, pp. 2171–2181, May 2018, doi:
10.1158/1078-0432.CCR-17-2210.

[23] A. Calcinotto et al., “Targeting TNF-α to Neoangiogenic Vessels Enhances Lymphocyte
Infiltration in Tumors and Increases the Therapeutic Potential of Immunotherapy,” The Journal of
Immunology, vol. 188, no. 6, pp. 2687–2694, Mar. 2012, doi: 10.4049/jimmunol.1101877.

[24] P. Probst, M. Stringhini, D. Ritz, T. Fugmann, and D. Neri, “Antibody-based Delivery of TNF
to the Tumor Neovasculature Potentiates the Therapeutic Activity of a Peptide Anticancer

77
Vaccine,” Clin Cancer Res, vol. 25, no. 2, pp. 698–709, Jan. 2019, doi: 10.1158/1078-0432.CCR-
18-1728.

[25] C. J. Kearney et al., “Tumor immune evasion arises through loss of TNF sensitivity,” Science
Immunology, vol. 3, no. 23, May 2018, doi: 10.1126/sciimmunol.aar3451.

[26] T. Habtetsion et al., “Alteration of Tumor Metabolism by CD4+ T Cells Leads to TNF-α-
Dependent Intensification of Oxidative Stress and Tumor Cell Death,” Cell Metabolism, vol. 28,
no. 2, pp. 228-242.e6, Aug. 2018, doi: 10.1016/j.cmet.2018.05.012.

[27] M. E. Keir, G. J. Freeman, and A. H. Sharpe, “PD-1 Regulates Self-Reactive CD8+ T Cell
Responses to Antigen in Lymph Nodes and Tissues,” The Journal of Immunology, vol. 179, no. 8,
pp. 5064–5070, Oct. 2007, doi: 10.4049/jimmunol.179.8.5064.

[28] D. G. Helou et al., “LAIR-1 acts as an immune checkpoint on activated ILC2s and regulates
the induction of airway hyperreactivity,” J Allergy Clin Immunol, pp. S0091-6749(21)00936–2,
Jun. 2021, doi: 10.1016/j.jaci.2021.05.042.

[29] B. P. Hurrell et al., “TNFR2 Signaling Enhances ILC2 Survival, Function, and Induction of
Airway Hyperreactivity,” Cell Rep, vol. 29, no. 13, pp. 4509-4524.e5, Dec. 2019, doi:
10.1016/j.celrep.2019.11.102.

[30] J. A. Moral et al., “ILC2s amplify PD-1 blockade by activating tissue-specific cancer
immunity,” Nature, vol. 579, no. 7797, Art. no. 7797, Mar. 2020, doi: 10.1038/s41586-020-2015-
4.

[31] S. Ishikawa, T. Ishikawa, C. Tezuka, K. Asano, M. Sunagawa, and T. Hisamitsu, “Efficacy
of Juzentaihoto for Tumor Immunotherapy in B16 Melanoma Metastasis Model,” Evidence-Based
Complementary and Alternative Medicine, vol. 2017, p. e6054706, Feb. 2017, doi:
10.1155/2017/6054706.

[32] P. Andersson et al., “Molecular mechanisms of IL-33–mediated stromal interactions in cancer
metastasis,” JCI Insight, vol. 3, no. 20, doi: 10.1172/jci.insight.122375.

[33] F. Yang et al., “IL-33/ST2 Axis Regulates Vasculogenic Mimicry via ERK1/2-MMP-2/9
Pathway in Melanoma,” DRM, vol. 235, no. 3, pp. 225–233, 2019, doi: 10.1159/000498857.

[34] S. Ostrand-Rosenberg and P. Sinha, “Myeloid-derived suppressor cells: linking inflammation
and cancer,” J Immunol, vol. 182, no. 8, pp. 4499–4506, Apr. 2009, doi:
10.4049/jimmunol.0802740.

[35] S. Trabanelli et al., “Tumour-derived PGD2 and NKp30-B7H6 engagement drives an
immunosuppressive ILC2-MDSC axis,” Nat Commun, vol. 8, Sep. 2017, doi: 10.1038/s41467-
017-00678-2.

78
[36] M. J. Schuijs et al., “ILC2-driven innate immune checkpoint mechanism antagonizes NK cell
anti-metastatic function in the lung,” Nat Immunol, vol. 21, no. 9, pp. 998–1009, Sep. 2020, doi:
10.1038/s41590-020-0745-y.

[37] V. Lucarini et al., “IL-33 restricts tumor growth and inhibits pulmonary metastasis in
melanoma-bearing mice through eosinophils,” OncoImmunology, vol. 6, no. 6, p. e1317420, Jun.
2017, doi: 10.1080/2162402X.2017.1317420.

[38] E. Metryka et al., “Lead (Pb) Exposure Enhances Expression of Factors Associated with
Inflammation,” Int J Mol Sci, vol. 19, no. 6, Jun. 2018, doi: 10.3390/ijms19061813.

[39] L. Maggi et al., “Human circulating group 2 innate lymphoid cells can express CD154 and
promote IgE production,” Journal of Allergy and Clinical Immunology, vol. 139, no. 3, pp. 964-
976.e4, Mar. 2017, doi: 10.1016/j.jaci.2016.06.032.

[40] H. Nishikado, K. Mukai, Y. Kawano, Y. Minegishi, and H. Karasuyama, “NK cell-depleting
anti-asialo GM1 antibody exhibits a lethal off-target effect on basophils in vivo,” J Immunol, vol.
186, no. 10, pp. 5766–5771, May 2011, doi: 10.4049/jimmunol.1100370.

79
Materials and Methods
Mouse experiments
Experimental protocols were approved by the USC institutional Animal Care and Use Committee
(IACUC) and conducted in accordance with the USC Department of Animal Resources’
guidelines. 5-10 week old age and sex matched mice were used in the studies. BALB/cByJ, RAG2
-
/-
(C.B6(Cg)-Rag2
tm1.1Cgn
/J) and Rag2
-/-
γc
-/-
(C;129S4-Rag2tm1.1Flv Il2rgtm1.1Flv/J) mice were
bred in our animal facility at the Keck School of Medicine, University of Southern California
(USC) (all the three strains are in BALB/c background). PD-1-deficient (Pdcd1
-/-
) BALB/c mice
were generated in the Sharpe laboratory as previously described[27]. PD-1
-/-
BALB/c mice were
backcrossed to RAG2
-/-
mice to create Rag2
-/-
PD-1
-/-
mice.

Cell Culture
B16 cells (graciously given by Dr. Weiming Yuan) were cultured in Dulbecco’s modified eagle
medium (DMEM, Sigma-Aldrich Co.) supplemented with 10% heat-inactivated fetal calf serum
and a penicillin/streptomycin cocktail. Cells were routinely cultured in a humidified atmosphere
of 5% Co2 at 37°C in DMEM.

In vivo experiments and tissue preparation
B16 cells (2.5 × 10
5
cells) were grown and injected by tail intravenously (i.v.) into recipient mice
in a volume of 100 µL phosphate-buffered saline (PBS). Where indicated, mice were
intraperitoneally (i.p.) injected with anti-asialo GM1 or isotype (Wako) every third day (day 0, 3,
6, 9, 12), and/or PD-1 blocking antibody or isotype (500µg/mouse, BioXCell) every fourth day
(day 4, 8, 12). After 14 days, mice were euthanized and numbers of tumor colonies on lung surfaces
were counted per field. Lungs were then collected and processed for the indicated readout as
previously described[28]. Briefly, the lungs were perfused with PBS and digested in Collagenase
IV (MP Biomedicals, LLC) for 1 hour at 37°C. Samples were then stained and ILC2s were isolated
based on the absence of common lineage markers (CD3, CD5, CD4, TCRβ, CD45R, Gr-1, CD56,
CD11c, CD11b, Ter119, FcεRIα, CD335), and the expression of CD45, ST2 and CD127.

Human ILC2 isolation and in vitro culture

80
Experimental protocols were approved by the USC Institutional Review Board (IRB) and
conducted in accordance with the principles of the Declaration of Helsinki. Human blood ILC2s
were isolated from total peripheral blood mononuclear cells (PBMCs) to a purity of > 95% on a
FACSARIA III system as described previously[29]. Briefly, human fresh blood was first diluted
1:1 in PBS 1X and transferred to SepMateTM-50 separation tubes (STEMCELL Technologies)
filled with 12mL Lymphoprep™. Samples were centrifuged for 10 minutes and PBMCs were
collected. CRTH2
+
cells were then isolated using the CRTH2 MicroBead Kit, used according to
the manufacturer’s conditions. Samples were then stained and ILC2s were isolated based on the
absence of common lineage markers (CD3, CD5, CD14, CD16, CD19, CD20, CD56, CD235a,
CD1a, CD123), and the expression of CD45, CRTH2 and CD127. Isolated ILC2s were cultured at
37°C (5x10
4
/mL) with rhIL-2 (10ng/mL) and rhIL-7 (10ng/mL) in complete RPMI (cRPMI). To
make cRPMI, RPMI (Gibco) was supplemented with 10% heat-inactivated FBS (Omega
Scientific), 100 units/mL penicillin and 100mg/mL streptomycin (GenClone). When indicated,
human ILC2s were activated with 50ng/mL rhIL-33 for the indicated times. Human PD-1 blocking
antibody or isotype was included in the culture (10µg/ml, BioXCell).

Flow Cytometry
The following murine antibodies were used: biotinylated anti-mouse lineage CD3ε (145-2C11),
CD4 (GK1.5), CD5 (53-7.3), TCRβ (H57-597), CD45R (RA3-6B2), Gr-1 (RB6-8C5), CD11c
(N418), CD11b (M1/70), Ter119 (TER-119), FcεRIα (MAR-1), CD335 (29A1.4), Streptavidin-
FITC, PE-Cy7 anti-mouse CD127 (A7R34), APCCy7 anti-mouse CD45 (30-F11), APC anti-
mouse CD49b (DX5), PE anti-mouse PD-1 (29F.1A12) were purchased from BioLegend. PerCP-
eFluor710 anti-mouse ST2 (RMST2-2) was purchased from Thermofisher. Intranuclear staining
was performed using the Foxp3 Transcription Factor Staining Kit (Thermofisher) per the
manufacturer’s instructions. PE anti-human/mouse RelA NFκB p65 (IC5078P) and Alexa Fluor
647 anti-human/mouse NFκB p52 (C-5) from Biolegend and Santa Cruz Biotechnology
respectively were used. Intracellular staining was performed using the BD Biosciences
Cytofix/Cytoperm kit. When indicated, cells were stimulated in vitro for 4 hours with 50µg/mL
PMA, 500µg/mL ionomycin (both Sigma) and 1µg/mL Golgi plug (BD Biosciences) before
cytokine assessment. BV510 anti-mouse TNF- α (MP6-XT22, Biolegend) was used. PE-Cy7
Annexin V and DAPI were used according to the manufacturer’s instructions. The following

81
human antibodies were used: FITC anti-human lineage cocktail including CD3 (UCHT1), CD14
(HCD14), CD16 (3G8), CD19 (HIB19), CD20 (2H7), CD56 (HCD56). Additional lineage markers
were added: FITC anti-human CD235a (HI264), FITC anti-human FCεRIα (AER-37), FITC anti-
human CD1a (HI149), FITC anti-human CD123 (6H6) and FITC anti-human CD5 (L17F12).
APCCy7 anti-human CD45 (HI30), PECy7 anti-human CD127 (A019D5). Live/dead fixable
violet cell stain kit was used to exclude dead cells (Thermofisher) and CountBright absolute
counting beads (Thermofisher) to calculate absolute cell numbers when indicated. Stained cells
were analyzed on FACSCanto II and/or FACSARIA III systems and the data were analyzed with
FlowJo version 10 software.

Murine ILC2 and in vitro culture
Murine ILC2s were FACS-sorted to a purity of >95% on a FACSARIA III gating system. Isolated
ILC2s were cultured at 37°C (5x10
4
/mL) for 48 hours as indicated with rmIL-2 (10ng/mL), rmIL-
7 (10ng/mL), +/- rmIL-33 (10ng/ml) purchased from BioLegend in cRPMI. In the indicated
experiments, cells were cultured for 48 hours with mouse PD-1 blocking antibody (10µg/ml) or
appropriate isotype purchased from BioXCell. In the experiments involving Transwell plates, B16
cells (10 x 10
3
) were cultured in the bottom well of a 0.4 µm pore Transwell plate (Corning) with
anti-TNF- α or isotype (BioXCell, 10µg/ml). Sorted activated PD-1
-/-
or WT ILC2s (80 x 10
3
) were
cultured in the insert as described above for 48 hours.

Cytokine measurements
The amounts of cytokines in culture supernatants were measured by Legendplex or ELISA MAX
deluxe kit according to the manufacturer’s instructions (Biolegend).

Adoptive Transfer
Experiments were performed as described previously[16]. Mice were first inoculated with B16
cells as described above. Briefly, ILC2 from IL-33 treated PD-1
-/-
or WT mice were sorted from
the lungs. 5.0 x 10
4
of the appropriate population of cells were transferred in PBS to Rag
-/-
γc
-/-

mice by i.v. injection every sixth day (day 0, 6, 12). Mice were euthanized on day 14 and lung
tumor burden was assessed and quantified as previously described. Lungs were then collected and
processed for the indicated readout.

82

Statistical analysis
Experiments were repeated at least three times (n=4-8 each). A student t-test was used for
comparisons between each group using Prism Software (GraphPad Software Inc.). Mean, Standard
Deviation and degree of significance was indicated as: *p<0.05, **p<0.01, ***p<0.001.

83

Figures

Figure 1
B16 melanoma induces PD-1 expression on pulmonary ILC2s.
(A) A cohort of PD-1
-/-
or WT mice was given an intravenous injection of B16 melanoma cells
(2.5 x 10
5
) on day 0. On day 14 tumor cell colonies on the lung surface were counted per field (B)
and presented with corresponding quantification.
(C) Lungs were then digested and ILC2s were identified and quantified. Representative gating
strategy identifying lung ILC2s and PD-1 expression by flow cytometry.
(D) Percentage of ILC2s expressing PD-1 in lungs of WT BALB/cByJ mice intravenously injected
with B16 melanoma or PBS
(E) Percentage and total number of ILC2s in the lungs of PD-1
-/-
or WT mice with B16 melanoma.
Error bars are the mean ± SEM. Data are representative of 3 individual experiments with n=5.
Student’s t-test, p < 0.05, **p < 0.01, ***p < 0.001.

84

Figure 2
PD-1 expression on pulmonary ILC2s promotes B16 tumor growth and drives melanoma-
induced fatality.
(A) A cohort of Rag2
-/-
or Rag2
-/-
PD-1
-/-
mice were intravenously injected with B16 melanoma
(2.5 x 10
5
) cells on day 0. Mice were also intraperitoneally injected with anti-asialo GM1 or isotype
every three days. On day 14, mice were euthanized and lung tumor burden per field was assessed
and quantified.

(B) Kaplan-Meier survival curves of Rag2
-/-
or Rag2
-/-
PD-1
-/-
mice intravenously injected with B16
melanoma.
(C) Whole lungs were isolated on day 14 and total tumors per field on lung surfaces were
quantified. Representative lungs for the cohorts are presented.
(D) Total number of lung ILC2 cells in Rag2
-/-
or Rag2
-/-
PD-1
-/-
mice on day 14.
(E) A cohort of Rag2
-/-
γc
-/-
mice were intravenously injected with B16 melanoma cells (2.5 x 10
5
)
on day 0. Activated PD-1
-/-
or WT ILC2s were adoptively transferred intravenously on days 0, 6
and 12. Mice were euthanized on day 14 and lung tumor burden was quantified (F).
Error bars are the mean ± SEM. Data are representative of 3 individual experiments with n=5.
Student’s t-test, p < 0.05, **p < 0.01, ***p < 0.001.

85

Figure 3
PD-1 deficiency on IL-33 stimulated ILC2s enhances TNF-a expression and phosphorylation
of canonical NF κB pathway.

86
(A) A cohort of WT mice were intranasally challenged with IL-33 (0.5µg) for three consecutive
days, days 0-2. On day 3, mice were euthanized and percentage of TNF- α
+
ILCs was quantified
in (B).
(C) A cohort of WT or PD-1
-/-
mice were intranasally challenged with IL-33 (0.5µg) for three
consecutive days, days 0-2. On day 3, mice were euthanized and percentage of TNF- α
+
ILCs was
quantified.
(D) Volcano plot comparison of whole transcriptome gene expression of sorted PD-1
-/-
ILC2s and
WT control, n=2. Differentially expressed genes (defined as statistically significant adjusted p-
value<0.05) with changes of at least 1.45 fold-change (FC) are shown in yellow (upregulated
genes) or blue (downregulated genes).
(E) Upregulated (yellow) genes in the NF κB pathway and corresponding heatmap representation
(F)
(G) A cohort of PD-1
-/-
or WT mice was given an intravenous injection of B16 melanoma (2.5 x
10
5
) on day 0. On day 14, lung ILC2s were identified as defined in Fig. 1C and transcription factors
involved in the NF κB pathway were measured for phosphorylation by intranuclear flow cytometry.
(H) Representative flow cytometry plots and corresponding quantification of phosphorylation
levels of transcription factor p52.
(I) Representative flow cytometry plots and corresponding quantification of phosphorylation
levels of transcription factor p65.
Error bars are the mean ± SEM. Data are representative of 3 individual experiments with n=5.
Student’s t-test, p < 0.05, **p < 0.01, ***p < 0.001.

87

Figure 4
Blocking PD-1 on ILC2s increases TNF- α

production and enhances cytotoxic properties.
(A) A cohort of WT mice was given an intravenous injection of B16 melanoma on day 0. On day
14 lung ILC2s were sorted by flow cytometry and cultured with isotype or PD-1 blocking antibody
for 48 hours.
(B) After 48 hours, ILC2 cells were harvested and measured for TNF- α

expression by intracellular
flow cytometry. Representative flow cytometry plots and corresponding quantification are
presented as percent TNF- α
+
ILC2s.
(C) B16 melanoma cells were cultured in the bottom well of 0.4 µm Transwell plate. PD-1
-/-
or
WT ILC2s were cultured in the top well insert at 8:1 ratio with B16 melanoma cells respectively
for 48 hours. In the condition specified, anti-TNF- α

or isotype (10µg/ml) was included in the
bottom well of the culture. After 48 hours, B16 cells were collected and stained with apoptosis kit
to assess viability. Representative flow cytometry plots and corresponding quantification are
presented.

88
Error bars are the mean ± SEM. Data are representative of 3 individual experiments with n=5.
Student’s t-test, p < 0.05, **p < 0.01, ***p < 0.001.

Figure 5
Blocking ILC2 PD-1 expression increases TNF- α production and inhibits tumor progression
in vivo.

89
(A) A cohort of PD-1
-/-
or WT mice was given an intravenous injection of B16 melanoma cells
(2.5 x 10
5
) on day 0. On day 14 lungs were digested and percentage of TNF- α
+
ILC2s was
quantified (B).
(C) A cohort of Rag2
-/-
or Rag2
-/-
PD-1
-/-
mice were intravenously injected with B16 melanoma
(2.5 x 10
5
) cells on day 0. Mice were also intraperitoneally injected with anti-asialo GM1 or isotype
every three days. On day 14, mice were euthanized and percentage of TNF- α
+
ILC2s was
quantified (D).
(E) A cohort of Rag2
-/-
mice were intravenously injected with B16 melanoma cells (2.5 x 10
5
) on
day 0. Mice were intraperitoneally injected with anti-asialo GM1 or isotype every three days. Mice
were also intraperitoneally injected with PD-1 blocking antibody or isotype (500µg/mouse) every
four days. On day 14, mice were euthanized and lung tumor burden per field was assessed and
quantified (F).

(G) Total number of lung ILC2 cells present on day 14.
(H) Representative flow cytometry plots and corresponding quantification of TNF- α
+
ILC2s
present in the lung on day 14.
Error bars are the mean ± SEM. Data are representative of 3 individual experiments with n=5.
Student’s t-test, p < 0.05, **p < 0.01, ***p < 0.001.

Figure 6
Blocking PD-1 engagement on human blood ILC2 cells increases TNF- α secretion.

90
(A) ILC2s from the blood were isolated from four human donors and cultured for 72 hours in
cRPMI with IL-33, survival cytokines, and PD-1 blocking antibody or isotype.
(B) Representative flow cytometry plots demonstrating gating strategy for ILC2 sorting protocol.
(C) Representative flow cytometry plots of PD-1 expression on ILC2s cultured with and without
IL-33 after 72 hours.
(D) Supernatant was collected after 72 hours and TNF- α

secretion was measured by Luminex.
Error bars are the mean ± SEM. Student’s t-test, p < 0.05, **p < 0.01, ***p < 0.001.

Supplemental Figures

Supplemental Figure 1. PD-1 expression on pulmonary ILC2s promotes B16 tumor growth
and drives melanoma-induced fatality.
(A) Representative flow cytometry plots of NK cell depletion in the spleen of Rag2
-/-
or Rag2
-/-
PD-1
-/-
mice after repeated intraperitoneal injections with anti-asialo GM1 or isotype as described
in Figure 2A. Corresponding quantification is represented as number of total spleen NK cells.
(B) Representative flow cytometry plots and corresponding quantification of total number of lung
ILC2 cells present on day 14.
Error bars are the mean ± SEM. Data are representative of 3 individual experiments with n=5.
Student’s t-test, p < 0.05, **p < 0.01, ***p < 0.001.

91

Supplemental Figure 2. PD-1 deficiency on IL-33 stimulated ILC2s enhances TNF-a
expression and phosphorylation of canonical NF κB pathway
(A) Secreted levels GM-SCF by PD-1
-/-
and WT ILC2s cultured for 48 hours as measured by
ELISA.
(B) Principal component analysis (PCA) of the normalized RNA seq data transcripts per million
(TPM) of activated PD-1
-/-
or WT ILC2s.

92
Chapter 4:
Orai inhibition modulates pulmonary ILC2 metabolism, effector
function and alleviates airway hyperreactivity

Paper under peer review

Abstract
Group 2 innate lymphoid cells (ILC2s) play an important role in a broad range of immune-
mediated disorders, including allergic asthma, through their secretion of type 2 cytokines. Ca
2+

entry via CRAC (Ca
2+
release-activated Ca
2+
) channels is a predominant mechanism of
intracellular Ca
2+
elevation in immune cells that are important players in airway inflammation.
Here we examine the function of Orai1 and Orai2, pore components of CRAC channels, present
on human and murine ILC2s that are upregulated by IL-33 activation. Inhibition of these channels
using a blocker, or genetic deletion significantly downregulates ILC2 effector function and
cytokine production. Mechanistically, inhibition alters ILC2 metabolic and mitochondrial
homeostasis that subsequently leads to the upregulation of reactive oxygen species production
through the uncoupling of the mitochondrial electron transport chain. Importantly, this therapeutic
inhibitor ameliorates the development of ILC2-mediated airway inflammation in multiple mouse
models. Similarly, we demonstrate translationally that Orai channels in human ILC2s are crucial
for the development of airway hyperreactivity in humanized mice. Thus, these findings have a
broad impact on the basic understanding of the function of Ca
2+
signaling in proinflammatory
ILC2s, as well as provides potential insight into the development of novel therapy to treat allergic
and atopic inflammatory diseases.

Introduction
Group 2 innate lymphoid cells (ILC2s) are innate immune cells that are rapidly activated
by epithelium alarmins, namely thymic stromal lymphopoietin (TSLP), interleukin (IL)-25 and IL-
33(1). ILC2s are the dominant innate lymphoid cell population in the lungs at steady state and
upon activation their release of type-2 cytokines, predominately IL-5 and IL-13, is a central driver
in responding eosinophil infiltration and the development of airway hyperreactivity independent

93
of adaptive immunity(2). Their non-redundant role in the lung is uniquely pro-inflammatory due
to their positional niche after allergen challenge. At steady state, ILC2s occupy the peripheral and
central sites of the lung, but cluster at the airway epithelium and alveolar space upon airway
inflammation(3). This allows the cells to be the first responders to allergens that enter the system,
driving eosinophilia, increased goblet cell activation and mucus production(4).
Due to their crucial role in the development of inflammation associated with allergic
asthma, delineating the intracellular signaling and cellular mechanisms that drive ILC2 effector
functions is critical for potential modulation and immunoregulatory control(5). We and others have
shown that CRAC (Ca
2+
release-activated Ca
2+
) channels represent a primary mode of extracellular
Ca
2+
entry in a variety of immune cells including T cells and mast cells that play an essential role
in immune cell proliferation, activation, and basic effector function(6–10). Three Orai proteins
have been previously described as major subunits forming the CRAC channels(11). Recently, it
has been shown that Orai1 and Orai2 can form heterotrimeric CRAC channels, and that inhibition
of both subunits will severely affect neutrophil and mast cell function(11,12). During an immune
response, depletion of the endoplasmic reticulum (ER) Ca
2+
stores activates ER-resident protein
Stromal Interaction Molecule 1 (STIM1) to gate Orai channels on the plasma membrane, allowing
for the initiation of the store-operated Ca
2+
entry (SOCE), a major mechanism utilized by non-
excitable cells to raise intracellular Ca
2+
concentration(13). Despite the well documented
importance of CRAC channels in immune effector function, the mechanisms by which they are
regulated and their role in pathogenic ILC2s, specifically in the context of the development of
allergic airway inflammation remains unresolved.
Recently, our group and others have detailed the importance of metabolism and
mitochondrial health in ILC2 effector function(14–16). It has been previously demonstrated that
cytokine IL-5 and IL-13 secretion by pathogenic ILC2s is dictated by the utilization of fatty acid
oxidation (FAO) to generate ATP(14,15). In the absence of FA or if the pathway is defective in a
way that limits ATP generation, ILC2s will metabolically adapt to utilize glycolysis(15). Emerging
evidence demonstrates that Ca
2+
entry via the CRAC channels may play a vital role in metabolism.
Impairments in Ca
2+
exchange have been shown to lead to high alterations in the metabolism of
neurons in models of Alzheimer’s Disease(17). Previously it has been shown in CD4+ Th17 cells
that deletion of STIM1 and STIM2 severely affected mitochondrial and metabolic health, resulting
in altered inner mitochondrial membrane architecture and reduced immune cell effector

94
function(18). Vaeth et al recently documented the necessity of STIM1 for functional oxidative
phosphorylation (OXPHOS) and glycolysis in non-pathogenic Th17 cells, while pathogenic Th17
cells operate glycolysis independent of Orai or CRAC channels(6). While it is clear that
metabolism has a critical role in ILC2 activation, currently the relationship between Ca
2+
entry in
ILC2s and metabolism, if any, has yet to be explored.
In this study, we show that Orai1 and Orai2 are expressed on murine and human ILC2s,
and that inhibition of the channels result in an amelioration of the development of humanized
ILC2-dependent airway hyperreactivity (AHR). We demonstrate that blocking the Orai channels
downregulate both the FAO/OXPHOS and glycolysis in proinflammatory ILC2s. Further the
mitochondrion is significantly affected, resulting in a downregulation of the mitochondrial electron
transport chain and an upregulation in reactive oxygen species (ROS) through an uncoupling of
the mitochondrial electron transport chain. Together, our findings give further insight into the
fundamental biology of ILC2s, opening avenues into potential downstream targets that may be
pharmacologically manipulated to specifically modulate human ILC2s for the treatment of a broad
array of inflammatory diseases. Specifically, our study can have a direct translational impact on
development of improved therapies for allergic asthma.

Results
Calcium channels Orai1 and Orai2 are expressed on murine pulmonary ILC2s and are
upregulated by IL-33 activation
The study of calcium channels in autoimmune and allergic diseases has recently gained
therapeutic interest. Here we have studied the pro-inflammatory role of calcium channels Orai1
and Orai2 in eosinophilic asthma mouse models. The role of the Ca
2+
-calcineurin-NFAT pathway
is well defined in immune cells, but the mechanism on how Ca
2+
ions enter cells and its
downstream effect on the cell health and survival has only recently begun to be understood. It has
been previously demonstrated that immune cells largely express Orai1, and that mucosal tissues
including the lung tend to express high levels of Orai2(12). To investigate whether pulmonary
ILC2s express the Orai channel isoforms, we first performed RNA-sequencing on FACs-sorted
ILC2s activated with recombinant mouse (rm)IL-33 (Figure 1A). Orai1 and Orai2 are highly
expressed in activated murine ILC2s, while Orai3 demonstrated basal expression comparatively
(Figure 1A). We then took advantage of publicly available single-cell RNA sequencing (scRNA-

95
seq) for naïve vs IL-33 activated pulmonary ILC2s to investigate further the individual cell
expression of Orai1 and Orai2 (Figure 1B). Gene counts for Orai1 and Orai2 isoforms were
elevated in both naïve and IL-33 induced ILC2s, suggesting these channels may play an important
role in Ca
2+
entry in pulmonary ILC2s. To explore the expression of the channel in vivo, we
challenged mice for three consecutive days with recombinant mouse (rm)IL-33 and assessed
expression level of the proteins in pulmonary ILC2s by flow cytometry on day four. Our data
demonstrate that activated pulmonary ILC2s express significant levels of Orai1 and Orai2 (Figure
1C-D). Interestingly, naive ILC2s expressed basal levels of Orai1 and Orai2, though IL-33
activation further enhanced the expression, especially of Orai1. The results indicated that Orai1
and Orai2 are expressed on activated pulmonary ILC2s and may play a role in the development of
airway hyperreactivity.

Blocking Orai channels on pulmonary ILC2s significantly downregulates pro-inflammatory
effector function
In order to determine the role of the expressed Orai channels in ILC2 effector function, we
next aimed at determining the most efficient and specific pan-Orai inhibitor on ILC2s. Previously,
using a chemical library screen, we identified compound 5D, N-[2,2,2-trichloro-1-(2-
naphthylamino)ethyl]-2-furamide, that showed an IC50 value of 195 nM in blocking SOCE in
primary T cells(19). Biologically, we have previously shown that 5D efficiently suppressed
effector T-cell function and IL-2 production, suggesting that it is a potent tool to study the function
of Orai channels in the context of functional immune cells(19). Given the essential role of ILC2s
in the pathology of allergic asthma and their dynamic expression of Orai channels, we examined
the effect of 5D on ILC2-mediated airway inflammation. To ensure 5D does indeed block Ca
2+

entry in pulmonary ILC2s, activated ILC2s were FACS-sorted and cultured with survival
cytokines recombinant mouse (rm)IL-2 and IL-7 overnight. After 18 hours, cells were loaded with
Fluo-8 AM and pretreated with thapsigargin to deplete the intracellular Ca
2+
stores. Fluorescence
of the Ca
2+
entry detected by the stain was then measured by flow cytometry before and after
calcium and the inhibitor 5D addition. SOCE measurement with and without addition of compound
5D showed a strong reduction in the presence of compound 5D (Figure 2A). To investigate further
the effect of blocking Ca
2+
in ILC2s, we examined the function of ILC2s after treatment with
compound 5D in vitro (Figure 2B). Briefly, pulmonary naïve ILC2s were FACS-sorted and

96
cultured in cRPMI for 24 hours with 5D or the vehicle. After 24 hours, IL-33 or saline were added
to the culture for an additional 48 hours. ILC2s cultured with 5D demonstrated lower levels of
proliferation marker Ki67 (Figure 2C), as well as reduction in secreted proinflammatory
cytokines, including IL-5 and IL-13 (Figure 2D, S2A) compared to the vehicle control.
Intracellular IL-5 and IL-13 levels were also significantly lower in cells cultured with 5D, further
suggesting an important role of SOCE in ILC2 pro-inflammatory effector function (Figure 2E).
To confirm our findings, we generated Orai1
fl/fl
; UBC-Cre/ERT2 and Orai1
fl/fl
; Orai2
-/-
; UBC-
Cre/ERT2 mice as described in the Methods and validation was performed to confirm the Orai2
deletion (Figure S1). We first sorted naïve pulmonary ILC2s from control, Orai1
fl/fl
; UBC-
Cre/ERT2 and Orai1
fl/fl
; Orai2
-/-
; UBC-Cre/ERT2 mice. Animals of all genotypes had similar
number of ILC2s in the lungs at steady state (Figure S2B). Cells were then cultured with IL-33 to
induce expansion. Further, cells were treated with 4-hydroxy tamoxifen (4-OHT) in vitro, to
induce Orai1 deletion. After expansion, supernatant and intracellular cytokines were measured by
ELISA and flow cytometry respectively (Figure 2F-G). And indeed, major effector function
cytokines secreted from ILC2s, including IL-5, IL-13, as well as IL-9, IL-6 are significantly
downregulated by the deletion of Orai channels (Figure 2F, S2C). Significantly, Orai1 deletion
downregulates cytokine production, though deletion of both Orai channels has a greater effect both
at the supernatant level and intracellularly (Figure 2F-G, S2D). Together, our findings provide
strong evidence that blocking Ca
2+
entry specifically through Orai channels play a crucial and
previously underrecognized role in the effector function of ILC2s.

Inhibiting Orai channels affects the transcriptomic landscape of pro-inflammatory
pulmonary ILC2s
To further explore the relationship of ILC2s and the role Orai channels play in their effector
function, we performed RNA sequencing on activated ILC2s cultured with and without 5D for 24
hours. Not surprisingly, a significant number of genes were differentially expressed between
vehicle-treated and compound 5D-treated cells (Figure 3A). 160 genes were significantly
upregulated, while 199 genes were significantly downregulated, totaling 359 genes differentially
expressed in ILC2s treated with 5D as compared to the control. We further analyzed the pathways
of differentially expressed genes by Ingenuity Pathway Analysis (IPA) (Figure 3B). In line with
our previous data, the most downregulated pathways in ILC2s cultured with 5D include Th2

97
pathway and activation, as well as metabolic pathways. Cholesterol biosynthesis and glycolysis
appear to be significantly downregulated in 5D-treated ILC2s, suggesting blocking Orai channels
may largely affect the mitochondria and metabolic health of the ILC2s(20). Importantly, we
analyzed the RNA sequencing (RNA-seq) data to focus specifically on mRNA related to ILC2
health and effector function. We found that Il9, Mki67, Il5, and Il6 were significantly
downregulated in ILC2s treated with 5D in confirmation that cytokines are downregulated without
Orai channels (Figure 3C-D). Il13 was also downregulated, though not statistically. Interestingly,
transcription factors important in ILC2 homeostasis were largely unaffected by Orai inhibition,
suggesting an alternative mechanism by which 5D regulates cytokine production (Figure 3D). To
confirm the glycolytic pathway was downregulated as identified by IPA, we investigated the
individual genes involved and indeed, glycolysis is largely decreased in Orai-inhibited ILC2s
(Figure 3E). Altogether, our data demonstrate that Orai channels support and improve ILC2 pro-
inflammatory cytokine production and effector function.

Inhibiting Orai channels significantly downregulates metabolic pathways in pulmonary
ILC2s
Orai channels and calcium entry has previously been shown to play a crucial role in
metabolic processes in proinflammatory Th17 cells(18). Our group and others have previously
established that ILC2s primarily rely on fatty acid oxidation for their cytokine production and
effector function(14,15). Though we saw little significant difference in our RNA-seq data
suggesting a transcriptomic difference in fatty acid oxidation, the decrease in cytokine production
witnessed by ILC2s cultured with 5D led us to investigate the functional use of the fatty acid
oxidation pathway in Orai-inhibited cells. We first measured fatty acid uptake in ILC2s cultured
with 5D or the control vehicle (Figure 4A). Interestingly, we found that 5D-treated ILC2s showed
decreased mean fluorescence intensity of BODIPY, suggesting that fatty acid uptake is inhibited
by Orai channel blocking ex vivo (Figure 4A). To confirm the differential fatty acid uptake and
to understand how the metabolomics results functionally, we measured oxygen consumption rate
(OCR), reflective of active oxidative phosphorylation (Figure 4B). ILC2s treated with the Orai
inhibitor demonstrated significantly lower basal respiration (Figure 4C), maximum respiration
(Figure 4D), and spare respiratory capacity (Figure 4E), demonstrating that inhibition of the Orai
channels has a significant detrimental effect on the mitochondrial pathways and function.

98
Previously, our group has demonstrated that in the absence of FAO, ILC2s preferentially switch
to glycolysis as an energy source(16). However, our previous RNA-seq data as demonstrated in
Figure 3B suggested that glycolysis is also downregulated in compound 5D-treated ILC2s. And
indeed, 2-NBDG levels, as well as the L-lactate in the supernatant of cultured compound 5D-
treated ILC2s versus the control were both significantly downregulated when compared to controls
(Figure 4F-G). To confirm these results, we measured glycoPER, the glycolytic capacity of the
cells utilizing Seahorse technology. As expected, compound 5D-treated ILC2s demonstrated
significantly less capacity to perform effective glycolysis than the control activated ILC2s (Figure
4H-I). Together, these results demonstrate calcium flux is indispensable for metabolism, and
suggests a possible defect in overall mitochondrial function after Orai inhibition.

Orai-dependent mitochondrial function is essential for ILC2 effector function
RNA-sequencing data, together with impaired metabolic pathways, point to a significant
difference in mitochondrial function after inhibition of Orai channels. It’s been demonstrated
previously that functional mitochondria are critical for metabolism, both aerobic OXPHOS/FAO
and anerobic glycolysis(21,22). Strikingly, our RNA-seq revealed a severe decrease in genes
associated with various protein complexes involved in the mitochondrial electron transport chain
(mtETC) when ILC2s were cultured with 5D (Figure 5A-B). Mitochondrial mass and respiration
were measured by flow cytometry as mean fluorescence of Mitotracker green and red, respectively
(Figure 5C-D), and showed a significant decrease in ILC2s cultured with 5D, further suggesting
a functional decrease in the mitochondria of the treated cells. Surprisingly, both total and
mitochondrial reactive oxygen species (ROS) levels were elevated in Orai-inhibited ILC2s (Figure
5E-F). Traditionally a lowered membrane potential coincides with lower ROS production.
However it has been shown that oxidative stress can occur at very high or low levels of membrane
potential and that an increase in ROS in these situations could point to mitochondrial uncoupling
of the ETC(23). To investigate whether the rise we see in ROS is the result of mitochondrial
uncoupling, we measured the ratio of NADH/NAD+ in cells treated with 5D (Figure 5G). We
found treated ILC2s did have higher ratios of NADH/NAD+ while producing less ATP through
oxidative phosphorylation than cells treated with the vehicle (Figure 5H), suggesting the cells are
indeed undergoing mitochondrial uncoupling. The elevated ROS resulted from this uncoupling
therefore likely points to a defect in the cell’s antioxidants. To investigate this, as well as whether

99
the increased ROS could explain the loss of effector function after Orai inhibition in vitro, we
utilized the addition of antioxidant N-Acetyl-L-cysteine (NAC). FACs-sorted ILC2s were cultured
for 48 hours with the vehicle, 5D, or the combination of 5D and NAC and were afterwards assessed
for effector function. And indeed, proliferation and pro-inflammatory intracellular cytokine
production were restored after addition of NAC in the culture (Figure 5I-K). Altogether, our data
demonstrate a previously unknown mechanism by which Ca
2+
entry through Orai channels
regulates pro-inflammation through modulation of cellular mitochondrial function.

Orai inhibition downregulates ILC2-mediated airway hyperreactivity in vivo
To explore the direct effect of Orai channels on ILC2 effector function in vivo, we first
subjected BALB/c genetic background wild-type mice to the physiologically relevant setting
utilizing Alternaria alternata (A. alternata), an allergen known to indirectly stimulate ILC2s(24)
with the addition of intraperitoneal (i.p.) injections of either 5D (0.02mg/mouse) or the vehicle
(Figure 6A). Mice given the injections of 5D had significantly lower levels of lung resistance and
the development of airway hyperreactivity (Figure 6B) as compared to those that received the
vehicle. Furthermore, mice treated with 5D harbored significantly fewer eosinophils infiltrating
the BAL (Figure 6C). Consistent with these results, we found fewer lung ILC2s in mice treated
with 5D, as well as a smaller percentage of ILC2s that produce IL-5, their primary
proinflammatory cytokine responsible for eosinophil recruitment (Figure 6D-E). To further focus
on the direct role of Orai channels in ILC2 effector function in vivo, we again utilized FACS-sorted
pulmonary ILC2s from inducible Orai1
-/-
and Orai1/2
-/-
mice, as well as control ILC2s. The Orai-
deleted and control ILC2s were adoptively transferred into cohorts of Rag2
-/-
γc
-/-
mice, mice that
lack all B, T, and NK cells, including ILC2s. Mice were then subjected to the A. alternata protocol
outlined above, and lung function was assessed on day 4 (Figure 6F). As expected, mice
adoptively transferred with Orai1
-/-
demonstrated significantly lower AHR and fewer eosinophils
than those given the control ILC2s (Figure 6G-H). Mice adoptively transferred the knock-out cells
also demonstrated fewer ILC2s found in the lung after day 4, though the same number of cells was
transferred day 0 (Figure 6I). Strikingly, Orai1/2
-/-
transferred mice had an even greater effect,
suggesting that Orai2 has the ability to partially rescue the deletion of Orai1 in the context of ILC2
effector function, as also seen in Figure 2F. Together, our data suggest that Orai1 and Orai2

100
support pathogenic ILC2 effector function in vivo, driving the development of ILC2-dependent
airway hyperreactivity.

Functional Orai channels enhance the induction of AHR in hILC2 recipient mice
We next investigated whether our results in murine models would translate to human
ILC2s. Peripheral blood ILC2s from healthy donors were FACS-sorted and cultured with IL-33 to
measure the expression of Orai channels (Figure 7A). Human ILC2s do indeed express Orai1 and
Orai2, both at the naïve state and after IL-33 stimulation (Figure 7B-C). The cells were then
cultured with 5D or vehicle and the supernatant was collected to measure cytokines by ELISA. As
expected, inhibition of the Orai channels significantly downregulated proinflammatory cytokine
production in the human ILC2s (Figure 7D). To investigate if this would translate therapeutically,
we utilized our previously described humanized mouse model(25). Human IL-5 has been shown
to activate murine eosinophils emphasizing the feasibility of using humanized mice in eosinophilic
inflammatory studies(26,27). Here, we isolated human peripheral ILC2s from healthy donors and
cultured them with recombinant human (rh)IL-7 and rhIL-2 for 48 hours (Figure 7E). The cells
were then adoptively transferred into Rag2
-/-
γc
-/-
mice and the mice were challenged for 3
consecutive days with rhIL-33 intranasally to induce an ILC2-dependent murine model of asthma.
Mice were also given i.p. injections of 5D or the vehicle and lung inflammation was measured on
day 3. Excitingly, mice given the Orai inhibitor demonstrated significantly lower AHR
development (Figure 7F), as well as fewer eosinophils in the BAL and human ILC2s found in the
lungs (Figure 7G-H). Altogether our results using humanized mice suggests this therapeutic drug
5D targeting the Orai pathway has the potential to alleviate asthma symptoms in patients.

Discussion
Overall this study introduces a novel mechanism by which Ca
2+
entry through Orai
channels modulate proinflammatory ILC2 effector function and the development of airway
hyperreactivity in a variety of pharmacological murine models of lung inflammation. We show
that specific inhibition of Orai1 and Orai2 on pulmonary ILC2s leads to a significant decline of
functional metabolic pathways as well as the direct downregulation of mtETC proteins.
Mechanistically, this causes an increase in mitochondrial ROS production and an uncoupling of
the mitochondrial electron transport chain. Further, Orai inhibition severely downregulates ILC2

101
effector function and cytokine production in vivo and in vitro, potentially offering a novel
therapeutic application of the CRAC channels in treating ILC2-dependent diseases.
To our knowledge, our group is the first to report the role of Orai channels and Ca
2+
entry
in the health and homeostasis of pulmonary ILC2s, specifically in the context of AHR. Currently,
therapeutic measures for asthma treatment involving calcium include the blockers for chloride
channels (e.g. sodium cromoglicate and nedocromil sodium) and the Ca
2+
-activated K
+
channel,
KCa3.1(28–30). KCa3.1 channels cause efflux of K
+
ions to maintain the driving force for Ca
2+

entry via the CRAC channels. These studies indicate that targeting the activation of CRAC
channels would be excellent drug targets for asthma therapy. The widely used small molecule
blockers of the Ca
2+
-NFAT pathway including cyclosporine A or FK506 have a broad range of
side effects because of their ubiquitous function(31,32). Unlike calcineurin, CRAC channels have
been shown to play a primary role in immune cell types including T cells, B cells and mast cells,
and thus Orai blockers are expected to be more specific with significantly fewer side effects.
Importantly, our humanized mouse model demonstrates the potential for 5D as a therapeutic
application against ILC2-dependent lung inflammation. One limitation of the study is that injection
of the drug in the humanized mouse model is not acting only on the transferred human ILC2s.
However this model, that is independent of adaptive immunity, in combination with the Orai1/2
-/-
adoptive transfer illustrate the exciting potential for the inhibition of Orai channels on ILC2s as a
means of treatment in context of human AHR. Although recent years have revealed that ILC2s
play an important role in a broad range of immune-mediated disorders, a major gap exists between
understanding ILC2 biology and modulating these cells for therapeutic application.
We observed in our study that pulmonary mouse and human ILC2s express both Orai1 and
Orai2 channels after IL-33 stimulation. Interestingly, while both isoforms are expressed at basal
levels in naïve ILC2s, Orai1 appears to be highly inducible after activation, while Orai2 is only
slightly enhanced in both murine and human ILC2s. Grimes et al recently detailed Orai isoform
expression on neutrophils and their effect on neutrophil activation(33). They also found basal
levels of Orai1 and Orai2 on resting neutrophils, expressed at a 1:1 ratio, while activation by
inflammatory agents raised the ratio to 30:1(33). They concluded the ratio change was mainly due
to the elevation of Orai1 expression after exposure(33). We believe a similar phenomenon is
occurring in activated ILC2s. Inhibition of both Orai1 and Orai2 through therapeutic 5D or
genetically engineered double knock-out mice leads to a significant downregulation in

102
proliferation and cytokine production, including IL-5, IL-13, GM-CSF, IL-9 and IL-6. The role of
Orai1 in the regulation of immune effector function has been well established, but investigation
into the alternative isoforms and their role is lacking. Recently Vaeth et al demonstrated that Orai1
deletion in T cells reduces effector function, while specific deletion of Orai2 enhances SOCE in
both T cells and macrophages(12). Alternatively, in neutrophils, deletion of Orai2 was shown to
reduce SOCE and negatively affect effector function(33). The effect of genetically deleting only
Orai2 on ILC2s requires further investigation. Regardless, both studies found that double deletion
of Orai1 and Orai2 on T cells or neutrophils reduces the cells’ ability to pathogenetically function
in their respective disease setting(12,33). Similarly in such settings, we discovered that blocking
Orai channels specifically in pulmonary ILC2s in vivo efficiently ameliorated the development of
ILC2 dependent AHR stimulated by A. alternata or IL-33 in both murine models and humanized
mouse models of lung inflammation. We observed lower levels of AHR, as well as fewer
infiltrating BAL eosinophils and IL-5-producing ILC2s in the lungs. Our results suggest Orai
channels play a crucial role in the development of lung inflammation in ILC2-dependent models
of asthma.
Our group here has demonstrated that Orai channel inhibition severely alters mitochondrial
health and function, potentially offering a mechanism by which Ca
2+
concentration controls
pathogenic immune function. Previously it has been shown in CD4+ Th17 cells that inhibition of
CRAC channels severely affected mitochondrial homeostasis, resulting in similar observations of
increased ROS production as well as reduced immune cell effector function(18). Ca
2+
entry is well
known to assist mitochondria in the production of ATP through the increase in production and
consumption of NADH involved in Complex I and V in the mtETC(34). Traditionally,
mitochondrial membrane potential and mass can be indicative of effective mtETC function and
ATP synthesis. Additionally, lower mitochondrial membrane potential classically correlates with
lower levels of ROS production(35). In our case however we see a decrease in membrane potential
in treated cells correlated with an increase in mtROS production. We suspect the cells with
inhibited Orai channels are undergoing mtETC uncoupling. In these cases, mitochondrial ETC
uncoupling, especially in the impairment of ADP-coupled oxidative phosphorylation, will lead to
an increase in ROS production primarily due to a defect in the cellular antioxidant system(23),
supported by our data demonstrating supplementation of antioxidants correlates with increased
ILC2 proliferation and cytokine production. The ETC uncoupling is confirmed by the increased

103
NADH/NAD+ ratio in conjunction with the decreased ATP production in cells treated with the
inhibitor, as mtETC uncoupling leads to the cell focusing on the generation of ROS rather than the
maintenance of ATP production(36). What remains unclear however is where specifically in the
pathway the inhibition of Ca
2+
flux is acting to induce such uncoupling. One option is that the
inhibition of the Ca
2+
flux itself results in defects in the antioxidant system, subsequently leading
to an uncontrolled rise in ROS production. Excessive ROS has been linked to mitochondrial DNA
damage, similar to that seen in our RNA-seq data, and defects in the complexes can lead to an
uncoupling of the mtETC(37). An alternative option is that the Orai inhibition is acting directly to
downregulate the mtETC genes, resulting in defects in the chain. Defects could lead to a lower
membrane potential as well as the uncoupling of the ETC, resulting to higher ROS levels that
overwhelm the antioxidant system. Regardless, excessive ROS has been well linked to insufficient
ATP production, impairment of the mitochondria, DNA damage and eventually cell death(38).
Though the details of how mtROS interferes with gene transcription is still being investigated,
mitochondrial (mt)DNA and mitochondrial reactive oxygen species (mtROS) have been shown to
be linked to cascades of activation and inhibition transcription factors, including the CRAC-
associated NFAT and inflammasome NLRP3, providing further evidence that metabolism and
immune responses are intimately dependent on one another(21).
In summary, in this report we reveal a previously unrecognized immunoregulatory role for
inhibition of Orai channels on ILC2s in the context of murine and humanized mouse AHR
development. Our results suggest a crucial role for Orai channels in the functional mitochondrial
electron transport chain and subsequent proinflammatory effector function. Further investigation
into the signaling pathways of how this defect affects cytokine production are warranted, as
mitochondrial control of nuclear genes is largely unknown. However, findings from this study
could lead to identification of therapeutic targets that can be pharmacologically manipulated to
specifically modulate ILC2s for the treatment of asthma. More broadly, this study provides a novel
paradigm for metabolically manipulating other complex cells as well. ILC2s are well established
to be present in mucosal tissue and their overproduction of proinflammatory cytokines has been
linked to a significant variety of autoimmune disease and disorders(4). The results of our studies
will therefore have potential implications in other diseases involving ILC2s, including allergic
diseases such as atopic dermatitis, allergic rhinitis and inflammatory bowel disease.

104
References
1. Barlow JL, Peel S, Fox J, Panova V, Hardman CS, Camelo A, et al. IL-33 is more potent than
IL-25 in provoking IL-13-producing nuocytes (type 2 innate lymphoid cells) and airway
contraction. J Allergy Clin Immunol. 2013 Oct;132(4):933–41.

2. Kim HY, Chang YJ, Subramanian S, Lee HH, Albacker LA, Matangkasombut P, et al. Innate
lymphoid cells responding to IL-33 mediate airway hyperreactivity independently of adaptive
immunity. J Allergy Clin Immunol. 2012 Jan;129(1):216-227.e1-6.

3. Mindt BC, Fritz JH, Duerr CU. Group 2 Innate Lymphoid Cells in Pulmonary Immunity and
Tissue Homeostasis. Front Immunol. 2018 Apr 30;9:840.

4. Bartemes KR, Kita H. Roles of innate lymphoid cells (ILCs) in allergic diseases: The 10-year
anniversary for ILC2s. Journal of Allergy and Clinical Immunology. 2021 May 1;147(5):1531–
47.

5. McKenzie ANJ. Type-2 Innate Lymphoid Cells in Asthma and Allergy. Ann Am Thorac Soc.
2014 Dec;11(Suppl 5):S263–70.

6. Vaeth M, Kahlfuss S, Feske S. CRAC Channels and Calcium Signaling in T Cell-Mediated
Immunity. Trends in Immunology. 2020 Oct 1;41(10):878–901.

7. Clemens RA, Lowell CA. CRAC channel regulation of innate immune cells in health and
disease. Cell Calcium. 2019 Mar;78:56–65.

8. Li X, Zeng Q, Wang S, Li M, Chen X, Huang Y, et al. CRAC Channel Controls the
Differentiation of Pathogenic B Cells in Lupus Nephritis. Frontiers in Immunology [Internet].
2021 [cited 2022 May 17];12.

9. Lioudyno MI, Kozak JA, Penna A, Safrina O, Zhang SL, Sen D, et al. Orai1 and STIM1 move
to the immunological synapse and are up-regulated during T cell activation. Proceedings of the
National Academy of Sciences. 2008 Feb 12;105(6):2011–6.

10. Feske S, Gwack Y, Prakriya M, Srikanth S, Puppel SH, Tanasa B, et al. A mutation in Orai1
causes immune deficiency by abrogating CRAC channel function. Nature. 2006 May
11;441(7090):179–85.

11. Di Capite JL, Bates GJ, Parekh AB. Mast cell CRAC channel as a novel therapeutic target in
allergy. Curr Opin Allergy Clin Immunol. 2011 Feb;11(1):33–8.

12. Vaeth M, Yang J, Yamashita M, Zee I, Eckstein M, Knosp C, et al. ORAI2 modulates store-
operated calcium entry and T cell-mediated immunity. Nat Commun. 2017 Mar 15;8(1):14714.

13. Srikanth S, Gwack Y. Orai1-NFAT Signaling Pathway Triggered by T Cell Receptor
Stimulation. Mol Cells. 2013 Mar 31;35(3):182–94.

105

14. Wilhelm C, Harrison OJ, Schmitt V, Pelletier M, Spencer SP, Urban JF, et al. Critical role of
fatty acid metabolism in ILC2-mediated barrier protection during malnutrition and helminth
infection. J Exp Med. 2016 Jul 25;213(8):1409–18.

15. Galle-Treger L, Hurrell BP, Lewis G, Howard E, Jahani PS, Banie H, et al. Autophagy is
critical for group 2 innate lymphoid cell metabolic homeostasis and effector function. Journal of
Allergy and Clinical Immunology. 2020 Feb 1;145(2):502-517.e5.

16. Howard E, Lewis G, Galle-Treger L, Hurrell BP, Helou DG, Shafiei-Jahani P, et al. IL-10
production by ILC2s requires Blimp-1 and cMaf, modulates cellular metabolism, and ameliorates
airway hyperreactivity. J Allergy Clin Immunol. 2021 Apr;147(4):1281-1295.e5.

17. Jadiya P, Kolmetzky DW, Tomar D, Di Meco A, Lombardi AA, Lambert JP, et al. Impaired
mitochondrial calcium efflux contributes to disease progression in models of Alzheimer’s disease.
Nat Commun. 2019 Aug 29;10(1):3885.

18. Kaufmann U, Kahlfuss S, Yang J, Ivanova E, Koralov SB, Feske S. Calcium Signaling
Controls Pathogenic Th17 Cell-Mediated Inflammation by Regulating Mitochondrial Function.
Cell Metabolism. 2019 May;29(5):1104-1118.e6.

19. Kim KD, Srikanth S, Tan YV, Yee MK, Jew M, Damoiseaux R, et al. Calcium signaling via
Orai1 is essential for induction of the nuclear orphan receptor pathway to drive Th17
differentiation. J Immunol. 2014 Jan 1;192(1):110–22.

20. Ribas V, García-Ruiz C, Fernández-Checa JC. Mitochondria, cholesterol and cancer cell
metabolism. Clin Transl Med. 2016 Jul 25;5:22.

21. Angajala A, Lim S, Phillips JB, Kim JH, Yates C, You Z, et al. Diverse Roles of Mitochondria
in Immune Responses: Novel Insights Into Immuno-Metabolism. Frontiers in Immunology.
2018;9:1605.

22. Berthiaume JM, Kurdys JG, Muntean DM, Rosca MG. Mitochondrial NAD+/NADH Redox
State and Diabetic Cardiomyopathy. Antioxidants & Redox Signaling. 2019 Jan 20;30(3):375–98.

23. Li X, Fang P, Mai J, Choi ET, Wang H, Yang X feng. Targeting mitochondrial reactive oxygen
species as novel therapy for inflammatory diseases and cancers. Journal of Hematology &
Oncology. 2013 Feb 25;6(1):19.

24. Hurrell BP, Howard E, Galle-Treger L, Helou DG, Shafiei-Jahani P, Painter JD, et al. Distinct
Roles of LFA-1 and ICAM-1 on ILC2s Control Lung Infiltration, Effector Functions, and
Development of Airway Hyperreactivity. Front Immunol [Internet]. 2020

25. Shafiei-Jahani P, Helou DG, Hurrell BP, Howard E, Quach C, Painter JD, et al. CD200-
CD200R immune checkpoint engagement regulates ILC2 effector function and ameliorates lung
inflammation in asthma. Nat Commun. 2021 May 5;12(1):2526.

106

26. Ingley E, Cutler RL, Fung MC, Sanderson CJ, Young IG. Production and purification of
recombinant human interleukin-5 from yeast and baculovirus expression systems. European
Journal of Biochemistry. 1991;196(3):623–9.

27. Tavernier J, Tuypens T, Verhee A, Plaetinck G, Devos R, Van der Heyden J, et al.
Identification of receptor-binding domains on human interleukin 5 and design of an interleukin 5-
derived receptor antagonist. Proc Natl Acad Sci U S A. 1995 May 23;92(11):5194–8.

28. Holgate ST, Polosa R. Treatment strategies for allergy and asthma. Nat Rev Immunol. 2008
Mar;8(3):218–30.

29. Duffy SM, Berger P, Cruse G, Yang W, Bolton SJ, Bradding P. The K+ channel iKCA1
potentiates Ca2+ influx and degranulation in human lung mast cells. Journal of Allergy and
Clinical Immunology. 2004 Jul 1;114(1):66–72.

30. Cruse G, Duffy SM, Brightling CE, Bradding P. Functional KCa3.1 K+ channels are required
for human lung mast cell migration. Thorax. 2006 Oct;61(10):880–5.

31. Olyaei AJ, de Mattos AM, Bennett WM. Nephrotoxicity of immunosuppressive drugs: new
insight and preventive strategies. Curr Opin Crit Care. 2001 Dec;7(6):384–9.

32. Naesens M, Kuypers DRJ, Sarwal M. Calcineurin inhibitor nephrotoxicity. Clin J Am Soc
Nephrol. 2009 Feb;4(2):481–508.

33. Grimes D, Johnson R, Pashos M, Cummings C, Kang C, Sampedro GR, et al. ORAI1 and
ORAI2 modulate murine neutrophil calcium signaling, cellular activation, and host defense.
PNAS. 2020 Sep 29;117(39):24403–14.

34. Marcu R, Wiczer BM, Neeley CK, Hawkins BJ. Mitochondrial Matrix Ca2+ Accumulation
Regulates Cytosolic NAD+/NADH Metabolism, Protein Acetylation, and Sirtuin Expression. Mol
Cell Biol. 2014 Aug;34(15):2890–902.

35. Ballinger SW. Mitochondrial dysfunction in cardiovascular disease. Free Radical Biology and
Medicine. 2005 May 15;38(10):1278–95.

36. Vardhana SA, Hwee MA, Berisa M, Wells DK, Yost KE, King B, et al. Impaired mitochondrial
oxidative phosphorylation limits the self-renewal of T cells exposed to persistent antigen. Nat
Immunol. 2020 Sep;21(9):1022–33.

37. Tan AS, Baty JW, Berridge MV. The role of mitochondrial electron transport in tumorigenesis
and metastasis. Biochimica et Biophysica Acta (BBA) - General Subjects. 2014 Apr
1;1840(4):1454–63.

38. Orrenius S, Gogvadze V, Zhivotovsky B. Mitochondrial Oxidative Stress: Implications for
Cell Death. Annual Review of Pharmacology and Toxicology. 2007;47(1):143–83.

107

39. Srikanth S, Woo JS, Wu B, El-Sherbiny YM, Leung J, Chupradit K, et al. The Ca2+ sensor
STIM1 regulates the type I interferon response by retaining the signaling adaptor STING at the
endoplasmic reticulum. Nat Immunol. 2019 Feb;20(2):152–62.

40. Akbari O, Stock P, Meyer EH, Freeman GJ, Sharpe AH, Umetsu DT, et al. ICOS/ICOSL
Interaction Is Required for CD4+ Invariant NKT Cell Function and Homeostatic Survival. The
Journal of Immunology. 2008 Apr 15;180(8):5448–56.

41. Rigas D, Lewis G, Aron JL, Wang B, Banie H, Sankaranarayanan I, et al. Type 2 innate
lymphoid cell suppression by regulatory T cells attenuates airway hyperreactivity and requires
inducible T-cell costimulator-inducible T-cell costimulator ligand interaction. J Allergy Clin
Immunol. 2017 May;139(5):1468-1477.e2.

42. Helou DG, Shafiei-Jahani P, Lo R, Howard E, Hurrell BP, Galle-Treger L, et al. PD-1 pathway
regulates ILC2 metabolism and PD-1 agonist treatment ameliorates airway hyperreactivity. Nat
Commun. 2020 Dec;11(1):3998.

43. Galle-Treger L, Suzuki Y, Patel N, Sankaranarayanan I, Aron JL, Maazi H, et al. Nicotinic
acetylcholine receptor agonist attenuates ILC2-dependent airway hyperreactivity. Nat Commun.
2016 Oct 18;7(1):13202.

44. Howard E, Hurrell BP, Helou DG, Quach C, Painter JD, Shafiei-Jahani P, et al. PD-1 Blockade
on Tumor Microenvironment-Resident ILC2s Promotes TNF-α Production and Restricts
Progression of Metastatic Melanoma. Front Immunol. 2021;12:733136.

45. Wallrapp A, Riesenfeld SJ, Burkett PR, Abdulnour REE, Nyman J, Dionne D, et al. The
neuropeptide NMU amplifies ILC2-driven allergic lung inflammation. Nature. 2017
Sep;549(7672):351–6.

108
Materials and Methods

Mouse experiments
Experimental protocols were approved by the USC institutional Animal Care and Use Committee
(IACUC) and conducted in accordance with the USC Department of Animal Resources’
guidelines. Age and sex matched mice were used in the studies. BALB/cByJ, and Rag2
-/-
γc
-/-
(C;129S4-Rag2tm1.1Flv Il2rgtm1.1Flv/J) mice were bred in the animal facility at the Keck School
of Medicine, University of Southern California (USC). Orai1-deficient mice have been previously
described(39). Orai2-deficient mice were generated using CRISPR/Cas9-mediated recombination.
Four sgRNAs targeting exon 2 (the first coding exon) of the mouse Orai2 gene were designed and
examined for recombination in situ, of which two sgRNAs were chosen for injection (PNABio
Inc.). The following sgRNA sequences were used: sgRNA1: 5’ gccgggttcagggcaggcagggg 3’ and
sgRNA2: 5’ gggcatggattaccgagactggg 3’. Purified sgRNAs together with Cas9 mRNA were
injected into day 0.5 single-cell embryos from Orai1
fl/fl
mice and transferred into the oviducts of
pseudopregnant recipient mice (Genome modification facility, University of Southern California).
The resulting pups were screened by PCR to identify those that had successful deletion of 43
nucleotides between the two sgRNA sequences and validated by sequencing (Figure S1). The
mice with desired deletion were bred to Tg(UBC-Cre/ERT2) mice (The Jackson Laboratory stock
no. 007001) to generate Orai1
fl/fl
; Orai2
-/-
; UBC-Cre/ERT2 mice.

In vivo experiments and tissue preparation
For induction of IL-33 stimulation, when indicated mice were challenged for 3 consecutive days
with carrier-free recombinant mouse (rm)IL-33 (0.5µg/mouse in 50µL, BioLegend) or PBS. On
day 4, lungs were collected and processed as previously described(40). Briefly, the lungs were
perfused with PBS and digested in Collagenase IV (400U/mL, MP Biomedicals, LLC) for 1 hour
at 37°C. The lungs were then stained with antibodies to identify ILC2s. ILC2s were gated as
lineage (CD3e, CD5, CD45R, Gr-1, CD11c, CD11b, Ter119, TCR γ δ, TCR β, CD335 and FC εRI α)
negative, CD45
+
, ST2
+
, CD127
+
cells.

Flow Cytometry

109
The following murine antibodies were used: anti-mouse lineage CD3e (145-2C11), CD5 (53-7.3),
TCR β (H57-597), CD45R (RA3-6B2), Gr-1 (RB6-8C5), CD11c (N418), CD11b (M1/70), Ter119
(TER-119), Fc εRI α (MAR-1), TCR γ δ (eBioGL3), CD56, CD335 (29A1.4), Streptavidin-FITC,
Streptavidin-APC, PE-Cy7 anti-mouse CD127 (A7R34), APCCy7 anti-mouse CD45 (30-F11),
PECy7 anti-mouse CD45 (30-F11), APCCy7 anti-mouse CD11c (N418), were purchased from
BioLegend. PE anti-mouse SiglecF (E50-2440) was purchased from BD Biosciences. PerCP-
eFluor710 anti-mouse ST2 (RMST2-2), eFluor450 anti-mouse CD11b (M1/70) were purchased
from ThermoFisher. Rabbit anti-mouse anti-human Orai1 and Orai2 antibodies were purchased
from Novus Biologicals. Intranuclear staining was performed using the Foxp3 Transcription
Factor Staining Kit (Thermofisher) per the manufacturer’s instructions. APC anti-mouse Ki67
(SolA15, Thermofisher) was used. Intracellular staining was performed using the BD Biosciences
Cytofix/Cytoperm kit. When indicated, cells were stimulated in vitro for 4 hours with 50µg/mL
PMA, 500µg/mL ionomycin (both Sigma) and 1µg/mL Golgi plug (BD Biosciences) before
cytokine assessment. APC anti-mouse IL-13 (85BRD, Thermofisher), PE anti-mouse IL-5
(TRFK5, BioLegend) were used. Live/dead fixable violet cell stain kit was used to exclude dead
cells (Thermofisher) and CountBright absolute counting beads (Thermofisher) to calculate
absolute cell numbers when indicated. For analysis of the mitochondria, cells were stained with 40
nM MitoTracker Green or Red (Life Technologies) for 20 minutes or MitoSox Red (Life
Technologies) for 10 minutes, respectively, at 37°C. For glucose uptake measurements, cells were
incubated in media containing 50μg/ml 2-NBDG (Thermo Fisher Scientific) for 20 min at 37°C
after surface antibody staining. To measure lipid droplet quantification, cells were incubated in
media containing 1000 ng/ml Bodipy or Bodipy FL C16 (Thermo Fisher Scientific) at 37°C for 30
min. Stained cells were analyzed on FACSCanto II and/or FACSARIA III systems and the data
were analyzed with FlowJo version 10 software.

Murine ILC2 and in vitro culture
Murine ILC2s were FACS-sorted to a purity of >95% on a FACSARIA III system. Isolated ILC2s
were cultured at 37°C (5-7x10
4
/mL) for 24 hours as indicated with N-{2,2,2-trichloro-1-
[(naphthalen-2-yl)amino]ethyl}furan-2-carboxamide (5D solubilized in DMSO, purchased from
Asinex) and rmIL-2 (10ng/mL), rmIL-7 (10ng/mL) survival cytokines. After 24 hours, +/- rmIL-
33 (10ng/ml) was added for an additional 48 hours in complete RPMI (cRPMI). For cRPMI, RPMI

110
(Gibco) was supplemented with 10% heat-inactivated FBS (Omega Scientific), 100 units/mL
penicillin and 100mg/mL streptomycin (GenClone). In the indicated experiments, N-Acetyl-L-
cysteine (NAC, 10µM, Sigma) was included in the culture.

Human ILC2 isolation and in vitro culture
Experimental protocols were approved by the USC Institutional Review Board (IRB) and
conducted in accordance with the principles of the Declaration of Helsinki. Human blood ILC2s
were isolated from total peripheral blood mononuclear cells (PBMCs) to a purity of > 95% on a
FACSARIA III system as described previously. Briefly, human fresh blood was first diluted 1:1
in PBS 1X and transferred to SepMateTM-50 separation tubes (STEMCELL Technologies) filled
with 12mL Lymphoprep™. Samples were centrifuged for 10 minutes and PBMCs were collected.
CRTH2
+
cells were then isolated using the CRTH2 MicroBead Kit, used according to the
manufacturer’s conditions. Samples were then stained and ILC2s were isolated based on the
absence of common lineage markers (CD3, CD5, CD14, CD16, CD19, CD20, CD56, CD235a,
CD1a, CD123), and the expression of CD45, CRTH2 and CD127. Isolated ILC2s were cultured at
37°C (5x10
4
/mL) with recombinant human (rh)IL-2 (10ng/mL) and rhIL-7 (10ng/mL) in cRPMi.

Culture measurements
Murine or human ILC2s were FACS-sorted and cultured in media containing survival cytokines
and 5D when indicated for 24-48 hours. The levels of cytokines present in culture supernatants
were measured by customized 7-panel mouse Legendplex kit or human cytokine ELISA kits
according to the manufacturer’s instructions (BioLegend). Supernatants were analyzed for L-
lactate levels using the Glycolysis Cell-Based Assay Kit (Cayman Chemicals). Total cellular ROS
was measured utilizing the Cellular Ros Assay kit (Red) from Abcam. NAD/NADH was measured
by NAD/NADH Assay Kit (Colorimetric) also from Abcam according to the manufacturer’s
protocol.

Metabolic flux analysis
The real-time extracellular acidification rate and oxygen consumption rate (OCR) were measured
with a Seahorse XF HS (Higher sensitivity) analyzer (Seahorse Bioscience, North Billerica, Mass).
Briefly, 50,000 activated lung ILC2s were plated in Seahorse medium supplemented with 1

111
mmol/L pyruvate, 2 mmol/L glutamine, and 10 mmol/L glucose. The Mito Stress Test Kit (Agilent
Technologies, San Diego Calif) with 1 mmol/L oligomycin, 4 mmol/L carbonyl cyanide-4-
(trifluoromethoxy)phenylhydrazone (FCCP), and 0.5 mmol/L rotenone (Rot) and antimycin A was
used, according to the manufacturer’s protocol. The Glycolysis Rate Assay with 0.5µM rotenone
(Rot) and antimycin A and 50mM 2-deoxy-D-glucose was used, according to the manufacturer’s
protocol.

Calcium flux analysis
WT ILC2s were FACS-sorted and cultured in media containing survival cytokines for 24 hours.
Cells were then loaded with 100µg/mL Fluo-8 AM for 30 min in calcium-free HBSS buffer. Cells
were then pretreated for 15 minutes with 1 mM thapsigargin to passively deplete Ca
2+
stores.
SOCE was measured by exchanging the Ca
2+
-free HBSS with that containing 2 mM CaCl2, with
the addition of 5D or the vehicle in the solution. Fluorescence was measured by flow cytometry.

Adoptive Transfer
Experiments were performed as described previously(41). Briefly, indicated populations of mouse
or human ILC2s were isolated as above and cultured for 24 hours with survival cytokines. 5.0 x
10
4
of the appropriate population of cells were transferred in PBS to Rag
-/-
γc
-/-
mice by tail
intravenous (i.v.). 24 hours later, 0.5 μg rmIL-33 i.n. in 50 μL was given once a day for 3 days. A
cohort was also given i.p. injections of 5D or vehicle. AHR was then measured on day 4.
Alternatively, 100 μg Alternaria Alternata extracts i.n. in 50 μl was given once a day for 4 days.
AHR was then measured on day 5.

Measure of Airway Hyperreactivity
Experiments were performed as described previously(42). Lung function was evaluated by direct
measurement of lung resistance using the FinePointe RC system (Buxco Research Systems,
Wilmington, NC) under general ketamine and xylazine anesthesia. AHR was measured by
exposure to an aerosol containing increasing doses of Methacholine (Sigma), following a baseline
measurement after the delivery of a PBS aerosol. Maximum lung resistance values were recorded
during a 3-minute period after each methacholine challenge.

112
BAL collection
BAL fluid was collected as previously described(43). The trachea was cannulated, the lungs
lavaged three times with 1.0 ml PBS and the collected fluid pooled. Eosinophils were gated as
CD45
+
CD11c
-
SiglecF
+
single cells.

RNA sequencing and data analysis
Transcriptomic analysis was performed as described previously(44). Briefly, RNA was extracted
from cultured cells using the MicroRNeasy kit, and cDNA was generated for library preparation
using 10ng of RNA. Samples were then amplified and sequenced on a NextSeq 500 system
(Illumina). Raw reads were aligned, normalized, and further analyzed with Partek Genomics Suite
software, version 7.0 (Partek, St Louis, Mo). Pathway analysis was performed by using Qiagen
Ingenuity Pathway Analysis software. Single-cell RNA sequencing data were downloaded and
reanalyzed as described previously(45).

Statistical analysis
Experiments were repeated at least three times (n=4-8 each) and data are shown as the
representative of >2 independent experiments. A student t-test for unpaired data was used for
comparisons between each group using Prism Software (GraphPad Software Inc.). Mean, Standard
Deviation and degree of significance was indicated as: *p<0.05, **p<0.01, ***p<0.001.

113
Figures
Figure 1
Murine pulmonary ILC2s express calcium release activated calcium channels Orai1 and
Orai2.
(A) Orai1, Orai2, Orai3 expression in FACs-sorted IL-33 activated pulmonary murine ILC2s by
bulk RNA-sequencing.
(B) Orai1 and Orai2 gene expression level in naïve and IL-33 activated ILC2s by single-cell RNA
sequencing. Bolded number represents the quantification of the number of expressing ILC2s (blue
dots) in each plot.
(C) Representative flow cytometry histogram of Orai1 expression on murine ILC2s with and
without IL-33 stimulation in vivo and corresponding quantification.
(D) Representative flow cytometry histogram of Orai2 expression on murine ILC2s with and
without IL-33 stimulation in vivo and corresponding quantification.

114
Error bars are the mean ± SEM. Student’s t-test, *p < 0.05, **p < 0.01, ***p < 0.001.
Figure 2
Pulmonary ILC2 effector function is dependent on Orai1 and Orai2.
(A) Block of endogenous SOCE in FACS-sorted pulmonary ILC2s after exposure to compound
5D measured by flow cytometry. Cells were pretreated with thapsigargin to deplete the
intracellular Ca
2+
stores before addition of Ca
2+
and 5D exposure.
(B) FACS-sorted pulmonary murine ILC2s were cultured for 24 hours with survival cytokines
rmIL-2, rmIL-7 and with rmIL-33 or PBS control for 24 hours. 5D or the vehicle control were then
added to the wells for an additional 48 hours.
(C) After 48 hours, cells were collected and Ki67 as a measure of proliferation was assessed and
quantified by flow cytometry.
(D) Supernatant was collected and secreted cytokines were measured by Legendplex (S2A).

115
(E) Representative flow cytometry plots of IL-5
+
and IL-13
+
ILC2s from ILC2s cultured with 5D
or the vehicle and corresponding quantitation.
(F) Pulmonary ILC2s were sorted from wild-type (WT), Orai1
-/-
and Orai1/2
-/-
mice and cultured
with and without tamoxifen for 48 hours. Cells were collected, counted, and cultured for an
additional 48 hours. Supernatant was collected and proinflammatory cytokines were measured by
Legendplex (S2B).
(G) ILC2s were collected and intracellular IL-5 and IL-13 production was measured and quantified
by flow cytometry (S2C).
Error bars are the mean ± SEM. Student’s t-test, *p < 0.05, **p < 0.01, ***p < 0.001. ILC2 cell
image provided with permission from Servier Medical Art.

Figure 3

116
Orai inhibition alters pro-inflammatory transcriptomic profile in pulmonary murine ILC2s.
(A) Volcano plot comparison of whole transcriptome gene expression of sorted ILC2s cultured
with 5D or the vehicle control. Differentially expressed genes (defined as statistically significant
adjusted p-value<0.05) with changes of at least 2.0 fold-change (FC) are shown in yellow
(upregulated genes) or blue (downregulated genes). Heat plot of selected differentially expressed
genes also included.
(B) Ingenuity Pathway Analysis (IPA) identifies pathways highly likely to be downregulated by
exposure to compound 5D. The -log p value is shown on the y-axis of the bar chart, and the fraction
of genes identified in the pathway that are differentially expressed between our conditions,
represented by the line graph right y-axis.
(C) Fold change of genes involved in ILC2 markers and activation in cells exposed to 5D as
compared to the control. Statistically significant differentially expressed genes are in red.
(D) Fold change of genes involved in ILC2 transcription factors in cells exposed to 5D as compared
to the control. Statistically significant differentially expressed genes are in red.
(E) Fold change of genes involved glycolysis in ILC2s exposed to 5D as compared to the control.
Statistically significant differentially expressed genes are in red.
Student’s t-test, *p < 0.05, **p < 0.01, ***p < 0.001.

117
Figure 4
Orai channels significantly affect metabolic function in pulmonary murine ILC2s.
(A) ILC2s were cultured with 5D or the vehicle. Lipid droplet uptake was measured by Bodipy FL
C16 staining and quantified.
(B) OCR was measured under basal conditions and in response to indicated drugs in 5D-treated or
untreated ILC2s.
(C) Basal respiration as a difference in OCR.
(D) Maximum respiration as a difference in OCR after FCCP treatment.
(E) Spare respiratory capacity presented as the difference in OCR after FCCP treatment and basal
respiration.
(F) Treated and untreated ILC2s were generated as previously described. Glucose uptake was
measured by 2-NBDG staining and quantified as 2-NBDG MFI.
(G) Enzymatic quantification of lactate accumulation in the supernatants of cultured cells from
(F).

118
(H) Glycolytic capacity in ILC2s cultured with 5D or the vehicle was measured in response to the
indicted drugs and quantified as glycoPER.
(I) Compensatory glycolysis ILC2s cultured with 5D or the vehicle was measured in response to
the indicted drugs.
Error bars are the mean ± SEM. Student’s t-test, *p < 0.05, **p < 0.01, ***p < 0.001.

Figure 5
Orai channels significantly downregulate mitochondrial function, affecting ILC2 effector
function and cytokine production.

119
(A) Schematic of the mitochondrial electron transport chain (mtETC) and the associated
complexes.
(B) Heatmap of statistically significant differentially expressed genes associated with the mtETC
in ILC2s cultured with 5D or the vehicle. Arrow colors coordinate with the associated complex in
(A).
(C) Flow cytometry histogram and quantification of mitochondrial mass measured by Mitotracker
Green in ILC2s cultured with 5D or the vehicle.
(D) Flow cytometry histogram and quantification of mitochondrial membrane potential measured
by Mitotracker Red in ILC2s cultured with 5D or the vehicle.
(E) Total cellular reactive oxygen species (ROS) in treated and untreated ILC2s measured by
microplate fluorescence.
(F) Flow cytometry histogram and quantification of mitochondrial ROS in ILC2s cultured with
5D or the vehicle as measured by Mitosox.
(G) NADH/NAD+ ratio quantification in ILC2s cultured with 5D or the vehicle measured by
microplate reader.
(H) ATP production presented as OCR measured by Seahorse Mitostress Test assay.
(I) Representative flow cytometry plots and quantification of intracellular IL-5 production in
ILC2s cultured with the vehicle, 5D, or 5D and the addition of antioxidant NAC.
(J) Representative flow cytometry plots and quantification of intracellular IL-13 production in
ILC2s cultured with the vehicle, 5D, or 5D and the addition of antioxidant NAC.
(K) Representative flow cytometry plots and quantification of proliferation in ILC2s cultured with
the vehicle, 5D, or 5D and the addition of antioxidant NAC.
Error bars are the mean ± SEM. Student’s t-test, *p < 0.05, **p < 0.01, ***p < 0.001.

120

Figure 6
Inhibition of Orai channels in pulmonary ILC2s significantly downregulates development of
airway inflammation.
(A) A cohort of BALB/cBYJ mice were challenged intranasally with Alternaria alternata (A.
alternata) for four consecutive days. Mice were also given injections of 5D or vehicle. On day 4,
AHR (B) was assessed. Additionally total number of eosinophils in the BAL (C), number of lung
ILC2s (D), and IL-5 producing ILC2s (E) is presented as mean numbers +/- SEM.
(F) Pulmonary ILC2s from WT control, Orai1
-/-
and Orai1/2
-/-
mice were sorted into separate
populations and deletion was induced as previously described. 5x10
4
ILC2s from the indicated
populations were intravenously injected to Rag
-/-
γc
-/-
host mice. 24 hours after transfer, mice were
challenged intranasally with Alternaria alternata for four consecutive days. On day 5, AHR (G)

121
was assessed. Additionally total number of eosinophils in the BAL (H) and total number of
pulmonary ILC2s (I) is presented as mean numbers +/- SEM.
Error bars are the mean ± SEM. Student’s t-test, *p < 0.05, **p < 0.01, ***p < 0.001. Mouse image
provided with permission from Servier Medical Art.

Figure 7
Inhibition of Orai channels ameliorates human ILC2-mediated AHR.
(A) FACS-sorted human blood ILC2s were cultured with rhIL-2, rhIL-7 and rhIL-33 for 24 hours.
5D or vehicle were added to the culture for an additional 48 hours.
(B) Flow cytometry histogram and quantification of Orai1 expression after 24 hours with survival
cytokines in (A).
(C) Flow cytometry histogram and quantification of Orai2 expression after 24 hours with survival
cytokines in (A).
(D) After 48 hours with 5D or vehicle, supernatant was collected and IL-5 and IL-13 levels were
measured by ELISA.

122
(E) FACS-sorted human blood ILC2s from healthy donors were adoptively transferred into Rag2
-
/-
γc
-/-
mice. Mice were intranasally challenged with rhIL-33 (1ug) or PBS and treated with i.p.
injections of 5D or vehicle days 0, 1, and 2. Mice were euthanized on day 3 and AHR (F) was
measured.
(G) Total number of eosinophils infiltrating the BAL.
(H) Total number of human ILC2s found in the lung.
Error bars are the mean ± SEM. Student’s t-test, *p < 0.05, **p < 0.01, ***p < 0.001. Mouse image
provided with permission from Servier Medical Art.

123
Chapter 5: Future Perspectives

As discussed in Chapter 1, ILC2s are implicated in a variety of diseases, the specific disease
determining whether the cells play a pathogenic or therapeutic role in the body. Understanding the
mechanisms by which ILC2s functionally operate allows for the potential for modulation in any
context necessary. While these chapters begin to identify important therapeutic targets, it is
inevitable that this research gives way to many potential future directions. Here I address naturally
occurring data-driven questions that arise from my work and how we propose to move forward.

Part 1: The regulatory role of ILC2
10
s in disease context

Several publications have described the physiological importance of IL-10 in asthma and
its function in the resolution of inflammation
1–3
. It is secreted by multiple cells, including but not
limited to monocytes, macrophages, mast cells, T and B lymphocytes, and dendritic cells
4–8
. In
line with this, our group and others have previously established the important immunoregulatory
function of secreted IL-10 from Treg cells in the lungs during asthmatic challenge
9
. Recently, we
and others have discovered an unlikely source of IL-10 from ILC2s in the lungs, and described
these cells as IL-10-producing ILC210s in both human and mice contexts
10–14
. In the OVA-
inflamed lungs, the proportion of ILC210s was remarkably comparable to that of regulatory T cells.
This therefore suggests that these cells have become important players in our understanding of the
complex mechanisms driving the pathophysiology of asthma. Of note, recently, Zhang et al.
identified transcription factors cMaf and Blimp1 as the primary proteins facilitating the
immunoregulatory IL-10 cytokine production in immune cells through combined temporal RNA
profiling and complex computational algorithms
15
. Many of the identified immune cells’ IL-10
responses rely on cMaf and Blimp-1, two transcription factors that we recently identified as crucial
for ILC2 IL-10 production. Therefore, scientific knowledge gained from these studies has the
potential to educate on the modulation of a vast number of IL-10 producing immune cells directly
in a variety of disease contexts.
The number of ILC210s mirroring the number of Tregs during OVA murine models of
asthma raises a number of questions regarding possible other similarities between the cell
populations. For example, inflammation occurs during an imbalance of the Treg: T effector cell

124
ratio. As this ratio decreases, inflammation in the body increases. This can transpire due to an
increase in the Treg cells through downregulated proliferation or a heightened level of cell death,
or it can occur through increased proliferation/recruitment of T effector cells or enhanced survival.
The two populations within a microenvironment also have a profound effect on the opposite
population and thereby influence the ratio. Recently, Dowling et al. detailed the suppressive effect
of Tregs on T effector cells as due to a decrease in “division destiny”, as opposed to direct
inducement of apoptosis or decreased proliferation
16
. Whether ILC210s have a similar effect on
proinflammatory ILC2 populations is unknown. Our published data demonstrating the ILC210-
derived IL-10’s ability to downregulate ILC2 effector function suggest the cells do have a direct
effect on the cells in vivo but what the IL-10 specifically effects to produce the downregulation of
proinflammation mechanistically within the cell is still to be answered. Further, it is probable that
the IL-10 produced by the ILC210s affects various cells throughout the lung microenvironment,
outside of the pathogenic ILC2s. How ILC210s specifically interact with the surrounding cells
requires further investigation.
Unlike Treg cells that express their identifying characteristic transcription factor FoxP3,
ILC210s are not a transcriptionally distinct population. Instead, they continue to express their ILC2
identifying transcription factor GATA3 while upregulating IL-10 production. In contrast, Wang et
al. demonstrated that IL-10 producing ILCs found in the intestines (termed ILCregs) did not
express traditional ILC1, ILC2 or ILC3 identifying markers
17
. They conclude these cells are a
completely distinct population, similar to Treg cells. However, as the ILC210s we observed in the
lungs were induced from a sorted, pure ILC2 population, it is clear that ILC210s require some
stimulation to begin producing IL-10. This induction of IL-10 from a pure population of cells under
stimulation has been demonstrated in a variety of immune cells, including B cells, Th17 cells,
macrophages, Th2 and Th1 cells
18–21
. What remains to be investigated is the commitment of these
inducible cells. Morita et al. demonstrated a similar IL-10 induction after ILC2 stimulation with
retinoic acid
13
. The group shows that the IL-10- population of ILC2s after stimulation do not
develop the ability to produce IL-10 after repeated stimulation
13
. We hypothesize therefore that
there is a distinct population of ILC2 cells primed with the ability to produce IL-10 under the
correct conditions. To test this, we will focus more closely on the chromosomal mechanisms by
which IL-10 is produced in some cells stimulated with IL-4 and those that do not. In a variety of
immune cells, reports have demonstrated that IL-10 production is stimulated by cMaf binding to

125
the MARE motif of the IL-10 promoter
22,23
. We have previously demonstrated the cMaf and
Blimp-1 play an important role in IL-10 production after IL-4 stimulation. Therefore, chromatin
immunoprecipitation (ChIP-seq) will reveal whether cMaf and Blimp-1 transcription factors
directly bind to the Il10 locus or a previously unrecognized motif crucial in ILC210 generation.
Furthermore, exploration into the 3D configuration of a gene of interest’s chromosome has
demonstrated that cis and trans chromosomal interactions will potently induce or repress
regulation of genes, as regulatory regions of the genome can be several kilobases away from the
gene they regulate. Transcription factors usually are responsible for anchoring and connecting the
separate genomic regions of interest. Park et al, for example demonstrated that the IL-6 stimulated
production of IL-21 in T cells is only possible because the transcription factor, Stat3, bound to
both the gene promoter and a previously unidentified enhancer located 49 kb upstream of the
gene
24
. Many cytokines are facilitated by the reorganization of chromosome 11, including IL-4,
IL-13, and IL-5
24
. Elucidating the chromatin landscape in ILC210 induction will generate a broader
picture of the genomic interactions and their role in cytokine production.

Part 2: Protective role of ILC2s in B16 melanoma

Our lab demonstrated that blocking checkpoint inhibitor PD-1 on pulmonary ILC2s in the
context of murine B16 melanoma stimulated the cells to begin producing high levels of cytotoxic
cytokine TNF- α. TNF- α in this context has the ability to directly induce apoptosis of the melanoma
cells in vitro, and in vivo the ILC2s delayed metastatic tumor progression. To investigate further
whether the TNF- α secreted by the ILC2 have this same effect in vivo, we will continue murine
models of B16 melanoma with and without the addition of anti-TNF- α intraperitoneal injections.
Similarly, we will sort pulmonary ILC2s from commercially available TNF- α knock-out mice and
adoptively transfer them into Rag2
-/-
γc
-/-
mice before the inoculation of the melanoma cells. These
two models will definitively determine whether the ILC2-derived TNF- α delays cancer
progression in vivo.
Additionally, we found in our model that knocking out PD-1 expression on ILC2s led to
an increased number of ILC2s found in the lungs during tumor metastasis, correlated with lower
tumor burden. While inhibiting PD-1 has been shown to increase the level of proliferation in

126
pulmonary ILC2 cells, it is also well known that tumor cells secrete various growth factors that
recruit various cells to the tumor site, including immunosuppressive Myeloid-Derived Suppressor
Cells (MDSCs)
25,26
. The recruitment of MDSCs from the bone marrow leads directly to tumor
growth. While ILC2s, unlike MDSCs, are thought to be primarily tissue resident cells that expand
and contract utilizing proliferation and cell death mechanisms, new research is highlighting the
importance of migration of ILC2s during adult lifetimes from both the bone marrow and between
tissues. These details are thoroughly outlined in Chapter 1, but it does lend the question of whether
the increase in pulmonary ILC2s found during the B16 melanoma model arises out of increased
proliferation on site or recruitment into the damaged tissue. We hypothesize that the increase in
ILC2 number is mainly the result of increased proliferation, primarily because the gating strategy
used to identify pulmonary ILC2s includes expression of ST2. While this is appropriate for the
lung ILC2s, ILC2s from other tissue sites such as the small intestine do not express high levels of
ST2. However, we can also measure expression levels of KLRG1, ICOS and CD25 to begin
addressing whether there are tissue migratory ILC2 populations in our model, as these tend to be
distinct between tissues. To measure proliferation, we can adoptively transfer fluorescently tagged
ILC2s into the B16 melanoma model. By observing the proliferation dye after euthanasia on day
14, we can begin to assess if PD-1
-/-
or WT ILC2 proliferate rapidly in this specific model in vivo.
One important concept that we explored in ILC2s in the context of airway inflammation
was metabolism and its effect on ILC2 effector function. An interesting area of research would be
to investigate the metabolism of ILC2s in this B16 melanoma context. Traditionally termed the
Warburg effect, tumor cells readily use glycolysis to quickly make ATP for their survival and
growth even when oxygen is available
27
. The result of this effect is two-fold: it enhances the
ability to produce large amounts of ATP for rapid and uncontrolled growth and effector function.
It also has the dual advantage of limiting potential anti-tumor immune responses. Traditional anti-
tumor defenses, such as NK cells and T effector cells also require glycolysis for their activation
28
.
Tumor cells therefore have the ability to outcompete surrounding immune cells for energy sources,
limiting the potential for any anti-tumor immune response. Our previous work with ILC2s in the
lungs has demonstrated that ILC2s, in contrast to T effector cells, primarily utilize fatty acid
oxidation and oxidative phosphorylation in the event of activation for their cytokine secretion
29,30
.
We hypothesize therefore that ILC2s may play a unique protective role in the context of melanoma
in that they may sustain the ability to pose an anti-tumor response where other immune cells lack

127
the resources to effectively respond. It must be noted, however, that our lab has also demonstrated
that knocking out PD-1 does indeed allow the ILC2s to switch to the glycolytic pathway during
allergic asthma
25
. It is known that ILC2s have the ability to compensate with different energy
pathways in situations with limited resources. We will therefore determine whether this is true for
ILC2s in the tumor. To test this hypothesis, we will begin by measuring the overall metabolic
preferences of ILC2s found in the tumor microenvironment, specifically with and without PD-1
engagement.
The experiments will be analyzed using our Seahorse XF HS (Higher Sensitivity) analyzer.
Briefly, WT ILC2s and PD-1
-/-
ILC2s sorted from the lungs after the melanoma model will be
settled in a plate while a series of inducers and inhibitors of specific metabolic pathways are added
to the samples. We plan on performing mainly 3 specific analyses: 1.) To measure ATP production,
we will run the ATP Real-Time rate assay, which quantifies the rate of ATP production from
glycolysis and mitochondria simultaneously using label-free technology in live cells. 2.) We will
then perform a Mito Stress test. By adding modulators of respiration into cell wells during the
assay, this will reveal the key parameters of mitochondrial function. The modulators included in
this assay are Oligomycin, Carbonyl cyanide-4 (trifluoromethoxy) phenylhydrazone (FCCP),
Rotenone, and Antimycin. This will allow us to measure the oxygen consumption (OCR), a
measurement of the rate the cell utilizes oxygen and will measure parameters such as basal
respiration, proton leak, maximal respiration, spare respiratory capacity, enabling a wide view of
mitochondrial function within the group of cells. 3.) We will finally run a Glycolysis Stress test to
directly measure glycolytic rates in the ILC2 populations. The modulators included in this assay
are Glucose, Oligomycin and glycolysis inhibitor 2-DG, as it will allow us to measure the actual
glycolytic capacity of the cells by recording modulation in ECAR as a result of lactate production.
Importantly, data will be normalized to the cell number. Simply, we will measure the capacity of
these cells to perform both glycolysis and oxidative phosphorylation on living cells in real time.

Part 3: Inhibiting calcium flux leading to attenuation of allergic asthma

In Chapter 4, we demonstrated the importance of calcium channels Orai1 and Orai2 on
ILC2 effector function and subsequent ILC2-dependent airway inflammation. As described above,
ILC2s play a central and rapid role in allergen-induced airway inflammation. The role of the Ca
2+
-

128
calcineurin-NFAT pathway is well defined in immune cells, but the mechanism on how Ca
2+
ions
enter cells has only recently begun to be understood. Our group and others have shown that CRAC
channels play an important role in immunity including in cells such as T cells and mast cells, with
their activation leading to allergic cytokines production
31–36
. Briefly, signaling molecule STIM1
was shown to detect endoplasmic reticulum (ER) Ca
2+
store depletion, leading to Orai channels
activation and slow increase of calcium in the ER
37
. The Ca
2+
influx leads to the activation of a
variety of transcription factors, including NF-kB and NFAT that subsequently allow for the
enhanced expression of genes. How this works specifically to control the production of IL-5, IL-
13 and proliferation in pulmonary ILC2s is unknown and requires further investigation. We
demonstrate that the mitochondria are highly regulated by Ca
2+
and this controls the cell’s effector
function, whether directly or indirectly. How this is modulated is also unknown. It has been
demonstrated in T cells that NFAT can interact with transcription factors GATA3 and MAF
38
.
GATA3 is well known in ILC2s to control development and certain effector functions.
Additionally, cMaf has been identified by our lab as being essential for IL-10 production,
suggesting NFAT activation may have an effect on the regulatory phenotype of ILC2s as well. We
will therefore test whether 5D after IL-4 stimulation blocks the cytokine production of IL-10 and
generation of ILC210s.
In our published data, we used Rag2
-/-
mice that are deficient in all adaptive B and T cells.
Since it is possible that 5D affect other immune cells, we will compare the results utilizing WT
mice and Rag2 deficient mice, and will closely monitor other immune cells including T cell
subsets, macrophages, and dendritic cells. By comparing the results between experimental groups
Rag2
-/-
and WT mice, we will be able to assess the effect of the Orai blocker in various immune
cells, including dendritic cells, CD8 and CD4 T cells, and macrophages.
Our lab also demonstrated that blocking the Orai channels significantly altered the
metabolism and mitochondrial function of the ILC2s. Kaufmann et al. showed a similar alteration
in Th17 cells after Oria inhibition, as well as mutated morphology of the mitochondrial structure.
We hypothesize that ILC2 mitochondria structure and function will similarly be affected. In order
to further define the mitochondrial function in relation to Orai calcium channels in ILC2s, we
propose to use a series of advanced microscopy techniques. Recent data have demonstrated the
shape of T cell mitochondria is dependent on the metabolic pathway being preferentially used by
the immune cell
39
. Higher mitochondrial activity suggests the cells use oxidative phosphorylation,

129
as opposed to glycolysis, and the mitochondria appear tubular in morphology. Conversely,
punctate morphology is seen with a higher level of glycolytic activity. Briefly, we will culture
freshly isolated naive ILC2s with 5D or vehicle and activate them with rmIL-33. Following 48
hours of culture, ILC2s will then be stained with Mitotracker Green (1uM) for 30 minutes at 37C
degrees and mounted on Superfrost microscopy slide using a cytospin in DAPI-containing
mounting medium. Cells will be visualized and analyzed by advanced airy scan confocal
microscopy
40
.
It has been suggested that cristae remodeling could alter the electron transport chain (ETC)
efficiency and redox balance, ultimately controlling metabolic adaptations in immune cells. To
test this hypothesis, we will assess cristae morphology in following 5D treatment by scanning
electronic microscopy (SEM). Cells will be processed for microscopy following a previously
published protocol
41
. Briefly, we will culture freshly isolated naive ILC2s with 5D or vehicle and
activate them with rmIL-33. Following 48 hours of culture, ILC2s will then be washed, fixed,
dehydrated in graded series of chilled ethanol and raising embedded to achieve optimal signal on
the microscope. 60nm ultrathin sections will then be cut and stained with uranyl acetate and lead
citrate using standard methods
42
.
Approaches that manipulate cellular metabolism can have profound effects on cell function
and disease development, as many mitochondrial complex inhibitors are in clinical trials for
diseases such as autoimmunity and cancer. As this fascinating area develops, novel and unexpected
therapeutic strategies arise to control metabolism and disease.

Together, these future directions will offer mechanisms that modulate ILC2 phenotypes,
trafficking, energy sources and activation. Although the last 10 years have revealed that ILC2s
play an important role in a broad range of immune-mediated disorders, a major gap exists between
understanding ILC2 biology and modulating these cells for therapeutic application. Our studies
and the proposed future directions will provide the means to help specifically design tools with the
capability to modulate ILC2 effector functions in a variety of disease contexts. The results of our
studies will therefore have potential implications in other diseases involving ILC2s, including
allergic diseases such as atopic dermatitis, allergic rhinitis and inflammatory bowel disease.
Specifically, we visualize the following ways these data have the potential to offer therapeutic

130
applications:
The role for ILC210s in suppressing unregulated inflammation: Identification of
transcription factors Blimp-1 and cMaf as being important for ILC210 IL-10 production offers a
wealth of innovative potential. Overexpression of the transcription factors, for example, may
upregulate IL-10 production to help regulate the local microenvironment. Continuous exposure to
high doses of global IL-10 dangerously predisposes patients to a variety of negative health
outcomes, including viral infections, the onset of chronic autoimmune disorders and the
development of cancer. The identification of cMaf and Blimp-1 allows us to characterize the
mechanisms regulating local, cell-specific IL-10-dependent inflammation in order to design
adapted therapeutics that will dramatically reduce side effects associated with global IL-10
therapy. Here, we have developed an ex vivo method of inducing ILC210 cells utilizing cytokine
stimulation. Transfusion of ILC210 cells may someday offer focused therapies for ILC2-dependent
diseases, including allergic asthma. Further, several receptors have been shown in the literature to
induce IL-10 production from ILC2s. The emerging picture is that the γc family likely stimulates
the signaling pathway necessary for IL-10 production. While there are currently no ILC2-specific
antibodies, should one become available in the future, a bispecific antibody targeting ILC2s and
their γ chain would be the most promising option for ILC2-targeted IL-10 induction.
The role for ILC2s’ anti-tumor activity after PD-1 inhibition: PD-1/PD-L1 blocking
antibodies have introduced a revolutionary cancer-treatment strategy. Unfortunately, one common
problem that has arisen from the treatment is an eventual resistance to the global PD-1 antibody,
resulting in renewed cancer growth. This occurs for a multitude of reasons, but one explanation is
the population of PD-1 expressing immunosuppressive cells, such as T regulatory (Treg) or
myeloid-derived suppressor cells (MDSC), also activate and expand in response to the PD-1
blocking antibody, effectively suppressing any anti-tumor activity potentially gained. If there were
an ILC2-specific PD-1 blocking antibody, however, or a method to block activity of PD-1 on
ILC2s, this suppressive expansion could be circumvented. Instead ILC2s would have the ability to
launch a local anti-tumor attack leading to an upregulation of secreted TNF- α with the ability to
induce apoptosis of the cancer cells. The ability of this to happen innately, even before T cell
activity, would mean the ILC2s could target the cancer cells immediately after metastatic
infiltration. Further, Liu et al. demonstrates that cMaf activity is partially responsible for resistance
to anti-PD-1 therapy
43
. Their studies show that inhibiting cMaf blocks tumor M2 macrophages

131
from projecting an immunosuppressive microenvironment, effectively allowing T cells to enhance
antitumor activity. Based on our studies here, we are now aware that cMaf also controls ILC210
activity and has the potential to contribute to a localized immunosuppressive environment.
Blocking cMaf activation specifically in ILC210s therefore may enhance antitumor activity by PD-
1 inhibited pulmonary ILC2s.
Nivolumab, an FDA-approved anti-PD-1 immunotherapy, has been used therapeutically
for the last ten years, while cyclosporine A has been used to indirectly influence Orai channels for
over twenty years. Moreover, cMaf and Blimp-1 as therapeutic targets are gaining an increasing
amount of popularity. These data provide valuable knowledge of novel mechanisms working in
ILC2s to influence previously unknown contributors to various immune related diseases. Insights
from these studies help to inform mechanisms behind various side effects, benefits, and off targets
affected by these conventionally used therapies. They also help to offer additional, potentially
more directed therapeutic targets in the world of drug development.

132
References

1. Couper, K. N., Blount, D. G. & Riley, E. M. IL-10: The Master Regulator of Immunity to
Infection. The Journal of Immunology 180, 5771–5777 (2008).

2.Kessler, B. et al. Interleukin 10 inhibits pro-inflammatory cytokine responses and killing of
Burkholderia pseudomallei. Sci Rep 7, 42791 (2017).

3.Lobo-Silva, D., Carriche, G. M., Castro, A. G., Roque, S. & Saraiva, M. Balancing the immune
response in the brain: IL-10 and its regulation. Journal of Neuroinflammation 13, 297 (2016).

4.Prasse, A. et al. IL-10–producing monocytes differentiate to alternatively activated macrophages
and are increased in atopic patients. Journal of Allergy and Clinical Immunology 119, 464–471
(2007).

5.Chung, E. Y. et al. Interleukin-10 Expression in Macrophages during Phagocytosis of Apoptotic
Cells Is Mediated by the TALE homeoproteins Pbx-1 and Prep-1. Immunity 27, 952–964 (2007).

6.Grimbaldeston, M. A., Nakae, S., Kalesnikoff, J., Tsai, M. & Galli, S. J. Mast cell–derived
interleukin 10 limits skin pathology in contact dermatitis and chronic irradiation with ultraviolet
B. Nat Immunol 8, 1095–1104 (2007).

7.Trinchieri, G. Interleukin-10 production by effector T cells: Th1 cells show self control. J Exp
Med 204, 239–243 (2007).

8.Llopiz, D. et al. IL-10 expression defines an immunosuppressive dendritic cell population
induced by antitumor therapeutic vaccination. Oncotarget 8, 2659–2671 (2016).

9.Rigas, D. et al. Type 2 innate lymphoid cell suppression by regulatory T cells attenuates airway
hyperreactivity and requires inducible T-cell costimulator-inducible T-cell costimulator ligand
interaction. J Allergy Clin Immunol 139, 1468-1477.e2 (2017).

10.Howard, E. et al. IL-10 production by ILC2s requires Blimp-1 and cMaf, modulates cellular
metabolism, and ameliorates airway hyperreactivity. J Allergy Clin Immunol 147, 1281-1295.e5
(2021).

11.Bando, J. K. et al. ILC2s are the predominant source of intestinal ILC-derived IL-10. J Exp
Med 217, e20191520 (2020).

12.Cao, Q. et al. Regulatory innate lymphoid cells suppress innate immunity and reduce renal
ischemia/reperfusion injury. Kidney International 97, 130–142 (2020).

13.Morita, H. et al. Induction of human regulatory innate lymphoid cells from group 2 innate
lymphoid cells by retinoic acid. Journal of Allergy and Clinical Immunology 143, 2190-2201.e9
(2019).

133
14.Seehus, C. R. et al. Alternative activation generates IL-10 producing type 2 innate lymphoid
cells. Nat Commun 8, 1900 (2017).

15.Zhang, H. et al. An IL-27-Driven Transcriptional Network Identifies Regulators of IL-10
Expression across T Helper Cell Subsets. Cell Reports 33, 108433 (2020).

16.Dowling, M. R. et al. Regulatory T Cells Suppress Effector T Cell Proliferation by Limiting
Division Destiny. Front Immunol 9, 2461 (2018).

17.Wang, S. et al. Regulatory Innate Lymphoid Cells Control Innate Intestinal Inflammation. Cell
171, 201-216.e18 (2017).

18.Zhu, B. et al. IL-4 and retinoic acid synergistically induce regulatory dendritic cells expressing
Aldh1a2. J Immunol 191, 3139–3151 (2013).

19.Neumann, C. et al. Role of Blimp-1 in programing Th effector cells into IL-10 producers.
Journal of Experimental Medicine 211, 1807–1819 (2014).

20.Xu, J. et al. c-Maf regulates IL-10 expression during Th17 polarization. J Immunol 182, 6226–
6236 (2009).

21.Mitchell, R. E. et al. IL-4 enhances IL-10 production in Th1 cells: implications for Th1 and
Th2 regulation. Sci Rep 7, 11315 (2017).

22.Mitchell, R. E. et al. IL-4 enhances IL-10 production in Th1 cells: implications for Th1 and
Th2 regulation. Scientific Reports 7, 11315 (2017).

23.Xu, J. et al. c-Maf Regulates IL-10 Expression during Th17 Polarization. J Immunol 182, 6226–
6236 (2009).

24.Park, J.-H. et al. Dynamic Long-Range Chromatin Interaction Controls Expression of IL-21 in
CD4+ T Cells. The Journal of Immunology 196, 4378–4389 (2016).

25.Helou, D. G. et al. PD-1 pathway regulates ILC2 metabolism and PD-1 agonist treatment
ameliorates airway hyperreactivity. Nat Commun 11, 3998 (2020).

26.Law, A. M. K., Valdes-Mora, F. & Gallego-Ortega, D. Myeloid-Derived Suppressor Cells as a
Therapeutic Target for Cancer. Cells 9, 561 (2020).

27.Liberti, M. V. & Locasale, J. W. The Warburg Effect: How Does it Benefit Cancer Cells?
Trends Biochem Sci 41, 211–218 (2016).

28.Chang, C.-H. et al. Metabolic Competition in the Tumor Microenvironment Is a Driver of
Cancer Progression. Cell 162, 1229–1241 (2015).

134
29.Galle-Treger, L. et al. Autophagy is critical for group 2 innate lymphoid cell metabolic
homeostasis and effector function. Journal of Allergy and Clinical Immunology 145, 502-517.e5
(2020).

30.Wilhelm, C. et al. Critical role of fatty acid metabolism in ILC2-mediated barrier protection
during malnutrition and helminth infection. J Exp Med 213, 1409–1418 (2016).

31.Vercelli, D. Discovering susceptibility genes for asthma and allergy. Nat Rev Immunol 8, 169–
182 (2008).

32.Holgate, S. T. & Polosa, R. Treatment strategies for allergy and asthma. Nat Rev Immunol 8,
218–230 (2008).

33.Barnes, P. J. Immunology of asthma and chronic obstructive pulmonary disease. Nat Rev
Immunol 8, 183–192 (2008).

34.Vig, M. & Kinet, J.-P. The long and arduous road to CRAC. Cell Calcium 42, 157–162 (2007).

35.Vig, M. & Kinet, J.-P. Calcium signaling in immune cells. Nat Immunol 10, 21–27 (2009).

36.Di Capite, J. L., Bates, G. J. & Parekh, A. B. Mast cell CRAC channel as a novel therapeutic
target in allergy. Curr Opin Allergy Clin Immunol 11, 33–38 (2011).

37.Srikanth, S. & Gwack, Y. Orai1-NFAT Signalling Pathway Triggered by T Cell Receptor
Stimulation. Mol Cells 35, 182–194 (2013).

38.Vaeth, M., Kahlfuss, S. & Feske, S. CRAC Channels and Calcium Signaling in T Cell-Mediated
Immunity. Trends in Immunology 41, 878–901 (2020).

39.Prasad, M. et al. Themis regulates metabolic signaling and effector functions in CD4+ T cells
by controlling NFAT nuclear translocation. Cell Mol Immunol 18, 2249–2261 (2021).

40.Srikanth, S., Kim, K.-D. & Gwack, Y. Methods to measure cytoplasmic and mitochondrial
Ca(2+) concentration using Ca(2+)-sensitive dyes. Methods Enzymol 543, 1–20 (2014).

41.Kaufmann, U. et al. Calcium Signaling Controls Pathogenic Th17 Cell-Mediated Inflammation
by Regulating Mitochondrial Function. Cell Metabolism 29, 1104-1118.e6 (2019).

42.Hurrell, B. P. et al. Distinct Roles of LFA-1 and ICAM-1 on ILC2s Control Lung Infiltration,
Effector Functions, and Development of Airway Hyperreactivity. Front. Immunol. 11, (2020).

43. Liu, Min et al. Transcription factor c-Maf is a checkpoint that programs macrophages in lung
cancer. The Journal of clinical investigation vol. 130, 2081-2096, (2020).

Abstract (if available)

Abstract ILC2s are the dominant innate lymphoid cell population in the lungs at steady state and their release of type-2 cytokines upon activation play a crucial role in the maintenance and homeostasis of body health. First described in 2001 as a population of non-B and non-T cells that produce cytokines IL-5 and IL-13, the improper activation or suppression of ILC2s has been implicated as a driving force behind the development and exacerbation of a variety of diseases throughout the human body. There are currently no treatments specifically targeting these cells, though they are increasingly identified as being responsible for the underlying pathological mechanisms behind diverse diseases. Understanding their biology and behavior therefore is essential for the development of ILC2-based therapeutic strategies.
Allergic asthma is an inflammatory disease of the upper airway that results from the inappropriate immune response to environmental allergens. In the context of this disease, ILC2s are the first responders to the subsequent secreted epithelial alarmins. ILC2s’ responding cytokine secretion is a central driver in responding eosinophil infiltration, increased airway hyperreactivity and associated lung tissue injury. In this study, we identified a subset of ILC2s (ILC210s) that actively produce and secrete IL-10, an anti-inflammatory cytokine with the ability to ameliorate allergic lung inflammation. We further identified key molecular and transcriptional requirements may be required for the induction of IL-10, with the potential for targeted modulation. Transcription factors Blimp-1 and cMaf are crucial for the induction of IL-10 production. Furthermore, glycolytic metabolism appears to be essential for the development of ILC210s, in contrast to the traditionally utilized fatty acid oxidation pathway that is important for the conventional pro-inflammatory phenotype. We address specific transcriptional and metabolic requirements for the modulation of pathogenic ILC2s with the intention of targeted conversion to ILC210s with the ability to regulate airway hyperreactivity.
In contrast to the uncontrolled ILC2 activation found in allergic asthma, we also studied the improper suppression of proinflammatory effector function in the context of metastatic melanoma. The role of ILC2s in cancer is highly debated and appears to be dependent on cancer location and type. We found that blocking PD-1 on pulmonary ILC2s inhibited the growth of metastatic melanoma through the enhancement of cytokine secretion TNF-a, intensifying the pro-inflammatory phenotype of ILC2s and subsequent anti-tumor activity.
Finally, we further addressed the urgent need of asthma patients by taking advantage of the recent breakthroughs in the field of Ca2+ signaling to benefit asthma therapy. We identified Orai1 and Orai2, long-sought pore component of calcium release activated calcium (CRAC) channels. We found that blocking the channels with the drug 5D inhibited the development of airway inflammation in both murine and humanized models. Moreover, inhibition of these channels severely affected metabolism and mitochondrial function, leading to the decrease in cytokine production and proliferation in ILC2s. This information will have a broad impact on basic understanding of the function of Ca2+ signaling in allergic ILC2s, and the development of novel therapy to treat severe forms of allergic and atopic inflammatory diseases
We believe that the results obtained from this study will provide insights into an important and understudied role of ILC2 modulation in diseases. Importantly, this dissertation identifies novel targets and knowledge related to the metabolism, development, and modulation of pulmonary ILC2 cells, offering new avenues for the development of therapeutic strategies in the treatment of ILC2-associated diseases.

Linked assets

University of Southern California Dissertations and Theses

Conceptually similar

PDF

Investigation of the role of ICOS in the regulation of ILC2 dependent airway hyperreactivity

PDF

IFN-α treatment inhibits induction of type-2 innate lymphoid cell dependent airway hyperresponsiveness

PDF

FoxP3⁺ regulatory T cells inhibit ILC2 dependent AHR

PDF

PD-1 is a regulator of ILC2-mediated airway hyperresponsiveness and inflammation

PDF

Immune regulation of ILC2 dependent asthma by sodium butyrate

PDF

Impact of cannabinoid receptor 2 stimulation on adipose tissue-resident ILC2s and its potential in alleviating obesity-induced type 2 diabetes mellitus

PDF

A novel therapeutic approach in asthma: depleting CD52-expressing leukocytes suppresses airway hyperreactivity and ameliorates lung inflammation

PDF

Airway inflammation and respiratory health in the Southern California children's health study

PDF

Common immune-related factors and risk of non-Hodgkin lymphomy

PDF

Quantity versus quality: how adipose tissue accumulation and immune cell profile associate with risk for type 2 diabetes in minority children and adults

PDF

Yeast-based nanobody identification for MCEMP1: mast-cell expressed membrane protein 1

PDF

The effect of inhibiting the NFkB pathway on myeloid derived suppressor cell cytokine production

PDF

Therapeutic potential of rhesus theta defensin-1 in the treatment of inflammatory lung diseases

PDF

Association of subclinical atherosclerosis with plasma B-vitamin, cysteine, homocysteine, and cysteinyl glycine in a cardiovascular disease-free population

PDF

Immunomodulatory and regenerative potential of amniotic fluid stem cells as a treatment strategy for pulmonary fibrosis

PDF

Ectopic fat and adipose tissue inflammation in overweight and obese African Americans and Hispanics

PDF

The essential role of histone H2A deubiquitinase MYSM1 in natural killer cell maturation and HSC homeostasis

PDF

The effects of hormone therapy on carotid artery intima-media thickness and serum lipids by ApoE4 genotype

PDF

Prenatal air pollution exposure, newborn DNA methylation, and childhood respiratory health

PDF

Stable expression of human B7-H4 in a mouse mammary tumor model as a target for cancer immunotherapy

Asset Metadata

Creator Howard, Emily (author)

Core Title Investigating the modulatory mechanisms of ILC2s as a therapeutic strategy in the context of health and disease

School Keck School of Medicine

Degree Doctor of Philosophy

Degree Program Infectious Diseases, Immunology and Pathogenesis

Degree Conferral Date 2022-12

Publication Date 04/25/2023

Defense Date 10/13/2022

Publisher University of Southern California (original), University of Southern California. Libraries (digital)

Tag allergic asthma,calcium,flow cytometry,health and disease,ILC2s,immune cell,immunology,lung inflammation,Melanoma,OAI-PMH Harvest

Format theses (aat)

Language English

Contributor Electronically uploaded by the author (provenance)

Advisor Machida, Kiego (committee chair), Akbari, Omid (committee member), Allayee, Hooman (committee member), Epstein, Alan (committee member), Golden, Lucy (committee member)

Creator Email eericson@usc.edu,emilyhoward745@gmail.com

Permanent Link (DOI) https://doi.org/10.25549/usctheses-oUC112195795

Unique identifier UC112195795

Identifier etd-HowardEmil-11286.pdf (filename)

Legacy Identifier etd-HowardEmil-11286

Document Type Dissertation

Format theses (aat)

Rights Howard, Emily

Internet Media Type application/pdf

Type texts

Source 20221026-usctheses-batch-988 (batch), University of Southern California (contributing entity), University of Southern California Dissertations and Theses (collection)

Access Conditions The author retains rights to his/her dissertation, thesis or other graduate work according to U.S. copyright law. Electronic access is being provided by the USC Libraries in agreement with the author, as the original true and official version of the work, but does not grant the reader permission to use the work if the desired use is covered by copyright. It is the author, as rights holder, who must provide use permission if such use is covered by copyright. The original signature page accompanying the original submission of the work to the USC Libraries is retained by the USC Libraries and a copy of it may be obtained by authorized requesters contacting the repository e-mail address given.

Repository Name University of Southern California Digital Library

Repository Location USC Digital Library, University of Southern California, University Park Campus MC 2810, 3434 South Grand Avenue, 2nd Floor, Los Angeles, California 90089-2810, USA

Repository Email cisadmin@lib.usc.edu