University of Southern California Dissertations and Theses

Development of novel immunosuppressant-based therapies to treat dacryoadenitis in a Sjögren’s syndrome mouse model

(USC Thesis Other)

Development of novel immunosuppressant-based therapies to treat dacryoadenitis in a Sjögren’s syndrome mouse model

PDF

Download a page range

Download transcript

Copy asset link

Request this asset

Transcript (if available)

Content DEVELOPMENT OF NOVEL IMMUNOSUPPRESSANT-BASED THERAPIES TO TREAT
DACRYOADENITIS IN A SJÖGREN’S SYNDROME MOUSE MODEL
by
Yaping Ju
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
(PHARMACEUTICAL SCIENCES)
December 2020
Copyright 2020 Yaping Ju
ii

Acknowledgements
First of all, I would like to thank my advisor, Dr. Sarah Hamm-Alvarez, for all the guidance,
support, encouragement and patience that you have given me during the past five and half years. I
have learned a great deal about both scientific research and teamwork as a result of the time spent
in the lab. These are experiences that I truly love and will always be grateful for.
Thank you to my other committee members, Dr. John Andrew MacKay and Dr. Curtis
Okamoto for the guidance on my research projects.
I am also extremely grateful to all the previous and current lab members, Frances Yarber,
Dr. Maria Edman, Srikanth Reddy Janga, Dr. Zhen Meng, Dr. Mihir Shah, Dr. Wannita Klinngam,
Dr, Zhe Li, Dr. Santosh Peddi, Dr. Changrim Lee, Dr. Mincheol Park, Christina Runzhong Fu,
Hao Guo, Minchang Choi, and Shruti Kakan for all their generous help.
I would also like to thank Dr. Junji Watanabe, Dr. Shuxing Li, and Dr. Jason Junge from
University of Southern California for all the help with the project.
Finally, I would like to thank my family. Thanks for my parents and my parents-in-law for
all the generous support. Thanks for my amazing husband Mr. Jong Hak Lee for all the love and
care.

iii

Table of Contents
Acknowledgements ......................................................................................................................... ii
List of Tables ................................................................................................................................. iv
List of Figures ..................................................................................................................................v
Abstract ........................................................................................................................................ viii
Introduction ......................................................................................................................................1
Chapter 1: Comparison of NOD versus NOR Mice as Mouse Models of
Autoimmune Dacryoadenitis ...........................................................................................................6
1.1 Introduction ................................................................................................................................6
1.2 Materials and Methods ...........................................................................................................8
1.3 Results ..................................................................................................................................13
1.4 Conclusion and Discussion ..................................................................................................25
Chapter 2: Development of a Molecularly Targeted Carrier of
Immunosuppressants Using Elastin-Like Polypeptide ..................................................................30
2.1 Introduction ..........................................................................................................................30
2.2 Material and Methods ...........................................................................................................33
2.3 Results ..................................................................................................................................39
2.4 Conclusion and Discussion ..................................................................................................59
Chapter 3: Intralacrimal Sustained Delivery of Rapa Shows Therapeutic Effects
Without Systemic Toxicity in NOD Mice .....................................................................................63
3.1 Introduction ..........................................................................................................................63
3.2 Material and Methods ...........................................................................................................66
3.3 Results ..................................................................................................................................74
3.4 Conclusion and Discussion ..................................................................................................88
Conclusion .....................................................................................................................................92
References ......................................................................................................................................93

iv

List of Tables
Chapter 2.
Table 1. Biophysical properties of ELP fusion proteins evaluated in this study ..........................45
Table 2. Thermodynamic parameters of ELP-FKBP interactions with Rapa ................................47
Table 3. Pharmacokinetic parameters of AF and IBPAF in NOD mice following
systemic administration .................................................................................................................49
Chapter 3.
Table 1. Phase transition characterization and Rapa binding dynamics of 5FA and
5FV ................................................................................................................................................76

v

List of Figures
Chapter 1.
Figure 1. Blood glucose levels in male BALB/c, NOD and NOR mice ........................................14
Figure 2. Basal and Stimulated Tear Production in male BALB/c, NOD and NOR
mice ................................................................................................................................................15
Figure 3. LG lymphocytic Infiltration in BALB/c, NOD and NOR mice ....................................17
Figure 4. CTSS activity in stimulated tears and its expression and distribution
levels in LG of male BALB/c, NOD and NOR mice ....................................................................20
Figure 5. Expression of inflammation-related genes in LG of NOD and NOR
mouse relative to age-matched BALB/c mouse LG ......................................................................21
Figure 6. Relative distributions of Rab3D in LG of male BALB/c, NOD and NOR
mice at 12 and 20 weeks of age .....................................................................................................22
Figure 7. SMA-enriched MEC distribution in NOD and NOR mouse LG relative
to age-matched BALB/c mouse LG ...............................................................................................24
Chapter 2.
Figure 1. ICAM-1 is increased in the inflamed LG of male NOD mice and can be
targeted by IBP ..............................................................................................................................42
Figure 2. AF and IBPAF have similar purity, molecular weight, hydrodynamic
radius, phase behavior, Rapa binding affinity and Rapa release profiles ......................................46
Figure 3. Cytotoxicity and biocompatibility of AF and IBPAF ....................................................48
vi

Figure 4. IBPAF binds efficiently to TNF-α stimulated bENd.3 cells relative to
AF ..................................................................................................................................................51
Figure 5. AF and IBPAF have similar pharmacokinetic profiles following
systemic administration .................................................................................................................54
Figure 6. Relative to AF, IBPAF accumulation is transiently enhanced in LG of
male NOD mice .............................................................................................................................56
Figure 7. Optical-imaging evaluation of biodistribution of ELP-FKBPs in NOD
mice after I.V. injection .................................................................................................................57
Figure 8. IBPAF shows greater accumulation relative to AF in lymphocytic
infiltration zones within the NOD LG ...........................................................................................59
Chapter 3.
Figure 1. Schematic illustration of intra-LG injection of 5FV-Rapa relative to
subcutaneous administration of a soluble Rapa carrier, FAF ........................................................66
Figure 2. Schematic illustration of the process of intraLG gland injection. ..................................70
Figure 3. Biophysical characterization of 5FA and 5FV ...............................................................75
Figure 4. Intra-LG injection does not promote tissue damage or fibrosis ....................................78
Figure 5. Lightsheet imaging of Rhodamine-labeled 5FV-Rapa following intra-
LG administration reveals formation of a depot ............................................................................79
Figure 6. Confocal fluorescence microscopy imaging of Rhodamine-labeled
5FV-Rapa injected intra-LG reveals a depot formation ................................................................80
vii

Figure 7. The 5FV-Rapa depot is retained in the LG for significantly longer than
its soluble control formulation (5FA-Rapa) ...................................................................................82
Figure 8. The surgery cut healed completely after 2weeks ...........................................................83
Figure 9. Intra-LG 5FV-Rapa significantly enhances basal and stimulated tear
secretion .........................................................................................................................................84
Figure 10. Both 5FV-Rapa and FAF-Rapa reduced lymphocytic infiltration ..............................85
Figure 11. Flow cytometry analysis of lymphocyte composition in the LG after
treatments .......................................................................................................................................86
Figure 12. 5FV-Rapa has significantly less systemic toxicity than FAF-Rapa ............................88

viii

Abstract
Sjögren’s syndrome (SS) is a complex autoimmune rheumatic disease that is characterized
by lymphocytic infiltration and inflammatory destruction of salivary glands and lacrimal glands.
The current therapies for SS-related dry eye all have significant disadvantages, thus novel therapies
that treat the underlying inflammation of SS while minimizing adverse effects are urgently needed.
Rapamycin (Rapa), is a potent immunomodulatory agent with undesirable pharmacokinetics and
systemic side effects. Using Elastin-like polypeptides (ELPs), which are thermo-responsive
protein polymers, we designed various drug carrier for Rapa that improves its pharmacokinetics
properties, enhances its therapeutic effects, and reduces its systemic toxicity. This work provides
a new future potential therapy for SS-related dacryoadenitis as well as possible treatment for other
autoimmune diseases.
1

Introduction
Sjögren’s syndrome (SS) is a complex autoimmune rheumatic disease that specifically
targets salivary glands (SG) and lacrimal glands (LG). The major pathological features are
lymphocytic infiltration and inflammatory destruction of SG and LG (Lemp, 2005; Nguyen &
Peck, 2009). These changes can lead to severe impairment of ocular surface health and oral health,
respectively. Dry eye caused by LG dysfunction is classified as aqueous-deficient as the LG
secretes the aqueous part of the tears which lubricate and moisten the ocular surface. SS patients
can develop additional systemic symptoms including inflammation of other internal organs such
as brain, lung and liver as well as, in a subset of patients, B-cell lymphoma(Nocturne & Mariette,
2015). SS is classified as primary (pSS) when the clinical manifestations occur alone, or as
secondary (sSS) when associated with another autoimmune disease, usually a connective tissue
disease. The prevalence of SS is estimated to be approximately 3% in subjects 50 years or older,
with a female to male ratio of 9:1 (Gaubitz, 2006), although the incidence and prevalence are likely
underestimated since many symptoms are nonspecific(Qin et al., 2015).
Mouse models are invaluable tools to understand disease pathogenesis as well as an ideal
subject for drug tests. Spontaneous mouse models for autoimmune diseases have been generated
through the crossing of mouse strains that have genetic susceptibility genes or loci followed by
careful monitoring of these mice for the development of disease phenotype. Extrinsically-induced
models, such as the murine cytomegalovirus induced model(Ohyama et al., 2006) and aromatase-
deficiency mice(Shim et al., 2004) have also been intensively studied to identify and address the
critical roles of these factors in the etiology of SS. In Chapter 1, I compared Non-Obese Diabetic
(NOD) mouse, a widely used spontaneous mouse model of SS-related dry eye disease, versus Non-
obese Diabetes-Resistant (NOR) mice, a derivative of NOD mice, in terms of SS-related
2

dacryoadenitis characteristics relative to a healthy control strain, BALB/c, to provide an additional
reference for mouse model selection for SS studies.
The first line therapy for aqueous-deficient dry eye disease is artificial tears, a replacement
therapy which increases the volume of the tear film and the residence time of tears on the ocular
surface. Providing only temporary relief of ocular symptoms, artificial tears fail to inhibit the
underlying inflammation related to autoimmunity that is present both on the ocular surface and in
the LG. Anti-inflammatory therapies such as topical corticosteroids, cyclosporine eyedrops, and
lifitegrast eyedrops are also used for clinical management of dry eye symptoms, though each has
disadvantages. Long term use of corticosteroids is complicated by the emergence of cataracts,
glaucoma, and infection(Foulks et al., 2015). Cyclosporine, approved as Restasis® by FDA in
2003 and Cequa® in 2019, can cause ocular surface pain, irritation and a burning sensation,
leading to low patient compliance and premature discontinuation of medication(Ames & Galor,
2015; J. Sheppard et al., 2020). Clinical trials of lifitegrast in dry eye patients have shown mixed
therapeutic effects(Holland et al., 2017; J. D. Sheppard et al., 2014; Tauber et al., 2015). Thus,
novel therapies that treat the underlying inflammation of SS while minimizing adverse effects are
urgently needed.
Rapamycin (Rapa), also known as Sirolimus, is a potent immunomodulatory agent
approved for renal allograft rejection (Kahan, 2000; MacDonald, 2001) and
lymphangioleiomyomatosis (Moir, 2016). It works by inhibiting the mammalian target of
Rapamycin (mTOR), a central regulator of cell growth, proliferation, translation, and autophagy
(Laplante & Sabatini, 2009; Wullschleger, Loewith, & Hall, 2006). Through mTOR inhibition,
Rapa has extensive immunosuppressive effects including inhibition of dendritic cell differentiation
and their ability to stimulate effector T cell responses (Hackstein et al., 2003; Turnquist et al.,
3

2007), block cell-cycle progression during T-cell activation, and sequester activated T cells in
lymphoid tissues (Gilbert & Weigle, 1993; Mondino & Mueller, 2007; Thomson, Turnquist, &
Raimondi, 2009). However, Rapa is highly hydrophobic with a low water solubility, a property
that limits its dosing options (Simamora, Alvarez, & Yalkowsky, 2001; Trepanier, Gallant, Legatt,
& Yatscoff, 1998). Currently, Rapa is only approved in oral solution formulation. Moreover, in
studies evaluating the partitioning of Rapa, it was found to have a plasma/whole blood ratio of
around 0.09, implying that when present in the blood, most drug will partition into red blood cells.
The free fraction of Rapa in plasma was only 2% and most of this drug was associated with
nonlipoprotein plasma proteins such as albumin (Trepanier et al., 1998; R. Yatscoff, LeGatt,
Keenan, & Chackowsky, 1993; R. W. Yatscoff, Wang, Chan, Hicks, & Zimmerman, 1995;
Zimmerman & Kahan, 1997). Rapa can also induce hyperglycemia, hyperlipidemia, insulin
resistance and increased incidence of new-onset type 2 diabetes (Morrisett et al., 2002; Salmon,
2015; Yang et al., 2012). Thus, an improved delivery method that prevents red blood cell extraction
of drug and targets its uptake into target sites could not only enhance the therapeutic effects of
Rapa, but also decrease its unwanted side effects.
Elastin-like polypeptides (ELPs) are an artificial, biomimetic class of protein polymers
inspired by the recurring hydrophobic motifs of tropoelastin (Dan E. Meyer & Chilkoti, 1999).
ELPs are thermo-responsive protein polymers consisting of different repeats of (Val-Pro-Gly-Xaa-
Gly)n pentamers where the guest residue, Xaa, specifies any amino acid and n determines the
number of pentapeptide repeats. ELPs can undergo a temperature-dependent reversible inverse
phase transition wherein they remain soluble below their transition temperature (Tt) and form a
coacervate above Tt (J. Despanie, J. P. Dhandhukia, S. F. Hamm-Alvarez, & J. A. MacKay, 2016;
D. E. Meyer & Chilkoti, 2002; Yeboah, Cohen, Rabolli, Yarmush, & Berthiaume, 2016). Taking
4

advantage of this property, ELPs can be easily purified by several rounds of heating and cooling
without addition of any extra purification tags. Phase separation for certain ELPs and their
derivatives may also be triggered by external factors such as alterations in ionic strength, pH
changes , electrical current, redox triggers, magnetism, and light(Ciofani, Genchi, Guardia, et al.,
2014; Shimoboji, Ding, Stayton, & Hoffman, 2002; Urry, 1993; Urry, Hayes, Gowda, Harris, &
Harris, 1992).
ELPs can be precisely genetically engineered and are biocompatible, biodegradable, of
low immunogenicity, and environmentally-responsive, which has enabled their diverse
applications in drug delivery(Ciofani, Genchi, Mattoli, Mazzolai, & Bandiera, 2014; J. Despanie
et al., 2016; Sarah R. MacEwan & Chilkoti, 2014). Tt can be finely tuned by adjusting the
hydrophobicity of the guest residue and the number of pentameric repeats. In general, ELPs with
high molecular weight and/or hydrophobic guest residues exhibit lower transition temperatures
than ELPs with low molecular weights and/or hydrophilic guest residues(Jordan Despanie, Jugal
P. Dhandhukia, Sarah F. Hamm-Alvarez, & J. Andrew MacKay, 2016). FK-506 binding protein
12 (FKBP), the cognate human receptor for Rapa, was genetically fused onto different ELPs to
generate a Rapa drug carrier, without impairing the thermo-responsiveness of the ELPs(Mihir Shah
et al., 2013).
Previously we have shown that when delivered systemically by FKBP-ELP, Rapa could
significantly inhibit LG inflammation and improve ocular surface symptoms in diseased male
NOD mice(C. Lee et al., 2019; Mihir Shah et al., 2013). However, the frequent dosing interval
used in these studies, as well as the systemic toxicity related to Rapa such as hyperglycemia and
hyperlipidemia were drawbacks in clinical translation of this strategy. In Chapter 2, I describe a
molecularly targeted drug carrier for Rapa based on the ELP system and shows that it can
5

specifically accumulate in the ICAM-1 enriched, inflamed LG in male NOD mice. In Chapter 3,
I describe a sustained local delivery method for Rapa, taking advantage of the thermal responsive
property of ELPs. Taken together, the delivery methods I have developed for Rapa could
significantly improve its pharmacokinetics properties, enhance its therapeutic effects and reduce
its systemic toxicity, thus broadening the application of this classical immunosuppressant in
autoimmune diseases as well as providing future potential therapies for SS-related dry eye diseases.

Graphic Illustration

6

Chapter 1: Comparison of NOD versus NOR Mice as Mouse Models of
Autoimmune Dacryoadenitis
1.1 Introduction
Sjögren’s syndrome (SS) is a systemic inflammatory autoimmune disease affecting more
than 4 million Americans, and is characterized by lymphocytic infiltration of lacrimal glands (LG)
and salivary glands (SG), reducing tear and saliva production and in severe cases, causing corneal
damage and compromised oral health (Lemp, 2005; Maciel, Crowson, Matteson, & Cornec, 2017).
In addition to exocrine gland dysfunction, SS patients progressively experience joint pain,
persistent dry cough, weight loss, fatigue, skin rashes, and in 5% of patients, non-Hodgkin
lymphoma (Nocturne & Mariette, 2015; Routsias et al., 2013). Although first described by the
Swedish ophthalmologist Henrik Sjögren in 1933, the disease mechanisms are still not fully
understood. Due to the lack of specific disease markers, early diagnosis of SS is challenging, which
makes it difficult to study the initiating biochemical and immunological events that occur prior to
manifestation of the hallmark clinical symptoms. In this regard, mouse models are invaluable tools
to understand disease pathogenesis as well as to develop promising new therapies.
Over the past decades, several spontaneous and experimentally-induced mouse models of
SS have been studied which have contributed significantly to our current understanding of the
disease. One of the more extensively studied SS mouse models is the Non-Obese Diabetic mouse
(NOD). NOD mice are derived from the outbred Jcl:ICR line of mice by Makino and colleagues
in 1980(Makino et al., 1980) and are prone to developing a series of autoimmune syndromes
including type 1 diabetes, autoimmune thyroiditis and autoimmune peripheral polyneuropathy
(Many, Maniratunga, & Denef, 1996; Salomon et al., 2001). Furthermore, they spontaneously
7

develop lymphocytic infiltration into the exocrine and endocrine glands (Hu, Nakagawa,
Purushotham, & Humphreys-Beher, 1992; Humphreys-Beher, 1996) including the LG which is
considered to be related to breakdown of multiple immune tolerance pathways (Anderson &
Bluestone, 2005; Nguyen & Peck, 2009). Interestingly, male NOD mice exhibit a more severe LG
disease, while the female NOD mice exhibit profound SG infiltration with lesser infiltration of LG
(Roescher et al., 2012).
The NOD mouse shares many disease characteristics with human SS with respect to
manifestations of disease. The LG of these mice and SS patients are infiltrated with T and B
lymphocytes as well as other immune cells, while tear flow is reduced (Kikutani & Makino, 1992;
Tsubota, Fujita, Tsuzaka, & Takeuchi, 2000). Common mechanistic pathologies are also shared
by these mice and SS patients associated with acinar secretory dysfunction. We have identified
elevated cathepsin S (CTSS), a cysteine endopeptidase implicated in autoimmunity and
inflammation (Conus & Simon, 2010; Reddy, Zhang, & Weiss, 1995), in lysosomes and secretory
vesicles in LG acinar cells in parallel with increased CTSS activity in tears of male NOD mice (Li
et al., 2010). Increased tear CTSS activity is also detected in SS patients relative to healthy controls,
patients with non-autoimmune dry eye and patients with other autoimmune diseases (Hamm-
Alvarez et al., 2014). A similar reduction in the expression and a redistribution to the basolateral
domain of the exocytosis effector, Rab3D, has been observed in acinar cells from SS patients
(Bahamondes et al., 2011) and male NOD mice (Meng et al., 2016). Atrophic myoepithelial cells
(MEC), multipolar stellate cells responsible for contraction to facilitate compression of acini to
assist in expulsion of tears and saliva, and are detected by their enrichment in α-smooth muscle
actin (SMA), are reduced in SS patients SG (Sisto et al., 2018) and in female NOD mice SG
8

(Nashida, Yoshie, Haga-Tsujimura, Imai, & Shimomura, 2013). These findings suggest that NOD
mice recapitulate characteristics of SS patients.
The incidence of spontaneous diabetes in the NOD mouse is 60% to 80% in females and
20% to 30% in males (Bach, 1994; Kikutani & Makino, 1992). Diabetes onset typically occurs
from 12 to 14 weeks of age in female mice and later in male mice. Previously, we have used male
NOD mice aged below 20 weeks, when most mice are diabetes-free, in studies of SS disease
biology as well as in development of potential SS therapeutics (Shah et al., 2017). However, the
onset of diabetes may constitute a confounding factor for prolonged drug studies on autoimmune
dacryoadenitis in male NOD mice. The Non-Obese Diabetes-Resistant (NOR) mice are a
diabetes-free major histocompatibility (MHC)-matched alternative to the NOD mice. The NOR
strain is the result of a genetic contamination of NOD with C57BL/KsJ. The C57BL/KsJ-derived
non MHC-linked diabetes-resistance factors in the NOR mice are reported not to affect the T-
lymphocyte accumulation characteristics of NOD mice. Although, NOR mice have been used in a
few studies of SS disease biology (H. S. Aluri et al., 2015; H. S. Aluri et al., 2017), there has not
been a side-by-side characterization of NOR mice relative to NOD mice as models for SS-
associated dacryoadenitis and other features of the disease process in the LG. Here we have
compared the development and progression of SS-like ocular symptoms in male NOD and NOR
mice relative to BALB/c healthy controls to provide an additional reference for mouse model
selection for SS studies related to LG dysfunction. Comparison of the features that are shared
between models to those that are distinct may also aid in understanding mechanisms responsible
for development of these signs of SS.
1.2 Materials and Methods
9

Mice. Male BALB/cJ (00065), NOD/ShiLtJ (001976) and NOR/LtJ (002050) mice were
purchased from The Jackson Laboratory (Bar Harbor, ME, USA). All animal use was in
compliance with policies approved by the University of Southern California Institutional Animal
Care and Use Committee and experiments were conducted in accordance with the ARVO
Statement for the Use of Animals in Ophthalmic and Vision Research.
Reagents. The rabbit anti-CTSS polyclonal antibody (Catalog No. 6686-100) and the
CTSS activity assay kit were from Biovision (Milpitas, CA, USA). Rabbit anti-Rab3D polyclonal
antibody was generated by Antibodies Inc. (Davis, CA) as previously reported (Evans et al., 2008).
Polyclonal rabbit anti-SMA antibody (Catalog No. ab5694) was from Abcam (Cambridge, UK).
Bovine serum albumin was from Jackson ImmunoResearch Laboratories (Westgrove, PA, USA).
Rhodamine phalloidin, Alexa-Fluor 488 goat anti-rabbit secondary antibody and ProLong Gold
Antifade Mounting Medium were from Invitrogen (Carlsbad, CA, USA). Ketamine (Ketaject) was
from Phoenix (St. Joseph, MO, USA) and xylazine (AnaSed) was from Akorn (Lake Forest, IL,
USA). Carbachol, used to stimulate tear production, was from Sigma-Aldrich (St. Louis, MO,
USA). O.C.T. Compound was obtained from Sakura Lifetek (Torrance, CA, USA). The reverse
transcription kit, TaqMan Universal PCR Master Mix, and primers for real-time (RT)-PCR
analysis of GAPDH (glyceraldehyde 3-phosphate dehydrogenase Mm99999915_g1), CTSS
(Mm01255859_m1), MHC II (Mm00439216_m1) and interferon-γ (IFN-γ) (Mm00801778_m1)
were from Life Technologies (Grand Island, NY, USA). The RNeasy plus Universal Mini Kit was
from Qiagen (Valencia, CA, USA). ZoneQuick phenol red threads were purchased from SHOWA
YAKUHIN KAKO CO., LTD (Tokyo, Japan). Free Style Lite test strips were from Abbott
Diabetes Care, Inc. (Alameda, CA, USA). The Bio-Rad protein assay kit was from Bio-Rad
10

(Hercules, CA, USA). All other chemicals were reagent grade and obtained from standard
suppliers.
Blood Glucose Measurements. Mice were anesthetized briefly using a continuous flow
of isoflurane through a nose cone. A tail nick was made and a drop of peripheral blood was
collected. Blood glucose was measured using the Free Style Lite test strips read with a glucose
meter (Abbott Diabetes Care, Inc., Alameda, CA, USA).
Basal and Stimulated Tear Measurements. Basal and stimulated tears were measured
as described (M. Shah et al., 2013; Shah et al., 2017). Briefly, for measurement of basal tear flow,
a ZoneQuick phenol red impregnated cotton thread was inserted under the lower eyelid for 10 sec
while mice were under anesthesia, and tear production was reported as a function of the length of
wetting of the thread in millimeters. The procedure was repeated in each eye and data presented
are an average of the wetting length of the threads from both eyes. For collection of stimulated tear
fluid, a small incision was made to expose the LG of anesthetized mice. A small piece of cellulose
mesh (Kimwipe; Fisher Scientific, Pittsburgh, PA, USA) was placed on the LG to capture added
secretagogue (3 μl of 50 μM carbachol solution). After stimulating the gland for 5 min, tears from
both eyes were collected with 2 μl microcapillary tubes (Drummond Scientific, Broomall, PA,
USA.) Stimulation was performed three times for a total collection time of 15 min, and the volume
of collected tears was recorded. Tears were placed on ice until further biochemical analysis.
CTSS Activity Analysis. CTSS activity in tears was measured using a protocol modified
from that described previously (Hamm-Alvarez et al., 2014; Li et al., 2010). Briefly, stimulated
mouse tear fluid was diluted in 200 μL of CTSS reaction buffer and divided into two 100 μL
reactions on a 96-well plate. A CTSS–specific inhibitor was added to one of the two wells for
11

each sample. After incubation at 37 oC for 1 hour, the plate was read using a SpectraMax Gemini
EM Microplate Reader with emission at 400nm and excitation at 505 nm. After the experiment,
the Bio-Rad protein assay was performed per the manufacturer’s protocol and CTSS activity was
normalized to 1 μg total tear protein.
Real-Time PCR. For measurement of gene expression, total RNA was isolated from
whole LG using the RNeasy plus Universal Mini Kit per the manufacturer’s protocol. DNA was
prepared using the reverse transcription kit per the manufacturer’s recommendations. Quantitative
(real-time) PCR was performed with the TaqMan gene expression assays on the ABI 7900HT real-
time PCR system, using the following probes: IFN-γ, MHC II, and CTSS. GAPDH was used as a
control housekeeping gene. Each reaction consisted of 1 μL cDNA from the reverse transcription
reaction mixed with 8 μL nuclease-free water and 1 μL assay primer mixed with 10 μL TaqMan
Universal PCR Master Mix. Each sample was run in triplicate. The thermal profile consisted of
preheating the samples at 95.8oC for 10 min followed by 40 repeats at 95.8oC for 15 sec each and
60.8oC for 1 min. The relative expression levels were calculated using the comparative CT method
(ΔΔCT method) on the default ABI software SDS 2.3 (ThermoFisher Scientific, Waltham, MA,
USA), as described previously (Wu et al., 2009).
Immunofluorescence Labeling and Confocal Microscopy. Mouse LG were processed
as previously described (Meng et al., 2016; Meng, Klinngam, Edman, & Hamm-Alvarez, 2017).
Briefly, mice were anesthetized by intraperitoneal injection with ketamine/xylazine (80-100
mg/kg/ 5-10 mg/kg),and euthanized by cervical dislocation. LGs were fixed, embedded in O.C.T.
compound and flash frozen on dry ice. Blocks were cryosectioned at 5 µm thickness and mounted
on glass slides. After being quenched and permeabilized as described (Meng et al., 2016) , tissue
sections were blocked with 1% BSA followed by incubation with primary and secondary
12

antibodies. Finally, samples were mounted with ProLong anti-fade mounting medium and imaged
using a ZEISS LSM 800 confocal microscope equipped with an Airyscan detector (Zeiss,
Thornwood, NY). Quantification of the signal intensity was done with ImageJ (1.49V,
http://imagej.nih.gov/ij/; provided in the public domain by the National Institutes of Health,
Bethesda, MD, USA). For quantification of Rab3D distribution, the drawing tool was used to
delineate the cell boundary and the apical half of the cells. The fluorescence intensity within the
apical half and the whole cell was measured and the apical/total ratio of fluorescence intensity was
calculated. Calculations were made in multiple acinar cells from 15 fields each from n=3 mouse
strain and age. For Quantification of SMA staining, SMA and DAPI intensity from 3 pictures per
mouse was determined using ImageJ. The fluorescence intensity for each stain was measured as
the Integrated Density over the entire image. For SMA staining, background fluorescence was
removed using the “subtract background” toll using a rolling ball diameter=10. The intensity of
SMA staining was normalized using DAPI staining.
Histology Analysis of LG. Quantification of lymphocytic inflammation in LG was
performed as described previously (M. Shah et al., 2013). Briefly, mice were euthanized and LGs
were removed and fixed in 10% neutral buffer formalin prior to embedding in paraffin blocks.
Paraffin sections were stained with hematoxylin-eosin (H&E) according to standard procedures
and photographed using a Nikon 80i microscope (Melville, NY, USA) equipped with a digital
camera. Images of three nonconsecutive whole gland cross sections were obtained for each LG.
The area of the LG occupied by lymphocytic foci was calculated using ImageJ software by a
blinded examiner.
Statistics. All statistical analyses were performed using GraphPad Prism software
(GraphPad, San Diego, CA, USA). Shapiro Wilk, KS, and D’Agostino & Pearson omnibus
13

normality tests were used to test for normal distribution in each data set. For normally distributed
data, a one-way ANOVA with the Tukey’s HSD test was used to compare differences between
groups, whereas the Mann-Whitney test was used for nonparametric analysis. A p < 0.05 was
considered as a significant difference.
1.3 Results
Blood glucose levels do not significantly differ in male NOD and NOR mice up to 20
weeks of age. The male NOD is less prone to diabetes than its female counterpart, but it has been
reported that 20%-30% of male NOD spontaneously develop diabetes by 30 weeks of age (Bach,
1994). In this study, at ages commonly associated with evaluation of symptoms of autoimmune
dacryoadenitis in this model, none of the strains exhibited peripheral blood glucose levels over
250 mg/dL, indicative of diabetes. Furthermore, there were no significant differences in peripheral
blood glucose levels between BALB/c, NOD and NOR mice at the ages measured from 10 to 20
weeks, suggesting that the male NOD mice as well as the NOR mice are relatively diabetes-free
at ages below 20 weeks (Fig 1).
14

Figure 1. Blood glucose levels in male BALB/c, NOD and NOR mice. Blood glucose levels of
BALB/c, NOD and NOR mice at 10, 13, 16 and 20 weeks of age were measured in peripheral
blood acquired by tail nick. A one-way ANOVA with the Tukey’s HSD test was used to compare
different time points within each strain; another one-way ANOVA with Tukey’s HSD test was
used to compare different strains at each measured time point. (N=10 per strain with the same
mice being measured at each age).

Reduction of stimulated tear flow and increased LG lymphocytic infiltration show different
patterns in male NOD relative to male NOR mice. Male NOD mice exhibit many SS-like
manifestations including lymphocytic infiltration of the LG in conjunction with LG dysfunction
and reduced tear flow (Doyle et al., 2007; Lavoie, Lee, & Nguyen, 2011; Park, Gauna, & Cha,
2015). The same disease characteristics have also been demonstrated in the male NOR mice (H.
S. Aluri et al., 2015). Here, we found no significant difference in basal tear secretion measured
using phenol red threads in male BALB/c, NOD or NOR mice at ages from 10 to 20 weeks of age
15

(Fig 2A). However, the stimulated tear volume (which was assessed only at 12 and 20 weeks of
age since the procedure is terminal) showed that at 12 weeks, NOD mice had significantly lower
stimulated tear volume than BALB/c (3.2 ± 0.6 μL versus 4.7 ± 0.8 μL), while there was no
significant difference between BALB/c and NOR mice although a trend to decreased secretion was
noted at 12 weeks (4.0 ± 1.2 μL). By 20 weeks of age, both NOD (3.0 ± 0.6 μL) and NOR (3.3 ±
0.7 μL) mice exhibited significantly lower stimulated tear volume than BALB/c mice (5.4 ± 0.6
μL) that was relatively indistinguishable (Fig 2B).

Figure 2. Basal and Stimulated Tear Production in male BALB/c, NOD and NOR mice. A)
Basal tear secretion in male BALB/c, NOD and NOR mice was measured using a phenol red thread
test from 10 weeks of age to 20 weeks of age. A one-way ANOVA with the Tukey’s HSD test was
used to compare different time points within each strain; another one-way ANOVA with Tukey’s
HSD test was used to compare different strains at each measured time point. (N=10 mice per strain
with the same mice measured at each age). B) Carbachol-stimulated tear volume was measured
in male BALB/c, NOD and NOR mice at 12 and 20 weeks of age with mice under general
anesthesia. A one-way ANOVA with the Tukey’s HSD test was used across strains at each time
point. (N=7 per strain per age) (*P ≤ 0.05; **P ≤ 0.01; ***P ≤ 0.001).

Lymphocytic infiltration was similarly assessed at 12 and 20 weeks of age by quantification,
in H&E-stained LG sections, of the percentage area per cross section of the gland covered by
lymphocytic foci. At 12 weeks, the percentages of LG area infiltrated by lymphocytes in male
16

NOD mice (10.8 ± 5.4%) and NOR mice (4.8 ± 3.3%,) were significantly higher than in male
BALB/c mice, while the lymphocytic infiltration was further enhanced by 20 weeks of age in male
NOD (17.5 ± 11.0%) and NOR (8.9 ± 6.1%) mice. At both ages, the extent of lymphocytic
infiltration in the NOD mice was significantly greater than that measured in the NOR mice,
however. In this study, BALB/c mice showed no detectable LG lymphocytic infiltration at either
age (Fig 3) although previous studies have shown low levels of lymphocytic infiltration (Li et al.,
2010; Schenke-Layland et al., 2010).
17

Figure 3. LG lymphocytic Infiltration in BALB/c, NOD and NOR mice. A) Representative
images of H&E staining in male NOD, NOR and BALB/c mouse LG at 12 and 20 weeks of age.
B) Quantification of the percentage of lymphocytic infiltration in multiple sections from each LG
as described in Methods. A one-way ANOVA with the Tukey’s HSD test was used across strains
at each time point. Scale bar=200 μm (N=10 mice per strain and age) ( *P ≤ 0.05; **P ≤ 0.01;
***P ≤ 0.001).

18

CTSS activity and expression is increased in male NOD and NOR mice compared
with BALB/c mice. Previously, our laboratory has demonstrated that CTSS activity in male NOD
mouse tear fluid was significantly increased when compared with age- matched BALB/c controls
(Li et al., 2010). We have further demonstrated that CTSS activity is significantly elevated in tears
of SS patients relative to healthy controls, to patients with non-SS dry eye and to patients with
other autoimmune diseases, suggesting that the increased tear CTSS activity might serve as a
potential biomarker for diagnosis of SS (Hamm-Alvarez et al., 2014). Significantly increased
CTSS activity in stimulated tear was detected from male NOD and NOR mice compared with
BALB/c mice at both 12 weeks and 20 weeks of age (Fig 4A). Besides CTSS activity in tears,
CTSS gene expression levels were examined in LG of male BALB/c, NOD and NOR mice. At 12
weeks and 20 weeks, NOD mouse LG showed a 7.6-fold (p<0.001) and 8.8-fold (p<0.05) increase
in CTSS gene expression compare to BALB/c; however, no difference in CTSS gene expression
was detected between NOR and BALB/c mouse LG (Fig 4B). LG CTSS protein levels were also
investigated by immunofluorescence labeling and confocal microscopy. A strong increase in CTSS
immunofluorescence was observed in NOD mouse LG compared with BALB/c mouse LG, while
the CTSS immunofluorescence labeling in the NOR mouse LG was less striking (Fig 4C).
However, CTSS immunofluorescence was also noticeably increased in NOR mouse LG relative
to BALB/c mouse LG mice, even if not as abundant as in the NOD mice. Arrows in images for
NOD and NOR mice point to subapical accumulations of CTSS which our previous studies suggest
may represent CTSS that reaches the tears through missorting into conventional secretory
pathways or through upregulation of unconventional secretory mechanisms (Meng et al., 2016).
These stores are not present in BALB/c mouse LG.
19

20

Figure 4. CTSS activity in stimulated tears and its expression and distribution levels in LG
of male BALB/c, NOD and NOR mice. A) CTSS activity levels in carbachol- stimulated tears
from male BALB/c, NOD and NOR mice was measured at 12 and 20 weeks of age. A one-way
ANOVA with the Tukey’s HSD test was used across the strains at each time point. (N=7). B) Gene
expression levels for LG CTSS in male BALB/c, NOD and NOR mice were measured at 12 and
20 weeks. A one-way ANOVA with Tukey’s HSD test was used across strains at each time point.
(N=5) (*P ≤ 0.05; **P ≤ 0.01; ***P ≤ 0.001). C) Intracellular distribution of CTSS protein in LG
of male BALB/c, NOD and NOR mice at 12 and 20 weeks of age was evaluated by
immunofluorescence labeling. Green: CTSS. Red: Actin. Blue: DAPI. Scale bar, 20 μm. Results
shown are typical of those seen from multiple images taken from LG from n=3 mice of each type
in each age group.

Expression of inflammation-related genes is elevated in NOD but not NOR mouse LG.
In addition to CTSS, gene expression levels of MHC II, a key molecule involved in antigen
presentation (Villadangos et al., 1999), was also tested by RT-qPCR. Similar to the changes in
gene expression of CTSS, there were large increases in gene expression of MHC II in NOD mouse
LG at 12 weeks and 20 weeks of 58.42-fold (P<0.001) and 94.87-fold (P=0.07) relative to BALB/c
mouse LG but no significant changes in NOR mouse LG relative to levels in age-matched BALB/c
mouse LG (Fig 5A). IFN-γ has been implicated in the development of ocular symptoms in LG
and ocular surface in SS (Brookes, Price, Venables, & Maini, 1995; Meng et al., 2017; Willeke et
al., 2009). Its gene expression level was significantly increased in the NOD mouse LG compared
with BALB/c mouse LG at both 12 weeks of age (3.29-fold, p<0.05) and 20 weeks of age (15.25-
fold, p<0.05)(Fig 5B). However, this cytokine was also not elevated in LG of NOR mice relative
to BALB/c mice at either time point.

21

Figure 5. Expression of inflammation-related genes in LG of NOD and NOR mouse relative
to age-matched BALB/c mouse LG. Gene expression levels of MHC II (A) and IFN-γ (B) were
measured by RT-PCR in LG from male BALB/c, NOD and NOR mice at 12 and 20 weeks of age.
A one-way ANOVA with Tukey’s HSD test was used across strains at each time point. (N=5 per
strain per age) ( *P ≤ 0.05 ; **P ≤ 0.01 ; ***P ≤ 0.001).

The cellular distribution of Rab3D is altered in male NOD and NOR mouse LG
relative to BALB/c mouse LG. We have previously shown that there is a significant depletion in
apical stores of Rab3D and a redistribution of immunofluorescence associated with this protein to
the basal domain in LG acinar cells (LGAC) of NOD mice relative to LGAC from BALB/c mice
(Meng et al., 2016). In LGAC from both NOD and NOR LG, depletion of subapical Rab3D
immunofluorescence and relocalization of some Rab3D immunofluorescence to the basal domain
(Fig 6A, arrows) could be detected compared to its largely subapical localization beneath the
lumenal actin in BALB/c mouse LGAC. Quantitative analysis of the percentage of total Rab3D
immunofluorescence in the apical half of the cell in LGAC from different mouse strains showed
an equivalent 50% loss of subapical Rab3D stores in LGAC from NOD and NOR mouse LG at 12
and 20 weeks (Fig 6B).
22

Figure 6. Relative distributions of Rab3D in LG of male BALB/c, NOD and NOR mice at 12
and 20 weeks of age. A) The distribution of Rab3D vesicles in LGAC in BALB/c, NOD and NOR
mice was assessed by immunofluorescence. Green: Rab3D. Red: Actin. Blue: DAPI. Asterisk:
Lumenal region. Arrows: Basal redistribution of Rab3D vesicles. Scale bar, 20 μm. Results shown
are typical of multiple images taken from LG from n=3 mice per strain per age. B) Quantification
of subapical Rab3D vesicle accumulation in LGAC from BALB/c, NOD and NOR mice. A one-
way ANOVA with Tukey’s HSD test was used across strains at each time point. (N=15 acini per
mouse, 3 mice per strain per age) ( *P ≤ 0.05;**P ≤ 0.01;***P ≤ 0.001).

23

Myoepithelial cell thinning was observed in both NOD and NOR. Myoepithelial cells
(MEC) are stellate-shaped cells with long slender processes that form a basket-like network around
LGAC. These cells express α-smooth muscle actin (SMA), a marker for MECs, which can
contract and thereby assist the secretory function of exocrine glands (Avci, Gunhan, Cakalagaoglu,
Gunal, & Celasun, 2012; Dartt, 2009; Gudjonsson, Adriance, Sternlicht, Petersen, & Bissell, 2005).
MECs express abundant neurotransmitter receptors such as muscarinic receptors, suggesting that
these cells are responsive to the same neural stimulation responsible for inducing secretion of LG
fluid (Lemullois, Rossignol, & Mauduit, 1996). It is also likely that an important role of MEC is
to support and maintain the structure of the LG (H. S. Aluri et al., 2015). Immunostaining of SMA
was utilized to evaluate the relative abundance of MECs in the LG of these mice. At 12 weeks
NOD and NOR mouse LG showed significantly reduced intensities of SMA staining compared
with LG from BALB/c mouse, suggestive of a parallel degeneration or thinning of MECs in these
mice. However, at 20 weeks no significant difference in SMA staining was detected between
BALB/c and NOR and only a trend of decrease was seen between BALB/c and NOD, suggesting
that degeneration of MECs are more apparent and consistent at early time point.

24

Figure 7. SMA-enriched MEC distribution in NOD and NOR mouse LG relative to age-
matched BALB/c mouse LG. A) Immunofluorescence labeling of alpha-SMA (green), a marker
for MEC, in LG at 12 and 20 weeks of age. Red: E-Cadherin. Blue: DAPI. Results shown are
representative of multiple images taken from LG of n=3 mice per strain and age. Scale bar, 25
μm. B) Quantification of integrated density of alpha-SMA staining normalized to DAPI staining.
A one-way ANOVA with Tukey’s HSD test was used across strains at each time point. (N=3
pictures per mouse, 3 mice per strain per age) ( *P ≤ 0.05;**P ≤ 0.01;***P ≤ 0.001).

A
25

1.4 Conclusion and Discussion
Despite differences in the immune system between human and mouse (Mestas & Hughes,
2004), murine models are invaluable tools for studying inflammatory disease pathology including
elucidation of the pathological mechanisms of SS. These models provide the opportunity to
observe biological changes at the tissue and molecular levels during the early disease state as well
as to study disease progression in detail, a feat difficult to achieve in human patients, but essential
for the development of new diagnostic and therapeutic strategies. Accumulation of new mutations
by inbreeding can produce significant substrain divergence over time, and potentially, differences
in substrain characteristics. Phenotypic differences and genetic drift between and within NOD
substrains have been observed (De Riva et al., 2013; Simecek et al., 2015) which may be due to
environmental factors including the role that the gut microbiome plays in priming T cell
immunoreactions (Markle, Frank, et al., 2013; Markle, Mortin-Toth, et al., 2013) which may
represent an essential part of the development of autoimmunity in SS (de Paiva et al., 2016). In
addition, no consensus has been reached about onset of diabetes in male NOD mice, probably due
to the divergence of different substrains housed in different environments (Bach, 1994; Kikutani
& Makino, 1992). Within the NOD/ShiLtJ substrain that we used here, we consistently observed
significant variance among individual mice with respect to LG lymphocytic infiltration that is
reflected in Fig 3. In comparison, NOR mice exhibited less variance in this measure which may
be an indicator of a more stable phenotype, albeit showing lesser lymphocytic infiltration. As for
the onset of diabetes, we did not detect any significant increase in blood glucose levels as late as
20 weeks in either NOD or NOR mice compared with BALB/c mice. However, we did not perform
the more sensitive diagnostic measurements such as oral glucose tolerance tests or HbA1c analysis,
instead acquiring random normal range blood glucose levels.
26

Overall assessment of the features of LG dysfunction measured here in NOD and NOR
mice suggest relatively parallel onset of the following symptoms: decreased stimulated tear
secretion, increased tear CTSS activity, increased subapical CTSS immunofluorescence,
redistribution of Rab3D to the basal pole of the cell, and thinning of MECs. Our assessment
suggests a relatively distinct pattern of other indicators of disease including gene expression of
MHC II, IFN-γ, and CTSS, which were significantly elevated in NOD mouse LG at 12 and 20
weeks but not in NOR mouse LG. Finally, our assessment showed that the onset of lymphocytic
infiltration in the LG was similar in both NOD and NOR mice but that the extent of infiltration in
NOD mice was twice that in NOR mice at both ages. It is possible that the increased gene
expression of pro-inflammatory factors creates an environment in which generalized inflammatory
processes are exacerbated in the NOD mouse background, relative to the NOR mouse.
Previous microarray analyses in LG of male NOD mice identified CTSS as a possible tear
biomarker for SS (Li et al., 2010) findings that were reinforced in a clinical study of SS patients
relative to patients with other autoimmune diseases and patients with dry eye due to other causes
(Hamm-Alvarez et al., 2014). Although we have not been able to investigate the source of
increased tear CTSS in the LG of patients at the molecular level, we have linked increased CTSS
secretion to decreased Rab3D activity and to its mislocalization in LGAC in mouse models (Meng
et al., 2016). While we detected upregulation of CTSS both at the gene expression level and protein
level as well as elevated CTSS activity in tears in NOD mice relative to BALB/c mice, in NOR
mice we detected only increased tear CTSS activity without any concomitant increased gene
expression and with only modestly increased cellular CTSS protein as assessed by
immunofluorescence. However we could detect increased subapical CTSS stores in LGAC of
NOR mice that may contribute to the increased secreted CTSS activity. Other mechanisms may
27

also be involved in the activation of CTSS released into tears in the NOR and NOD mice, which
may include reduced expression/functionality of CTSS activity inhibitors such as cystatin C in
tears (Riese et al., 1998; Saegusa et al., 2002), altered tear pH and/or enhanced proenzyme
activation processes in the tears associated with protease inhibitor imbalance.
However, a unifying factor shared by both NOD and NOR mice in exhibiting increased
tear CTSS seems to be mislocalization of Rab3D, occurring to the same extent and within the same
time frame, suggesting that this process may be instrumental in initiating some of the changes in
protein composition and loss of tear quality noted in the tears of SS patients. Interestingly,
comparable changes have been observed in exocrine glands of SS patients (Bahamondes et al.,
2011; Kamoi et al., 2012). Rab3D, a member of the Rab protein family, is associated with mature
secretory granules or secretory vesicles in diverse cell types including those of neural, endocrine,
exocrine, and immune origin (Fukuda, 2008). Because of its localization on secretory granules and
its redistribution upon stimulated secretion in a number of exocrine cells including LGAC (X.
Chen, Edwards, Logsdon, Ernst, & Williams, 2002; Lemp, 2005), Rab3D is thought to play a key
role during regulated exocytosis in exocrine secretion.
Previously we demonstrated using in vitro studies in cultured LGAC that IFN-γ exposure
reduced Rab3D gene expression levels and increased CTSS secretion (Meng et al., 2017). While
we have reconfirmed the significant upregulation of IFN-γ in the LG of NOD mice, in comparison,
no statistically significant increase in gene expression of IFN-γ was detected in NOR mice relative
to BALB/c mice, despite the relatively parallel increases in CTSS secretion and redistribution of
Rab3D. This suggests that the changes in the secretory pathway and composition of the tears seen
in vivo in these mice may be elicited by other factors than pro-inflammatory cytokines such as
IFN-γ. One possible mechanism could be the hypofunction of the bHLH transcription factor,
28

MIST1, which is a developmentally regulated and highly cell lineage-specific transcription factor
in diverse tissues. MIST1 binds to highly conserved CATATG E-boxes to directly activate
transcription of RAB3D, thus affecting the secretory phenotype of exocrine cells (Pin, Bonvissuto,
& Konieczny, 2000; Tian et al., 2010) The evidence that Mist1 knockout mice exhibit cellular
disorganization and functional defects in the exocrine pancreas also suggests a crucial role for
MIST1 in regulating exocytosis in exocrine cells (Kowalik et al., 2007). However, little is known
about the expression level and functionality of MIST1 in NOD and NOR mice, nor the relationship
between autoimmune inflammation and MIST1 expression.
Reduced stimulated tear production was equivalent in both NOD and NOR by 20 weeks
and was decreased in NOD at 12 weeks with a trend to a decrease in the NOR mice also at 12
weeks. This reduction is associated with notable loss of SMA in MEC at both timepoints in these
mice. It is also associated with a significant increase in lymphocytic infiltration of the LG at both
time points. It is intriguing, however, that the extent of lymphocytic infiltration, which was greater
in NOD mouse LG than in NOR mouse LG, was not correlated with either basal tear secretion
(which was unchanged) nor carbachol-stimulated tear secretion (which was equivalently reduced
by 20 weeks of age). This may suggest that secretory dysfunction is largely driven by other causes
such as altered intracellular trafficking including redistribution of Rab3D (Meng et al., 2016),
decreased neurological responsiveness (Waterman, Gordon, & Rischmueller, 2000), loss of
extracellular matrix (Li et al., 2010), and/or loss of MEC (Nashida et al., 2013). It is also possible
that lymphocytic infiltration above a certain threshold in the LG is sufficient to create an
environment that allows or enhances other changes required for secretory dysfunction to occur,
and that more inflammation of the LG does not necessarily denote more severe SS-like disease.
29

This is a critical issue, since many drug development studies rely on analyses of lymphocytic
infiltration of LG and SG as a key endpoint indicative of therapeutic effect.
In conclusion, in this study, we compared two commonly-used spontaneous mouse models
for SS-related autoimmune dacryoadenitis that are genetically related at both the molecular and
functional levels, relative to male BALB/c healthy control mice. Male NOD and NOR mice
exhibited parallel development of many key features of disease including reduced stimulated tear
production, increased lymphocytic infiltration of LGs, increased tear CTSS activity, increased
redistribution of Rab3D to the basal pole of LGAC and loss of MEC. Although lymphocytic
infiltration occurred on a parallel time course, it was more pronounced in the NOD mouse relative
to NOR, while other indicators of disease were present to approximately the same extent in both
strains. Other features which have been previously associated with disease, such as increased
expression of pro-inflammatory genes in the LG, were seen only in the NOD mouse.
Sjogren's syndrome is a systemic autoimmune disease with sophisticated mechanisms
which may involve both inflammation and secretory dysfunction. In male NOD LG, both
autoimmune inflammations indicated by significantly increased inflammation markers and
secretory deficit were observed. This model could be useful for the study of autoimmune-related
dry eye disease. In contrast, NOR mice appear to have a secretory deficit without a pronounced
increase in inflammation. This model could be advantageous for study of the role of secretory
dysfunction in dry eye without the effect of inflammation. Thus, which mouse model would be
used would depend upon the type of dry eye being studied or the question being asked and could
be used to determine if inflammation plays a role in SS-like dry eye.

30

Chapter 2: Development of a Molecularly Targeted Carrier of
Immunosuppressants Using Elastin-Like Polypeptide
2.1 Introduction
Rapamycin (Rapa), also known as Sirolimus, is a potent immunomodulatory agent
approved for renal allograft rejection (Kahan, 2000; MacDonald, 2001) and
lymphangioleiomyomatosis (Moir, 2016). It works by inhibiting the mammalian target of
Rapamycin (mTOR), a central regulator of cell growth, proliferation, translation, and autophagy
(Laplante & Sabatini, 2009; Wullschleger et al., 2006). Through mTOR inhibition, Rapa has
extensive immunosuppressive effects including inhibition of dendritic cell differentiation and their
ability to stimulate effector T cell responses (Hackstein et al., 2003; Turnquist et al., 2007), block
cell-cycle progression during T-cell activation, and sequester activated T cells in lymphoid tissues
(Gilbert & Weigle, 1993; Mondino & Mueller, 2007; Thomson et al., 2009). However, Rapa is
highly hydrophobic with a low water solubility, a property that results in low and unpredictable
oral bioavailability and limits dosing options (Simamora et al., 2001; Trepanier et al., 1998).
Moreover, in studies evaluating the partitioning of Rapa, it was found to have a plasma/whole
blood ratio of around 0.09, implying that when present in the blood, most drug will partition into
red blood cells. The free fraction of Rapa in plasma was only 2% and most of this drug was
associated with nonlipoprotein plasma proteins such as albumin (Trepanier et al., 1998; R.
Yatscoff et al., 1993; R. W. Yatscoff et al., 1995; Zimmerman & Kahan, 1997). Rapa can also
induce hyperglycemia, hyperlipidemia, insulin resistance and increased incidence of new-onset
type 2 diabetes (Morrisett et al., 2002; Salmon, 2015; Yang et al., 2012). Thus, an improved
delivery method that prevents red blood cell extraction of drug and targets its uptake into targets
31

sites could not only enhance the therapeutic effects of Rapa, but also decrease its unwanted side
effects.
ELPs are unique protein polymers consisting of pentameric repeats of (Val-Pro-Gly-Xaa-
Gly)n where the guest residue, Xaa, specifies any amino acid and n determines the number of
pentapeptide repeats. An interesting characteristic of ELPs is their ability to undergo a reversible
inverse phase transition wherein they remain soluble below their transition temperature (Tt) and
form a secondary liquid phase ‘coacervate’ above Tt (J. Despanie et al., 2016; D. E. Meyer &
Chilkoti, 2002; Yeboah et al., 2016). Taking advantage of this property, ELPs can be easily
purified by several rounds of heating and cooling without addition of any extra purification tags.
The ELP phase transition is triggered by heating above the Tt, which allows ELP coacervates to
be collected via centrifugation. Then, the ELP pellet is resolubilized in cold buffer and centrifuged
below the Tt, which removes insoluble impurities and cell debris. 3-5 cycles of heating/cooling
can generate a final product with higher than 95% purity (S. R. MacEwan, Hassouneh, & Chilkoti,
2014). ELPs can be precisely genetically engineered and are biocompatible, biodegradable, of low
immunogenicity, and environmentally-responsive, which enables diverse applications in drug
delivery (Ciofani, Genchi, Mattoli, et al., 2014; J. Despanie et al., 2016; Sarah R. MacEwan &
Chilkoti, 2014). In this study, we chose A192 with an amino acid sequence of (VPGAG)192, which
is stable, is soluble at physiological temperatures at relevant concentrations, and has molecular
weight above the cutoff for glomerular filtration (Janib et al., 2014).
Intercellular adhesion molecule-1 (ICAM-1, or CD54) is a cell-surface glycoprotein
member of the immunoglobulin (Ig) superfamily that plays a vital role in inflammatory and
immune responses. ICAM-1 is involved in lymphocyte migration, co-activation of T- and B-cells,
and leukocyte extravasation into lymphoid and inflamed non-lymphoid tissues through the
32

interaction with β2 integrin lymphocyte function-associated antigen-1 (LFA-1, αLβ2, or
CD11a/CD18) and macrophage 1 antigen (Mac-1) (Long, 2011). Overexpression of ICAM-1 is
common in diseases where inflammation and immune cells are involved, such as ischemia-
reperfusion injury, transplant rejection, and autoimmune diseases (Bloemen et al., 1995; Koning,
Schiffelers, & Storm, 2002; Muro & Muzykantov, 2005). Thus, ICAM-1 constitutes a promising
target for delivery of drugs directly to inflamed tissues.
We used the male Non-Obese Diabetic (NOD) mouse model to test features of ICAM-1
based targeting. These mice spontaneously develop the autoimmune dacryoadenitis
(inflammation of lacrimal gland, LG) that is characteristic of Sjögren’s Syndrome (SS). SS is a
systemic autoimmune disease associated with lymphocytic inflammation and loss of function of
LG and salivary glands (SG), leading to severe dry eye and dry mouth and development of
systemic symptoms (B. H. Lee, Tudares, & Nguyen, 2009; Nikolov & Illei, 2009). The male NOD
mouse shares many disease characteristics with SS with respect to LG inflammation and
dysfunction which manifest between 3-4 months, prior to development of diabetes (Anderson &
Bluestone, 2005; Y. Ju et al., 2018; Many et al., 1996). Female NOD mice do not develop LG
inflammation and loss of function; instead females develop a later and profound salivary gland
infiltration and loss of function concurrent with manifestation of extensive diabetes between 4-5
months (Humphreys-Beher, 1996; Lavoie et al., 2011; Many et al., 1996; Takahashi et al., 1997).
Due to the different patterns of organ inflammation, which affect ICAM-1 tissue expression, and
the concurrent development of diabetes, we did not use female mice for this study. Here we report
on the successful production of an ICAM-1 targeted ELP based on the A192 backbone that binds
selectively to cells expressing ICAM-1, and which has significantly higher accumulation in the
LG, a site of ICAM-1 enrichment in the NOD mouse model and in SS.
33

2.2 Material and Methods
Material and Reagents. bEnd.3 cells (ATCC® CRL-2299™) were from ATCC
(Manassas, VA). SHuffle® T7 Competent E. coli was from New England BioLabs Inc. (Ipswich,
MA). DNA oligos encoding ICAM-1 Binding Peptide with BseRI sticky ends
(/5Phos/GATTACCGACGGCGAAGCGACCGATAGCGGCGG,
/5Phos/GCCGCTATCGGTCGCTTCGCCGTCGGTAATCCC) was synthesized by Integrated
DNA Technologies (Coralville, IA). The plasmid expressing mICAM-1 turboGFP was from
Origene (Rockville, MD). Rapamycin was from LC Laboratories (Woburn, MA, USA). NHS-
Fluorescein, NHS-Rhodamine and Zeba™ Spin Desalting Columns, 7K MWCO (10 mL), and
LysoTracker™ Green DND-26 were from ThermoFisher Scientific Inc. (Rockford, IL). Sulfo-
Cyanine 7.5 NHS ester was from Lumiprobe Corp (Hallandale Beach, FL). Goat anti-mouse
ICAM-1 polyclonal antibody was from R&D Systems (Minneapolis, MN). The CytoSelect  Cell
viability and Cytotoxicity assay kit was from Cell Biolabs (San Diego, CA). Other reagents were
from standard suppliers.
ELP Design and Cloning. The modified pET25b(+) containing A192 was synthesized
using plasmid recursive directional ligation as described previously (J. Despanie et al., 2016;
McDaniel, Mackay, Quiroz, & Chilkoti, 2010). AF was cloned by fusing FKBP, the cognate
receptor for Rapa, to the carboxy terminus of A192 using two-step cloning as previous described
(J. P. Dhandhukia et al., 2017). For IBPAF cloning, two complementary DNA oligos encoding
IBP with BseRI sticky ends were annealed. The annealed product was inserted into the 5’ end of
a modified pET25b(+) vector containing AF digested with BseRI. For the cloning of IBPA192,
DNA oligos encoding IBP with BseRI sticky ends were fused onto the 5’ end of A192 at the BseRI
34

cutting site, respectively. The in-frame amino acid sequence was confirmed by diagnostic
digestion and DNA sequencing.
ELP Purification, Biophysical Characterization and Fluorescent labeling. AF and
IBPAF were purified according to standard protocols published previously(S. Aluri, Pastuszka,
Moses, & MacKay, 2012; D. E. Meyer & Chilkoti, 2002). Fusion protein polymers were analyzed
for purity by Coomassie blue SDS-PAGE gel analysis. The molecular weights of AF and IBPAF
were confirmed by Electrospray ionisation time-of-flight mass spectrometry (ESI-TOF). The
temperature-concentration phase diagram and hydrodynamic radius (Rh) at 37°C was determined
as previously described (S. Aluri et al., 2012; Guo et al., 2018). For fluorescent visualization in
both in vitro and in vivo studies, AF and IBPAF were covalently modified with NHS-Fluorescein,
NHS-Rhodamine, or Sulfo-Cyanine7.5 (Cy7.5) NHS esters, respectively, as previously described
(Peddi, Pan, & MacKay, 2018; Sreekumar et al., 2018).
Isothermal Titration Calorimetry The drug binding affinity between Rapa and AF or
IBPAF was evaluated by MicroCal PEAQ Isothermal Titration Calorimetry according to
previously reported methods (J. P. Dhandhukia et al., 2017). Briefly, titrations were performed at
37°C with Rapa at a concentration of 8 µM, in a volume of 280 µl in the calorimeter cell and with
protein at a concentration of 100 µM in the titration syringe, while 12 injections of the protein into
the drug were made. The drug and the protein were solubilized in the same buffer (1% v/v DMSO
in PBS) to prevent background heat of release due to differences in buffer composition.
Cell Surface Binding and Cellular Uptake. For cellular uptake of ELPs by 293T cells,
293T cells were transfected with mouse ICAM-1. Cells were then treated with 50 μM Rhodamine-
labeled A192, IBP-A192 prepared at 4°C for 30 min followed by extensive washing and 2 h
35

incubation at 37°C. Cells were processed for live cell imaging using a ZEISS LSM 800 confocal
microscope equipped with an Airyscan detector (Zeiss, Thornwood, NY).
For detection of cell surface-bound ELPs, bEnd.3 cells were first treated with 200 U/ml
mouse recombinant TNF-α at 37°C for 18 h, then washed and treated with 50 μM AF, IBPAF or
PBS respectively at 4°C for 30 min. After washing with PBS to remove unbound ELPs, cells were
fixed, permeabilized and labeled with mouse anti-ELP antibody and Alexa Fluor® 488 goat anti-
mouse secondary antibody followed by imaging by confocal fluorescence microscopy.
For detection of ELP internalization after surface binding, TNF-α treated bEnd.3 cells were
treated with 50 μM AF or IBPAF, respectively, at 4°C for 30 min. After extensive washing, cells
were incubated at 37°C for 2 h followed by imaging of live cells by confocal microscopy. 1 μl of
LysoTracker™ Green DND-26 was added to the cell culture media to visualize lysosomes
immediately prior to imaging. For ICAM-1 inhibition experiments, cells were pretreated with
5 µg/mL goat anti-mouse ICAM-1 antibody prior to IBPAF treatments. Cells without TNF-α
stimulation served as a control. Fluorescence intensity was calculated by ImageJ (1.49V,
http://imagej.nih.gov/ij/; available in the public domain by the National Institutes of Health,
Bethesda, MD, USA) as integrated fluorescence intensity within each cell divided by the area of
the cell. Results are plotted as fold-change fluorescence intensity relative to the AF group.
Flow cytometry. TNF-α stimulated bEnd.3 cells were trypsinized into single-cell
suspensions prior to incubation with Fluorescein-labeled AF, IBPAF or PBS at 4°C for 30 min
with mild shaking. Cells were washed with ice-cold PBS and processed for further analysis using
an LSRII Flow Cytometer (BD Biosciences, San Jose, CA).
36

Real-Time PCR. Total RNA was isolated from different organs of NOD and BALB/c mice
and converted to cDNA. Quantitative (real-time) PCR (qPCR) was performed with TaqMan gene
expression assays using probes for ICAM-1 (Mm00516024_g1) and GAPDH (Mm99999915_g1).
The relative expression levels were calculated using the comparative CT method (ΔΔCT method)
as described previously (Wu et al., 2009).
Western Blotting. Whole cell lysates or tissue lysates were prepared using RIPA buffer
containing protease inhibitors as previously described (Li et al., 2010). 40 μg of total protein was
resolved on PAGEr EX 10% gradient gels and then transferred to nitrocellulose membranes. The
membranes were incubated with anti-ICAM-1 and anti-actin primary antibodies, followed by
secondary antibodies and imaged using a Li-Cor Odyssey Scanning Infrared Fluorescence Imaging
System (Lincoln, NE).
Immunofluorescence Labeling. Mouse LG were processed as previously described
(Meng et al., 2016; Meng et al., 2017). Briefly, LGs were fixed, embedded in O.C.T. compound
and flash frozen on dry ice. Blocks were cryosectioned at 5 µm thickness and mounted on glass
slides. After being quenched and permeabilized as described, tissue sections were blocked with 1%
BSA followed by incubation with primary and secondary antibodies. Finally, samples were
mounted with ProLong anti-fade mounting medium and imaged using a ZEISS LSM 800 confocal
microscope equipped with an Airyscan detector (Zeiss, Thornwood, NY).
Drug Release in Plasma. Rapa was first encapsulated onto AF or IBPAF using a two-
phase solvent evaporation method as previously described (Mihir Shah et al., 2013). Briefly, 200
μM AF or IBPAF in PBS was mixed with a 2X molar access of Rapa dissolved in 70% hexane/30%
ethanol. The vial was stirred under dry nitrogen to drive evaporation of hexane/ethanol. Excess
drug was removed by centrifugation at 13.2K RPM, 20 min followed by filtration through a 0.2
37

µm syringe filter. Residual organic solvent was further removed by overnight dialysis with cut-off
molecular weight of 10kDa. To determine the amount of encapsulated drug, the sample was
injected into a C-18 reverse phase HPLC column and eluted in an water: methanol gradient from
40 % to 100 %. Rapa concentration was calculated using a standard curve.
To mimic the in vivo drug release environment, 500 μL of 160 μM AF-Rapa or IBPAF-
Rapa was mixed with equal volume of mouse plasma respectively follow by dialysis under sink
condition against PBS at 37°C. 50 μL of the mixture from dialysis cassette was collected 2 h, 4 h,
8 h, 24 h and 48 h after dialysis and proceeded with high-performance liquid chromatography to
measure Rapa concentration. Another 100 μL of the mixture was also collected at the same time
points and ELPs was separated by doing one round of ITC. ELP pellets were resolubilized into
100 μL of cold PBS and proceed with HPLC to measure the amount of Rapa retained on ELP.
Data is plotted as percentage of initial concentration versus time.
Animals. Male BALB/c controls and NOD mouse breeders were purchased from Taconic
Farms (Hudson, NY). Animals were bred in the University of Southern California (USC) Vivarium.
All animal procedures were approved by the USC Institutional Animal Care and Use Committee
and performed in accordance with the Guide for the Care and Use of Laboratory Animals 8th Ed
(2011).
Pharmacokinetic Analysis. Male NOD mice aged 14-16 weeks were injected
intravenously with 150 μl of 200 μM rhodamine-labeled ELPs (n=5). 20 μL of blood was collected
from a tail nick at 5 min, 30 min, 1, 2, 4, 8, 12, 24, 36, 48, and 72 h after injection and immediately
diluted into 80 μL of heparinized PBS. The blood was centrifuged at 16,100 rcf for 10 min at 4°C,
the supernatant was loaded onto a 384-well plate, and fluorescence was read using a SpectraMax
iD3 Multi-Mode Microplate Reader (Molecular Devices, San Jose, CA) with excitation and
38

emission wavelengths of 542 nm and 585 nm, respectively. After subtracting autofluorescence
measured from the plasma of non-injected mice, the fluorescence intensity values were converted
to concentration using a linear standard curve.
In a separate study, male NOD mice aged 14-15 weeks old were injected with 150 μl of
150 μM free Rapa (formulated in 90% PBS + 50% PEG400 + 5% Tween20), AF-Rapa, IBPAF-
Rapa through tail vein. Plasma was collected by cardiac puncture when the mice were under
general anesthesia and subjected to Liquid chromatography–mass spectrometry (LC-MS) to
measure Rapa concentration.
Liquid Chromatography–Mass Spectrometry. 50ul of plasma was mixed with 50ul of
500 ng/mL Tacrolimus which served as internal control. Both Rapa and internal control was
extracted by adding 500 μl Acetonitrile followed by incubation at -20 for 30 min. Supernatant was
collected after spinning at 13,000rpm for 5min and dried. The pellet was reconstituted in 80%
acetonitrile and further analyzed by Agilent 1200 linked onto API4000. The analytes were
separated using a C18 Kinetex column (Phenomenex) with the following dimension 50 ×3.0 ×2.6
μm. The mobile phase utilized a gradient system consisting of two components, where component
A was water with 0.5 % formic acid and component B was acetonitrile with 0.5 % formic acid.
The program consisted of 50 % Component B at time 0), where the concentration increased to 90%
after two minutes and held for another five minutes. Rapa and Tacrolimus were quantified using
MRM of 931.7 → 865.0 and 821.5 → 768.7, respectively.
In Vivo Biodistribution. Male NOD mice aged 14-16 weeks were injected I.V. with 150
μl of 200 μM Cy7.5-labeled ELPs and euthanized at the designated time points. Major organs were
collected, weighed and scanned for fluorescence by the IVIS In Vivo Imaging System (Perkin
Elmer, Waltham, MA). Fluorescence of individual organs was quantified and plotted as Radiant
39

Efficiency per milligram of organ by subtracting background from organs from an untreated mouse,
and then normalized to the weight of the organ. In a separate experiment, male NOD mice aged
14-15 weeks old were intravenously injected with 150 μl of 185 μM free Rapa (formulated in 90%
PBS + 50% PEG400 + 5% Tween20), AF-Rapa, IBPAF-Rapa. Both LG were collected and
scanned for fluorescence by the IVIS In Vivo Imaging System and quantified with the same method
as above.
Tissue lysates were also obtained and assessed for fluorescence intensity. Briefly, a small
fraction of each organ was weighed and homogenized in RIPA buffer. The supernatant was
collected after centrifugation at 20,000 x g, 4°C for 5 min. The fluorescence intensity was read
using the SpectraMax iD3 Multi-Mode Microplate Reader (Molecular Devices, San Jose, CA)
with excitation and emission wavelength of 780 nm and 845 nm, respectively. Fluorescence
intensity was analyzed after subtracting the autofluorescence in homogenate of the respective
organs of a mouse without injection, and normalized to weight of the organ fraction.
Statistics. Data presented are representative curves or mean ± SD. All experiments were
repeated at least three times. Statistical analyses were performed using a Student’s t-test or a one-
way ANOVA followed by Tukey’s post-hoc test using Prism (GraphPad Software, La Jolla, CA).
A p value of<0.05 was considered statistically significant.
2.3 Results
Evaluation of ICAM-1 levels in LG and other tissues in male NOD and BALB/c mice.
ICAM-1, a cell surface glycoprotein that regulates cellular adhesion and immunoreaction, is
upregulated in inflamed organs in many autoimmune conditions such as rheumatoid arthritis
(Ishikawa et al., 1994), systemic lupus erythematosus (Belmont, Buyon, Giorno, & Abramson,
40

1994), multiple sclerosis (Girgrah et al., 1991), as well as SS (St Clair, Angellilo, & Singer, 1992).
We confirmed increased ICAM-1 tissue expression in the LG by Western blotting and
immunofluorescence microscopy in tissues from the male NOD mouse, a model of the LG
inflammation in SS, relative to healthy age-matched male BALB/c mice. An apparent increase in
total LG ICAM-1 content was observed in NOD mice compared to BALB/c mice by Western
blotting (Fig 1A), consistent with previous findings in SG and LG biopsy samples from SS patients
(Saito et al., 1993). It should be noted that while male NOD mice share many pathological and
symptomatic similarities in the LG with SS patients, they do not develop the characteristic
autoimmune sialoadenitis seen in SS (Toda et al., 1999; Verheul, Verveld, Hoefakker, & Schuurs,
1995). Immunofluorescence labeling showed that ICAM-1 was localized to the vascular
endothelium, infiltrating lymphocytes, and the basolateral membranes of LG acinar cells in the
male NOD mouse LG at both early (12 week) and later (20 week) stages of autoimmune
dacryoadenitis (Fig 1C). ICAM-1 gene expression levels in LG, SG and other major organs in
male NOD mice relative to healthy control mice were also compared by RT-PCR. There was a
significant elevation in ICAM-1 gene expression in the LG of NOD mice compare with LG of
BALB/c mice, however, no significant differences were seen in salivary gland, lung, liver, kidney,
spleen and pancreas between NOD and BALB/c mice (Fig 1B). A similar trend was also seen for
protein levels of ICAM-1 assessed in different tissue of BALB/c and NOD mice by Western
blotting (Fig 1D,E).
41

42

Figure 1. ICAM-1 is increased in the inflamed LG of male NOD mice and can be targeted by
IBP. A) Western blotting of ICAM-1 levels in LG lysates from 12 week old male BALB/c and
NOD mice. Result shown is representative of n=3 experiments. B) Gene expression of ICAM-1
normalized to GAPDH in different organs from 14 week old male NOD and age-matched male
BALB/c mice obtained by qPCR (n=3, data are mean ± SD, *p<0.05. C) Immunofluorescence
labeling for ICAM-1 in LG sections from 12 week (12W) and 20 week (20W) old male BALB/c
and NOD mice (results shown are representative of n=3 mice each, Green: ICAM-1, Red: F-Actin,
Blue: DAPI. Scale bar=20 μm). D) A representative Western blot of ICAM-1 in different tissues
from BALB/c and NOD mice. E) Quantification of ICAM-1 level normalized to GAPDH in
different tissues of BALB/c and NOD mice (n=3).F) HEK-293T cells were transfected with
mICAM-1-turboGFP plasmid (green), then treated with 50 μM of either Rhodamine-labeled A192
(control) or IBPA192 at 4°C for 30 min. Unbound ELPs were removed by washing with ice-cold
PBS and cells were incubated in culture medium for 2 hr at 37°C before imaging by live cell
confocal fluorescence microscopy. Red: Rhodamine-labeled ELPs, Green: mICAM-1-turboGFP,
Blue: DAPI. Asterisk: mICAM-1-turboGFP transfected cells, Arrows: adjacent non-transfected
cells. Scale bar=20 μm.

Identification of a peptide-ELP fusion that targets murine ICAM-1. Several targeting
peptides have been described for human ICAM-1 (Chittasupho et al., 2009; Feng et al., 1998;
Manikwar et al., 2011; Tibbetts, Seetharama Jois, Siahaan, Benedict, & Chan, 2000); however, it
was unclear whether these peptides would have affinity for murine ICAM-1 since most were tested
on human cell lines. Amino acid sequence alignment reveals only 52.69% identity between human
and mouse ICAM-1, reinforcing this concern, which would limit in vivo analysis in our mouse
model. A previously published peptide, ICAM-1 binding peptide (IBP) drew our attention due to
its ability to bind to murine ICAM-1 and to inhibit ICAM-1-mediated intercellular adhesion
(Belizaire, Tchistiakova, St-Pierre, & Alakhov, 2003). IBP also retained its binding specificity
toward ICAM-1 when fused to the N-terminus of another protein without compromising the
biological function of that protein (Belizaire et al., 2003). Here, IBP was fused to the N-terminus
of A192 (IBPA192), and this construct was tested on mouse ICAM-1 (mICAM-1)-GFP transfected
HEK 293T cells. IBPA192 was specifically internalized into the mICAM-1=GFP overexpressing
cells, while no obvious internalization was seen in cells exposed to the control, A192 (Fig 1F).
43

Noticeably, IBPA192 was only internalized into ICAM-1 overexpressing cells, which appear green
in the images. In contrast, IBPA192 was not observed in adjacent non-transfected cells, which
confirms IBP has specificity for mouse ICAM-1. Based on this data, we chose IBP as an mICAM-
1 targeting peptide for further development in our mouse model.
Expression and Characterization of ELP fusions. The 12 kDa FK506-binding protein
(FKBP) is the human cognate receptors for Tacrolimus and Rapa. Our group has previously fused
FKBP to ELPs for the purpose of binding and carrying Rapa in several pre-clinical therapeutic
studies. (J. P. Dhandhukia et al., 2017; M. Shah et al., 2013) A bi-functional ELP fusion protein,
IBP-A192-FKBP (IBPAF) could theoretically target ICAM-1 with the IBP moiety at the N-
terminus, as well as carry Rapa by utilizing an FKBP moiety at the C-terminus. This construct
was cloned and purified with a yield of 40 mg/L in transiently transfected Shuffle T7 E.coli.
Another non-targeted control, A192-FKBP (AF), containing FKBP only at the C-terminus, was
expressed in the same cell line. The purity and molecular weights of these fusion protein polymers
were determined using SDS-PAGE (Fig 2A). The purity of the AF and IBPAF was 99.6 ± 0.3%
and 99.0 ± 0.1% across different stocks, respectively; these values were typical of stocks utilized
throughout subsequent studies. The molecular weights of AF and IBPAF were confirmed as 85.1
kDa and 86.6k Da, respectively, by electrospray ionisation time-of-flight mass spectrometry,
which agrees with their predicted molecular weights of 85.4 kDa and 87.1 kDa, respectively. These
and subsequent studies extensively utilized fluorescently-labeled IBPAF and AF constructs--for
all such studies with fluorescently-labeled ELPs, SDS-PAGE was utilized to confirm there were
minimal levels of excess dye left in the material. Purified fusion protein polymers were analyzed
using Dynamic Light Scattering (DLS) to determine the hydrodynamic radii (Rh) at physiological
temperature. The Rh of IBPAF and AF (25 μM each) at 37°C was 7.1 nm and 6.6 nm, respectively
44

(Table 1). No significant protein aggregation was detected by DLS, suggesting that both constructs
were soluble at physiological temperatures (Fig 2B). Binding of Rapa did not significantly change
the hydrodynamic radius of the carrier, nor cause its aggregation or precipitation of drug (Fig 2C).
The hydrodynamic radius of the ELP-Rapa did not change significantly after 24 h and 48 h
incubation at 37 °C, indicating that the carrier-drug complex was relatively stable at 37 °C. Optical
density measurements at 350 nm of AF and IBPAF were used to characterize the temperature-
concentration phase diagram for both constructs. Transition temperature (Tt) follows an inverse
relationship with logarithm of the ELP concentration (Fig 2D) with R
2
equal to 0.95 for AF and
0.96 for IBPAF, respectively. Using the fit parameters (Table 1) for AF and IBPAF, it was possible
to estimate their solubility profiles at different temperatures and concentrations, and to extrapolate
that both would remain completely soluble at physiological temperature at relevant concentrations
(1-500 μM). Cytotoxity and hemolysis assays were also performed to evaluate the safety of AF
and IBPAF ( Fig 3). No significant cytotoxicity and hemolysis was seen for either AF or IBPAF
for either short or long time treatment, suggesting good safety and biocompatibility.

45

Table 1. Biophysical properties of ELP fusion proteins evaluated in this study
Constructs Amino Acid Sequence

Predicted M.W.
b

Observed M.W.
c

Temperature-Concentration
Phase Diagram
d

R
h
e
at 37 °C

[nm]
Slope, m
[°C
Log(µM)]
Intercept, b
[°C]
AF MG(VPGAG)
192
-FKBP
a
-Y 85.4 85.1 -5.6 ± 0.3 65.8 ± 0.7 7.1 ± 0.1
IBPAF MFEGFSFLAFEDFVSSIG(VPGAG)
192
-FKBP
a
-Y 87.1 86.6 -5.3 ± 0.2 65.1 ± 0.4 6.6 ± 0.2
A192
MG(VPGAG)
192
-Y
73.6 N.A. -8.3 ± 0.2 74.7 ± 0.8 7.7 ± 0.9
IBPA192
MFEGFSFLAFEDFVSSIG(VPGAG)
192
-Y
75.4 N.A. -8.9 ± 0.9 73.0 ± 1.5 7.7 ± 0.1
a: FKBP amino acid sequence:
GVQVETISPGDGRTFPKRGQTCVVHYTGMLEDGKKFDSSRDRNKPFKFMLGKQEVIRGWEEGVAQMSVGQRAKLTISP
DYAYGATGHPGIIPPHATLVFDV ELLKLE
b: predicted M.W. based on amino acid sequence
c: experimental M.W. determined by running samples on ESI-TOF
d: Phase diagrams for assembly (Fig 3D) were fit with Equation 2. Values represent mean ± 95% CI.
e: Rh, hydrodynamic radius of 25 µM samples determined by Dynamic Light Scattering (n=3, mean ± SD).

46

Figure 2. AF and IBPAF have similar purity, molecular weight, hydrodynamic radius, phase
behavior, Rapa binding affinity and Rapa release profiles. A) Coomassie blue staining of
purified AF and IBPAF resolved by 10% SDS-PAGE. B) Hydrodynamic radii versus percentage
by mass plotted for AF and IBPAF as obtained by Dynamic Light Scattering. C) Hydrodynamic
radii versus percentage by mass plotted for AF-Rapa and IBPAF-Rapa complex. D) Concentration
versus Transition Temperature (Tt) phase diagrams of AF and IBPAF measured using optical
density at 350 nm. The 95% confidence interval around each best-fit line is indicated with dashed
lines. (n=3, mean ± SD). Isothermal titration calorimetry (ITCal) was used to confirm the proper
protein folding and Rapa-binding activity of the FKBP domain on both proteins. The heat pulse
generated by serial injection of E) AF and H) IBPAF into a sample cell containing Rapa at 37 °C
with respect to time is shown. Stoichiometry and thermodynamic curves of Rapa binding to F) AF
and I) IBPAF at 37°C. A representative figure is shown from n=3 independent experiments. Drug
release in plasma was evaluated by mixing 500 μL of 160 μM G) AF-Rapa or J) IBPAF-Rapa with
an equal volume of mouse plasma followed by dialysis under sink conditions against PBS at 37°C.
The amount of Rapa remaining in the dialysis cassette and retained on the ELP was measured by
HPLC at 2 h, 4 h, 8 h, 24 h and 48 h (n=3 at each time point).

47

Drug binding and release of AF and IBPAF. To study the binding affinity and
thermodynamics between Rapa and ELP-FKBPs, Isothermal Titration Calorimetry (ITCal) was
performed. This technique directly measures the heat that is either released or absorbed during a
biomolecular binding event, thus precisely determining the reaction stoichiometry (n), binding
constants (Kd), enthalpy (∆H) and entropy (ΔS). Rapa was titrated against successive injections of
ELP-FKBPs until saturation. The heat released from each injection was measured for all ELP-
FKBPs (Fig 2E,2H) and the data was fitted to a ‘one set of sites’ binding model (Fig 2F, 2I),
which enabled estimation of binding stoichiometry and thermodynamic parameters (Table 2). The
stoichiometry of Rapa bound to AF and IBPAF was 1, confirming that nearly all of the FKBP
domains were functionally able to bind drug. The binding affinity at 37 °C between Rapa and AF
and Rapa and IBPAF was 5.1 nM and 2.7 nM, respectively, close to the values reported previously
for other N-terminal FKBP-ELP fusion proteins using this method (J. P. Dhandhukia et al., 2017).
Table 2. Thermodynamic parameters of ELP-FKBP interactions with Rapa
N(sites) Kd (nM) ∆H (kJ/mol) ∆G(kJ/mol) -T∆S (kJ/mol)
AF 1.000±0.00 5.1±0.1 -75.8±0.8 -49.3±0.8 26.5
IBPAF 1.040±0.005 2.7±0.3 -69.3±1.0 -50.9±1.0 18.4
All experiments were performed at 37°C. Binding isotherms were fitted to a ‘one set of
sites’ binding model to generate binding stoichiometry and thermodynamic parameters. Data is
presented as mean ± SD.

Rapa is known to exhibit extensive protein binding with plasma protein such as albumin,
which leads to low plasma free drug concentration (S. R. MacEwan et al., 2014; R. W. Yatscoff et
al., 1995). Here we designed an in vitro release assay to test whether the ELP-Rapa complexes
would lose the drug to plasma protein in a short period of time. ELP-Rapa constructs were mixed
48

with mouse plasma and dialysed against PBS under sink conditions at 37°C to mimic in vivo
release and elimination conditions. The amount of Rapa in each ELP-plasma mixture (Fig 2G-
Plasma,2J-Plasma) versus that retained on each ELP (Fig 2G-AF,2J-IBPAF) was measured at
different time points. The amount of Rapa in the mixture was unchanged up to 48 h, suggesting no
significant loss of Rapa from the mixture. Both AF and IBPAF retained around 80% of Rapa with
only a small fraction of Rapa lost to plasma protein up to 48 h (Fig 2G,2J). Altogether, these data
suggest that both AF and IBPAF are able to bind Rapa with high affinity and retain the majority
of the drug when mixed with mouse plasma.

Figure 3. Cytotoxicity and biocompatibility of AF and IBPAF. A) An MTT assay was done to
determine the cytotoxicity of AF and IBPAF at the indicated doses of 10 and 20 M after 2h or
24h incubation with the mouse hepatocyte cell line 644 at 90% confluency. No significant
differences were seen in any ELP treatment groups versus PBS treatment, while 0.1% saponin
provided a positive control for cell death (n=3). B) A hemolytic assay was performed to test
whether ELP treatments caused hemolysis of mouse red blood cells. Water was used as a positive
control to promote lysis, and PBS was used as a negative control for no lysis. No significant
differences were seen in any ELP treatment group versus PBS (n=3). C) Weights of liver, kidney
and spleen were measured 24h after mice were injected with 150 μl of 200 μM AF, IBPAF or PBS.
No significant differences were seen between AF, IBPAF and PBS (n=5 for AF and IBPAF, n=2
for PBS).

49

Binding and Internalization of IBPAF in TNF-α stimulated bEnd.3 cells. BEnd.3 cells
are a mouse endothelial cell line that endogenously expresses ICAM-1, and ICAM-1 expression
can be further significantly induced in these cells by treatment with TNF-α (Hahne, Jager,
Isenmann, Hallmann, & Vestweber, 1993; McHale, Harari, Marshall, & Haskard, 1999). Increased
expression of ICAM-1 on bEnd.3 cells after TNF-α stimulation was confirmed by Western blotting
(Fig 4A). Surface binding of fluorescein-labeled AF and IBPAF to ICAM-1 on bEnd.3 cells was
examined by flow cytometry after 30 min of binding at 4°C. A 4-fold increased surface binding of
IBPAF was seen compared to AF (Fig 4B-C). Furthermore, immunofluorescence labeling with
anti-ELP antibody confirmed the specific cell surface binding of IBPAF, detected as puncta on the
cell membrane, while minimal labeling was detected in the absence of ELP or with AF (Fig 4D).
Based on these results, we concluded that IBPAF binds to mICAM-1 on the cell surface.
Table 3. Pharmacokinetic parameters of AF and IBPAF in NOD mice following systemic
administration

Analysis Method
One Compartment Model Non-Compartmental
AF IBPAF AF IBPAF
AUC(μm
h)
121±12 168±23
*
184±16 196±34
AUMC
(μM h
2
)
nd nd 1917±469 1538±254
CL (mL
h
-1
)
0.28±0.03 0.20±0.03
#
0.18±0.02 0.17±0.03
Vd (mL) 2.3±0.3 1.9±0.6 2.3±0.4 1.3±0.7
MRT (h) nd nd 9.8±3.1 8.4±0.7
50

T
1/2
,
Elimination (h)
5.7±0.4 6.6±1.4 8.9±3.2 5.5±2.9
kelimination
(h
-1
)
0.12±0.01 0.11±0.02 0.08±0.01 0.13±0.07
*A statistically significant difference was detected between AF and IBPAF. (*p=0.0037,
#p=0.0022, n=5, data is mean ± SD)

51

Figure 4. IBPAF binds efficiently to TNF-α stimulated bENd.3 cells relative to AF. A)
Western blotting for ICAM-1 in bENd.3 cells without (nontreated) and with TNF-α stimulation.
B) Flow cytometry was used to quantify surface binding of AF and IBPAF in TNF-α stimulated
bENd.3 cells. Cells were treated with 50 μΜ Fluorescein-labeled AF or IBPAF at 4°C for 30 min
before flow cytometry analysis. C) Quantification of the percentage of fluorescence-positive cells
after AF- and IBPAF-treatment. An unpaired t-test was performed to compare the difference
between AF and IBPAF. (*p ≤ 0.05, n=4, data is mean ± SD) D) Detection of surface bound ELP-
FKBPs using an anti-ELP antibody, AK1. Cells were treated with 50 μM AF or IBPAF at 4 °C for
30 min, respectively. Then cells were processed for immunofluorescence labeling using anti-ELP
antibody and an appropriate secondary antibody as described in Methods. Green: ELP, Blue: DAPI.
Scale bar=20 μm. E) TNF-α stimulated bENd.3 cells were treated with 50 μM of Rhodamine-
labeled AF or IBPAF at 4°C for 30 min, washed with PBS to remove unbound ELPs, and incubated
in fresh medium at 37°C for 2 h followed by live cell confocal laser scanning microscopy. For
anti-ICAM-1 antibody treatment experiments, cells were pretreated with 5 µg/mL anti-ICAM-1
antibody at 4°C for 30 min prior to ELP treatment. Uptake of IBPAF into bENd.3 cells without
TNF-α stimulation was also assessed. Scale bar=20 μm. F) Quantification of internalized
rhodamine signal normalized to AF treated cells. Data is presented as mean ± SD. A one-way
ANOVA with Tukey’s multiple comparison test was used to calculate the significance. (n=10
images per treatment per preparation from 3 separate preparations) (***p ≤ 0.001).

We further assessed whether surface-bound IBPAF was internalized into the cells.
Rhodamine labeled IBPAF was colocalized with Lysotracker® after 2 h of incubation, while no
significant internalization of AF was seen (Fig 4E), likely due to lack of surface binding as shown
in Fig 3. Moreover, in the cells without TNF-α stimulation, no significant internalization of IBPAF
was detected, likely also due to low levels of surface ICAM-1 expression and thus, minimal surface
binding (Fig 4E). To test the specificity of the IBPAF-ICAM-1 interaction, excess ICAM-1
antibody was used to inhibit the binding site of IBPAF prior to addition of IBPAF to the cells.
Antibody pre-incubation significantly decreased fluorescence associated with IBPAF inside the
cells (Fig 4E).Quantification of fluorescence intensity normalized to the area of the cell showed a
significantly higher value for IBPAF in TNF-α treated cells, relative to its accumulation in non-
treated cells or in anti-ICAM-1 antibody pre-treated cells, and relative to AF accumulation under
52

all conditions (Fig 4F). Collectively, these results suggest IBPAF specifically binds to ICAM-1
and is internalized into the cell after binding.
Pharmacokinetics of AF and IBPAF. The pharmacokinetics of AF and IBPAF were
compared by I.V. administration of rhodamine-labeled ELPs to 14-16 week male NOD mice and
collection of blood samples at various time intervals after administration. The ELP plasma
concentration versus time curve for AF and IBPAF are depicted in Fig 5B and Fig 5C, where both
AF and IBPAF show one-phase decay behavior. Two different methods were used to analyze the
concentration-time curve of each individual mouse: compartmental model by SAAM II and non-
compartmental analysis.
With compartmental model analysis, both AF and IBPAF fit into the one-compartment
model with I.V. bolus shown in Fig 5A, while the relevant pharmacokinetic parameters are shown
in Table 3. The one compartment model assumes that the entire body acts like a single, uniform
compartment, drug distributes/equilibrates instantaneously and rapidly throughout the
compartment, and drug elimination from the compartment also begins to occur immediately after
the I.V. bolus injection. The volume of distribution of both AF and IBPAF was close to the plasma
volume of a mouse, indicating that the ELP did not rapidly distribute to specific organs and tissues
directly after administration. A significantly higher AUC (168±23 vs. 121±12 μM h) and lower
clearance (0.20±0.03 vs. 0.28±0.03 mL h
-1
) was observed for IBPAF when compared with AF;
however, no statistical difference in half-life (6.6±1.4 vs. 5.7±0.4 h) or elimination rate constant
(0.11±0.02 vs. 0.12±0.01 h
-1
) was observed between AF and IBPAF (Table 3).
In contrast, when analyzed with the non-compartmental model, no statistical difference
was detected in any parameters between AF and IBPAF, which indicates that they have similar in
vivo distribution and elimination properties. These pharmacokinetic parameters, such as a long
53

half-life (5-9 h) and low volume of distribution (i.e., nearly equal to the plasma volume), are
considered favorable for delivery of therapeutics. Adding the targeting peptide neither
significantly changed the pharmacokinetic properties of ELP-FKBP nor led to premature clearance
from plasma.
To test whether ELP carrier could better retain Rapa in the systemic circulation, the
concentration of Rapa was measured in plasma 2 h after I.V. injection of free Rapa, AF-Rapa, or
IBPAF-Rapa to NOD mice. There was a significant increase in plasma concentration of Rapa when
it was delivered by AF or IBPAF compared with free Rapa (Fig 5D), which suggests that the
FKBP/Rapa interaction remains intact even after two hours in circulation.

54

Figure 5. AF and IBPAF have similar pharmacokinetic profiles following systemic
administration. 14-16 week old male NOD mice were administered I.V. with ELP-FKBPs (150
uL, 200 uM Rh-ELP). A) A one compartment model fit the observed concentration-time profiles
for rhodamine-labeled ELP-FKBPs. B-C) Plasma concentration versus time profiles for B) AF or
C) IBFAF were estimated in each subject using diluted samples of plasma (n=5 mice per treatment,
mean ± SD). D) To confirm that the FKBP remains bound with Rapa, the drug was dosed I.V. (150
uL, 150 uM Rapa) either free or bound to ELP-FKBPs. After 2 h, the plasma concentration of
Rapa in each subject was evaluated using a calibrated LC/MS assay. AF-Rapa and IBPAF-Rapa
retained significantly more drug in the plasma than did free-Rapa, consistent with the drug
remaining complexed with both carriers at this timepoint. (n=5 mice per group, mean ± SD, ***P
≤ 0.001).

In Vivo Biodistribution of AF and IBPAF. The biodistribution of AF and IBPAF given
by I.V. injection was assessed at 2 h, 8 h, and 24 h after I.V. injection (Fig 6). A near infrared
radiation dye, Cy7.5, was used to label the ELPs for its better tissue penetration and longer
emission wavelength where a minimal interference is given by tissue autofluorescence. The
amount of ELP per milligram of organ was measured from LG, SG and other major organs. The
autofluorescence of each organ was calculated from an uninjected mouse and subtracted from the
measured fluorescence intensity. At the 2 h time point, there was significantly more IBPAF
accumulation in the LG compared with AF (Fig 6A), likely related to the higher ICAM-1 levels in
the LG of NOD mice of this age. No significant differences between AF and IBPAF were seen in
other organs. There was also a trend to higher accumulation of both AF and IBPAF in liver and
spleen when compared with other organs, perhaps due to the discontinuous capillaries in these two
organs leading to higher retention. At 24 h, the increased IBPAF accumulation in LG was
diminished and there was no significant difference between AF and IBPAF. There was
significantly less IBPAF in the LG 24 h after injection compared with 2 h after injection, however,
no difference was seen in AF between any two points, which suggests intracellular degradation of
IBPAF may follow its ICAM-1 mediated endocytosis (Fig 6A).
55

56

Figure 6. Relative to AF, IBPAF accumulation is transiently enhanced in LG of male NOD
mice. Fluorescently-labeled ELPs were administered I.V. to 14-16 week old male NOD mice.
Tissues were obtained after 2, 8, or 24 h, homogenized, and quantified for total fluorescence. The
percentage of the initial dose per gram of tissue was then plotted in the: A) LG; B) SG; C) lung;
D) liver; E) kidney; F) spleen; and G) pancreas. A t-test was done to compare AF and IBPAF
within each organ at each time point. Data are mean ± SD (n=5 mice per group per time point, *p
≤ 0.05 , **P ≤ 0.01)
In accord with the fluorescence intensity measured in tissue lysates, IVIS imaging also showed
significantly higher accumulation of IBPAF in the LG 2 h after injection compared with AF (Fig
7A). Similarly, less IBPAF in the LG 24 h after injection compared with 2 h after injection was
observed (Fig 7A), whereas no difference was seen in AF between 2h and 24h, consistent with the
results from the tissue lysates (Fig 7A). Significantly higher accumulation of IBPAF in the lung
and pancreas was also seen when compared with AF at 2h, likely due to high levels of ICAM-1 in
these two tissues (Fig 1B). Lower IBPAF in the liver compared with AF at 2h may be due to the
higher accumulation of IBPAF in the lung and pancreas which decreases circulating IBPAF levels.
No significant differences between AF and IBPAF were detected by IVIS in any organ at 8 h or
24h post-injection (Fig 7). The low levels of the carriers found in other organs may be due to the
presence of both in circulation.

57

Figure 7. Optical-imaging evaluation of biodistribution of ELP-FKBPs in NOD mice after
I.V. injection. Optical IVIS imaging was used to quantify fluorescence of major organs after mice
were euthanized at H) 2 h; I) 8 h; and J) 24 h after I.V. injection of PBS, AF or IBPAF to 14-16
week old male NOD mice. Quantification of fluorescence signals of AF and IBPAF per mg of
organ in A) LG; B) SG; C) lung; D) liver; E) kidney; F) spleen; and G) pancreas. A t-test was
done to compare AF and IBPAF within each organ at each time point. Data is presented as mean
± SD (n=5 mice per group per time point, **P ≤ 0.01,*p ≤ 0.05).

Accumulation of IBPAF in the LG. Immunofluorescence labeling with an anti-ELP
antibody was conducted to assess the accumulation of ELP-FKBPs in the LG of NOD mice. Higher
levels of IBPAF were seen in the LG by 2 h after injection when compared with AF, which showed
labeling comparable to the secondary antibody control (Fig 8A). IBPAF could be detected
primarily associated with the plasma membranes of acinar cells and lymphocytes 2 h after injection,
rather than within cytoplasm, as indicated by the strong plasma membrane staining.
In a separate experiment, accumulation of AF-Rapa or IBPAF-Rapa in the LG of NOD
mice was also assessed to confirm that binding of the drug does not compromise the enhanced
accumulation of IBPAF in the LG. IVIS images of LG of NOD mice ex vivo were taken 2 h after
the NOD mice were injected with Cy7.5 labeled AF or IBPAF loaded with Rapa and quantified
(Fig 8B,C). There was a significant increase in radiant efficiency in LG from mice receiving
IBPAF-Rapa compared with AF-Rapa, confirming that binding of Rapa to the carrier does not
compromise the targeting effect of IBPAF. However, we were not able to assess the amount of
drug in the LG due to limited sensitivity of the assay as well as the small volume of LG.
58

59

Figure 8. IBPAF shows greater accumulation relative to AF in lymphocytic infiltration zones
within the NOD LG. A) LG were removed and processed 2 h after I.V. injection of unlabeled AF
or IBPAF for immunofluorescence staining using an anti-ELP antibody (AK1). To control for
nonspecific staining, a control group lacking the primary antibody is also shown (secondary
control). Arrow: plasma membrane of LG acinar cells. Arrowhead: plasma membrane of
lymphocytes. Blue: DAPI, Green: ELP, Red: F-actin. Scale bar=20 μm. Results are representative
of n=3 mice per group. B) IVIS images of ex vivo LG were taken 2 h after I.V. injection of Cy7.5
labeled AF or IBPAF loaded with Rapa. C) Quantification of radiant efficiency per mg of organ
from IVIS images. Data is presented as mean ± SD. (n=5 *P ≤ 0.05).

2.4 Conclusion and Discussion
We have successfully developed a bifunctional protein carrier, IBPAF, which can carry
Rapa and accumulate at sites of inflammation in a model of autoimmune dacryoadenitis which
mimics major features of SS, including increased ICAM-1 expression in the LG. IBPAF is a stable
protein in solution, with a hydrodynamic radius of around 7 nm, and binds Rapa with a K d of
approximately 3 nM. IBPAF binds specifically to ICAM-1-enriched bEND.3 cells in vitro and
appears to be internalized and trafficked to acidic compartment, possibly though ICAM-1 mediated
endocytosis. Previous studies have suggested that lysosomes may degrade FKBP-ELP constructs,
which could constitute one mechanism of cellular Rapa release (Peddi et al., 2018). This release
could spatially enhance mTOR inhibition by Rapa, since many studies have shown that mTOR is
localized and activated at lysosomes (Betz & Hall, 2013; Sancak et al., 2008). Non-targeted
FKBP-ELP constructs would not exhibit enhanced uptake by receptor-mediated endocytosis,
unlike the ICAM-1 targeted construct which would be expected to increase accumulation in the
endo-lysosomal pathway, based on knowledge of the receptor’s endocytosis. Studies have shown
that both monomeric and multimeric ligands targeted to ICAM-1 are internalized via the same
clathrin- and caveolin- independent pathway, namely cell adhesion molecule (CAM)-medicated
endocytosis, even though multimeric ligands show significantly higher accumulation in the
60

lysosomes than monomeric ligands (Muro, Schuchman, & Muzykantov, 2006; Muro et al., 2003).
However, due to the usage of different cell types, different targeting moieties (antibody vs. peptide),
and different physico-chemico properties of drug carriers, the mechanisms of intercellular
trafficking of ICAM-1 targeting moieties remains controversial.
In the pharmacokinetics analysis, both compartment and non-compartment models were
used to analyze the pharmacokinetics of AF and IBPAF in NOD mice. A minor difference was
seen between the two methods. The non-compartment model showed no statistically-significant
differences between AF and IBPAF in AUC, clearance, volume of distribution, and half-life.
However, the one-compartment model analysis by SAAM II showed a significantly lower AUC
and a higher clearance for AF. One possible reason for this is that the single exponential decay
model utilized by SAAM II underestimated the AUC of AF, as the two time points obtained at 8
h and 12 h are obviously above the curve (Fig 5B). The plasma concentration of AF seems to
plateau before 12 h, and then to follow a single exponential decay, which has been seen in other
FKBP-ELP constructs(C. Lee et al., 2019). Despite the differences in AUC and clearance, there
was no statistically significant difference in half-life between AF and IBPAF when analyzed by
the compartment model, which is consistent with the results obtained with the non-compartment
model. This further confirms that adding an ICAM-1 binding peptide will not lead to premature
elimination of the drug carrier from the body.
As seen for other ELP constructs, high levels of both AF and IBPAF were seen in the liver
and spleen(J. P. Dhandhukia et al., 2017; Jugal P. Dhandhukia et al., 2017) despite the relatively
low ICAM-1 expression levels in these two tissues. One explanation for this is the different
capillary structures in these organs. Fenestrated discontinuous sinusoids which are devoid of
organized basement membranes are more permeable to big molecules like nanoparticles and are
61

found in the liver and spleen (Wisse, De Zanger, Charels, Van Der Smissen, & McCuskey, 1985).
In addition, internalization of nanoparticles by Kupffer cells, hepatic B cells, sinusoidal endothelial
cells in the liver and splenic macrophages may also contribute to the sequestration of nanoparticles
in the liver and spleen (Cataldi, Vigliotti, Mosca, Cammarota, & Capone, 2017; Tsoi et al., 2016).
Significantly higher accumulation of IBPAF compared with AF was also seen in the lung 2h after
injection, which is likely due to the high level of endogenous ICAM-1. Several studies have used
targeting moieties to ICAM-1 to specifically deliver therapeutic agents to the lung, taking
advantage of the high endogenous ICAM-1 expression (Ferrer et al., 2014; Murciano et al., 2003;
Rossin, Muro, Welch, Muzykantov, & Schuster, 2008), which provides another potential
application for IBPAF.
It is well-known that the ICAM-1/LFA-1 interaction plays an important role in T-cell
activation and migration, which is crucial in autoimmune inflammation. Specifically, in the
scenario of inflammatory dry eye disease, the ICAM/LFA-1 interaction is involved in T cell
activation in the lymph nodes, T cell migration to the ocular surface, and T cell activation on the
ocular surface (Pflugfelder, Stern, Zhang, & Shojaei, 2017), which makes it an ideal target for an
anti-inflammatory reagent development. Lifitegrast, an LFA-1 antagonist targeting the LFA-
1/ICAM1 interaction, was recently approved for the topical treatment of dry eye disease (Murphy
et al., 2011; Semba et al., 2011). Several other drugs that modulate the LFA1/ICAM-1 interaction
have also been investigated systemically as anti-autoimmune inflammation agents, including
efalizumab for psoriasis (Kallen et al., 1999) and lovastatin for rheumatoid arthritis (Kallen et al.,
1999). The ICAM-1 targeting peptide used in this study, namely IBP, has been shown to block the
interaction between ICAM-1 and LFA-1 and to inhibit ICAM-1/LFA-1 dependent T-cell activation
by antigen presenting cells (Belizaire et al., 2003). In this study, we have mainly focused on
62

utilizing IBP as a targeting moiety for ICAM-1 to specifically demonstrate proof-of-principle
ability to delivery the ELP carrier (and ultimately the drug) to the site of inflammation. However,
it is possible that the IBPAF carrier itself may have anti-inflammatory effects through modulation
of ICAM-1/LFA-1 interactions. Further investigations will be necessary to explore this and other
potential effects.

63

Chapter 3: Intralacrimal Sustained Delivery of Rapa Shows Therapeutic
Effects Without Systemic Toxicity in NOD Mice
3.1 Introduction
Sjögren’s syndrome (SS) is a systemic autoimmune disease characterized by lymphocytic
infiltration of lacrimal glands (LG) and salivary glands (SG) and the consequent reduction in tear
and saliva production, respectively(Lemp, 2005; Nguyen & Peck, 2009). These changes can lead
to severe corneal damage and compromised oral health. SS patients can develop additional
systemic symptoms including inflammation of other internal organs such as brain, lung and liver
as well as, in a subset of patients, B-cell lymphoma(Nocturne & Mariette, 2015). SS is classified
as primary (pSS) when the clinical manifestations occur alone, or as secondary (sSS) when
associated with another autoimmune disease, usually a connective tissue disease. The estimated
incidence rate of pSS is 7 /100,000 person-year, and the overall prevalence is 61 cases/100,000
inhabitants, although the incidence and prevalence are likely underestimated since many
symptoms are nonspecific(Qin et al., 2015). The LG produces the aqueous component of the tear
film, as a result, inflammatory damage of the LG leads to aqueous-deficient dry eye disease,
treatment of which is our experimental focus.
The first line therapy for aqueous-deficient dry eye disease is artificial tears, a replacement
therapy which increases the volume of the tear film and the residence time of tears on the ocular
surface. Providing only temporary relief of ocular symptoms, artificial tears fail to inhibit the
underlying inflammation related to autoimmunity that is present both on the ocular surface and in
the LG. Anti-inflammatory therapies such as topical corticosteroids, cyclosporine eyedrops, and
lifitegrast eyedrops are also used for clinical management of dry eye symptoms, though each has
64

disadvantages. Long term use of corticosteroids is complicated by the emergence of cataracts,
glaucoma, and infection(Foulks et al., 2015). Cyclosporine can cause ocular surface pain, irritation
and a burning sensation, leading to low patient compliance and premature discontinuation of
medication(Ames & Galor, 2015). Clinical trials of lifitegrast in dry eye patients have shown
mixed therapeutic effects(Holland et al., 2017; J. D. Sheppard et al., 2014; Tauber et al., 2015).
Previously, our group has shown that Rapamycin (Rapa), an FDA-approved immunosuppressant,
when formulated as an eyedrop for twice daily administration significantly inhibited LG
inflammation and increases tear production over a 12 week treatment window in male Non-obese
Diabetic (NOD) mice, a commonly-used mouse model which manifests the ocular symptoms of
SS(Shah et al., 2017). However, topical administration of Rapa resulted in inefficient recovery in
the LG, the primary target for reducing autoimmune exocrinopathy. Here, we have developed a
sustained-release formulation of Rapa for local administration to the inflamed LG with the help of
Elastin-Like Polypeptides (ELPs).
ELPs are thermo-responsive protein polymers consisting of different repeats of (Val-Pro-
Gly-Xaa-Gly)n pentamers where the guest residue, Xaa, specifies any amino acid and n determines
the number of pentapeptide repeats. ELPs can undergo a temperature-dependent reversible inverse
phase transition wherein they remain soluble below their transition temperature (Tt) and form a
coacervate above Tt (J. Despanie et al., 2016; D. E. Meyer & Chilkoti, 2002; Yeboah et al., 2016).
Taking advantage of this property, ELPs can be easily purified by several rounds of heating and
cooling without addition of any extra purification tags. ELPs can be precisely genetically
engineered and are biocompatible, biodegradable, of low immunogenicity, and environmentally-
responsive, which has enabled their diverse applications in drug delivery(Ciofani, Genchi, Mattoli,
et al., 2014; J. Despanie et al., 2016; Sarah R. MacEwan & Chilkoti, 2014). Tt can be finely tuned
65

by adjusting the hydrophobicity of the guest residue and the number of pentameric repeats. In
general, ELPs with high molecular weight and/or hydrophobic guest residues exhibit lower
transition temperatures than ELPs with low molecular weights and/or hydrophilic guest
residues(Jordan Despanie et al., 2016). FK-506 binding protein 12 (FKBP), the cognate human
receptor for Rapa, was genetically fused onto different ELPs to generate a Rapa drug carrier,
without impairing the thermo-responsiveness of the ELPs(Mihir Shah et al., 2013). Previously we
have shown that when delivered systemically by FKBP-ELP, Rapa could significantly inhibit LG
inflammation and improve ocular surface symptoms in diseased male NOD mice(C. Lee et al.,
2019; Mihir Shah et al., 2013). However, the frequent dosing interval used in these studies, as well
as the systemic toxicity related to Rapa such as hyperglycemia and hyperlipidemia were drawbacks
in clinical translation of this strategy.
Here, taking advantage of the temperature responsive property of ELPs, we have developed
a high capacity ELP carrier for Rapa with a Tt below physiological temperature which forms a
depot loaded with Rapa at the injection site. We then tested its efficacy in SS-related autoimmune
dacryoadenitis in the male NOD mouse model. To accomplish this, we selected valine, a
hydrophobic amino acid, as the guest residue and generated [FKBP-(VPGVG)24]4-FKBP(5FV),
which is theoretically capable of carrying five Rapa molecules per carrier and also has a relatively
low Tt. We also made [FKBP-(VPGAG)24]4-FKBP(5FA) as a soluble control carrier for
comparison of tissue retention times after intra-LG delivery. For this control construct, the less
hydrophobic amino acid, alanine, was selected as the guest residue, which has a relatively high Tt.
We hypothesized that direct injection of 5FV-Rapa into the LG would generate a depot which
could be monitored utilizing diverse imaging approaches, and which would achieve sustained
66

release of Rapa into the LG to achieve therapeutic affects at a reduced dosing interval, while
minimizing the systemic toxicity of Rapa.

Figure 1. Schematic illustration of intra-LG injection of 5FV-Rapa relative to subcutaneous
administration of a soluble Rapa carrier, FAF. A) Graphical illustration of depot formation by
5FV-Rapa but not 5FA-Rapa. B) Schematic illustration of the study evaluating the therapeutic
efficacy of a single intra-LG injection of 5FV-Rapa (0.44 mg Rapa/kg x 1 dose) versus
subcutaneous injection every other day of FAF-Rapa (1mg Rapa/kg x 7 doses), a systemic
administration control, over 14 days in diseased male NOD mice. A control treatment was intra-
LG with PBS on Day 1 of the two week study, also shown. Created with biorender.com.

3.2 Material and Methods
Reagents. BLR(DE3) competent E. coli was from New England BioLabs Inc. (Ipswich,
MA, USA). Rapamycin (Rapa) was from LC Laboratories (Woburn, MA, USA). TO-PRO™-3
Iodide (642/661) and NHS-Rhodamine were from ThermoFisher Scientific Inc. (Rockford,
IL,USA). Carbachol, used to stimulate tear production, was from Sigma-Aldrich (St. Louis, MO,
USA). Ketamine (Ketaject) was from Phoenix (St. Joseph, MO, USA) and xylazine (AnaSed) was
from Akorn (Lake Forest, IL, USA). Zeba™ Spin Desalting Columns, 7K MWCO (10 mL) were
from ThermoFisher Scientific Inc. (Rockford, IL, USA). Sulfo-Cyanine 7.5 NHS ester was from
67

Lumiprobe Corp (Hallandale Beach, FL, USA). ZoneQuick phenol red threads were from
SHOWA YAKUHIN KAKO CO., LTD (Tokyo, Japan). Free Style Lite test strips were from
Abbott Diabetes Care, Inc. (Alameda, CA, USA). Other reagents were from standard suppliers.
Molecular Cloning. 5FA and 5FV were cloned using standard restriction enzyme-based
cloning. The pIDTsmart vector with the FKBP oligonucleotide sequence was ordered from
Integrated DNA Technologies (Coralville, IA,USA) with three restriction cut sites flanking the
FKBP gene: NdeI, BserI at the 5’ end and BamHI at the 3’ end. This vector was digested with
NdeI and BamHI and the gel-purified FKBP gene was inserted into a pET25b (+) vector digested
with the same set of enzymes. Next, pET25b (+) containing the FKBP gene and a modified
pET25b (+) vector (mPET) containing the (VPGAG)24 (A24) were double-digested with BserI and
BssHII. The appropriate gene containing fragments was gel purified and ligated to generate an
intermediate vector containing the FKBP-A24 (FA) fusion gene. Following double digestion of
this intermediate vector with NdeI and BamHI, FA was purified and inserted into an empty mPET
vector that was digested with the same set of enzymes. FA (mPET) was double-digested with two
sets of enzymes: BserI + BssHII and AcuI + BssHII. The appropriate fragments containing FA
were purified and ligated to generate FAFA (in the mPET vector). This was repeated using FAFA
(mPET) to generate FAFAFAFA. Finally, addition of FKBP at the 3’ end of FAFAFAFA was
performed as previously described(J. P. Dhandhukia et al., 2017) to generate FAFAFAFAF(5FA).
The same protocol was used to clone FVFVFVFVF (5FV) by using the (VPGVG)24 (V24)
sequence instead of A24. FAF was cloned as previously described(J. P. Dhandhukia et al., 2017).
ELP Purification, Characterization and labeling. 5FA, 5FV and FAF were purified
using three rounds of Inverse Transition Cycling (ITC) as described previously(C. Lee et al., 2019).
Fusion protein polymers were analyzed for purity by SDS-PAGE gel analysis and Coomassie blue
68

staining. The temperature-concentration phase diagram was determined as previously
described(Jugal P. Dhandhukia et al., 2017; Yaping Ju et al., 2019). Briefly, absorbance at 350
nm was measured in a DU800 UV−vis spectrophotometer (Beckman Coulter, CA,USA) under a
temperature gradient of 0.3°C/min. The transition temperature at each concentration of ELP was
defined as the temperature at which the maximum first derivative was achieved within each optical
density profile with respect to the temperature. The transition temperature from each concentration
was used to plot the phase diagram and fit with Equation 1:
Tt = b − mlog10[CELP] (1)
For fluorescent visualization after intra-LG injection, 5FA or 5FV were covalently
modified with Sulfo-Cyanine7.5 (Cy7.5) NHS esters or NHS-Rhodamine. Excessive dye was
removed by Zeba™ Spin Desalting Columns, 7K MWCO (ThermoFisher Scientific) per the
manufacturer’s protocol.
Surface Plasmon Resonance . 5FA and 5FV were immobilized on the surface of a Series
S Sensor Chip CM5 (GE Healthcare, Pittsburgh, PA) at pH 4.5 by amine coupling using 1-ethyl-
(3-dimethylaminopropyl)-carbodiimide hydrochloride (EDC) and NHS to a final density of 15000
response units (RU). Residual sites on the dextran were blocked with 1 M ethanolamine
hydrochloride. A control flow cell was blocked with ethanolamine for reference subtraction.
Experiments were performed in running buffer (0.1 M HEPES, 1.5 M sodium chloride, 30 mM
EDTA and 0.5% Surfactant P20, 1% DMSO, pH 7.4). Dilutions of Rapa dissolved in running
buffer were injected over the chip at 20 mL/min for 2 min. The proteins were then dissociated
from the chip for 2.5 min with running buffer. The remaining protein was removed from the chip
by 0.05% SDS at 30 mL/min for 30 s followed by 1 mM sodium hydroxide at 30 ml/min for 30 s.
Sensorgrams were generated and analyzed using Biacore T100 Evaluation Software (version 2.0.2).
69

The equilibrium RU observed for each injection was plotted against the concentration of protein.
The equilibrium Ka, Kd and KD values were derived by analysis of the plots using the steady-state
affinity model.
Rapamycin Encapsulation. 5FA or 5FV in PBS was mixed with a 10X molar access of
Rapa dissolved in 100% ethanol. The vial was stirred at 4°C for 2-3 hrs. Excess drug was removed
by centrifugation at 13.2K RPM, 20 min followed by filtration through a 0.2 µm syringe filter.
Residual organic solvent was further removed by overnight dialysis with a cut-off molecular
weight of 10 kDa. To determine the amount of encapsulated drug, the sample was injected into a
C-18 reverse phase HPLC column and eluted in a water: methanol gradient from 40% to 100%.
Rapa concentration was calculated using a standard curve.
Animal Procedures. C57BL/6J breeding pairs and male NOD mice aged 12 weeks old
were purchased from The Jackson Laboratory (Bar Harbor, ME, USA). Animals were bred in the
University of Southern California (USC) vivarium. All animal procedures were approved by the
USC Institutional Animal Care and Use Committee and performed in accordance to the guide for
care and use of laboratory animals, 8
th
ed (2011). For intra-LG injection (Figure 2), fur was
removed from the cheek using a pet razor and the area cleaned with three changes of betadine and
alcohol after mice were anesthetized with an i.p. injection of xylaxine/ ketamine and placed on a
heating pad. A small incision (5 mm) was made to visualize the LG. ELP-rapamycin or ELP were
injected into the LG on both sides using a 35-gauge blunt needle. The incision was closed with
tissue glue. Buprenorphine SR (0.5-1mg mg/kg) was given subcutaneously to relieve any pain
from the surgery. The mice were placed on a heating pad and monitored until they were fully
recovered from anesthesia. The animals were monitored daily post-op for any signs of pain stress
or infection. For LG collection, LGs were removed after mice were euthanized via intraperitoneal
70

injection with 55 mg Ketamine and 14 mg Xylazine per kilogram of body weight, followed by
cervical dislocation.

Figure 2. Schematic illustration of the process of intraLG gland injection.

Tissue Clearing and Confocal Microscopy. A single injection of 5 μl of Rhodamine-
labeled 5FA-Rapa or 5FV-Rapa were injected into both LG of C57 mice aged 13 weeks old for
optimized visualization of the depot. The LG were collected the next day and fixed in 4%
paraformaldehyde at 4°C overnight. Tissue clearing was done with an adapted protocol from “3D
imaging of solvent cleared organs” (3DISCO)(Ertürk & Bradke, 2013). In brief, LGs were washed
3 times in PBS before permeabilization with 1% Triton X for 1 hr at room temperature the next
day. Then the LG were stained with TO-PRO™-3 Iodide (642/661) at 4°C overnight. LGs were
washed 3 times with PBS the next day followed by dehydration in series of ethanol solutions
(ranging from 50% to 100% ethanol in water). After that, LGs were placed in 100% hexane for 2
hr at room temperature to remove any trace amounts of water. Then LGs were placed in BABB
solution (benzyl alcohol: benzyl benzoate=1:2) for clearing with close monitoring. After the LG
became transparent, they were stored in ethyl cinnamate for further imaging. Cleared glands were
71

imaged using a ZEISS LSM 800 confocal microscope equipped with an Airyscan detector (Zeiss,
Thornwood, NY). A Z-stack of 25 pictures at intervals of 3 μm was taken and converted into a
video format.
Lightsheet Microscopy and Image Processing. Cleared glands were also imaged using a
lightsheet microscope. Lightsheet data were collected on an M-Squared (M-2) Aurora microscope
(https://www.m2lasers.com/microscopy-aurora). The samples and objectives were immersed in
ethyl-cinnamate for index matching with the BABB cleared tissue preparation. Samples were
imaged with 594 nm (2% laser power, 10 ms exposure) and Hoechst Far-red with 647 nm (1%
laser power, 4 ms exposure) diode laser lines. LG with Rhodamine-labeled 5FV-Rapa was
collected as a 5X5 tile scan and LG with Rhodamine-labeled 5FA-Rapa as a 4X6 tile scan. The
Airy beam design of the Aurora requires deconvolution to produce isotropic resolution. A
deconvolution standard was produced by embedding multi-color silica beads in 2% low-melt
agarose. This hydrogel was dehydrated in washes (2 hrs each) of increasing ethanol concentration
(1x25%, 1x50%, 1x75%, 2x95%, 2x100%) and the final dehydration steps were in 2x 2 hr washes
in 100% methanol. The dehydrated gel was washed once in ethyl-cinnamate overnight and
transferred into 100 ml of fresh ethyl-cinnamate for storage in the immersion medium for imaging
the calibration standard. M-2 Deconvolution software processed both samples using 100 iterations
of the Richardson-Lucy deconvolution algorithm. The deconvolution step also prepares the 3-
dimensional tiles for stitching. 3-D, 2 channel tiles were converted into .ims file format in
Bitplane’s Imaris File Converter for stitching in Imaris Stitcher and all animated movies/snapshots
were produced in Imaris v9.6.0.
Safety Evaluation of Intra-LG Injection. Male C57 mice aged 12-14 weeks old were
injected intra-LG with 32 μl of 5FA or 5FV (366 μM) in total, 16 μl per side with 4 injections of
72

4 μl each . Mice without injections were used as nontreated controls. Mice were euthanized 2
weeks after injection and the LG were collected for immunofluorescence staining, H&E staining
and Trichrome staining, per established procedures.
Intra-LG Pharmacokinetics Study after Intra-LG Injection. C57 mice aged 12-14
weeks were injected intra-LG with 32 μl of Cy7.5-labeled 5FA or 5FV (1.2 mM Rapa, 366 μM
Cy7.5 labeled ELP) in total, 16 μl per side with 4 injections of 4 μl each. Each group included 2
male mice and 2 female mice. LG were collected after mice were euthanized at 1 or 2 weeks post-
injection. An IVIS image of the LG was taken to quantify the Cy7.5 fluorescence intensity. The
fluorescence intensity of each LG, after subtracting the background from a non-injected LG and
normalizing to the weight of the LG, was determined and plotted. The amount of Rapa in the LG
was also assessed at 2 days, 1 week and 2 weeks after injection using LC-MS following the
protocol previously reported(Yaping Ju et al., 2019). Mean residence time (MRT) is calculated
from Equation 2:
MRT=AUMC/AUC (2)
Where AUMC and AUC stand for Area under the Moment Curve and Area under the Curve,
respectively, and are estimated with the trapezoidal method.
Therapeutic Study. The protocol for this study is shown in Figure 1. 12 μl of 5FV-Rapa
(1.2 mM Rapa, 0.44 mg/kg, 366 μM 5FV) was injected using intra-LG injection into 13 week
diseased male NOD mice in total, 6 μl per each LG in 3 injections of 2 μl each. This lower volume
was chosen to avoid excessive swelling of the LG. The same volume of PBS was injected into
using the identical procedure as a negative control. FAF-Rapa (1 mg/kg) was injected
subcutaneously (s.c.) in the thigh every other day for two weeks as a systemic administration
73

control which was previously shown to be effective and sufficient to reduce lymphocytic
infiltration of the LG(Y. Ju et al., 2018). Mice receiving FAF-Rapa s.c. injections also received a
single PBS intra-LG injection comparably to the 5FV-Rapa group to control for any intra-LG
injection effects. The accumulated doses for 5FV-Rapa and FAF-Rapa were 0.44 mg/kg and 7
mg/kg, respectively. Basal tear production and blood glucose was measured before and two weeks
after treatment. All the mice were euthanized as described above for further analysis.
Basal and Stimulated Tear Measurements. Basal and stimulated tears were measured as
described(Y. Ju et al., 2018). Briefly, for measurement of basal tear flow, a ZoneQuick phenol red
impregnated cotton thread was inserted under the lower eyelid for 10 sec while mice were under
anesthesia, and tear production was reported as a function of the length of wetting of the thread in
millimeters. Data is plotted by values obtained from each individual eye, so each mouse
contributes two thread measurements. For collection of stimulated tear fluid, a small incision was
made to expose the LG of anesthetized mice. A small piece of cellulose mesh (Kimwipe; Fisher
Scientific, Pittsburgh, PA, USA) was placed on the LG to capture added secretagogue (3 μl of 50
μM carbachol solution). After stimulating the gland for 5 min, tears from both eyes were collected
with 2 μl microcapillary tubes (Drummond Scientific, Broomall, PA, USA.) Stimulation was
performed three times for a total collection time of 15 min, and the volume of collected tears was
recorded. Tears were stored on ice until further biochemical analysis.
Histology Analysis of LGs. Mice were euthanized and LGs were removed and fixed in
10% neutral buffered formalin prior to embedding in paraffin blocks. Paraffin sections were
stained with hematoxylin-eosin (H&E) according to standard procedures and photographed using
a Nikon 80i microscope (Melville, NY, USA) equipped with a digital camera. Images from three
74

nonconsecutive whole gland cross sections were obtained for each LG. The area of the LG
occupied by lymphocytic foci was calculated using ImageJ software by a blinded examiner.
Statistics. Statistical analyses were performed using a Student’s t-test or a one-way
ANOVA followed by Tukey's post-hoc test using Prism (GraphPad Software, La Jolla, CA). A p-
value <0.05 was considered statistically significant.
3.3 Results
Purification and characterization of 5FA and 5FV. 5FA and 5FV were successfully
purified from BLR(DE3) competent E. coli via inverse transition cycling with a similar yield of
about 80 mg/L. Coomassie blue SDS-PAGE gel analysis showed that both 5FA and 5FV had purity
above 90% (Figure 3A). Optical density measurements at 350 nm of purified 5FA and 5FV at a
series of different concentrations was used to characterize the temperature-concentration phase
diagram for both constructs (Figure 3B, 3C). Transition temperature (Tt) follows an inverse
relationship with logarithm of the ELP concentration (Figure 3D). Using the fit parameters (Table
1) for 5FA and 5FV, the estimated Tt for 5FA at 300 μM and 400 μM was 51.5°C and 51.2°C,
respectively, whereas the estimated Tt for 5FV at the same concentrations was 21.0 °C and 19.6°C.
By extrapolation to the estimated concentration used for the in vivo studies, which is between 300
μM and 400 μM, we anticipated that 5FV will form a coacervate whereas 5FA will remain soluble.
Surface plasmon resonance (SPR) was also conducted to study the binding properties between
Rapa and 5FA (Figure 3E) or 5FV (Figure 3F). The association (Ka), dissociation (Kd) and
equilibrium (KD) constants derived from the sensorgram between Rapa and 5FA or Rapa and 5FV
are summarized in Table 1. Both 5FA and 5FV have similar binding affinity for Rapa at 18°C,
which is around 60 nM. Notably, this value is significantly higher than the value of other FKBP-
ELP constructs we have reported previously at 37°C, which has been about 5 nM(J. P. Dhandhukia
75

et al., 2017). Since 5FV is not soluble at 37°C, it was not possible to measure the binding properties
between 5FV and Rapa at 37°C; therefore, no direct comparison can be made to compare the
binding properties with our previously reported constructs at this temperature.

Figure 3. Biophysical characterization of 5FA and 5FV. A) Coomassie blue staining of purified
5FA and 5FV resolved by SDS-PAGE. B) Optical density at 350 nm of 5FA was monitored as a
function of temperature at different concentrations. C) Optical density at 350 nm of 5FV was
monitored as a function of temperature at different concentrations. D) Phase transition temperature
was plotted vs. concentration as a phase diagram for 5FA and 5FV. The 95% confidence interval
around each best-fit line is indicated with dashed lines. E) Surface plasmon resonance (SPR)
sensorgram of different concentrations of Rapa monitored on a surface with immobilized 5FA at
18°C. F) SPR sensorgram of indicated concentrations of Rapa monitored on a surface with
immobilized 5FV at 18°C.

76

Table 1. Phase transition characterization and Rapa binding dynamics of 5FA and 5FV
Constructs
Temperature-Concentration Phase
Diagram
a

Ka (M
-1
s
-1
) Kd (s
-1
) KD(M)
Slope, m [°C
Log(µM)]
Intercept, b [°C]
5FA 2.4±1.9 57.5±2.1 (2.9±1.0)x10
4

(1.5±0.8)x10
-
3

(6.3±5.0)x10
-
8

5FV 11.5±0.9 49.5±1.1 (1.7±0.1)x10
4

(1.1±1.0)x10
-
3

(6.4±0.6)x10
-
8

Intra-LG injection does not elicit significant damage to LGs. A safety evaluation was
done to assess the potential damage, such as inflammation or fibrosis, caused by intra-LG injection
to the LG. Male C57 mice aged 12-14 weeks old were injected intra-LG with 32 μl of 5FA or 5FV
(366 μM) in total, 16 μl per side with 4 injections of 4 μl each. Two weeks after injection, LG from
mice were collected for further analysis. H&E staining did not show any significant immune cell
infiltration of the LG in 5FA and 5FV treated groups compared with nontreated controls (Figure
4A). Trichrome staining, which visualizes collagen deposition, a marker for fibrosis(J. Chen et al.,
2011), was also conducted to detect any potential fibrosis in the LG (Figure 4B). The ratio of the
area of collagen deposition to the area of the whole LG was calculated for all three groups. No
significant differences were seen among the three groups, which indicates that intra-LG injection
a: Phase diagrams for assembly (Fig 2D) were fit with Equation 1. Values represent mean ±
SD.
77

does not cause significant fibrosis in the LG (Figure 4C). Immunofluorescence staining with
HSP47 antibody, a marker for active fibroblasts which is elevated in several fibrotic
diseases(Razzaque & Taguchi, 1999), was also conducted to further detect fibrosis of LG (Figure
4D). The number of HSP47 positive cells were quantified and compared between the three groups.
No significant differences were seen among the three groups, confirming that intra-LG injection
does not cause significant fibrosis in the LG (Figure 4E). Based on the above results, we
concluded that intra-LG injection is a safe procedure and does not cause significant or long term
damage to the glands.

78

Figure 4. Intra-LG injection does not promote tissue damage or fibrosis. Male C57 mice
were either used as controls (CON) or administered with 32 μl of 5FA or 5FV (366 μM) in total,
16 μl per side with 4 injections of 4 μl each. LG were obtained after 2 weeks. A) H&E staining
of a representative LG section from each group (n=3 mice in each group, 2 sections per mouse).
Scale bar: 200 um. B) Trichrome staining of a representative LG from each group). Scale bar: 200
um. C) Quantification of the percentage of the area stained blue with Trichrome. (n=3 mice in
each group, 2 sections per mouse) D) HSP47 immunofluorescence staining of LG representative
of each group. White: F-actin, Blue: DAPI, Green: HSP47. Scale bar: 20 um. E) Quantification of
the number of HSP47+ cells in each picture (n=3 mice each group, 3 pictures from each mouse).
Data is presented as mean ± SD.

5FV-Rapa forms a depot in the LG after intra-LG injection. To confirm that 5FV-
Rapa does form a depot at the injection site, lightsheet imaging evaluation of the cleared whole
LG was done one day after intra-LG injection of Rhodamine labeled ELP-Rapa (Figure 5). No
significant depot formation was detected in the LG injected with Rhodamine labeled 5FA-Rapa
(Figure 5A,B,C,D,). In sharp contrast, a major depot formed in the 5FV-Rapa treated gland
(Figure 5E,F,G,H,), as shown by the asterisk in Figure 5E. Rhodamine signals were also detected
in the peripheral area of the LG as shown by the arrows in Figure 5E, suggesting that some 5FV-
Rapa was trapped beneath the LG capsule.
The cleared LG was also imaged by confocal fluorescence microscopy in order to evaluate
the localization of the ELP-Rapa inside the LG. No significant signal for 5FA-Rapa was detected
by confocal microscopy (Figure 6A,). However, a strong signal from the interlobular area as well
as the interstitial area between the acini was detected in the LG injected with Rhodamine labeled
5FV-Rapa (Figure 6B,). No significant signal was detected inside the LG acinar cells. Based on
these data, we concluded that 5FV-Rapa forms a depot in the LG, primarily in the interlobular area,
after intra-LG injection, whereas the soluble control 5FA-Rapa does not form a depot.
79

Figure 5. Lightsheet imaging of Rhodamine-labeled 5FV-Rapa following intra-LG
administration reveals formation of a depot. Male C57 mice were administered with a single 5
μl injection of rhodamine-labeled 5FV-Rapa or rhodamine-labeled 5FA-Rapa (366 μM of ELP)
and LG were collected 1 day after injection. A) A reconstructed 3D image of LG injected with
rhodamine-labeled 5FA-Rapa. B) A cross-section of the 3D image in panel A showing three
orthogonal planes. C) Serial sections of LG injected with rhodamine-labeled 5FA-Rapa. D)
Images of each cross section in panel C. E) A reconstructed 3D image of LG injected with intra-
LG rhodamine-5FV-Rapa. F) A cross-section of the 3D image in panel F showing three orthogonal
planes at the site of injection. G) Serial sections of LG injected with rhodamine-labeled 5FV-Rapa.
H) Images of each cross section in panel G. Magenta: Rhodamine-ELP, Cyan: nuclear stain, Scale
bar: 500um. Asterisk: depot formed after intra-LG delivery of 5FV-Rapa. Arrows: 5FV-Rapa
trapped beneath the capsule after intra-LG injection.

80

Figure 6. Confocal fluorescence microscopy imaging of Rhodamine-labeled 5FV-Rapa
injected intra-LG reveals a depot formation. Male C57 mice were injected with a single intra-
LG injection of 5 μl containing rhodamine-labeled 5FV-Rapa or rhodamine-labeled 5FA-Rapa
(366 μM of ELP). LG were collected after 1 day. Serial sections were acquired in the cleared
tissue at 15 um intervals following following A) intra-LG injection with Rhodamine-labeled 5FA-
Rapa and B) intra-LG injection with Rhodamine-labeled 5FV-Rapa. 5FV-Rapa is retained
throughout the LG, especially in the interlobular (arrowhead) and interacinar (arrows) areas after
24 hrs. Red: Rhodamine-ELP, White: nuclear stain, Scale bar: 100 um.

5FV-Rapa exhibits longer retention time in the LG than 5FA-Rapa. A
pharmacokinetics study was conducted to evaluate the retention time of ELPs and Rapa following
intra-LG administration associated with either formulation. Both male and female C57 mice aged
12-14 weeks old were used for the PK study so that any sex differences in terms of elimination
rate could be studied. Mice were injected intra-LG with Cy7.5 labeled 5FA-Rapa or 5FV-Rapa on
both sides of the LG as described in Methods, and LGs were collected after different time point
for evaluation of ELP fluorescence utilizing IVIS analysis in parallel with the determination of
Rapa concentration. One week post-injection, no significant differences in fluorescence levels
detected by IVIS in the LG were detected between 5FA and 5FV (Figure 7A, 7B). However, two
weeks after injection, there was significantly higher fluorescence detected in the LG injected with
5FV group compared to 5FA (Figure 7A, 7B), indicating that 5FV is retained in the LG for longer
period than the soluble carrier 5FA. The amount of Rapa retained in the LG following intra-LG
81

injection of 5FA-Rapa and 5FV-Rapa was also assessed by LC-MS. There was a significantly
higher amount of Rapa retained in the LG in the 5FV-Rapa injected group relative to the 5FA-
Rapa injected group at 2 days and 7 days post injection, which demonstrates that when delivered
by the depot forming 5FV, Rapa is retained in the LG for a much longer period than when delivered
by the soluble 5FA (Figure 7C). Fourteen days after injection, no detectable amounts of Rapa
were seen in either 5FA-Rapa or 5FV-Rapa groups. The MRT of Rapa estimated by Equation 2
for 5FA-Rapa and 5FV-Rapa was 11.7 hr and 75.6 hr, respectively. No sex differences were seen
in fluorescence or Rapa levels, implying that the pharmacokinetics following intra-LG injection
was similar between males and females. Taken together, intra-LG 5FV-Rapa injection sustains
drug levels for around one week making it superior to 5FA-Rapa which releases all the drug in
less than a week.

82

Figure 7. The 5FV-Rapa depot is retained in the LG for significantly longer than its soluble
control formulation (5FA-Rapa). A) Ex vivo whole LG optical imaging (IVIS) from male and
female C57 mice aged 12-14 weeks old at one week and two weeks after intra-LG injection of 32
μl of Cy7.5-labeled 5FA or 5FV (1.2 mM Rapa, 366 μM Cy7.5 labeled ELP) in total, 16 μl per
side with 4 injections of 4 μl each . (M, male mice; F, female mice). B) Quantification of radiant
efficiency per milligram of organ of C57 LG one week and two weeks after injection of Cy7.5
labeled 5FA-Rapa and 5FV-Rapa (n=4). C) Amount of Rapa in the LG at 30 min, 2 days, 7 days
and 14 days after intra-LG injection of 5FA-Rapa or 5FV-Rapa assessed by LC-MS (n=4). *p<0.05,
**p<0.01.

5FV-Rapa increases basal tear secretion and stimulated tear volume. Due to its
superior PK features, we advanced 5FV-Rapa treatment to a therapeutic study for evaluation side
by side with systemic administration of FAF-Rapa, which previously showed activity in reducing
autoimmune exocrinopathy in the male NOD disease model(C. Lee et al., 2019). The dosing
regimen for 5FV-Rapa was 0.44 mg/kg given at study onset (total 0.22 mg/kg administered to each
side of the LG in 3 different injections of 1 μl ), whereas FAF-Rapa was given subcutaneously at
1 mg/kg every other day for 7 doses in total. The surgical cut required for intra-LG injection was
clean after surgery and 5FV injection, and was completely healed without complications after 2
weeks (Figure 8). 5FV-Rapa showed significant improvement in basal tear production relative to
either intra-LG PBS (0.9±2.7 mm vs -1.0±1.6 mm, p<0.01) or subcutaneous FAF-Rapa (0.9±2.7
mm vs -0.5±2.1 mm), shown as the difference between post-treatment and pre-treatment tear
production (Figure 9A). Noticeably, the 5FV-Rapa group was the only group that showed an
increased tear production after treatment whereas both PBS and FAF-Rapa showed reduced basal
tear production after treatment as disease progresses when mice age. This suggested that 5FV-
Rapa may prevent and/or reverse autoimmune dacryoadenitis. Topical carbachol stimulation of
the LG reflects the physiological process of muscarinic stimulation of the LG by innervating
nerves in response to a stimulus and is thus considered a stimulated secretory response by the LG.
83

The 5FV-Rapa group also showed significantly increased carbachol-stimulated tear secretion
when comparing with FAF-Rapa group (4.7±0.9 μl vs 3.2±1.2 μl, p<0.05) (Figure 9B). This
further confirms that local administration of 5FV-Rapa has stronger effects than subcutaneous
FAF-Rapa in increasing tear production in male NOD mice. This is all the more remarkable given
that the cumulative dose given with FAF-Rapa was 16 times greater than the single dose given
with single intra-LG injection with 5FV.

Figure 8. The surgery cut healed completely after 2weeks.

84

Figure 9. Intra-LG 5FV-Rapa significantly enhances basal and stimulated tear secretion. 13
week (diseased) male NOD mice were treated with the dose regimen in Figure 1B. Tear
production was evaluated 2 weeks after the initial treatments at the conclusion of the protocol. A)
The differences in basal tear production as measured by phenol red thread tests for values post-
treatment (mm) and pre-treatment (mm) were plotted. Each data point represents one individual
eye (n=28). B) After treatments, tear production after topical carbachol stimulation of the LG was
plotted. Each data point represents one individual mouse (n=7). *p<0.05, **p<0.01.

Both 5FV-Rapa and FAF-Rapa reduce LG inflammation in male NOD mice.
Lymphocytic infiltration of the LG quantified as the percentage of the LG area occupied by
lymphocytes and determined histologically is a major indicator of disease severity in SS-related
dacryoadenitis(Janga et al., 2019). We examined the lymphocytic infiltration of the LG after
different treatments using H&E staining. As previously reported(C. Lee et al., 2019), the
subcutaneous FAF-Rapa regimen (5.6%±2.0 vs 24.3%±7.6%, p<0.001) was able to significantly
decrease lymphocytic infiltration of the LG compared with PBS. However, a single intra-LG
injection of 5FV-Rapa (16.5%±6.5% vs 24.3%±7.6%, p<0.01) also suppressed lymphocytic
infiltration (Figure 10). Further flow cytometry studies were done to evaluate the composition of
lymphocytes remaining the LG under each conditions, revealing no difference in percentage of
85

infiltrating CD3+ cells, CD3+CD4+ cells, CD3+CD8+ cells, CD45R+ cells, and CD11b+ cells
between the three treatments (Figure 11) and suggesting that these infiltrating lymphocytes are
decreased proportionally after treatment.

Figure 10. Both 5FV-Rapa and FAF-Rapa reduced lymphocytic infiltration. Representative
H&E staining of NOD mouse LG after treatment as in Figure 1 with A) a single treatment with
intra-LG PBS; B) a single treatment of intra-LG 5FV-Rapa (0.44 mg/kg with 0.22mg/kg to each
gland.) C) subcutaneous FAF-Rapa injections (1mg/kg x 7doses). D) Quantification of the
percentage of area of LG covered by lymphocytes after different treatments (n=14).
**p<0.01***p<0.001.

86

Figure 11. Flow cytometry analysis of lymphocyte composition in the LG after treatments.
A) Percentage of CD3+ cells in total lymphocytes isolated. B). Percentage of CD4+ cells in CD3+
cells. C). Percentage of CD8+ cells in CD3+ cells. D). Percentage of CD45R+ cell in total
lymphocytes isolated. E) Percentage of CD11b+ cells in total lymphocytes isolated. n=4 mice for
each group.

5FV-Rapa exhibits significantly less systemic toxicity than FAF-Rapa. It is well
documented that Rapa treatment can cause onset of hyperglycemia or exacerbate pre-existing
insulin resistance, as well as disrupt normal lipid metabolism in both human and experimental
animals(M. Fraenkel et al., 2008; Lamming et al., 2012; Larsen et al., 2006; Morrisett et al., 2002).
As previously reported, 20%-30%of male NOD mice spontaneously develop type I diabetes as a
result of lymphocytic infiltration of pancreas by 5 months(Bach, 1994; Kikutani & Makino, 1992).
Here and in previous work, we have used male NOD mice younger than 5 months, when most
mice are diabetes free, in studies of SS disease biology as well as in development of potential SS
87

therapeutics(C. Lee et al., 2019; Shah et al., 2017). However, their predisposition to
hyperglycemia represents a good model with which to test the effects of Rapa across different
treatment groups. The FAF-Rapa group showed significant body weight reduction when compared
with the intra-LG PBS group (-1.0±3.6% vs +2.7±2.3%, p<0.01), whereas no significant difference
was observed between 5FV-Rapa and PBS (+1.4±2.9% vs +2.7±2.3%)(Figure 12A). Blood
glucose in mice was also assessed before and after treatment regimens. The FAF-Rapa group
showed significantly increased blood glucose compared with 5FV-Rapa (+120.9±71.5 mg/dl vs
+6.2±23.0 mg/dl, p<0.001)(Figure 12B). Serum cholesterol levels after treatment were also
significantly increased in the FAF-Rapa group when compared with both intra-LG PBS and 5FV-
Rapa, respectively (148.8±11.3 mg/dl vs 79.7±10.1 mg/dl, p<0.001; 148.8±11.3 mg/dl vs 82.7±9.8
mg/dl, p<0.001), whereas no significant difference was seen between 5FV-Rapa and intra-LG
PBS(Figure 12C). There was also a trend to increased serum triglycerides post-treatment in the
FAF-Rapa group when compared with PBS and 5FV-Rapa (Figure 12D).
88

Figure 12. 5FV-Rapa has significantly less systemic toxicity than FAF-Rapa. Body weight
and serum chemistry were evaluated 2 weeks after the initial injections in 13 week male NOD
mice at study endpoint as shown in Figure 1. A) Percentage of body weight (BW) change post-
treatment compared with pre-treatment (n=14). B) Change in blood glucose post-treatment
compared with pre-treatment(n=14). C) Serum cholesterol level after different treatments (n=6).
D) Serum triglycerides levels after different treatments (n=6). **p<0.01***p<0.001.

3.4 Conclusion and Discussion
ELPs are thermo-responsive biopolymers consisting of pentameric repeats of (Val-Pro-
Gly-Xaa-Gly)n that can undergo reversible inverse phase transition wherein they remain soluble
below their transition temperature (Tt) and form insoluble coacervates above Tt. The Tt is tunable
by the number of pentameric repeats and the biophysical properties of the Xaa. As the number of
repeats or hydrophobicity of Xaa increases, Tt decreases as a result. Thus it is possible to engineer
89

an ELP with any desired Tt by adjusting the number of repeats and choosing different Xaa. Here,
by choosing a hydrophobic Xaa, Valine, we successfully engineered an ELP with Tt below
physiological temperature, which instantly formed an insoluble coacervate upon injection. The use
of Valine as Xaa to create an ELP with low Tt has been reported previously as a strategy to deliver
a sustained released depot into mice(Wang et al., 2015). Another soluble control, 5FA, was also
developed where a less hydrophobic amino acid, Alanine, was selected as Xaa for comparison with
5FV. With the lightsheet imaging system and 3D reconstruction process, we were able to visualize
and confirm that 5FV forms a depot inside the LG, whereas 5FA does not form a comparable depot
in the LG. As a consequence, 5FV-Rapa showed significantly longer retention time and was able
to maintain Rapa in LG for longer time after intra-LG injection compared with 5FA, which
suggests a promising future for using ELPs with low transition temperature for local sustained
release.
Mouse LG are small organs normally weighing 20-40 mg per pair, which limits the volume
that can be injected. The practical volume for intra-LG injection reported by previous studies
ranges from 2 μl to 50 μl (Lin, Chen, Fan, Chuck, & Zhou, 2015; Wang et al., 2015; Zoukhri,
Macari, & Kublin, 2007). In order to deliver sufficient Rapa dosage through direct intra-LG
injection, we developed a high-capacity carrier for Rapa which could carry a large amount of Rapa
per carrier molecule. Compared with the Rapa carriers we developed previously such as FSI(Mihir
Shah et al., 2013) or FAF(J. P. Dhandhukia et al., 2017) which can only carry 1 or 2 Rapa per
carrier molecule, the high capacity carrier 5FV can carry 5 Rapa per carrier molecule, greatly
increasing the dosage of Rapa that can be administered into LG with a limited injection volume.
Systemic administration of FAF-Rapa appeared to elicit a more robust inhibition of
lymphocytic infiltration of the NOD mice LG when compared with intra-LG delivery of 5FV-Rapa
90

(Figure 8D). This may be due to the notable difference in Rapa dose in between the Rapa
treatment regimens. In the FAF-Rapa group, the NOD mice received 7 doses of 1 mg/kg of Rapa
while in the 5FV group, the NOD mice received only one dose of 0.44mg/kg of Rapa in 5FV-Rapa
group due to the limited volume which could be administered to the LG. While the significantly
higher dose of FAF-Rapa group may contribute to its stronger inhibition of LG lymphocytic
infiltration; other mechanisms can also play a role. Systemic FAF-Rapa could possibly lead to
universal immunosuppression as well as decreased circulating lymphocytes, thus enabling fewer
available to migrate into LG, whereas intra-LG 5FV-Rapa might not have had a comparable effect
due to limited access to the systemic circulation. Within this context, it is notable that intra-LG
5FV but not systemic FAF was able to improve both basal and stimulated tear production (Figure
9) suggesting that local effects may be more important in eliciting a practical improvement in
aqueous tear production. Further evaluation of blood and spleen lymphocyte counts as well as
additional evaluation of the precise relationship between lymphocytic infiltration and loss of tear
production is required to resolve these issues.
It is well-documented that prolonged systemic Rapa treatment can impair the function of
pancreatic β-cells and cause glucose intolerance and hyperlipidemia(Merav Fraenkel et al., 2008;
Houde et al., 2010; Schindler, Partap, Patchen, & Swoap, 2014). Metabolic syndrome is a known
risk factor for dry eye diseases due to abnormal tear dynamics, tear film dysfunction, and altered
LG function(Moss, Klein, & Klein, 2000; Zhang, Zhao, Deng, Sun, & Wang, 2016), which may
explain the lack of improvement in tear production in the FAF-Rapa treatment group since
systemic Rapa administration elicits significant metabolic dysfunction. In contrast, intra-LG
delivery of Rapa by 5FV could inhibit inflammation of LG without causing metabolic syndrome,
which may contribute to its ability to significantly improve tear production relative to FAF-Rapa.
91

In conclusion, we have successfully developed a sustained local delivery method for Rapa,
taking advantage of the thermal responsive property of ELPs. Intra-LG delivery is safe and does
not cause severe inflammation or fibrosis of the LG. When delivered by 5FV, Rapa inhibit local
inflammation and improves tear production without causing any systemic side effects, which
provides a new future potential therapy for SS-related dacryoadenitis. This method of depot-
based delivery of Rapa may be suitable for local delivery to sites of inflammation in other
autoimmune and inflammatory disorders.

92

Conclusion
In this thesis, we first compared two commonly used spontaneous mouse model for SS-
related dacryoadenitis at both molecular and functional level. While NOR and NOD mice share
features of the autoimmune dacryoadenitis characteristic of SS, the NOD mouse exhibits a stronger
disease phenotype and earlier disease onset compared with NOR mouse. This provides guidance
for choice of the mouse model for different questions to be studies. We then developed an ICAM-
1 targeted protein-polymer carrier for Rapa that specifically binds to ICAM-1 in vitro and
accumulates in ICAM-1 overexpressing tissue in vivo, which may be useful for molecular
targeting in diverse inflammatory diseases where ICAM-1 is elevated. At last, we developed a
depot-forming formulation of Rapa that can be administered intralacrimally. This formulation
could achieve sustained release of Rapa into the LG to achieve therapeutic affects at a reduced
dosing interval, while minimizing the systemic toxicity of Rapa. Taken together, we believe these
works could provide extensive potential application for developing novel therapies for SS-related
dry eye disease.

93

References
(2011). In th (Ed.), Guide for the Care and Use of Laboratory Animals. Washington (DC).
Aluri, H. S., Kublin, C. L., Thotakura, S., Armaos, H., Samizadeh, M., Hawley, D., . . . Zoukhri,
D. (2015). Role of Matrix Metalloproteinases 2 and 9 in Lacrimal Gland Disease in Animal
Models of Sjogren's Syndrome. Invest Ophthalmol Vis Sci, 56(9), 5218-5228.
doi:10.1167/iovs.15-17003
Aluri, H. S., Samizadeh, M., Edman, M. C., Hawley, D. R., Armaos, H. L., Janga, S. R., . . .
Zoukhri, D. (2017). Delivery of Bone Marrow-Derived Mesenchymal Stem Cells Improves
Tear Production in a Mouse Model of Sjogren's Syndrome. Stem Cells Int, 2017, 3134543.
doi:10.1155/2017/3134543
Aluri, S., Pastuszka, M. K., Moses, A. S., & MacKay, J. A. (2012). Elastin-like peptide
amphiphiles form nanofibers with tunable length. Biomacromolecules, 13(9), 2645-2654.
doi:10.1021/bm300472y
Ames, P., & Galor, A. (2015). Cyclosporine ophthalmic emulsions for the treatment of dry eye: a
review of the clinical evidence. Clinical investigation, 5(3), 267-285.
doi:10.4155/cli.14.135
Anderson, M. S., & Bluestone, J. A. (2005). The NOD mouse: a model of immune dysregulation.
Annu Rev Immunol, 23, 447-485. doi:10.1146/annurev.immunol.23.021704.115643
Avci, A., Gunhan, O., Cakalagaoglu, F., Gunal, A., & Celasun, B. (2012). The cell with a thousand
faces: detection of myoepithelial cells and their contributions in the cytological diagnosis
of salivary gland tumors. Diagn Cytopathol, 40(3), 220-227. doi:10.1002/dc.21544
Bach, J. F. (1994). Insulin-dependent diabetes mellitus as an autoimmune disease. Endocr Rev,
15(4), 516-542. doi:10.1210/edrv-15-4-516
94

Bahamondes, V., Albornoz, A., Aguilera, S., Alliende, C., Molina, C., Castro, I., . . . Gonzalez, M.
J. (2011). Changes in Rab3D expression and distribution in the acini of Sjogren's syndrome
patients are associated with loss of cell polarity and secretory dysfunction. Arthritis Rheum,
63(10), 3126-3135. doi:10.1002/art.30500
Belizaire, A. K., Tchistiakova, L., St-Pierre, Y., & Alakhov, V. (2003). Identification of a murine
ICAM-1-specific peptide by subtractive phage library selection on cells. Biochem Biophys
Res Commun, 309(3), 625-630. doi:10.1016/j.bbrc.2003.08.050
Belmont, H. M., Buyon, J., Giorno, R., & Abramson, S. (1994). Up-regulation of endothelial cell
adhesion molecules characterizes disease activity in systemic lupus erythematosus. The
Shwartzman phenomenon revisited. Arthritis Rheum, 37(3), 376-383.
doi:10.1002/art.1780370311
Betz, C., & Hall, M. N. (2013). Where is mTOR and what is it doing there? J Cell Biol, 203(4),
563-574. doi:10.1083/jcb.201306041
Bloemen, P. G., Henricks, P. A., van Bloois, L., van den Tweel, M. C., Bloem, A. C., Nijkamp, F.
P., . . . Storm, G. (1995). Adhesion molecules: a new target for immunoliposome-mediated
drug delivery. FEBS Lett, 357(2), 140-144. doi:10.1016/0014-5793(94)01350-a
Brookes, S. M., Price, E. J., Venables, P. J., & Maini, R. N. (1995). Interferon-gamma and
epithelial cell activation in Sjogren's syndrome. Br J Rheumatol, 34(3), 226-231.
Cataldi, M., Vigliotti, C., Mosca, T., Cammarota, M., & Capone, D. (2017). Emerging Role of the
Spleen in the Pharmacokinetics of Monoclonal Antibodies, Nanoparticles and Exosomes.
Int J Mol Sci, 18(6). doi:10.3390/ijms18061249
95

Chen, J., Lee, S. K., Abd-Elgaliel, W. R., Liang, L., Galende, E.-Y., Hajjar, R. J., & Tung, C.-H.
(2011). Assessment of Cardiovascular Fibrosis Using Novel Fluorescent Probes. PLOS
ONE, 6(4), e19097. doi:10.1371/journal.pone.0019097
Chen, X., Edwards, J. A., Logsdon, C. D., Ernst, S. A., & Williams, J. A. (2002). Dominant
negative Rab3D inhibits amylase release from mouse pancreatic acini. J Biol Chem,
277(20), 18002-18009. doi:10.1074/jbc.M201248200
Chittasupho, C., Xie, S. X., Baoum, A., Yakovleva, T., Siahaan, T. J., & Berkland, C. J. (2009).
ICAM-1 targeting of doxorubicin-loaded PLGA nanoparticles to lung epithelial cells. Eur
J Pharm Sci, 37(2), 141-150. doi:10.1016/j.ejps.2009.02.008
Ciofani, G., Genchi, G. G., Guardia, P., Mazzolai, B., Mattoli, V., & Bandiera, A. (2014).
Recombinant human elastin-like magnetic microparticles for drug delivery and targeting.
Macromol Biosci, 14(5), 632-642. doi:10.1002/mabi.201300361
Ciofani, G., Genchi, G. G., Mattoli, V., Mazzolai, B., & Bandiera, A. (2014). The potential of
recombinant human elastin-like polypeptides for drug delivery. Expert Opin Drug Deliv,
11(10), 1507-1512. doi:10.1517/17425247.2014.926885
Conus, S., & Simon, H. U. (2010). Cathepsins and their involvement in immune responses. Swiss
Med Wkly, 140, w13042. doi:10.4414/smw.2010.13042
Dartt, D. A. (2009). Neural regulation of lacrimal gland secretory processes: relevance in dry eye
diseases. Prog Retin Eye Res, 28(3), 155-177. doi:10.1016/j.preteyeres.2009.04.003
de Paiva, C. S., Jones, D. B., Stern, M. E., Bian, F., Moore, Q. L., Corbiere, S., . . . Pflugfelder, S.
C. (2016). Altered Mucosal Microbiome Diversity and Disease Severity in Sjogren
Syndrome. Sci Rep, 6, 23561. doi:10.1038/srep23561
96

De Riva, A., Varley, M. C., Bluck, L. J., Cooke, A., Deery, M. J., & Busch, R. (2013). Accelerated
turnover of MHC class II molecules in nonobese diabetic mice is developmentally and
environmentally regulated in vivo and dispensable for autoimmunity. J Immunol, 190(12),
5961-5971. doi:10.4049/jimmunol.1300551
Despanie, J., Dhandhukia, J. P., Hamm-Alvarez, S. F., & MacKay, J. A. (2016). Elastin-like
polypeptides: Therapeutic applications for an emerging class of nanomedicines. J Control
Release, 240, 93-108. doi:10.1016/j.jconrel.2015.11.010
Despanie, J., Dhandhukia, J. P., Hamm-Alvarez, S. F., & MacKay, J. A. (2016). Elastin-like
polypeptides: Therapeutic applications for an emerging class of nanomedicines. Journal of
Controlled Release, 240, 93-108. doi:https://doi.org/10.1016/j.jconrel.2015.11.010
Dhandhukia, J. P., Li, Z., Peddi, S., Kakan, S., Mehta, A., Tyrpak, D., . . . MacKay, J. A. (2017).
Berunda Polypeptides: Multi-Headed Fusion Proteins Promote Subcutaneous
Administration of Rapamycin to Breast Cancer In Vivo. Theranostics, 7(16), 3856-3872.
doi:10.7150/thno.19981
Dhandhukia, J. P., Shi, P., Peddi, S., Li, Z., Aluri, S., Ju, Y., . . . MacKay, J. A. (2017). Bifunctional
Elastin-like Polypeptide Nanoparticles Bind Rapamycin and Integrins and Suppress Tumor
Growth in Vivo. Bioconjugate Chemistry, 28(11), 2715-2728.
doi:10.1021/acs.bioconjchem.7b00469
Doyle, M. E., Boggs, L., Attia, R., Cooper, L. R., Saban, D. R., Nguyen, C. Q., & Peck, A. B.
(2007). Autoimmune dacryoadenitis of NOD/LtJ mice and its subsequent effects on tear
protein composition. Am J Pathol, 171(4), 1224-1236. doi:10.2353/ajpath.2007.070388
97

Ertürk, A., & Bradke, F. (2013). High-resolution imaging of entire organs by 3-dimensional
imaging of solvent cleared organs (3DISCO). Experimental Neurology, 242, 57-64.
doi:https://doi.org/10.1016/j.expneurol.2012.10.018
Evans, E., Zhang, W., Jerdeva, G., Chen, C. Y., Chen, X., Hamm-Alvarez, S. F., & Okamoto, C.
T. (2008). Direct interaction between Rab3D and the polymeric immunoglobulin receptor
and trafficking through regulated secretory vesicles in lacrimal gland acinar cells. Am J
Physiol Cell Physiol, 294(3), C662-674. doi:10.1152/ajpcell.00623.2006
Feng, Y., Chung, D., Garrard, L., McEnroe, G., Lim, D., Scardina, J., . . . Endemann, G. (1998).
Peptides derived from the complementarity-determining regions of anti-Mac-1 antibodies
block intercellular adhesion molecule-1 interaction with Mac-1. J Biol Chem, 273(10),
5625-5630.
Ferrer, M. C., Shuvaev, V. V., Zern, B. J., Composto, R. J., Muzykantov, V. R., & Eckmann, D.
M. (2014). Icam-1 targeted nanogels loaded with dexamethasone alleviate pulmonary
inflammation. PLoS One, 9(7), e102329. doi:10.1371/journal.pone.0102329
Foulks, G. N., Forstot, S. L., Donshik, P. C., Forstot, J. Z., Goldstein, M. H., Lemp, M. A., . . .
Jacobs, D. S. (2015). Clinical Guidelines for Management of Dry Eye Associated with
Sjögren Disease. Ocul Surf, 13(2), 118-132. doi:https://doi.org/10.1016/j.jtos.2014.12.001
Fraenkel, M., Ketzinel-Gilad, M., Ariav, Y., Pappo, O., Karaca, M., Castel, J., . . . Leibowitz, G.
(2008). mTOR Inhibition by Rapamycin Prevents β-Cell Adaptation to Hyperglycemia and
Exacerbates the Metabolic State in Type 2 Diabetes. Diabetes, 57(4), 945-957.
doi:10.2337/db07-0922

98

Fraenkel, M., Ketzinel-Gilad, M., Ariav, Y., Pappo, O., Karaca, M., Castel, J., . . . Leibowitz, G.
(2008). mTOR inhibition by rapamycin prevents beta-cell adaptation to hyperglycemia and
exacerbates the metabolic state in type 2 diabetes. Diabetes, 57(4), 945-957.
doi:10.2337/db07-0922
Fukuda, M. (2008). Regulation of secretory vesicle traffic by Rab small GTPases. Cell Mol Life
Sci, 65(18), 2801-2813. doi:10.1007/s00018-008-8351-4
Gaubitz, M. (2006). Epidemiology of connective tissue disorders. Rheumatology (Oxford), 45
Suppl 3, iii3-4. doi:10.1093/rheumatology/kel282
Gilbert, K. M., & Weigle, W. O. (1993). Th1 cell anergy and blockade in G1a phase of the cell
cycle. The Journal of Immunology, 151(3), 1245-1254.
Girgrah, N., Letarte, M., Becker, L. E., Cruz, T. F., Theriault, E., & Moscarello, M. A. (1991).
Localization of the CD44 glycoprotein to fibrous astrocytes in normal white matter and to
reactive astrocytes in active lesions in multiple sclerosis. J Neuropathol Exp Neurol, 50(6),
779-792.
Gudjonsson, T., Adriance, M. C., Sternlicht, M. D., Petersen, O. W., & Bissell, M. J. (2005).
Myoepithelial cells: their origin and function in breast morphogenesis and neoplasia. J
Mammary Gland Biol Neoplasia, 10(3), 261-272. doi:10.1007/s10911-005-9586-4
Guo, H., Lee, C., Shah, M., Janga, S. R., Edman, M. C., Klinngam, W., . . . MacKay, J. A. (2018).
A novel elastin-like polypeptide drug carrier for cyclosporine A improves tear flow in a
mouse model of Sjogren's syndrome. J Control Release, 292, 183-195.
doi:10.1016/j.jconrel.2018.10.026

99

Hackstein, H., Taner, T., Zahorchak, A. F., Morelli, A. E., Logar, A. J., Gessner, A., & Thomson,
A. W. (2003). Rapamycin inhibits IL-4--induced dendritic cell maturation in vitro and
dendritic cell mobilization and function in vivo. Blood, 101(11), 4457-4463.
doi:10.1182/blood-2002-11-3370
Hahne, M., Jager, U., Isenmann, S., Hallmann, R., & Vestweber, D. (1993). Five tumor necrosis
factor-inducible cell adhesion mechanisms on the surface of mouse endothelioma cells
mediate the binding of leukocytes. J Cell Biol, 121(3), 655-664.
Hamm-Alvarez, S. F., Janga, S. R., Edman, M. C., Madrigal, S., Shah, M., Frousiakis, S. E., . . .
Stohl, W. (2014). Tear cathepsin S as a candidate biomarker for Sjogren's syndrome.
Arthritis Rheumatol, 66(7), 1872-1881. doi:10.1002/art.38633
Holland, E. J., Luchs, J., Karpecki, P. M., Nichols, K. K., Jackson, M. A., Sall, K., . . . Shojaei, A.
(2017). Lifitegrast for the Treatment of Dry Eye Disease: Results of a Phase III,
Randomized, Double-Masked, Placebo-Controlled Trial (OPUS-3). Ophthalmology,
124(1), 53-60. doi:10.1016/j.ophtha.2016.09.025
Houde, V. P., Brûlé, S., Festuccia, W. T., Blanchard, P.-G., Bellmann, K., Deshaies, Y., & Marette,
A. (2010). Chronic Rapamycin Treatment Causes Glucose Intolerance and Hyperlipidemia
by Upregulating Hepatic Gluconeogenesis and Impairing Lipid Deposition in Adipose
Tissue. Diabetes, 59(6), 1338. doi:10.2337/db09-1324
Hu, Y., Nakagawa, Y., Purushotham, K. R., & Humphreys-Beher, M. G. (1992). Functional
changes in salivary glands of autoimmune disease-prone NOD mice. Am J Physiol, 263(4
Pt 1), E607-614. doi:10.1152/ajpendo.1992.263.4.E607
Humphreys-Beher, M. G. (1996). Animal models for autoimmune disease-associated xerostomia
and xerophthalmia. Adv Dent Res, 10(1), 73-75. doi:10.1177/08959374960100011501
100

Ishikawa, H., Hirata, S., Nishibayashi, Y., Kubo, H., Nannbae, M., Ohno, O., & Imura, S. (1994).
Role of adhesion molecules in the lymphoid cell distribution in rheumatoid synovial
membrane. Rheumatology International, 13(6), 229-236. doi:10.1007/bf00290200
Janga, S. R., Shah, M., Ju, Y., Meng, Z., Edman, M. C., & Hamm-Alvarez, S. F. (2019).
Longitudinal analysis of tear cathepsin S activity levels in male non-obese diabetic mice
suggests its potential as an early stage biomarker of Sjögren’s Syndrome. Biomarkers,
24(1), 91-102. doi:10.1080/1354750X.2018.1514656
Janib, S. M., Pastuszka, M., Aluri, S., Folchman-Wagner, Z., Hsueh, P. Y., Shi, P., . . . Mackay, J.
A. (2014). A quantitative recipe for engineering protein polymer nanoparticles. Polym
Chem, 5(5), 1614-1625. doi:10.1039/c3py00537b
Ju, Y., Guo, H., Yarber, F., Edman, M. C., Peddi, S., Janga, S. R., . . . Hamm-Alvarez, S. F. (2019).
Molecular Targeting of Immunosuppressants Using a Bifunctional Elastin-Like
Polypeptide. Bioconjugate Chemistry, 30(9), 2358-2372.
doi:10.1021/acs.bioconjchem.9b00462
Ju, Y., Janga, S. R., Klinngam, W., MacKay, J. A., Hawley, D., Zoukhri, D., . . . Hamm-Alvarez,
S. F. (2018). NOD and NOR mice exhibit comparable development of lacrimal gland
secretory dysfunction but NOD mice have more severe autoimmune dacryoadenitis. Exp
Eye Res, 176, 243-251. doi:10.1016/j.exer.2018.09.002
Kahan, B. D. (2000). Efficacy of sirolimus compared with azathioprine for reduction of acute renal
allograft rejection: a randomised multicentre study. The Rapamune US Study Group.
Lancet, 356(9225), 194-202.
101

Kallen, J., Welzenbach, K., Ramage, P., Geyl, D., Kriwacki, R., Legge, G., . . . Hommel, U. (1999).
Structural basis for LFA-1 inhibition upon lovastatin binding to the CD11a I-domain. J
Mol Biol, 292(1), 1-9. doi:10.1006/jmbi.1999.3047
Kamoi, M., Ogawa, Y., Nakamura, S., Dogru, M., Nagai, T., Obata, H., . . . Tsubota, K. (2012).
Accumulation of secretory vesicles in the lacrimal gland epithelia is related to non-
Sjogren's type dry eye in visual display terminal users. PLoS One, 7(9), e43688.
doi:10.1371/journal.pone.0043688
Kikutani, H., & Makino, S. (1992). The murine autoimmune diabetes model: NOD and related
strains. Adv Immunol, 51, 285-322.
Koning, G. A., Schiffelers, R. M., & Storm, G. (2002). Endothelial cells at inflammatory sites as
target for therapeutic intervention. Endothelium, 9(3), 161-171.
Kowalik, A. S., Johnson, C. L., Chadi, S. A., Weston, J. Y., Fazio, E. N., & Pin, C. L. (2007). Mice
lacking the transcription factor Mist1 exhibit an altered stress response and increased
sensitivity to caerulein-induced pancreatitis. Am J Physiol Gastrointest Liver Physiol,
292(4), G1123-1132. doi:10.1152/ajpgi.00512.2006
Lamming, D. W., Ye, L., Katajisto, P., Goncalves, M. D., Saitoh, M., Stevens, D. M., . . . Baur, J.
A. (2012). Rapamycin-induced insulin resistance is mediated by mTORC2 loss and
uncoupled from longevity. Science, 335(6076), 1638-1643. doi:10.1126/science.1215135
Laplante, M., & Sabatini, D. M. (2009). mTOR signaling at a glance. Journal of Cell Science,
122(20), 3589.
Larsen, J. L., Bennett, R. G., Burkman, T., Ramirez, A. L., Yamamoto, S., Gulizia, J., . . . Hamel,
F. G. (2006). Tacrolimus and sirolimus cause insulin resistance in normal sprague dawley
rats. Transplantation, 82(4), 466-470. doi:10.1097/01.tp.0000229384.22217.15
102

Lavoie, T. N., Lee, B. H., & Nguyen, C. Q. (2011). Current concepts: mouse models of Sjogren's
syndrome. J Biomed Biotechnol, 2011, 549107. doi:10.1155/2011/549107
Lee, B. H., Tudares, M. A., & Nguyen, C. Q. (2009). Sjögren's syndrome: An old tale with a new
twist. Arch Immunol Ther Exp (Warsz), 57(1). doi:10.1007/s00005-009-0002-4
Lee, C., Guo, H., Klinngam, W., Janga, S. R., Yarber, F., Peddi, S., . . . MacKay, J. A. (2019).
Berunda Polypeptides: Biheaded Rapamycin Carriers for Subcutaneous Treatment of
Autoimmune Dry Eye Disease. Molecular Pharmaceutics, 16(7), 3024-3039.
doi:10.1021/acs.molpharmaceut.9b00263
Lemp, M. A. (2005). Dry eye (Keratoconjunctivitis Sicca), rheumatoid arthritis, and Sjogren's
syndrome. Am J Ophthalmol, 140(5), 898-899. doi:10.1016/j.ajo.2005.06.031
Lemullois, M., Rossignol, B., & Mauduit, P. (1996). Immunolocalization of myoepithelial cells in
isolated acini of rat exorbital lacrimal gland: cellular distribution of muscarinic receptors.
Biol Cell, 86(2-3), 175-181.
Li, X., Wu, K., Edman, M., Schenke-Layland, K., MacVeigh-Aloni, M., Janga, S. R., . . . Hamm-
Alvarez, S. F. (2010). Increased expression of cathepsins and obesity-induced
proinflammatory cytokines in lacrimal glands of male NOD mouse. Invest Ophthalmol Vis
Sci, 51(10), 5019-5029. doi:10.1167/iovs.09-4523
Lin, B.-w., Chen, M.-z., Fan, S.-x., Chuck, R. S., & Zhou, S.-y. (2015). Effect of 0.025% FK-506
Eyedrops on Botulinum Toxin B–Induced Mouse Dry Eye. Invest Ophthalmol Vis Sci,
56(1), 45-53. doi:10.1167/iovs.13-12925
Long, E. O. (2011). Intercellular Adhesion Molecule 1 (ICAM-1): Getting a Grip on Leukocyte
Adhesion. J Immunol, 186(9), 10.4049/jimmunol.1100646.
doi:10.4049/jimmunol.1100646
103

MacDonald, A. S. (2001). A worldwide, phase III, randomized, controlled, safety and efficacy
study of a sirolimus/cyclosporine regimen for prevention of acute rejection in recipients of
primary mismatched renal allografts. Transplantation, 71(2), 271-280.
MacEwan, S. R., & Chilkoti, A. (2014). Applications of elastin-like polypeptides in drug delivery.
J Control Release, 190, 314-330. doi:10.1016/j.jconrel.2014.06.028
MacEwan, S. R., Hassouneh, W., & Chilkoti, A. (2014). Non-chromatographic purification of
recombinant elastin-like polypeptides and their fusions with peptides and proteins from
Escherichia coli. J Vis Exp(88). doi:10.3791/51583
Maciel, G., Crowson, C. S., Matteson, E. L., & Cornec, D. (2017). Prevalence of Primary Sjogren's
Syndrome in a US Population-Based Cohort. Arthritis Care Res (Hoboken), 69(10), 1612-
1616. doi:10.1002/acr.23173
Makino, S., Kunimoto, K., Muraoka, Y., Mizushima, Y., Katagiri, K., & Tochino, Y. (1980).
Breeding of a non-obese, diabetic strain of mice. Jikken Dobutsu, 29(1), 1-13.
doi:10.1538/expanim1978.29.1_1
Manikwar, P., Tejo, B. A., Shinogle, H., Moore, D. S., Zimmerman, T., Blanco, F., & Siahaan, T.
J. (2011). Utilization of I-domain of LFA-1 to Target Drug and Marker Molecules to
Leukocytes. Theranostics, 1, 277-289.
Many, M. C., Maniratunga, S., & Denef, J. F. (1996). The non-obese diabetic (NOD) mouse: an
animal model for autoimmune thyroiditis. Exp Clin Endocrinol Diabetes, 104 Suppl 3, 17-
20. doi:10.1055/s-0029-1211673

104

Markle, J. G., Frank, D. N., Mortin-Toth, S., Robertson, C. E., Feazel, L. M., Rolle-Kampczyk,
U., . . . Danska, J. S. (2013). Sex differences in the gut microbiome drive hormone-
dependent regulation of autoimmunity. Science, 339(6123), 1084-1088.
doi:10.1126/science.1233521
Markle, J. G., Mortin-Toth, S., Wong, A. S., Geng, L., Hayday, A., & Danska, J. S. (2013).
gammadelta T cells are essential effectors of type 1 diabetes in the nonobese diabetic mouse
model. J Immunol, 190(11), 5392-5401. doi:10.4049/jimmunol.1203502
McDaniel, J. R., Mackay, J. A., Quiroz, F. G., & Chilkoti, A. (2010). Recursive directional ligation
by plasmid reconstruction allows rapid and seamless cloning of oligomeric genes.
Biomacromolecules, 11(4), 944-952. doi:10.1021/bm901387t
McHale, J. F., Harari, O. A., Marshall, D., & Haskard, D. O. (1999). Vascular endothelial cell
expression of ICAM-1 and VCAM-1 at the onset of eliciting contact hypersensitivity in
mice: evidence for a dominant role of TNF-alpha. J Immunol, 162(3), 1648-1655.
Meng, Z., Edman, M. C., Hsueh, P. Y., Chen, C. Y., Klinngam, W., Tolmachova, T., . . . Hamm-
Alvarez, S. F. (2016). Imbalanced Rab3D versus Rab27 increases cathepsin S secretion
from lacrimal acini in a mouse model of Sjogren's Syndrome. Am J Physiol Cell Physiol,
310(11), C942-954. doi:10.1152/ajpcell.00275.2015
Meng, Z., Klinngam, W., Edman, M. C., & Hamm-Alvarez, S. F. (2017). Interferon-gamma
treatment in vitro elicits some of the changes in cathepsin S and antigen presentation
characteristic of lacrimal glands and corneas from the NOD mouse model of Sjogren's
Syndrome. PLoS One, 12(9), e0184781. doi:10.1371/journal.pone.0184781
Mestas, J., & Hughes, C. C. (2004). Of mice and not men: differences between mouse and human
immunology. J Immunol, 172(5), 2731-2738.
105

Meyer, D. E., & Chilkoti, A. (1999). Purification of recombinant proteins by fusion with thermally-
responsive polypeptides. Nat Biotechnol, 17(11), 1112-1115. doi:10.1038/15100
Meyer, D. E., & Chilkoti, A. (2002). Genetically encoded synthesis of protein-based polymers
with precisely specified molecular weight and sequence by recursive directional ligation:
examples from the elastin-like polypeptide system. Biomacromolecules, 3(2), 357-367.
Moir, L. M. (2016). Lymphangioleiomyomatosis: Current understanding and potential treatments.
Pharmacol Ther, 158, 114-124. doi:10.1016/j.pharmthera.2015.12.008
Mondino, A., & Mueller, D. L. (2007). mTOR at the crossroads of T cell proliferation and
tolerance. Semin Immunol, 19(3), 162-172. doi:10.1016/j.smim.2007.02.008
Morrisett, J. D., Abdel-Fattah, G., Hoogeveen, R., Mitchell, E., Ballantyne, C. M., Pownall, H.
J., . . . Kahan, B. D. (2002). Effects of sirolimus on plasma lipids, lipoprotein levels, and
fatty acid metabolism in renal transplant patients. J Lipid Res, 43(8), 1170-1180.
Moss, S. E., Klein, R., & Klein, B. E. K. (2000). Prevalence of and Risk Factors for Dry Eye
Syndrome. Archives of Ophthalmology, 118(9), 1264-1268.
doi:10.1001/archopht.118.9.1264
Murciano, J. C., Muro, S., Koniaris, L., Christofidou-Solomidou, M., Harshaw, D. W., Albelda, S.
M., . . . Muzykantov, V. R. (2003). ICAM-directed vascular immunotargeting of
antithrombotic agents to the endothelial luminal surface. Blood, 101(10), 3977-3984.
doi:10.1182/blood-2002-09-2853
Muro, S., & Muzykantov, V. R. (2005). Targeting of antioxidant and anti-thrombotic drugs to
endothelial cell adhesion molecules. Curr Pharm Des, 11(18), 2383-2401.
106

Muro, S., Schuchman, E. H., & Muzykantov, V. R. (2006). Lysosomal enzyme delivery by ICAM-
1-targeted nanocarriers bypassing glycosylation- and clathrin-dependent endocytosis. Mol
Ther, 13(1), 135-141. doi:10.1016/j.ymthe.2005.07.687
Muro, S., Wiewrodt, R., Thomas, A., Koniaris, L., Albelda, S. M., Muzykantov, V. R., & Koval,
M. (2003). A novel endocytic pathway induced by clustering endothelial ICAM-1 or
PECAM-1. J Cell Sci, 116(Pt 8), 1599-1609.
Murphy, C. J., Bentley, E., Miller, P. E., McIntyre, K., Leatherberry, G., Dubielzig, R., . . . O'Neill,
C. A. (2011). The pharmacologic assessment of a novel lymphocyte function-associated
antigen-1 antagonist (SAR 1118) for the treatment of keratoconjunctivitis sicca in dogs.
Invest Ophthalmol Vis Sci, 52(6), 3174-3180. doi:10.1167/iovs.09-5078
Nashida, T., Yoshie, S., Haga-Tsujimura, M., Imai, A., & Shimomura, H. (2013). Atrophy of
myoepithelial cells in parotid glands of diabetic mice; detection using skeletal muscle actin,
a novel marker(). FEBS Open Bio, 3, 130-134. doi:10.1016/j.fob.2013.01.009
Nguyen, C. Q., & Peck, A. B. (2009). Unraveling the pathophysiology of Sjogren syndrome-
associated dry eye disease. Ocul Surf, 7(1), 11-27.
Nikolov, N. P., & Illei, G. G. (2009). Pathogenesis of Sjögren's syndrome. Curr Opin Rheumatol,
21(5), 465-470. doi:10.1097/BOR.0b013e32832eba21
Nocturne, G., & Mariette, X. (2015). Sjogren Syndrome-associated lymphomas: an update on
pathogenesis and management. Br J Haematol, 168(3), 317-327. doi:10.1111/bjh.13192
Ohyama, Y., Carroll, V. A., Deshmukh, U., Gaskin, F., Brown, M. G., & Fu, S. M. (2006). Severe
focal sialadenitis and dacryoadenitis in NZM2328 mice induced by MCMV: a novel model
for human Sjögren's syndrome. J Immunol, 177(10), 7391-7397.
doi:10.4049/jimmunol.177.10.7391
107

Park, Y. S., Gauna, A. E., & Cha, S. (2015). Mouse Models of Primary Sjogren's Syndrome. Curr
Pharm Des, 21(18), 2350-2364.
Peddi, S., Pan, X., & MacKay, J. A. (2018). Intracellular Delivery of Rapamycin From FKBP
Elastin-Like Polypeptides Is Consistent With Macropinocytosis. Frontiers in
pharmacology, 9, 1184-1184. doi:10.3389/fphar.2018.01184
Pflugfelder, S. C., Stern, M., Zhang, S., & Shojaei, A. (2017). LFA-1/ICAM-1 Interaction as a
Therapeutic Target in Dry Eye Disease. Journal of Ocular Pharmacology and Therapeutics,
33(1), 5-12. doi:10.1089/jop.2016.0105
Pin, C. L., Bonvissuto, A. C., & Konieczny, S. F. (2000). Mist1 expression is a common link
among serous exocrine cells exhibiting regulated exocytosis. Anat Rec, 259(2), 157-167.
Qin, B., Wang, J., Yang, Z., Yang, M., Ma, N., Huang, F., & Zhong, R. (2015). Epidemiology of
primary Sjogren's syndrome: a systematic review and meta-analysis. Ann Rheum Dis,
74(11), 1983-1989. doi:10.1136/annrheumdis-2014-205375
Razzaque, M. S., & Taguchi, T. (1999). The possible role of colligin/HSP47, a collagen-binding
protein, in the pathogenesis of human and experimental fibrotic diseases. Histol
Histopathol, 14(4), 1199-1212. doi:10.14670/hh-14.1199
Reddy, V. Y., Zhang, Q. Y., & Weiss, S. J. (1995). Pericellular mobilization of the tissue-
destructive cysteine proteinases, cathepsins B, L, and S, by human monocyte-derived
macrophages. Proc Natl Acad Sci U S A, 92(9), 3849-3853.
Riese, R. J., Mitchell, R. N., Villadangos, J. A., Shi, G. P., Palmer, J. T., Karp, E. R., . . . Chapman,
H. A. (1998). Cathepsin S activity regulates antigen presentation and immunity. J Clin
Invest, 101(11), 2351-2363. doi:10.1172/jci1158
108

Roescher, N., Lodde, B. M., Vosters, J. L., Tak, P. P., Catalan, M. A., Illei, G. G., & Chiorini, J.
A. (2012). Temporal changes in salivary glands of non-obese diabetic mice as a model for
Sjogren's syndrome. Oral Dis, 18(1), 96-106. doi:10.1111/j.1601-0825.2011.01852.x
Rossin, R., Muro, S., Welch, M. J., Muzykantov, V. R., & Schuster, D. P. (2008). In vivo imaging
of 64Cu-labeled polymer nanoparticles targeted to the lung endothelium. J Nucl Med, 49(1),
103-111. doi:10.2967/jnumed.107.045302
Routsias, J. G., Goules, J. D., Charalampakis, G., Tzima, S., Papageorgiou, A., & Voulgarelis, M.
(2013). Malignant lymphoma in primary Sjogren's syndrome: an update on the
pathogenesis and treatment. Semin Arthritis Rheum, 43(2), 178-186.
doi:10.1016/j.semarthrit.2013.04.004
Saegusa, K., Ishimaru, N., Yanagi, K., Arakaki, R., Ogawa, K., Saito, I., . . . Hayashi, Y. (2002).
Cathepsin S inhibitor prevents autoantigen presentation and autoimmunity. J Clin Invest,
110(3), 361-369. doi:10.1172/JCI14682
Saito, I., Terauchi, K., Shimuta, M., Nishiimura, S., Yoshino, K., Takeuchi, T., . . . Miyasaka, N.
(1993). Expression of cell adhesion molecules in the salivary and lacrimal glands of
Sjogren's syndrome. J Clin Lab Anal, 7(3), 180-187.
Salmon, A. B. (2015). About-face on the metabolic side effects of rapamycin. Oncotarget, 6(5),
2585-2586. doi:10.18632/oncotarget.3354
Salomon, B., Rhee, L., Bour-Jordan, H., Hsin, H., Montag, A., Soliven, B., . . . Bluestone, J. A.
(2001). Development of spontaneous autoimmune peripheral polyneuropathy in B7-2-
deficient NOD mice. J Exp Med, 194(5), 677-684.
109

Sancak, Y., Peterson, T. R., Shaul, Y. D., Lindquist, R. A., Thoreen, C. C., Bar-Peled, L., &
Sabatini, D. M. (2008). The Rag GTPases bind raptor and mediate amino acid signaling to
mTORC1. Science, 320(5882), 1496-1501. doi:10.1126/science.1157535
Schenke-Layland, K., Xie, J., Magnusson, M., Angelis, E., Li, X., Wu, K., . . . Hamm-Alvarez, S.
F. (2010). Lymphocytic infiltration leads to degradation of lacrimal gland extracellular
matrix structures in NOD mice exhibiting a Sjogren's syndrome-like exocrinopathy. Exp
Eye Res, 90(2), 223-237. doi:10.1016/j.exer.2009.10.008
Schindler, C. E., Partap, U., Patchen, B. K., & Swoap, S. J. (2014). Chronic rapamycin treatment
causes diabetes in male mice. American journal of physiology. Regulatory, integrative and
comparative physiology, 307(4), R434-R443. doi:10.1152/ajpregu.00123.2014
Semba, C. P., Swearingen, D., Smith, V. L., Newman, M. S., O'Neill, C. A., Burnier, J. P., . . .
Gadek, T. R. (2011). Safety and pharmacokinetics of a novel lymphocyte function-
associated antigen-1 antagonist ophthalmic solution (SAR 1118) in healthy adults. J Ocul
Pharmacol Ther, 27(1), 99-104. doi:10.1089/jop.2009.0105
Shah, M., Edman, M. C., Janga, S. R., Shi, P., Dhandhukia, J., Liu, S., . . . Hamm-Alvarez, S. F.
(2013). A rapamycin-binding protein polymer nanoparticle shows potent therapeutic
activity in suppressing autoimmune dacryoadenitis in a mouse model of Sjogren's
syndrome. J Control Release, 171(3), 269-279. doi:10.1016/j.jconrel.2013.07.016
Shah, M., Edman, M. C., Janga, S. R., Shi, P., Dhandhukia, J., Liu, S., . . . Hamm-Alvarez, S. F.
(2013). A rapamycin-binding protein polymer nanoparticle shows potent therapeutic
activity in suppressing autoimmune dacryoadenitis in a mouse model of Sjögren's
syndrome. J Control Release, 171(3), 269-279. doi:10.1016/j.jconrel.2013.07.016
110

Shah, M., Edman, M. C., Reddy Janga, S., Yarber, F., Meng, Z., Klinngam, W., . . . Hamm-Alvarez,
S. F. (2017). Rapamycin Eye Drops Suppress Lacrimal Gland Inflammation In a Murine
Model of Sjögren's Syndrome. Invest Ophthalmol Vis Sci, 58(1), 372-385.
doi:10.1167/iovs.16-19159

Sheppard, J., Kannarr, S., Luchs, J., Malhotra, R., Justice, A., Ogundele, A., . . . Bacharach, J.
(2020). Efficacy and Safety of OTX-101, a Novel Nanomicellar Formulation of
Cyclosporine A, for the Treatment of Keratoconjunctivitis Sicca: Pooled Analysis of a
Phase 2b/3 and Phase 3 Study. Eye Contact Lens, 46 Suppl 1, S14-s19.
doi:10.1097/icl.0000000000000636
Sheppard, J. D., Torkildsen, G. L., Lonsdale, J. D., D'Ambrosio, F. A., Jr., McLaurin, E. B.,
Eiferman, R. A., . . . Semba, C. P. (2014). Lifitegrast ophthalmic solution 5.0% for
treatment of dry eye disease: results of the OPUS-1 phase 3 study. Ophthalmology, 121(2),
475-483. doi:10.1016/j.ophtha.2013.09.015
Shim, G. J., Warner, M., Kim, H. J., Andersson, S., Liu, L., Ekman, J., . . . Gustafsson, J. A. (2004).
Aromatase-deficient mice spontaneously develop a lymphoproliferative autoimmune
disease resembling Sjogren's syndrome. Proc Natl Acad Sci U S A, 101(34), 12628-12633.
doi:10.1073/pnas.0405099101
Shimoboji, T., Ding, Z. L., Stayton, P. S., & Hoffman, A. S. (2002). Photoswitching of Ligand
Association with a Photoresponsive Polymer−Protein Conjugate. Bioconjugate Chemistry,
13(5), 915-919. doi:10.1021/bc010057q
Simamora, P., Alvarez, J. M., & Yalkowsky, S. H. (2001). Solubilization of rapamycin. Int J Pharm,
213(1-2), 25-29.
111

Simecek, P., Churchill, G. A., Yang, H., Rowe, L. B., Herberg, L., Serreze, D. V., & Leiter, E. H.
(2015). Genetic Analysis of Substrain Divergence in Non-Obese Diabetic (NOD) Mice.
G3 (Bethesda), 5(5), 771-775. doi:10.1534/g3.115.017046
Sisto, M., Lorusso, L., Ingravallo, G., Tamma, R., Nico, B., Ribatti, D., . . . Lisi, S. (2018). Reduced
myofilament component in primary Sjogren's syndrome salivary gland myoepithelial cells.
J Mol Histol, 49(2), 111-121. doi:10.1007/s10735-017-9751-2
Sreekumar, P. G., Li, Z., Wang, W., Spee, C., Hinton, D. R., Kannan, R., & MacKay, J. A. (2018).
Intra-vitreal alphaB crystallin fused to elastin-like polypeptide provides neuroprotection in
a mouse model of age-related macular degeneration. J Control Release, 283, 94-104.
doi:10.1016/j.jconrel.2018.05.014
St Clair, E. W., Angellilo, J. C., & Singer, K. H. (1992). Expression of cell-adhesion molecules in
the salivary gland microenvironment of Sjogren's syndrome. Arthritis Rheum, 35(1), 62-
66.
Takahashi, M., Ishimaru, N., Yanagi, K., Haneji, N., Saito, I., & Hayashi, Y. (1997). High
incidence of autoimmune dacryoadenitis in male non-obese diabetic (NOD) mice
depending on sex steroid. Clin Exp Immunol, 109(3), 555-561.
Tauber, J., Karpecki, P., Latkany, R., Luchs, J., Martel, J., Sall, K., . . . Semba, C. P. (2015).
Lifitegrast Ophthalmic Solution 5.0% versus Placebo for Treatment of Dry Eye Disease:
Results of the Randomized Phase III OPUS-2 Study. Ophthalmology, 122(12), 2423-2431.
doi:10.1016/j.ophtha.2015.08.001
Thomson, A. W., Turnquist, H. R., & Raimondi, G. (2009). Immunoregulatory functions of mTOR
inhibition. Nat Rev Immunol, 9(5), 324-337. doi:10.1038/nri2546
112

Tian, X., Jin, R. U., Bredemeyer, A. J., Oates, E. J., Blazewska, K. M., McKenna, C. E., & Mills,
J. C. (2010). RAB26 and RAB3D are direct transcriptional targets of MIST1 that regulate
exocrine granule maturation. Mol Cell Biol, 30(5), 1269-1284. doi:10.1128/mcb.01328-09
Tibbetts, S. A., Seetharama Jois, D., Siahaan, T. J., Benedict, S. H., & Chan, M. A. (2000). Linear
and cyclic LFA-1 and ICAM-1 peptides inhibit T cell adhesion and function. Peptides,
21(8), 1161-1167.
Toda, I., Sullivan, B. D., Rocha, E. M., Da Silveira, L. A., Wickham, L. A., & Sullivan, D. A.
(1999). Impact of gender on exocrine gland inflammation in mouse models of Sjogren's
syndrome. Exp Eye Res, 69(4), 355-366. doi:10.1006/exer.1999.0715
Trepanier, D. J., Gallant, H., Legatt, D. F., & Yatscoff, R. W. (1998). Rapamycin: distribution,
pharmacokinetics and therapeutic range investigations: an update. Clin Biochem, 31(5),
345-351.
Tsoi, K. M., MacParland, S. A., Ma, X. Z., Spetzler, V. N., Echeverri, J., Ouyang, B., . . . Chan,
W. C. (2016). Mechanism of hard-nanomaterial clearance by the liver. Nat Mater, 15(11),
1212-1221. doi:10.1038/nmat4718
Tsubota, K., Fujita, H., Tsuzaka, K., & Takeuchi, T. (2000). Mikulicz's disease and Sjogren's
syndrome. Invest Ophthalmol Vis Sci, 41(7), 1666-1673.
Turnquist, H. R., Raimondi, G., Zahorchak, A. F., Fischer, R. T., Wang, Z., & Thomson, A. W.
(2007). Rapamycin-conditioned dendritic cells are poor stimulators of allogeneic CD4+ T
cells, but enrich for antigen-specific Foxp3+ T regulatory cells and promote organ
transplant tolerance. J Immunol, 178(11), 7018-7031.
113

Urry, D. W. (1993). Molecular Machines: How Motion and Other Functions of Living Organisms
Can Result from Reversible Chemical Changes. Angewandte Chemie International Edition
in English, 32(6), 819-841. doi:10.1002/anie.199308191
Urry, D. W., Hayes, L. C., Gowda, D. C., Harris, C. M., & Harris, R. D. (1992). Reduction-driven
polypeptide folding by the ΔTt mechanism. Biochem Biophys Res Commun, 188(2), 611-
617. doi:https://doi.org/10.1016/0006-291X(92)91100-5
Verheul, H. A., Verveld, M., Hoefakker, S., & Schuurs, A. H. (1995). Effects of ethinylestradiol
on the course of spontaneous autoimmune disease in NZB/W and NOD mice.
Immunopharmacol Immunotoxicol, 17(1), 163-180. doi:10.3109/08923979509052727
Villadangos, J. A., Bryant, R. A., Deussing, J., Driessen, C., Lennon-Dumenil, A. M., Riese, R.
J., . . . Ploegh, H. L. (1999). Proteases involved in MHC class II antigen presentation.
Immunol Rev, 172, 109-120.
Wang, W., Jashnani, A., Aluri, S. R., Gustafson, J. A., Hsueh, P.-Y., Yarber, F., . . . MacKay, J.
A. (2015). A thermo-responsive protein treatment for dry eyes. J Control Release, 199,
156-167. doi:10.1016/j.jconrel.2014.11.016
Waterman, S. A., Gordon, T. P., & Rischmueller, M. (2000). Inhibitory effects of muscarinic
receptor autoantibodies on parasympathetic neurotransmission in Sjogren's syndrome.
Arthritis Rheum, 43(7), 1647-1654. doi:10.1002/1529-0131(200007)43:7<1647::aid-
anr31>3.0.co;2-p
Willeke, P., Schluter, B., Schotte, H., Domschke, W., Gaubitz, M., & Becker, H. (2009).
Interferon-gamma is increased in patients with primary Sjogren's syndrome and Raynaud's
phenomenon. Semin Arthritis Rheum, 39(3), 197-202.
doi:10.1016/j.semarthrit.2008.04.002
114

Wisse, E., De Zanger, R. B., Charels, K., Van Der Smissen, P., & McCuskey, R. S. (1985). The
liver sieve: considerations concerning the structure and function of endothelial fenestrae,
the sinusoidal wall and the space of Disse. Hepatology, 5(4), 683-692.
Wu, K., Joffre, C., Li, X., MacVeigh-Aloni, M., Hom, M., Hwang, J., . . . Hamm-Alvarez, S. F.
(2009). Altered expression of genes functioning in lipid homeostasis is associated with
lipid deposition in NOD mouse lacrimal gland. Exp Eye Res, 89(3), 319-332.
doi:10.1016/j.exer.2009.03.020
Wullschleger, S., Loewith, R., & Hall, M. N. (2006). TOR signaling in growth and metabolism.
Cell, 124(3), 471-484. doi:10.1016/j.cell.2006.01.016
Yang, S.-B., Lee, H. Y., Young, D. M., Tien, A.-C., Rowson-Baldwin, A., Shu, Y. Y., . . . Jan, L.
Y. (2012). Rapamycin induces glucose intolerance in mice by reducing islet mass, insulin
content, and insulin sensitivity. Journal of molecular medicine (Berlin, Germany), 90(5),
575-585. doi:10.1007/s00109-011-0834-3
Yatscoff, R., LeGatt, D., Keenan, R., & Chackowsky, P. (1993). Blood distribution of rapamycin.
Transplantation, 56(5), 1202-1206.
Yatscoff, R. W., Wang, P., Chan, K., Hicks, D., & Zimmerman, J. (1995). Rapamycin: distribution,
pharmacokinetics, and therapeutic range investigations. Ther Drug Monit, 17(6), 666-671.
Yeboah, A., Cohen, R. I., Rabolli, C., Yarmush, M. L., & Berthiaume, F. (2016). Elastin-like
polypeptides: A strategic fusion partner for biologics. Biotechnol Bioeng, 113(8), 1617-
1627. doi:10.1002/bit.25998
Zhang, X., Zhao, L., Deng, S., Sun, X., & Wang, N. (2016). Dry Eye Syndrome in Patients with
Diabetes Mellitus: Prevalence, Etiology, and Clinical Characteristics. Journal of
ophthalmology, 2016, 8201053-8201053. doi:10.1155/2016/8201053
115

Zimmerman, J. J., & Kahan, B. D. (1997). Pharmacokinetics of sirolimus in stable renal transplant
patients after multiple oral dose administration. J Clin Pharmacol, 37(5), 405-415.
Zoukhri, D., Macari, E., & Kublin, C. L. (2007). A single injection of interleukin-1 induces
reversible aqueous-tear deficiency, lacrimal gland inflammation, and acinar and ductal cell
proliferation. Exp Eye Res, 84(5), 894-904. doi:10.1016/j.exer.2007.01.015

Abstract (if available)

Linked assets

University of Southern California Dissertations and Theses

Conceptually similar

PDF

Cathepsin L changes in development of autoimmune dacryoadenitis

PDF

Targeting LFA-1/ICAM-1 interaction using ICAM-1 binding elastin-like polypeptide

PDF

Development of an elastin-like polypeptide-based cyclosporine A formulation that improves autoimmune-mediated dry eye characteristic of Sj鰃ren抯 syndrome

PDF

The trafficking and pathogenesis of cathepsin S in Sjögren’s syndrome

PDF

Characterization of cathepsin S as potential biomarkers of Sjögren’s syndrome in mouse models

PDF

Development and therapeutic assessment of multivalent protein polymers for cancer and eye diseases

PDF

Temperature triggered protein assembly enables signaling switching and peptide drug delivery

PDF

Cognate receptor fusion proteins enable parenteral delivery of challenging rapalogues

PDF

Increased cathepsin S plays an important role in ocular surface manifestations in Sjögren’s syndrome

PDF

Exploring serum and tear micro-RNA as biomarkers for early diagnosis of Sjögren’s Syndrome

PDF

Secretion of exosomes through the potential endolysosomal pathway in Rab3DKO and Non-Obese Diabetic murine models

PDF

Three advancements in biotechnology: new tools for synthetic biology and next generation sequencing

PDF

Investigation of mechanisms underlying development of autoimmune dacryoadenitis in lacrimal gland

PDF

Effects of particle architecture on in-vivo pharmacokinetics and bio-distribution of therapeutic nanostructures

PDF

Integrin-mediated targeting of protein polymer nanoparticles carrying a cytostatic macrolide

PDF

Tubulin-based fusion proteins as multifunctional tools

PDF

Development of protein polymer therapeutics for the eye

PDF

Comparing drug release kinetics of rapamycin bound FKBP-ELP fusion proteins

PDF

Cellular uptake mechanism of elastin-like polypeptide fusion proteins

PDF

Controlled ocular drug delivery using peptide-mediated phase separation

Asset Metadata

Creator Ju, Yaping (author)

Core Title Development of novel immunosuppressant-based therapies to treat dacryoadenitis in a Sjögren’s syndrome mouse model

School School of Pharmacy

Degree Doctor of Philosophy

Degree Program Pharmaceutical Sciences

Publication Date 12/06/2020

Defense Date 10/20/2020

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

Tag drug delivery,elastin-like polypeptides,non-obese diabetic mice,OAI-PMH Harvest,rapamycin,Sjögren’s syndrome

Language English

Contributor Electronically uploaded by the author (provenance)

Advisor Hamm-Alvarez, Sarah (committee chair), MacKay, John Andrew (committee member), Okamoto, Curtis (committee member)

Creator Email juyaping1992@outlook.com,yju@usc.edu

Permanent Link (DOI) https://doi.org/10.25549/usctheses-c89-399983

Unique identifier UC11668216

Identifier etd-JuYaping-9181.pdf (filename),usctheses-c89-399983 (legacy record id)

Legacy Identifier etd-JuYaping-9181.pdf

Dmrecord 399983

Document Type Dissertation

Rights Ju, Yaping

Type texts

Source 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 a...

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