University of Southern California Dissertations and Theses

Protein deamidation mediated metabolic reprogramming during KSHV lytic replication

(USC Thesis Other)

Protein deamidation mediated metabolic reprogramming during KSHV lytic replication

PDF

Download a page range

Download transcript

Copy asset link

Request this asset

Transcript (if available)

Content

PROTEIN DEAMIDATION MEDIATED METABOLIC REPROGRAMMING
DURING KSHV LYTIC REPLICATION

by

Mao Tian

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
(MEDICAL BIOLOGY)

May 2021

Copyright 2021
Mao Tian
ii

ACKNOWLEDGEMENTS
First and foremost, I would like to thank my esteemed supervisors, Prof. Pinghui Feng
for his invaluable supervision, continuous support, and patience during my PhD study. His
immense knowledge and experience in science have encouraged me during my academic
research, and his kind-heartedness has helped me adjust my daily life in USC. I am extremely
grateful to my committee chair, Dr. James Ou, my committee member, Dr. Weiming Yuan, and
Dr. Jian Xu for their generous help and advice during my PhD. I would also like to thank my lab
colleagues, Dr. Jun Zhao, Dr. Jessica Carriere, Dr. Na Xie, Dr. Qizhi Liu, and Dr. Xiaoxi Lin for
their technical support on my study. It is also their kindness that have made my study and life in
the US a wonderful time. Finally, I would like to express my gratitude to my family and my
friends. Without their tremendous trust and encouragement in the past few years, it would be
impossible for me to complete my research.
iii

TABLE of CONTENTS
ACKNOWLEDGEMENTS ............................................................................................................. ii
LIST OF TABLES .......................................................................................................................... v
LIST OF FIGURES ...................................................................................................................... vi
ABSTRACT ................................................................................................................................. vii
Chapter 1: Introduction ................................................................................................................. 1
1. Oncogenic Herpesvirus: KSHV ............................................................................................. 1
1.1 Herpesvirus family ........................................................................................................... 1
1.2 Oncogenic virus .............................................................................................................. 2
1.3 Kaposi’s Sarcoma ........................................................................................................... 3
1.4 Two phases of KSHV: Lytic Replication and Latent Infection ......................................... 5
1.5 Major ORFs and microRNA in KSHV .............................................................................. 9
1.6 Oxidative Stress and Oncogenesis in KSHV ................................................................ 14
1.7 Crosstalk between KSHV and other pathogens ............................................................ 16
2. Central Carbon Metabolism Reprogramming during virus-infection ................................... 18
2.1 Central Carbon Metabolism .......................................................................................... 18
2.2 Pentose Phosphate Pathway ........................................................................................ 20
2.3 PPP and Glycolysis Balance ......................................................................................... 22
2.4 G6PD ............................................................................................................................ 23
2.6 G6PD and other PPP enzymes in cancers ................................................................... 25
2.7 G6PD in other diseases ................................................................................................ 26
3. Protein deamidation and deamidase .................................................................................. 28
4. Multi-omics Analysis............................................................................................................ 32
4.1 Transcriptomics Analysis: RNA-sequencing and single cell-sequencing ...................... 32
4.2 Post-translational modification (PTM) Analysis ............................................................. 33
4.3 Metabolomics and Metabolic Flux ................................................................................. 35
4.4 Multi-omics integration methods ................................................................................... 35
Chapter 2 Protein Deamidation in KSHV-mediate Metabolic Reprogramming ........................... 37
1.Introduction ...................................................................................................................... 37
2. Result .............................................................................................................................. 38
Chapter 3 Method ....................................................................................................................... 80
Data and Code Availability .................................................................................................. 80
Experimental Model and Subject Details ............................................................................ 80
iv

Experiment Method ............................................................................................................. 81
Quantification and Statistical Analysis ................................................................................ 90
Chapter 4 Discussion and Perspective ....................................................................................... 91
Discussion ........................................................................................................................... 91
Perspective ......................................................................................................................... 93
References .................................................................................................................................. 95

v

LIST OF TABLES
TABLE 1 HERPESVIRUS FAMILY AND ASSOCIATED DISEASE, TARGET CELLS .............................................. 2
TABLE 2 TYPES OF KAPOSI’S SARCOMA ............................................................................................................... 3
TABLE 3 OXIDATIVE PPP REACTIONS ................................................................................................................. 20
TABLE 4 NON-OXIDATIVE PPP REACTIONS ........................................................................................................ 21
TABLE 5 GLUTAMINE AMIDOTRANSFERASE ........................................................................................................ 31
TABLE 6 COMMON PTMS ....................................................................................................................................... 34

vi

LIST OF FIGURES
FIGURE 1- 1 KSHV EPISOME .......................................................................................................... 9
FIGURE 1- 2 CENTRAL CARBON METABOLISM ................................................................................ 19
FIGURE 1- 4 G6PD IN OXIDATIVE RESPONSE ................................................................................. 23
FIGURE 1- 3 HISTORY OF G6PD .................................................................................................... 24
FIGURE 1- 5 METABOLOMICS ANALYSIS WORKFLOW ...................................................................... 35
FIGURE 2- 1 METABOLOMICS ANALYSIS SHOWED KSHV ................................................................. 41
FIGURE 2- 2 CENTRAL CARBON METABOLISM IS UP REGULATED DURING KSHV LYTIC REPLICATION . 47
FIGURE 2- 3 PROTEOMICS ANALYSIS OF KSHV LYTIC REPLICATION ............................................... 53
FIGURE 2- 4 PPP IS REQUIRED FOR KSHV LYTIC REPLICATION ...................................................... 61
FIGURE 2- 5 THE FUNCTIONAL ROLE OF G6PD DEAMIDATION ........................................................ 68
FIGURE 2- 6 CAD DEAMIDATES G6PD TO PROMOTE KSHV REPLICATION ....................................... 76
SUPPLEMENTARY FIGURE 2- 1 ...................................................................................................... 43
SUPPLEMENTARY FIGURE 2- 2 ...................................................................................................... 49
SUPPLEMENTARY FIGURE 2- 3 ...................................................................................................... 55
SUPPLEMENTARY FIGURE 2- 4 ...................................................................................................... 57
SUPPLEMENTARY FIGURE 2- 5 ...................................................................................................... 58
SUPPLEMENTARY FIGURE 2- 6 ...................................................................................................... 63
SUPPLEMENTARY FIGURE 2- 7 ...................................................................................................... 64
SUPPLEMENTARY FIGURE 2- 8 ...................................................................................................... 70
SUPPLEMENTARY FIGURE 2- 9 ...................................................................................................... 72
SUPPLEMENTARY FIGURE 2- 10 .................................................................................................... 78

vii

ABSTRACT
Deamidation is a post-translational modification of glutamine and asparagine, which is
catalyzed by cellular glutamine aminotransferase (GATs). Deamidation regulates basic
biological functions, such as the immune response and transcription activity. Recently, our
group identified that deamidated RelA upregulates anaerobic glycolysis in cancers. Kaposi’s
sarcoma-associated herpesvirus (KSHV) is an oncogenic virus that causes Kaposi’s Sarcoma
(KS) and B cell lymphoma in immune-compromised individuals. Previous studies have shown
that latent infection of KSHV reprograms host metabolism. However, the mechanism of how
KSHV lytic replication reprograms host cells remains rudimentary. Here, we reported that KSHV
activates CAD, a de novo pyrimidine synthesis enzyme, to deamidate G6PD, the pentose
phosphate pathway (PPP) rate-limiting enzyme, to upregulate carbon metabolism to favor viral
lytic replication. We applied both metabolomics and proteomics as a system approach to study
the reprogramming process with biochemical validation for the regulatory mechanism. We found
a robust parallel activation in central carbon metabolism and deamidation in key central carbon
metabolism enzymes by analyzing proteomics and metabolomics analyses. We applied
13
C-
glucose labeled tracing for metabolic fluxes and chemical inhibitor screening to reveal that PPP
is critical for KSHV replication. Then we identified G6PD is deamidated during KSHV infection,
and G6PD’s deamidation is coupled with its increased activity during KSHV infection.
Reconstructed G6PD deamidation mutants have higher activities than wild type and promote
KSHV replication. We screened mammalian GATs and identified CAD deamidated G6PD during
KSHV infection, where inhibition of CAD can impair G6PD deamidation and reduce KSHV
replication. Our work uncovers the key role of protein deamidation directly regulating metabolic
enzymes and metabolic reprogramming in general, expanding the functional repertoire of CAD
as a responsive deamidase that regulates PPP to reprograming the central carbon metabolism.
The system approach we developed will provide a new pipeline to research the metabolism
reprogramming in virus-associated disease.
1

Chapter 1: Introduction
1. Oncogenic Herpesvirus: KSHV
1.1 Herpesvirus family
Herpesvirus family is a widely spread DNA virus family with more than 130 members
and infects different species, including mammals, birds, fishes, etc. Herpes virus is formed by
four parts: a dsDNA genome, a spherical nucleocapsid, a tegument layer, and an envelope.
(Mettenleiter et al., 2009). Herpesviruses contain a double-strand DNA (dsDNA) genome from
50 kb to 200 kb and have two replication phases: latent infection and lytic replication. Lytic
replication produces infectious virus by several steps: 1) viral attachment and entry; 2) capsid
transportation to the nucleus; 3) viral gene expression; 4) viral DNA replication; 5) capsid
assembly and maturation; 6) envelop formation and egress (Kukhanova et al., 2014). During
latent infection, the viral genome circularizes as double-stranded circular DNA and exists as an
episome in the host nucleus after nucleocapsid transports from cell membrane to the nucleus.
Lytic genes remain silent in latent infection, while latent genes consistently express.
Herpesvirus is causing diseases in humans and can be categorized into three sub-
families: alphaherpesviruses, betaherpesviruses, and gammaherpesviruses. Herpes simplex 1
(HSV-1) belongs to the alpha herpesviruses family and has a high prevalence rate among
populations. Lytic replicated HSV-1 can cause cold sores, while neurons can establish latency
infection. Latent infected HSV-1 in neurons is correlated with neurodegeneration disease,
including Alzheimer’s and Parkinson’s diseases (Harris and Harris, 2018). Cytomegalovirus
(HCMV) has the largest genomes among the herpesvirus families, with a 235 kb genome
encoding 208 proteins (Stern-Ginossar et al., 2012). Gamma herpesviruses has two well-known
human oncogenic viruses, Epstein-Barr virus (EBV) and Kaposi’s sarcoma-related virus
(KSHV). Both KSHV and EBV can establish life-long persistent infection in a human in the form
2

of latent infection or lytic replication (Feng et al., 2013). Table 1 summarizes the common
mammalian herpesviruses and their viral disease, target cells, and latency sites.
Virus Disease Primary target cells Site of latency
Alpha herpesviruses

Herpes simplex 1 (HSV-1) Cold sores Mucoepithelial cells Neurons
Herpes simplex 2 (HSV-2) Genital ulcers Mucoepithelial cells Neurons
Beta herpesviruses

Cytomegalovirus(HCMV) Mononucleosis,
birth defects
Monocytes, epithelia,
lymphocytes,
Monocytes,
lymphocytes
Gamma herpesviruses

Epstein-Barr virus (EBV) Mononucleosis,
lymphoma
B cells B cells
Kaposi’s sarcoma-related virus
(KSHV)
Kaposi’s Sarcoma Lymphocytes B cells
Table 1 Herpesvirus Family Members, Associated Disease, Target Cells, and Latency Sites

1.2 Oncogenic virus
Oncogenic viruses are the causing agent of 20% of cancers in human and include both
RNA and DNA viruses, such as hepatitis B virus (HBV), hepatitis C virus (HCV), human
papillomavirus (HPV), EBV, KSHV, Merkel cell polyomavirus (MCPyV), and human T-
lymphotropic virus (HTLV-1) (Luo and Ou, 2015). The first oncogenic virus, Rous Sarcoma virus
(RRV), was identified in 1910 in chicken, where Dr. Rous found a healthy chicken developed
sarcoma after infecting with supernatant of grinded sarcoma tissue (Weiss and Vogt, 2011).
After that, few oncogenic viruses including murine leukemia virus (MuLV), feline leukemia virus
(FeLV) using similar methods. Oncogenic viruses share similar features. For instance,
oncogenic virus-infected cells have transformed cell properties in vitro, including loss of contact
inhibition, ability to proliferate permanently. The reagent’s exposure time is used to distinguish
3

the oncogenic virus and carcinogens. Oncogenic virus has the transient effect, which means
once cells infected with oncogenic virus, it can be transformed into cancer cells. But the
carcinogens have continuous effects on cells and need longer exposure time to transform
normal cells. Oncogenic virus is usually temperature-sensitive compares to carcinogens.
Among the herpesvirus, EBV is the first identified oncogenic virus in Burkitt’s lymphoma
cell lines in 1964 (Middeldorp et al., 2003), while KSHV is identified at 1994. Kaposi’s sarcoma
was initially described along with the AIDS epidemic together with non-Hodgkin’s lymphomas in
1981. However, it was until 1994 that KSHV was first discovered and then isolated by Patrick S.
Moore et al (Chang et al., 1994). KSHV can be transmitted through sexual transmission,
mother-to-child transmission, and saliva transmission. The seroprevalence is 10% in the U.S
(Simpson et al., 1996). But, the seroprevalence of KSHV varies across different regions globally
(Antman and Chang, 2009), the highest seroprevalence can go up 50% in some areas of
African (Gao et al., 1996), and the lowest is 0.2% in Japan (Fujii et al., 1999).
1.3 Kaposi’s Sarcoma
Kaposi’s sarcoma is the malignant lesion in soft tissues, such as skin, mucous membranes, or
lymph nodes, which is usually found in immune-deficient patients. KS lesion was formed by
purple, red, or brown blotches or tumors on the skin. There are four types of Kaposi’s sarcoma
originated from different resources.
Class Name Affected Population
Class I Epidemic (AIDS-associated) HIV Patients
Class II Classic (Mediterranean)
Mediterranean, Eastern European, and Middle Eastern
heritage
Class III Endemic (African) Equatorial Africa
Class IV Iatrogenic (transplant-related) Transplant patients
Table 2 Types of Kaposi’s Sarcoma
4

AIDS-associated KS is the most common KS that developed from HIV patients and can
be alleviated with highly active antiretroviral therapy (Tirelli and Bernardi, 2001). Classic KS
develops in people with weak immune systems in Mediterranean areas and can worsen when
the patients age. Endemic KS is associated with high prevalence rates of KSHV in African.
Endemic KS occurs in younger patients than other KS types and usually originate from lymph
nodes. Endemic KS cases are increasing with the HIV epidemic becoming more common in
African and are among the most common cancers in sub-African. Iatrogenic KS develops in
organ-transplantation patients when they take immune-suppressing drugs, and the KS
symptoms can decrease when stop taking the immune-suppressing drugs. Currently, there are
six treatments are used in KS: highly active antiretroviral therapy (HAART), radiation therapy,
surgery, cryosurgery, chemotherapy, immunotherapy. HAART combines multiple antiretroviral
drugs to that are used to alleviate the immune system damage by HIV infection. Therefore,
HAART is used to treat the epidemic KS. A median follow-up research of HIV
+
KSHV
+
patients
who used HAART has shown that HAART has high response rate in almost all the patients
(Cattelan et al., 2001). In a clinical response research, both alone and in combination of HAART
and chemotherapy and HAART alone can significantly reduce the KS (Pellet et al., 2001).
Radiation therapy is a traditional cancer treatment used for hundreds of years. Radiation, using
high-energy X-rays, targets and damages cancer cells, eventually leads to programmed cell
death. There are two types of radiation therapy, one uses the external machine that target the
specific area of body, another injects the radioactive substance into the cancer tissues directly.
Surgery procedures on the skin used both local excision and electrodesiccation to physical cut
the tumor tissue. Since KS normally develops visible lesion on skin or soft tissue, the surgery
curettage will excise the tumor directly with electrode to stop bleeding. Regarding the tumor
lesion area, the electrodesiccation process can be applied several times. Cryosurgery, also
called cryotherapy, freezes and destroys the abnormal skin cancer tissue. KS’s chemotherapy
commonly uses a drug called liposomal doxorubicin, known as Doxil. As KS is a systematic
5

disease that originated from different tissue simultaneously, blood injection of Doxil can kill the
cancer cells in different organs. For KS originated from the skin, topical agents can be applied
as a gel, for local lesions in the mouth, drugs are injected into the lesion, a treatment called
intralesional chemotherapy. Immunotherapy uses substance that can boost or restore the
body's natural defenses against KS. Commonly used immunotherapy drugs include IFN-α and
interleukin-12 (IL-12) (Couzin-Frankel, 2013).

1.4 Two phases of KSHV: Lytic Replication and Latent Infection
KSHV-associated malignancies, such as KS or PEL, are majorly composed of cells that
are KSHV latently infected, whereas only 1-5% of these lesions are cells supporting lytic
replication (Cesarman, 2014). Lytic replication promotes both de novo infection and cell
transformation through stimulation of cell proliferation and evasion of host cell innate immune
system. There are two phases of KSHV infections, lytic replication and latent infection. Upon
KSHV infection, most of cells will enter the latent infection by default and can be reactivated
under stress, such as oxidative stress and pathogen infection. Some cells can undergo lytic
replication upon high titer infection in vitro, such as human oral keratinocyte or lymphatic
endothelial cells.
Lytic Replication and Reactivation
KSHV contains two functional lytic origins that controls KSHV replication: the left-end
oriLyt (oriLyt-L) and the right-end oriLyt (oriLyt-R). OriLyt-L locates in the region between K4.2
and K5, while oriLyt-R is between open reading frame (ORF) 69 and vFLIP. oriLyt-L contains
various transcription factor (TF) binding sites with AT-rich regions and GC repeat regions.
oriLyt-R is an inverted duplication of oriLyt-L that is similar to the RRV oriLyt (AuCoin et al.,
2002). OriLyt region contains general TF binding sites, such as AT-rich region, AP1 TF-binding
sites, and a consensus TATA box motif; and viral TF response element, such as ORF50
response element. ORF50 is the major transcription regulator of KSHV lytic replication, also
6

called as replication and transcription activator (referred as RTA in the following text) (Papp et
al., 2019). OriLyt responds to KSHV’s core replication proteins, kRTA and other trans-acting
factors, including K8 (K-bZIP), ORF6, ORF9, ORF40, ORF41, ORF44, ORF56, and ORF59.
(AuCoin et al., 2004). In the transcription regulation core, RTA is the major trans activator of
KSHV, ORF6 is ssDNA binding protein; ORF9 is the DNA polymerase; ORF40 and ORF41 are
primase-associated factors; ORF44 is the helicase; ORF56 is the primase; and ORF59 is a
polymerase processivity factor. KSHV lytic replication requires the virus-specified origin binding
protein (OBP) to bind the lytic origin (ori-lyt) sequence and recruit the core replication
machinery. K8, a bZip protein with high similarity with EBV Zta protein, can bind to Ori-lyt
regions(Lin et al., 2003).
RTA is sufficient and essential for reactivation of KSHV latent infected cells (Kaul et al.,
2019), and forms tetramer in its transactivation and viral activation(Bu et al., 2007). The nuclear
translocalization of RTA was required from KSHV reactivation (Bu et al., 2008). As mentioned,
stress balance is critical for KSHV latency maintenance. Hypoxia stress can stimulate the KSHV
latent protein LANA to activate HIF-1α. HIF-1α then activates RTA’s promoter region by binding
the putative hypoxia response elements (HRE-1 and -6) inside (Cai et al., 2006).
Mammalian TF signaling also participate in KSHV lytic replication. For instance, RBP-Jκ,
a Notch-mediated transcription factor, interact with RTA when RTA is recruited to responsive
elements, which suggesting that RTA might activate cellular Notch signal transduction. When
human Notch (hNIC) was activated in KSHV latent infected BCBL1 , lytic genes, such as vIL-6,
K3 and K5, are induced, but the full repertoire of lytic viral gene expression cannot be evoked
(Chang et al., 2005a). Host topoisomerases I and II (Topo I and II) participates in KSHV lytic
DNA replication (González-Molleda et al., 2012). Treatment of Topo I and Topo II inhibitors can
inhibit KSHV DNA replication (González-Molleda et al., 2012). Cellular transcription factors are
required in KSHV latency maintenance and gene expression. During reactivation, CCCTC-
binding factor (CTCF) and cohesion complex is recruited for KSHV gene expression, but later
7

suppress the lytic gene expression, suggesting a complex regulatory process (Li et al., 2014).
RTA’s DNA binding ability can be modified by polyADP-ribosylation and phosphorylation by
poly(ADP-ribose) polymerase 1 (PARP-1) and kinase hKFC, thus repressing KSHV lytic
replication(Gwack et al., 2003a). During the KSHV lytic replication process, PARP-1 inhibit lytic
replication though ADP-ribosylation of RTA, whereas the overexpression of PF-8 can degrade
PARP-1 and enhance lytic tran-activation (Cheong et al., 2015).
Toll-like receptors (TLRs) are a group of transmembrane proteins in host innate immune
response. They are conserved with an extracellular leucine-rich repeats (LRRs) domain and a
conserved cytoplasmic Toll/IL-1 receptor (TIR) domain. TLRs family are critical for pathogens
recognition and innate immune pathways activation/ A screening of TLR agonists in their ability
to reactivate KSHV in PEL found that TLR7/8 agonists can activate the KSHV, as well as
vesicular stomatitis virus (VSV), a bona fide activator of TLR7/8. Activation of Toll-like receptor
can reactivate KSHV from latency (Gregory et al., 2009).
Chemical treatment can switch latently infected cells to lytic forms. The histone
deacetylation reagents, such as sodium butyrate and TPA, can activate latent infected KSHV
suggesting the reactivation of KSHV latent maintenance and lytic reactivation is not only a
transcriptional, but also an epigenomic regulation process(Candido et al., 1978). In the TPA-
induced reactivation of KSHV, cellular kinases are activated, including protein kinase C (PKC),
mitogen-activated protein kinase (MAPK), and extracellular signal-regulated kinase (ERK)
pathway, making kinase inhibitors a possible therapy for acute reactivation (Cohen et al., 2006).
Interestingly, Inflammatory cytokines have contradictory effects on lytic replication. IFN-ᵧ
can activate KSHV replication but IFN-α inhibit the ionomycin-induced KSHV lytic replication.
Other inflammatory cytokines, such as TNFα, TNFβ, IL-1, IL-2, and IL-6 has no effect on KSHV
replication (Chang et al., 2000).

Latent infection of KSHV
8

Different from the lytic replication that requires trans-activator RTA and K8, KSHV latent
protein expression requires a viral trimeric complex formed by ORF24, ORF31, and ORF34. In
the complex, ORF34 serves as the bridge that connects ORF24 and ORF31. The complex
binding to the K8.1 late promoter to regulate late protein expression. (Brulois et al., 2015).
LANA is a KSHV latent protein that critical for KSHV viral epitome maintenance and lytic
gene silencing. LANA has a proline-rich domain, three poly-Gln domains, a SUMO-2-interacting
motif (SIM), and a DNA-binding domain. The SIM can specifically recruit transcription repressor,
Kruppel-associated box domain-associated protein-1 (KAP-1) to form the KAP-1 and Sin3A
repressions in the LANA-associated complex (Cai et al., 2013). A research that infects multiple
proliferation cell lines with KSHV found KSHV establish stable latent infection in epithelial,
endothelial, and mesenchymal cell lines, suggesting the receptor and entry process of KSHV is
common(Bechtel et al., 2003). Latency establishment is a multi-step process with a combination
of trans-acting factors and epigenetic changes (Grundhoff and Ganem, 2004). RTA activates
the LANA promoter in a dose-dependent manner while LANA activates Notch pathway by
interacting with the downstream effector, RBP-Jκ, to suppress the RTA expression (Lan et al.,
2004) (Lan et al., 2005a) (Lan et al., 2005b).
Epigenomic regulation is reprogrammed during KSHV latent infection. Using de novo
infection in SLK cells, Grundhoff et al. found that latent-specific epigenetic patterns are
established soon after infection. Upon infection, the global DNA methylation patterns are
increased characterized by increased CpG methylation, while the heterochromatin marker,
H3K9-me3, is distributed among latent genome (Günther and Grundhoff, 2010). Other than the
common transcription regulators, chromatin remodeling regulates the latent and lytic replication
balance as well. Treatment of sodium butyrate and TSA can decrease the histone acetylation
and increase the accessibility of transcription core to lytic genes, such as the Ini1/Snf5 to
ORF50 promoter(Lu et al., 2003). Sodium butyrate treatment promotes the dissociation of LANA
9

from RTA promoter(Lu et al., 2006). Depletion of KAP-1, a transcriptional repressor controls
chromosomal remodeling, can suppress the LANA-associated complex and activate RTA-
mediated lytic gene expression (Chang et al., 2009). Also, epinephrine and norepinephrine
reactivate KSHV in PEL cells via cAMP/PKA signaling, suggesting the adrenergic activation
might control KSHV reactivation in patients (Chang et al., 2005b). High STAT3 expression is
correlated with the low lytic replication of KSHV latent infected cells. TRIM28 is suppressed after
knockdown of STAT3 thus leads to KSHV lytic replication (King et al., 2015).

1.5 Major ORFs and microRNA in KSHV

KSHV encodes at least 87 ORFs and 17 microRNAs, including 14 ORFs as cellular
orthologues such as viral interleukin-6 (vIL-6), vCyclin, vFLIP, vGPCR. Specifically, ORFs are
Figure 1- 1 KSHV Episome
Figure is adapted from (Avey et al., 2015)
10

categorized into lytic and latent genes, in which the lytic genes consist of early lytic,
intermediate lytic, and latent lytic genes. 14 miRNAs are co-expressed in latency-associated
region. Early lytic genes normally encode DNA replication-required proteins that promote
nucleotide synthesis and viral gene expression. Late lytic genes encode envelope and capsid
proteins for virion assembly and egress. (Avey et al., 2015)
RTA
ORF50 (RTA), a homolog of the EBV R protein, that can trigger KSHV reactivation by
over-expression and required for KSHV positive B cell reactivation was identified as lytic
replication activator in 2000 (Lukac et al., 1998) (Lukac et al., 1999) (Gradoville et al., 2000).
ORF57 and K-bZip (K8) both have the RTA binding region that consist a 12-bp palindromic
sequences and response to RTA activation(Lukac et al., 2001). RTA activates the KSHV lytic
gene translation by two mechanisms, one is binding the promoter region of lytic genes, another
is activating the host TF that promotes the lytic gene expression.
RTA has 691 amino acid (a.a.) with two disordered regions, two polar regions, and one
proline-rich regions. RTA has a multifunctional regulatory region, 520 – 535 a.a., that regulates
DNA binding affinity and protein stability. Deletion of that regulatory region can lead to the
abundant expression of variant, ORF50B, and increased DNA binding affinity (Chang and Miller,
2004). The DNA-binding inhibitory sequence (DBIS) is located at 490 – 535 a.a., while the
protein abundance regulatory signal (PARS) are divided into two parts, one is located between
590 and 650 a.a., another is overlap with the DBIS region. Intriguingly, the fusion of two
components of PARS can interact with glutathione S-transferase and degrade thereafter (Chang
et al., 2008). RTA contains SIMs and degrades SUMO-2/3 modified proteins(Izumiya et al.,
2013). A 3kb length RNA transcribed from the reserves strand of ORF50 can encode a small
peptide, vSP-1, and protect RTA from ubiquitin-mediated degradation (Jaber and Yuan, 2013).
A global screening of RTA-bound promoter screening identified more than 30 proteins
can be up-regulated by RTA. However, this process can be block with expression of NF-kB
11

subunit RelA though the binding of RTA coactivator RBP-Jκ (Izumiya et al., 2009). The few
RBP-Jκ recognition sites on KSHV, including ORF57 promoter and SSB promoter, mediated the
interaction between RBP-Jκ and RTA. Deletion of RBP-Jκ recognition sites leads to lower RTA
responsiveness, suggesting the Notch pathway activation during the KSHV reactivation (Liang
et al., 2002) (Liang and Ganem, 2003)
Demethylation of RTA promoter can activate the lytic transcription (Chen et al., 2001).
RTA can bind to cellular protein octamer-1(Oct-1) and DNA to transactivate the intermediate
lytic gene(Carroll et al., 2007). RTA’s promoter region contains an octamer motif that binds to
Oct-1 to enhance its transactivation while Oct-2 can competitively inhibit the RTA transcription
(Di Bartolo et al., 2009). Cellular chromatin-binding factor HMGB1 acts synergistically with RTA
to activate lytic gene expression without direct interaction between HMGB1 and RTA. This
synergistic response might be mediated though Oct-1 by forming a ternary complex (Harrison
and Whitehouse, 2008). RTA can also recruit chromatin remodeling proteins to transactivate
viral promoters, including CBP, SWI/SNF complex, and TRAP/Mediator.(Gwack et al., 2003b).
The cellular transcriptional repressor protein Hey1 is a target for RTA-mediated ubiquitination,
because RTA blocks the interaction between Hey1 and mSin3A (Gould et al., 2009). Since
proline residues in RTA are high conserved and RTA is enriched with phosphorylation in vivo,
RTA might be controlled by peptidyl-prolyl cis/trans isomerases (PPIases). Enhanced
expression of cellular Pin1 can promote RTA function but inhibit late gene synthesis and virion
production(Guito et al., 2014).
vIL-6
vIL-6 is the homolog of mammalian IL6 and has higher expression in patients with
KSHV-related diseases, such as KS, MCD, and PEL, as shown in serum samples. In KSHV
+

patients, HIV
+
patients also have higher vIL6 expression compared to HIV
-
patients, suggesting
vIL6 relates to malignancy development (Aoki et al. 2001). vIL-6 is a lytic gene but can be
12

expressed in latency in B cell lines, suggesting some lytic genes might expressed in a context-
dependent manner (Chandriani and Ganem, 2010).
vFLIP
Viral FLICE inhibitory protein (vFLIP) is a KSHV latent protein that activate NFκB during
latent infection for maintaining latent infection. During lytic replication, RTA recruits a novel
cellular ubiquitin E3 ligase targeting vFLIP degradation, leading to NFκB inhibition (Ehrlich et al.,
2014). In addition, vFLIP has more potent effects compared to cellular FLIP in B
cells. (Guasparri et al., 2004).
LANA
Latent nuclear antigen (LANA), encoded by ORF73, is the major latency regulator in
KSHV. Deletion of LANA can increase lytic gene expression, such as RTA, ORF57, vIL-6, and
increase virus titer (Li et al., 2008). LANA also regulate cellular transcription activity though
direct inhibition of NF-kB or upregulation of cellular Zinc finger protein, Staf. (Renne et al.,
2001).
Others
KSHV lytic replication requires the virus-specified origin binding protein (OBP) to bind
the lytic origin (ori-lyt) sequence and recruit the core replication machinery. Ori-lyt regions
distribute between K4.2 and K5, between K12 and ORF71 have similar sequences. K8, a bZip
protein with high similarity with EBV Zta protein, can bind to Ori-lyt regions(Lin et al., 2003).

miRNAs of KSHV
KSHV genome encodes 17 known miRNAs during latent infection, and the deletion of
miRNA clusters can reactivate KSHV in latent infected cells. A viral miRNA targetome shows
that more than 2,000 cellular mRNAs are targeted by KSHV miRNA, among which transcription
process is the major target (Gottwein et al., 2011). KSHV miRNA can target host TFs, such as
NF-κB, IκB-α, MYB, C/EBPα and ETS-1. The deletion of miRNA cluster of KSHV leads to a
13

reduction of NF-κB response. KSHV miR-K1 target the 3’ UTR of IκB-α resulting the decrease of
IκB-α protein expression. Overexpression of miR-K1 inhibit KSHV lytic replication by activating
NF-κB signaling (Lei et al., 2010). KSHV miR-K12-11 miRNA is a cellular miR-155 ortholog,
thereby regulating cell growth. Considering the miR-115 involved in B cell proliferation, miR-
K12-11 might induce the transformation of KSHV-positive B-cell tumors (Gottwein et al., 2007).
KSHV miRNA K12-1 is a viral oncogene that can activate NF-κB and STAT3(Chen et al., 2016).
KSHV miR-K10 target the 3’UTR of TGF-β type II receptor (TβRII) to inhibit the TGF-β
signaling, which might promote latency maintenance and tumor transformation(Lei et al., 2012).
miR-K3+1 and miR-K3 both share high homology with cellular miR-23 and can regulate miR-23
targets, including caspase 3, caspase 7, TRAIL receptor, p130, RBL2, JARID2, and HMMR
(Manzano et al., 2013).
Other miRNAs can directly target KSHV lytic genes. For instance, miR-K9 targeting the
3’UTR of RTA, inhibiting the RTA transcription and enhancing the spontaneous lytic replication
(Bellare and Ganem, 2009). miR-K3 target the nuclear factor I/B thus suppressing RTA
expression (Lu et al., 2010a). miR-K5 and miR-K6-3P can attenuates the ORF56 protein
expression by targeting the ORF56 3’UTR region(Lin and Ganem, 2011).
KSHV miRNA-deletion virus has higher lytic gene expression compared to wild type
virus. This is not only due to the direct targets in lytic gene promoter regions, but also the
chromatin remodeling regulators. miRK12-5 increase the histone H3 acetylation and decrease
of K9 methylation, while K12-4-5p can target Rbl2 to increase DNMT1 expression(Lu et al.,
2010b).
On the contrary, miR-K12-3 and miR-K12-11 promote lytic replication. The depletion of
miR-K12-3 and miR-K12-11 up-regulate lytic gene expression and reactivate KSHV replication
in latent infected cells. This was conducted by inhibiting host transcription factors, such as MYB,
C/EBPα and ETS-1, to activate RTA indirectly (Plaisance-Bonstaff et al., 2014).

14

1.6 Oxidative Stress and Oncogenesis in KSHV
Oxidative Stress in KSHV infection
Oxidative stress is the imbalance between pro-oxidants and antioxidants, which will
activate the redox signaling. Different from the oxidative stress, hypoxia is the physiological
condition caused by lack of oxygen and can lead to oxidative stress in cells. Hypoxia activate
the HIF-1α and NF-κB signaling, which can stimulate the inflammatory cytokines and metabolic
reprogramming. Oxygen sensors, including oxygen-dependent prolyl hydroxylases 1–3 (PHD1–
3) and asparagine hydroxylase factor inhibiting HIF (FIH), can transfer the oxidative stress
signals to HIF-1α. HIF-1α translocate to nucleus to form the heterodimer with HIF-1β and
binding to cofactor p300/CBP. HIF-1α can activate the inflammatory response with induction of
IFNγ and IL-17(McGarry et al., 2018). Oxidative stress induces KSHV reactivation in PEL by
treating the PEL cells with Diethyl maleate (DEM) that will deplete intracellular GSH and create
oxidative stress.
Excessive ROS in cell can lead to the oxidative damage in nucleotide, lipids, and
proteins. ROS can activate the ROS scavengers in cytoplasm and mitochondria, including
SODs, GSH, GPx, and PRx. Additionally, ROS can stimulate PI3K, MAPK, HIF-1α and NF-κB
pathway to support cell survival and proliferation(Schieber and Chandel, 2014). Activated NF-κB
inhibits KSHV lytic replication, thereby supporting the latency establishment (Brown et al.,
2003). However, inhibition of NF-kB can induce ROS and reactivate KSHV in PEL cells, results
in cell death (Li et al., 2011). Back to 2009, the first reverse-transfected cell microarrays
identified that ORF75 and K13 are NF-kB activator, while K10.5 and K1 can inhibit ORF75 and
K13-mediated NF-kB activation(Konrad et al., 2009), suggesting the complicated regulatory
mechanism of NF-kB in KSHV latent and lytic switch. A previous research has shown that
deamidation of NF-kB will shunt the NF-kB to promote aerobic glycolysis(Zhao et al., 2020).
While KSHV has a pseudo-enzyme GATs, kGAT, that might involve in this post-translational
regulation.
15

Hypoxia conditions, such as 1% O
2
increases KSHV reactivation (Davis et al., 2001).
Hypoxia can activate the response element of TF X-box binding protein 1 (XBP-1) in RTA’s
promoter region and trigger KSHV reactivation in PEL(Dalton-Griffin et al., 2009). The promoter
region of ORF34-37, genes involved in lytic gene transcription core, has multiple hypoxia-
responsive elements (HREs) and can be controlled by HIF-1α (Haque et al., 2006) Oxidative
and nitrative stress is correlated with AIDS-related KS as shown in immunohistochemical
staining of MnSOD and NOS2 (Mallery et al., 2004).

Oncogenesis of KS
From the biopsy sample in a nodular KS lesion, various cell lines are presented in the
KS spindle cells, including KSHV
+
spindle cells, CD34
+
vascular cells, D2-40
+
lymphatic
endothelial cells, macrophages, lymphocytes, plasma cells and red blood cells. The
inflammatory infiltration in KS lesion features the oncogenic role of paracrine angiogenic and
inflammatory signals in the microenvironment of KS lesions. (Mesri et al., 2010) Inflammatory
cytokine can lead to the chronic inflammation and lead to oncogenesis. KSHV has several viral
genes that are homologs of human chemokines or cytokines, including vIL-6, viral macrophage
inflammatory protein (vIMP), vCCL1, vCCL2, and vCCL3. In B cells, vIL-6 are consistently
expressed under latent replication. Exogenous expression of vIMP-1A, vIMP-1B, and vIL-6 can
upregulate the VEGF expression and promote cell proliferation (Liu et al., 2001). KSHV also
express vBCL-2, a suppressor of cell apoptosis, to inhibit cell death and promote proliferation.
Latent protein vGPCR (ORF74), a chemokine receptor, also served as a critical role in
KSHV-mediated oncogenesis. Both transgenic mice and exogenous expression of vGPCR can
induce cancer in mice, in which 30% of vGPCR transgenic mice can develop tumors (Guo et al.,
2003). The expression of Exogenous expression of KSHV vGPCR in vivo can induce tumors
that resembles KS (Montaner et al., 2003). The molecular mechanism of vGPCR-mediated
oncogenesis involves both transcription regulation and activation of signaling pathways, such as
16

Pi3K/AKT/mTOR pathway (Bhatt and Damania, 2012) (Cannon and Cesarman, 2004). vGPCR
activates NF-κB and NFAT, which are synergistically increased by HIV-1 Tat and vGPCR-
induced tumorigenesis (Guo et al., 2004). Activation of NF-kB regulated pathway is a hallmark
of KS or KSHV vGPCR-induced neoplasia(Martin et al., 2008). K1 transgenic mice can develop
malignancy with aberrant signaling including increase immune response and activation of NF-
kB, Oct2, and Src (Prakash et al., 2002). In addition, a clinical study found that KS+, KSHV+,
HIV+ patients has lower neutralizing antibodies and CD4 T cell counts comparing to the KS-,
KSHV+, HIV+ patients (Kimball et al., 2004). KSHV specific CD8 T cells is rare in patients
progressed to KS (Lambert et al., 2006).
1.7 Crosstalk between KSHV and other pathogens
Both KSHV and EBV remain silent in people lifelong, but in immune-deficient patients,
the reactivation of KSHV and EBV can lead to severe diseases. Coinfection of EBV and KSHV
is common in AIDS-associated primary effusion lymphoma (PEL). However, EBV and KSHV in
co-infected cell response differently towards the stimulation (Miller et al., 1997). Whereas, the
KSHV lytic regulator RTA interacts with the EBV-Z protein to achieve the joint inhibition of lytic
replication(Jiang et al., 2008).
HIV-1 infection can trigger KSHV reactivation(Merat et al., 2002).Early research on the
co-infection of KSHV and HIV-1 in KS patients has found the KSHV DNA loads in PBMCs is
correlated with lower CD4 cell counts and high HIV-1 RNA (Min and Katzenstein, 1999). Later a
longitude in HAART-treated HIV-1 patients identified that higher CD4 cell counts is related to the
lower KS incidence rates (Mocroft et al., 2004). KSHV latency regulator LANA interacts with
HIV-1 TAT and other enhancer elements to activate HIV-1 LTR (Hyun et al., 2001).
KSHV ORF45 is necessary but not sufficient to activate the integration of the HIV-1 LTR
(Karijolich et al., 2014). KSHV protein KIE2 activate the HIV-1 LTR, while counterintuitively, HIV-
1 Tat increased the KSHV capsid protein expression (Huang et al., 2001). Besides Tat, HIV-1
can infect T cells to produce the cytokines that trigger KSHV reactivation(Mercader et al., 2000)
17

(Jr et al., 1997). Few epitopes have been reported to trigger CD8+ T cells (Lepone et al., 2010).
Among KSHV positive patients and EBV positive patients, patients developed cancer have
much lower T cells counts than asymptomatic carriers.(Guihot et al., 2006).
Both KSHV and EBV, entails the endosomal sorting complex required for transport
(ESCRT) machinery for its maturation and release. The knockdown of ESCRT adaptor protein
Alix will accumulate the viral DNA and capsid proteins of EBV and KSHV (Lee et al., 2012). The
proline-rich region of RTA is known as the ESCRT binding region, acting in virus transportation
in cytoplasm. In 2008, both the structure of KSHV and murine gammaherpesvirus-68 (MHV-68)
were identified. However, the comparison of two structures shows a different pattern in
interaction between capsid and inner tegument proteins compares to alpha and beta
herpesviruses(Dai et al., 2018) (Dai et al., 2008).

18

2. Central Carbon Metabolism Reprogramming during virus-infection
2.1 Central Carbon Metabolism
Central carbon metabolism utilizes glucose or glycogen as resources to generate energy
or synthesize the macromolecules that serves as substrates of other metabolism, such as
amino acid metabolism or nucleotide metabolism. Central carbon metabolism is composed by
glycolysis/gluconeogenesis, citric acid (TCA) cycle, pentose phosphate pathway, and glycogen
metabolism. Central carbon metabolism is critical for cell viability because it generates ATPs as
energy resource, NAD(P)H as oxidation-reduction agents, and macromolecules that are used in
synthesis of nucleotide and amino acids.
The disrupted central carbon metabolisms are related to physiological conditions, such
as cancers, cardiovascular diseases, and neurodegenerative diseases. Cancer cells are more
active than normal cells with faster glucose consumption and fermentation to lactate. The
increased carbon metabolism was found by Warburg in 1920s and is called Warburg effect
(Warburg, 1925). Neurodegenerative diseases, such as Parkinson’s disease and Huntingdon’s
disease, are related with alteration in central carbon metabolism, redox homeostasis and energy
failure (Burns and Manda, 2017).
The irregular central carbon metabolism can be used for disease diagnosis and
treatment. The positron emission tomography (PET) scanning using a special dye containing
radioactive tracers to measure blood flow, oxygen usage, and glucose metabolism. Due to the
hyperactive glucose metabolism in cancer cells, PET scanning can detect cancer cells in
different organs (Ter-Pogossian, 1983). Some well-known cancer drugs, such as metformin, will
reduce the central carbon metabolism intermediates (Liu et al., 2016).
Central carbon metabolism is conserved in most organisms. Glycolysis and TCA
consume glucose and produce energy, ATP, while PPP synthesize the nucleotide metabolism
intermediates, ribose-5-phosphate, and antioxidants, NADPH. Glycolysis can be divided into
two process, ATP consumption process and ATP production process. After glucose being
19

transported from extracellular by glucose transporters
(GLUTs), hexokinase (HK) will catalyze the
phosphorylation of glucose to glucose-6-phosphate
(G6P). Then the rate-limiting enzyme of glycolysis,
phosphofructokinase (PFK) will convert the fructose-6-
phosphate to fructose-1,6-bisphosphate. Fructose-1,6-
biphosphate will be catalyzed into two three-carbon
substrates: dihydroxyacetone phosphate (DHAP) and
glycerol-3-phosphate (G3P/GAP), by fructose-
bisphosphate aldolase (ALDOA). DHAP is not stable
and will be continuously converted to G3P by glycerol-
3-phosphate dehydrogenase (GPD1). Glucose will be
converted into two G3Ps and utilize 2 ATP in the ATP
consumption process. During the ATP utilization
process, glyceraldehyde 3-phosphate dehydrogenase (GAPDH) will convert G6P to 1,3-
biphosphoglycerate and produce two NADPH. Phosphoglycerate kinase (PGK) catalyzes 1,3-
biphophoglycerate to 3-phosphoglycerate (3PG) and produce 2 ATPs. 3PG will be converted
subsequentially by phosphoglycerate-mutase (PGM) and enolase (ENO) to
phosphoenolpyruvate (PEP). Pyruvate kinase will catalyze the PEP to pyruvate and produce
two ATPs. Therefore, in total, glycolysis converts glucose to pyruvate and synthesize two ATPs.
Pyruvate enter the TCA cycle and electron transport chain (ETC) to generate 34 ATPs.
Central carbon metabolism is critical not only for its ability to generate ATPs, but also
because the end products and by products of the central carbon metabolism can be used to
synthesis nucleotide and one-carbon amino acid. The end products of PPP, ribose-5-
phosphate, is the substrate of PRPS in de novo purine synthesis and will be converted into
PRPP, then AICAR and IMP.
Figure 1- 2 Central Carbon Metabolism
20

2.2 Pentose Phosphate Pathway
PPP is the branch pathway of glycolysis and formed by two branches, oxidative PPP
and non-oxidative PPP. Otto Warburg first found G6PD by observing the catalyzation from G6P
to 6PG needs NADPH and then revealed the whole pathway in next few years. PPP is well
known for two roles, one is producing the nucleotide intermediate, ribose-5-phosphate, another
one is serving as the major role in produce NADPH for antioxidative response. Deficiency of
PPP enzymes, such as G6PD, TALDO, TKT, and RPI, can lead to severe diseases. (Reviewed
in (Perl et al., 2011)).
Oxidative PPP
The oxidative PPP is formed by three irreversible reactions including three enzymes,
G6PD, 6PGD and 6PGL. The first step of ox-PPP use glucose-6-phosphate to form 6-
phosphogluconate (6-PG) by G6PD. G6PD is the rate-limiting enzyme of PPP pathway/and
6PGD is the third enzyme in the oxidative PPP, which converts 6-phosphogluconate (6-PG) to
Ru-5-P and produces NADPH.
Enzymes EC Direction Substrates Products
6-phosphogluconic
dehydrogenase
1.1.1.44 Irreversible 6-phospho gluconate+ NADP D-ribulose 5-phosphate + CO
2
+
NADPH + H
+

glucose-6-phosphate
dehydrogenase
1.1.1.49 Irreversible D-glucopyranose-6-phosphate +
NADP
6-phospho-D-glucono-1,5-lactone
+ NADPH + PROTON
6-phosphogluconolactonase 3.1.1.31 Irreversible 6-phospho-D-glucono-1,5-
lactone + H
2
O
PROTON + 6-phospho-D-
gluconate
Table 3 Oxidative PPP Reactions
Oxidative PPP has a fast response to ROS and is the major one in some conditions. UV
or hydrogen peroxide treatment on skin fibroblast and keratinocyte can activate G6PD in few
seconds (Kuehne et al., 2015). G6PD-deficient embryonic stem cell is similar with the wild type
cells but undergo programmed cell death in response to the oxidative stress (FILOSA et al.,
2003).
21

Non-oxidative PPP
The non-oxidative PPP branch is formed by five reversible reactions that enable the
exchange of intermediates between glycolysis and PPP. In addition, the non-oxidative PPP is
critical for its role in recycle of ribose-5P though partnering with gluconeogenesis to produce
glucose-6P for produce NADPH under oxidative stress. Enzymes catalyze the non-oxidative
PPP includes: Ru5P isomerase (RPI), Ru5P epimerase (RPE), Transketolase (TKT), and
Transaldolase (TALDO)
Enzymes EC Direction Substrates Products
ribose-5-phosphate
isomerase
5.3.1.6 Reversible ribose 5-phosphate ribulose 5-phosphate
erythrose-4-
phosphate isomerase
5.1.3.1 Reversible ribulose 5-phosphate xylulose 5-phosphate
transketolase 2.2.1.1 Reversible sedoheptulose 7-phosphate +
glyceraldehyde 3-phosphate
ribose 5-phosphate + xylulose 5-
phosphate
transketolase 2.2.1.1 Reversible erythrose 4-phosphate + xylulose 5-
phosphate
fructose 6-phosphate +
glyceraldehyde 3-phosphate
transaldolase 2.2.1.2 Reversible sedoheptulose 7-phosphate +
glyceraldehyde 3-phosphate
erythrose 4-phosphate + fructose 6-
phosphate
Table 4 Non-oxidative PPP reactions
TKT catalyzes sedoheptulose-7P and GAP to R5P and xylulose-5P, which produce the
nucleotide intermediates from the glycolysis product. TKT forms a homodimer and has higher
expression in kidney, small intestine, and liver (Heinrich et al., 1976). In specific tissues, such as
testis, where TKT barely expressed, the isoform TKTL1 or TKTL2 will expressed to
compensate. RPI structure has a α/β/(α/β)/β/α fold with similar structure of alcohol
dehydrogenase family. RPI forms a homodimer that consist two subunits with different
conformation representing the open/close cleft (Zhang et al., 2003). Besides producing
antioxidant and nucleotide synthesis intermediate, PPP’s activity is higher in S/G2 phase, while
the deficiency of G6PD can lead to slower cell proliferation (Vizán et al., 2009).
22

2.3 PPP and Glycolysis Balance
PPP and glycolysis both utilized glucose and are connected though the non-oxidative
PPP. The PPP product R5P can be converted to glycolysis product G3P, and then catalyzed by
gluconeogenesis, the reverse pathway of glycolysis, to form G6P, which could be used by PPP
produce more NADPH.
Aerobic glycolysis and PPP are higher in cancer tissues compared to normal cells.
Inhibition of glycolysis can inhibit tumor growth in multiple cancers, including head and neck
squamous cells(Li et al., 2017), liver cancers (Geschwind et al., 2002), and ovarian cells (Zhang
et al., 2006). This might due to the increased cell cycle arrest and apoptosis (Zhu et al., 2016) or
the inhibition of autophagy(Wang et al., 2018a). Glycolysis inhibitor treatment also has the
synergistic effect of other cancer therapy, including metformin (Liu et al., 2019), Adriamycin, and
Paclitaxel (Maschek et al., 2004).
The ratio between glycolysis and PPP is 1:5 in oral keratinocyte as shown in Chapter 2
Fig. 2-2F. But in neurons, the ratio is higher and ~14% glucose is metabolized though PPP
(Rodriguez-Rodriguez et al., 2013). The difference between neurons and normal cells is due to
the degradation of PFKFB3 in neurons, where PFKFB3 is consistently degradation by
ubiquitination (Bolaños et al., 2010). Inhibition of glycolysis can induce PPP, vice versa.
Inhibition of glycolysis by PFKFB3 inhibitor PFK15 shuts glycolysis to PPP (Bolaños and
Almeida, 2010).Inhibition of GAPDH can also activate PPP (Ralser et al., 2007). The equilibrium
between PPP and glycolysis shifts in brain tumor and brain injury, suggesting the protective role
of PPP in stress conditions. Traumatic brain injury (TBI) patients have up-regulated
glycolysis/PPP rate (Jalloh et al., 2015). Glioblastoma cells also higher glycolysis/PPP ratio
regulated by hypoxia, while inhibition of PPP reduces cell proliferation and inhibition of
glycolysis decreased migration (Kathagen-Buhmann et al., 2016) (Kathagen et al., 2013).
23

The balance between glycolysis and PPP provides a fast response towards oxidative
stress, and their complementary roles to each other making them a difficult target for therapy.
However, the consistent need for metabolism in cancers or neurodegeneration diseases make it
a distinguishable biomarker. PET-CT scan can identify multiple types of cancers, heart
diseases, or neurodegeneration, but the radioactive materials used in PET-CT scan are harmful
and not sensitive to all types of cancers.

2.4 G6PD
Pentose phosphate pathway was
discovered in 1930s by Warburg laboratory
in Germany. Later, they found this enzyme
catalyzes the oxidation of G6P to 6PG and
named it Zwischenferment, intermediate
enzymes in German, which is named as
glucose-6-phosphate dehydrogenase (G6PD) now. G6PD is the rate-limiting enzyme of
pentose phosphate pathway. G6PD contains 515 amino acid and is formed by a N-terminal
domain with dinucleotide binding site, a link domain with substrate binding site, and a C-terminal
domain. G6PD is conventionally considered being active by forming a tetramer with NADP+
binding to the dimer-interface. The gene encodes G6PD locates at X chromosome, so the
deficiency of G6PD is an X-linked, hereditary genetic diseases. G6PD-deficience can cause
neonatal jaundice and acute hematolytic anemias. Dosage compensation in mammals
inactivates one X-chromosome in female cells, G6PD locus on the inactive X chromosome is
derepressed during dosage compensation.
G6PD has two domains, in which N-terminal domain contains the NADP binding site,
substrate binding site, and catalytic site. The C-terminal domain of G6PD contains the second
NADP binding site, which serves as structure function that stabilize the structure of G6PD. PPP
Figure 1- 3 G6PD in Oxidative Response
24

pathway is not only the metabolism pathway produce energy and nucleotide metabolism
intermediates but also producing the antioxidant. Previous research has proven that cancer
stem cells are proven to have higher PPP than glycolysis. CSC’s increased PPP pathway might
due to the induction of ROS.
2.5 G6PD variant deficiency and related diseases
G6PD is first identified in 1956 as its deficiency leads to digestive diseases. In 1966,
WHO standardized the procedures for measuring G6PD enzymatic activities. G6PD gene was
first cloned and sequenced in 1986, and since then, about 140 natural variants of G6PD have
been identified. G6PD enzymatic activities have been classified into five classes by WHO since
1989 based on the severity of G6PD deficiency. Class I is severely deficient G6PD-deficiency
and can cause chronic non-spherocytic hemolytic anemia. Class II is less severely deficient with
1–10% residual G6PD activity and associated with acute hemolytic anemia. Class III is
moderately deficient with 10–60% activity, and Class IV represents normal activity with 60–
150% G6PD activity. Different from Class I – IV, Class V indicates increased G6PD activity
(>150%) (1989). A global comprehensive research of G6PD before has shown that G6PD has
140 reported mutations, most of the Class I deficiency mutations occur between Exon 8 to Exon
10, suggesting the critical role of C-terminal in G6PD structure.
Red blood cells (RBCs) are a group of uniform and small cells that can be extracted
easily from peripheral blood mononuclear cells (PBMCs). RBCs are generated from myeloid
Figure 1- 4 History of G6PD
25

stem cells and lost nuclei and cellular organs such as mitochondria, Golgi apparatus, and
endoplasmic reticulum during red cell maturation. Each red blood cell can carry 270 million
hemoglobin molecules, which is capable of transferring oxygens in the human blood. Enzyme
deficiency can lead to critical diseases in human, G6PD-deficiency in red blood cells leads to
hemolytic anemia with 300 million people affected according to a research in 1973.
According to a global prevalence research of G6PD, the highest prevalence of G6PD
deficiency is in certain area of African, up to 20%. This is also hypothesized related to the
malaria prevalence in those African area. In addition, a report in 1991 has found that G6PD-
deficiency patient has higher prevalence in diabetes patients comparing to healthy population.
G6PD- deficiency can lead to neonatal jaundice, while the G6PD can protect the embryo from
oxidative stress related DNA damage.
2.6 G6PD and other PPP enzymes in cancers
G6PD expression is correlated with prostatic carcinoma comparing to benign
hyperplasia, suggesting a potential clinical indicator role(Zampella et al., 1982). Additionally,
G6PD-deleted cancer cells has a slower growth rate and a smaller tumor size than wild type
cells, which is correlated with cell cycle gene expression. The expression of G6PD varies in
tumor tissue and control cancer growth. (Hu et al., 2013) G6PD is required or clear-cell renal
cell carcinoma (Lucarelli et al., 2015). The post-translational modification of G6PD regulates
cancer cell growth. In hepatocellular carcinoma, O-GlcNAcylation of G6PD promote cell
proliferation on serine 84(Rao et al., 2015). As mentioned before, the expression of TKTL1 was
low in normal tissues. But multiple malignant tissue have detected high TKTL1 expression, such
as breast cancer (Földi et al., 2007), metastatic papillary thyroid carcinomas (Zerilli et al., 2008),
gastric cancer(Staiger et al., 2006) , nasopharyngeal carcinoma(Zhang et al., 2008) ,and
astrocytic gliomas(Völker et al., 2008).
Cancer cells develop resistance towards chemical treatment. In cancer cells that
developed chemoresistance to neoadjuvant chemotherapy has overexpressed Rac1, a GTP
26

binding protein, which upregulates non-oxidative pentose phosphate pathway(Li et al., 2020).
Inhibiting TKT using oxythiamine (OT), a thiamine antagonist, can sensitize hepatocellular
carcinoma (HCC) cells to sorafenib treatment (Xu et al., 2016). In addition, PPP also interacts
with TCA cycle. 6PGD directly interacts with malic enzyme (ME1). ME1 activates 6PGD and
enhance NADPH production (Lucarelli et al., 2015). G6PD was also reported to regulate
circadian clock as shown by the correlated transcriptional peaks and motif enrichment(Rey et
al., 2016). Dehydroepiandrosterone (DHEA) is a lipid-form hormone that produced in the
adrenal gland. The physiological function is DHEA is helping generating testosterone and
estrogen. DHEA is also a non-competitive inhibitor of G6PD. Another well-known inhibitor of
G6PD is 6-AN, which can be converted into 6-ANADP to competitively inhibit NADP
+
binding of
G6PD. The treatment of DHEA in HeLa cells decreased cancer cell migration and
proliferation(Fang et al., 2016).
KSHV latent infected cell, lymphoma, can be reactivated using a p53 activator, Nutlin-3a,
an inhibitor blocks the interaction between MDM2-p53 interaction. P53 has been reported to
activate G6PD though phosphorylation of G6PD. DNA damage markers, such as
phosphorylated CHK1/2 or phosphorylated p53 can be induced by oxidative stress. Inhibition of
cdk1 can induce apoptosis and reactivate KSHV in PEL. Beside p53, the major protein of
double strand breaks (DSBs) repair, ATM, promotes NADPH production and regulates G6PD
though Hsp27.

2.7 G6PD in other diseases
Oxidative balance is critical for neurons. For instance, neuron cells suffered from
nitrosative stress that required PPP to maintain cell redox balance. G6PD is required for the
DSB repair and maintain dNTPs pool in neuron cells(Cosentino et al., 2011). PPP inhibitor
sensitized neuron cells to cell death, therefore, inhibiting glycolysis shift neurons to utilize PPP
27

can partially rescue (Bolaños and Almeida, 2010). RPI deficiency can cause slowly progressive
leukoencephalopathy, a disorder of brain white matter (Huck et al., 2004).
Oxidative stress has been reported favors herpesvirus infection in vertebrates according
to a meta-analysis. Antioxidant treatment can reduce the virus load. Protein oxidation is usually
indicated by protein carboxylation (PC). An HPV research have suggested that HPV infection
induce the cellular ROS, however, when the HPV-infected cell transformed to cancer cell, the
oxidative stress was reduced. A novel inhibitor, polydatin, inhibits G6PD to increase redox
response in cells which in turn leads to ER stress and cell death. Additionally, polydatin can
reduce cancer cell invasiveness in vitro and in vivo (Mele et al., 2018). A compendium study
using the genome-wide CRISPR screening reveals that G6PD is synthetic lethal when
mitochondria is damaged, which might due to the organelle stress. (Jiang et al., 2020)
From a cancer biology view, the chronic inflammation or oxidative stress, causing by
consistent virus infection or carcinogen exposure, can increase the tumorigenicity.
Overexpression of Nox1 can increase the free radical generation and cell growth in the non-
tumorigenic NIH 3T3 cells and this process can be block by co-expressed catalase. However,
overexpression of G6PD in cells can increase the anchorage-independent growth, suggesting
G6PD might have other critical roles in cell growth, such as producing intermediates for de novo
nucleotide synthesis. An intrigue finding is the treatment of DHEA can decrease the abnormal
inflammation and restore the hormone production in vivo, which might due to the hormone
regulation by DHEA or abnormal function of G6PD in diseases.

28

3. Protein deamidation and deamidase
Protein deamidation is the modification that remove one amino group from asparagine
and glutamine and turn into aspartic acid and glutamic acid. Deamidation will lead to one Da
mass change and isoelectric shift.
Deamidation was identified at 19 century and considered as an automated process
related with protein aging. In life-long proteins, deamidation accumulates and denatured,
providing a possible reason of aging-related protein malfunction (Hains and Truscott, 2010).
Deamidation is emerging as a crucial regulatory mechanism, but the molecular basis and
function of protein deamidation are poorly understood (Massière and Badet-Denisot, 1998).
Feng Shao’s group published on Science in 2010 showing a bacterial effector CHBP can
potently inhibit ubiquitin and NEDD8 ubiquitination pathway though deamidation (Cui et al.,
2010). There are two types of deamidation: automated deamidation and enzymatic-specific
deamidation. Enzymatic deamidation is catalyzed by proteins with glutamine amido-transferase
(GAT) domain. Cellular GATs catalyze the synthesis of nucleotides, amino acids, glycoproteins
and other metabolites (Zhao et al., 2016a). Our group’s previous search among the human
protein using the similarity search of glutamine aminotransferase domain (GAT domain) have
identified 25 proteins that can potentially has the deamidase abilities as shown in Table 5.
Work from our lab demonstrates key regulatory roles of protein deamidation in innate
immune defense and implicate cellular GAT in protein deamidation. Our lab’s previous research
on murine gamma HV68 found that a pseudo-enzyme, vGAT, recruit its cellular counterpart,
PFAS, to deamidate RIG-I. RIG-I is an RNA sensor that can recognize the virus’s 5’-ppp RNA
and form oligomers to activate innate immunity. Deamidated RIG-I loss it’s activity to activate
the IFN-betta pathway that helps viral immune evasion process(He et al., 2015). Similar
regulation role of viral pseudo-enzyme was the HSV-1 tegument protein, UL37. We found UL37
deamidates the helicase domain of RIG-I, results in inability to sense viral dsRNA and
suppression of innate immune response (Zhao et al., 2016b). This research identified that UL37
29

is a bona fide deamidase. Our lab also revealed that UL37 can deamidate cGAS for virus
immune evasion on an evolutionally specific site. cGAS is a DNA sensor that catalyze cGAMP
synthesis to activate downstream STING innate immune signaling (Zhang et al., 2018). Cellular
GATs also participate in transcription regulation. We showed that the enzyme carbamoyl-
phosphate synthetase, aspartate trans-carbamoylase, and dihydroorotase (CAD) deamidates
RelA, a subunit of the pro-inflammatory mediator NF-κB, to shunt the cell’s gene expression
from those encoding inflammatory mediators to glycolytic enzymes (Zhao et al., 2020). In my
research as presented in Chapter 2, we did a global deamidation proteomics, which
characterized deamidation-regulated pathways, including central carbon metabolism, ER protein
processing, ribosome function, and cell death. The global screening identified that deamidation
not only regulate innate immune response, but also other physiological pathways. Here, I listed
all the possible cellular GATs in table.5.

30

Symbol Names Size
(kDa)
substrate product GATase
domain
ASNS asparagine synthetase (glutamine-
hydrolyzing)
64.3 Asp + Gln + ATP +
H2O
Asn + Glu + AMP
+ PPi
002-191
ASNSD1 asparagine synthetase domain
containing 1
72.1

002-184
CAD carbamoyl-phosphate synthetase
2, aspartate transcarbamylase,
and dihydroorotase
243 HCO3 + Gln +
2ATP + H2O
CP + Glu + 2ADP
+ Pi
180-354
CPS1 carbamoyl-phosphate synthase 1 165.6 HCO3 + NH3 +
2ATP
CP + 2ADP + Pi 222-395
CTPS1 CTP synthase 1 66.7 UTP + Gln + ATP CTP + Glu + ADP
+ Pi
300-554
CTPS2 CTP synthase 2 65.7 UTP + Gln + ATP CTP + Glu + ADP
+ Pi
300-554
DDOST dolichyl-
diphosphooligosaccharide-protein
glycosyltransferase non-catalytic
subunit
50.7 sugar-lipid +
protein(-Asn-X-
Ser/Thr)
lipid + glycoprotein
(-Asn(sugar)-X-
Ser/Thr-)
none
F13A1 coagulation factor XIII A chain 83.3 protein(-Glu-) +
protein(-Lys-)
protein(-Glu-Lys-)
+ NH3
192-502
F13B coagulation factor XIII B chain 75.5

none
GFPT1 glutamine-fructose-6-phosphate
transaminase 1
78.8 Gln + Fru-6-P Glu + GlcN-6-P 002-305
GFPT2 glutamine-fructose-6-phosphate
transaminase 2
76.9 Gln + Fru-6-P Glu + GlcN-6-P 002-288
GLS glutaminase 73.5 Gln + H2O Glu + NH3 244-523
GLS2 glutaminase 2 66.3 Gln + H2O Glu + NH3 177-463
GLUL glutamate-ammonia ligase 42.1 ATP + Glu + NH3 /
Glu
ADP + Pi + Gln /
GABA + CO2
114-280
31

Table 5 Glutamine Amidotransferase

GMPS guanine monophosphate synthase 76.7 ATP + XMP + Gln +
H2O
AMP + PPi + GMP
+ Glu
030-209
NADSYN1 NAD synthetase 1 79.3 ATP + NaAD + Gln
+ H2O
AMP + PPi + NAD
+ Glu
003-289,
339-603
NARS asparaginyl-tRNA synthetase 62.9 ATP + Asn +
tRNA(Asn)
AMP + PPi + Asn-
tRNA(Asn)
none
NARS2 asparaginyl-tRNA synthetase 2,
mitochondrial
54.1 ATP + Asn +
tRNA(Asn)
AMP + PPi + Asn-
tRNA(Asn)
none
NTAN1 N-terminal asparagine amidase 34.7

037-304
PARK7 Parkinsonism associated
deglycase
19.9 protein(-
Arg/Lys/Cys(OXO)-
) + H2O
protein(-
Arg/Lys/Cys-) +
lactate
003-186
PFAS phosphoribosylformylglycinamidin
e synthase
144.7 ATP + FGAR + Gln
+ H2O
ADP + Pi +FGAM
+ Glu
1064-
1332
PPAT phosphoribosyl pyrophosphate
amidotransferase
57.4 PRPP + Glu 5PRA + PPi + Gln 012-261
TGM1 transglutaminase 1 89.8 protein(-Gln-) +
protein(-Lys-)
protein(-Gln(-
Lys-)-) + NH3
369-462
TGM2 transglutaminase 2 77.3 protein(-Gln-) +
protein(-Lys-)
protein(-Gln(-
Lys-)-) + NH3
272-359
WDYHV1 WDYHV motif containing 1 23.7 Gln-protein + H2O Glu-protein + NH3
32

4. Multi-omics Analysis
Multi-omics analysis aims to integrate several dimensions molecular datasets for a
system analysis. Common multi-omics datasets include genomics, transcriptomics,
epigenomics, proteomics, and metabolomics. Although multi-omics is a straightforward concept
and has been proposed for years, there is no mature pipelines. Currently, there are a few
common approaches for multi-omics analysis. Combination of CHIP-sequencing and RNA-
sequencing can improve the differential gene expression profile as both the activated
transcription factor and regulated gene can be identified. when doing multi-omics analysis,
different omics datasets are processed independently for pathway enrichment analysis and then
integrate. Here, I will introduce omics analysis methods including transcriptomics analysis, post-
translational modification proteomics analysis, network analysis, and metabolomics analysis,
which are all applied in my dissertation research.

4.1 Transcriptomics Analysis: RNA-sequencing and single cell-sequencing
RNA sequencing
The first-generation RNA sequencing was developed by Sanger and Coulson 50 years
ago that utilizes chain termination reaction (Ozsolak and Milos, 2011). The first-generation RNA-
sequencing was modified and applied for the first assembly of human genome in 2001(Lander
et al., 2001) (Venter et al., 2001). Then microarray analysis was developed by hybridizing
DNA/RNA sequence in a microchip before sequencing. However, microarray sequencing is
expensive and has disadvantages including cross-hybridization and poor readings for low
expression genes.
The next-generation sequencing (NGS) uses the fragmented complimentary DNA
(cDNA) to coat the chips then sequence by synthesizing cDNA using fluorescence labelled
dNTPs for detection (Metzker, 2010). Therefore, the NGS raw file contains the fragmented
sequence and needs to be mapped into the reference genome and assembled to complete
33

gene transcripts. Genes transcripts will then be normalized by reads per million (RPM) or
transcripts per million (TPM) and compared between control and experimental groups. The
analysis workflow of NGS has been developed for decades and has mature pipelines in different
platforms. The pipeline used in this research is the new tuxedo pipeline developed at John
Hopkins (Pertea et al., 2016) inclused HISAT2, StringTie and Ballgown open-source packages.
Single cell sequencing
Single cell sequencing (sc-seq) is developed based on NGS and microfluid chip. Sc-seq
uses beads-captured single cell for cDNA amplification and quantification. Comparing to NGS’s
bulk sequencing, the sc-seq offers single cell level expression that provide heterogeneity
information for the sample. Currently, sc-seq is widely used in peripheral blood mononuclear cell
(PBMC) and cancer tissue research to investigate the cell heterogeneity (Baslan and Hicks,
2017) (Papalexi and Satija, 2018). However, the depth of sc-seq is lower than NGS, which
means the low expressed genes are hard to detected. Sc-seq conducted though 10x genomics
will output an expression matrix that can be subjected to Seurat package in R for differential
gene expression and clustering (Butler et al., 2018). The clustering analysis is an unsupervised
analysis and can divide the single cell population into different groups for biomarker
identification. In cancer research, sc-seq can distinguish cancer cell, cancer stem cell,
peripheral cells, and surrounding immune cells for differential expression profile.

4.2 Post-translational modification (PTM) Analysis
Post-translational modifications (PTMs) regulate protein activities by addition of
modifying groups or alteration of amino acids. PTM can activate a protein, subject the protein to
the proteasome degradation, or change the subcellular localization. Mass spectrum measure
the mass to charge ratio which enables the detection of one Da changes and precise positions
of the modification. Therefore. MS has become the major method to detect the PTM. The global
screening using MS has enable the quantitation of proteomics by heavy isotope labeling and
34

PTM-proteomics combining with the specific antibody purification. Phosphopeptide enrichment
strategies have been developed before the MS analysis and are adopted to MS-based
phosphoproteomics. Those enrichment methods include antibody-based enrichment, hydrophilic
interaction chromatography, and metal affinity chromatography(Hoffman et al., 2015). Ubiquitin
affinity purifications and MS analysis have revealed the function of conventional and
unconventional ubiquitin-mediated proteasomal degradation(Xu et al., 2009)(Kim et al., 2011).
Here is the feature table of some common post-translational modification:
PTM Type Site ∆Mass (Da) Stability
Phosphorylation Tyr +80 +++
Ser, Thr +80 ++
Acetylation Lys +42 +++
Glycosylation N-linked +>800 +/++
O-linked +203, >800 +/++
Ubiquitination Lys, Met +>1,000 +/++
Deamidation Asn, Gln +1 +++
Table 6 Common PTMs
Stability: + labile in tandem MS; ++ moderately stable; +++ stable
As a powerful tool to identify and quantify the peptides, LC/MS can be used to quantify
the protein abundance in complicated samples. Currently, there are unlabeled and labeled
proteomics, in which labeled proteomics including SILAC, iTRAQ. Stable isotope labeling of
amino acids in cell culture (SILAC)-based MS/MS label two groups with light/heavy isotope
amino acid separately and analyzed together (Wang et al., 2018b). Isobaric Tags for Relative
and Absolute Quantification (iTRAQ) labels samples with different isotope tags and can
separated up to 16 samples (Luo and Zhao, 2012). Labelled proteomics provides a more
accurate quantification than unlabeled proteomics but is more time-consuming and expensive.
35

As the limit of detection (LOD) of peptides varies, current normalization methods for
RNA-sequencing and metabolomics are not suitable for post-translational modification. For
instance, rare or unstable PTMs might have one or two peptides in one group and no detected
peptide in another group. But the missing value in another group will be normalized as one
peptide, causing comparison bias. In addition, due to the limitation of the statistic modeling
fitting of different peptides, both spectral count and ion abundance could be used to quantify the
protein abundance with limitations.

4.3 Metabolomics and Metabolic Flux
Metabolomics use mass spectroscopy (MS) or nuclear magnetic resonance (NMR)
spectroscopy to perform high-throughput metabolites screening in biological samples, including
plasma, blood, intracellular/extracellular contents. Metabolomics characterizes and quantifies
small molecules (<1,500 Da) to study the correlation between metabolism and certain
pathological conditions(Cambiaghi et al., 2017). The whole process of a metabolomics analysis
including four steps, 1) sample preparation; 2) data acquisition; 3) data processing; 4) data
interpretation (Figure 1-5). Besides the general analysis, kinetic models of metabolites and
network-integration analysis are used to measure metabolism flux changes.

Figure 1- 5 Metabolomics Analysis Workflow

4.4 Multi-omics integration methods
Integrating gene expression molecular layers, such as transcriptomics and epigenomics,
has been applied in gene regulatory research. For instance, the chromosome conformation
36

capture analysis uses the epigenomics analysis to identify the histone modification patterns,
such as methylation or acetylation, to identify the activated genome region. Then analysis
results will be integrated with gene expression profile for identification(Merelli et al., 2015).
Molecular level datasets can also be combined with the patient information, such as
diets, symptom severity, and disease state. In a longitudinal irritable bowel syndrome study,
researchers used patients’ information, fecal metabolome, fecal metagenome and mucosal
transcriptome data to identify the pattern of IBS diseases and possible therapeutic target, purine
starvation (Mars et al., 2020).
Bayesian probabilistic causal network (BN) was used to integrate multi-omics results to
calculate probability significance. Therefore, using BN for multi-omics analysis can identify key
regulatory factors. For late-onset Alzheimer’s disease (LOAD), BNs help researchers identify
ATP6V1A as a top regulator of neuronal subnetwork and it’s deficiency can be rescued by NCH-
51(Wang et al., 2021).
For metabolomics related researches, combining metabolite pathway enrichment
analysis and gene pathway enrichment can help identify targets and mechanism (Aon et al.,
2020).A NASA study about spaceflight health risk using multi-omics to find and validate that
mitochondrial dysfunction could be the potential risk of space travel(Silveira et al., 2020).
Recently, spatial genomics has also been developed to combine the spatial information, scRNA-
seq and spatial expression for a high-resolution gene expression profile(Liu et al., 2020).
Additionally, the datasets generate from both the host and pathogen can be analyzed together
using a holo-omic system biology pipeline (Nyholm et al., 2020). For a specific organism, Zea
mays L., multi-dimensional information is collected and pre-analyzed in a database called
ZEAMAP as a tool to help researchers, showing that agriculture and evolution have also started
using multi-omics tools(Gui et al., 2020).
37

Chapter 2 Protein Deamidation in KSHV-mediate Metabolic Reprogramming
1.Introduction
Kaposi's sarcoma-associated herpesvirus (KSHV), also known as HHV8, is a human
pathogen that can cause morbidity and mortality by diseases such as Kaposi's sarcoma (KS),
multicentric Castleman disease (MCD), and primary effusion lymphoma (PEL). After de novo
infection, KSHV has two replication cycles – lytic and latent – determined by host cell response.
Most infected host cells are in latent replication and are reactivated only under stress stimulus,
so KSHV-positive patients can establish life-long persistent infection without symptoms.
However, immune-compromised KSHV-positive patients develop severe malignancies such as
KS and PEL. In KS lesions, KSHV lytic replication is a central component. 1 – 5 % KS cells
supports lytic replication by expressing oncogenic lytic genes and producing KSHV for
malignant lesion development.
Viruses, such as HIV-1 and HPV, funnel cellular anabolic metabolism to favor their
infection (Valle-Casuso et al., 2019) (Fleming et al., 2019),but the regulatory mechanism
remains rudimentary. Reactivation of KSHV from latent replication cells activates anabolic
metabolism, including glutamine metabolism, glycolysis, lipid metabolism(Sychev et al., 2017)
(Zhu et al., 2017). However, previous research is limited to KSHV latent established cells, where
both the genomic and epigenomics changes have primed the host cell. To study the metabolic
reprogramming in de novo infection response, we conducted the metabolomics analysis on
KSHV de novo infected cells.
Post-translational modification (PTM) regulates protein activities by the addition of
modifying groups or alteration of amino acids, which response rapidly and has reversible
regulation. How PTM regulates metabolic enzymes have been studied recently. For example,
glycosylation of PFK has been found critical for cancer development (Yi et al. 2012).
Deamidation, catalyzed by enzymes with an aminotransferase (GAT) domain, remove an amino
group from glutamine or asparagine and crucial regulatory mechanism(Zhao et al., 2016a).
38

Deamidation of a bacterial effector CHBP can potently inhibit ubiquitin and NEDD8
ubiquitination pathway(Cui et al., 2010). Work from our lab demonstrates the key regulatory
roles of protein deamidation in innate immune defense and transcriptional regulation(He et al.,
2015) (Zhang et al., 2018). Cellular GATs catalyze the synthesis of nucleotides, amino acids,
glycoproteins, and other metabolites. Specifically, a de novo pyrimidine synthesis enzyme,
CAD, deamidate transcription factor RelA to regulate cell cycle and metabolism(Zhao et al.,
2020). Integrating metabolomics and proteomics analysis of cells supporting KSHV lytic
replication, we discovered a paralleled increase in glycolysis and deamidation of glycolic
enzymes, implying the regulatory roles of deamidation in metabolism. We identified KSHV
utilizes deamidation to promote the pentose phosphate pathway (PPP) though CAD and G6PD,
thereby enhancing KSHV lytic replication.
2. Result
KSHV Infection Upregulate Central Carbon Metabolism
Like all herpesviruses, KSHV has two phases of infection, lytic replication and latency.
Upon de novo infection, KSHV undergoes latent infection by default via circularizing its linear
genome in the nucleus. Under certain circumstances, such as high MOI infection, cells can
support lytic replication to produce infectious virions. To profile the overall metabolic changes,
we collect intracellular metabolites for mass spectrometry profiling from KSHV-infected cells at
24 hours post-infection (hpi) and 48 hpi. We used a recombinant KSHV, rKSHV219, to infect
HOK16B cells, a human oral keratinocyte cell line that supports lytic replication. rKSHV219
expresses RFP under the PAN promoter's control that marks lytic replication and GFP under
globin promoter marking latent infection. At 48 hpi, 90% of HOK16B cells, when infected at
multiplicity of infection (MOI) of 30, supported KSHV lytic replication as indicated by RFP
expression (Fig. 2-S1A, B). Metabolomics analysis identified ~560 metabolites, which shows a
global change in intracellular concentration, including 230 consistently metabolites, 180
increased and 50 decreased (Fig. 2-1A, B,). Principal component analysis (PCA) of metabolites
39

showed that mock-infected and KSHV-infected at 24 hpi clustered together, validating the
quality and reproducibility of metabolomics analysis (Fig. 2-1C). However, cells infected with
KSHV at 48 hpi were somewhat scattered likely due to the unstable lysed cells during late
stages of KSHV infection. Furthermore, a heatmap of metabolites showed that KSHV infection
remarkably elevated the intracellular concentration of a large panel of metabolites that
encompass diverse types of intermediates, including those of nucleotides, central carbon
metabolism, lipids, and glycoproteins (Fig. 2-S1E). Pathway enrichment analysis highlighted
significant increases in metabolites of sphingolipid metabolism, glycolysis, TCA cycle, pentose
phosphate pathway (PPP) and nucleotide metabolism, e.g., purine synthesis (Fig. 2-1D).
Ceramide can induce apoptosis in KSHV infected cells, while lipid synthesis is critical for virus
egress, so Because majority of identified metabolites belongs to lipid metabolism, so the
sphingolipid metabolism has been enriched, which correlates with previous finding (Ref.Mich
paper). A heatmap of the top 30 significantly changed metabolites with pathway annotation
showed carbohydrate and nucleotide metabolites were major components of this metabolic
reprogramming (Fig. 2-1E). When metabolites of the glycolysis pathway were analyzed in the
pathway network, we found that all detected metabolites in carbon metabolism are increased,
indicating the up-regulation of central carbon metabolism during KSHV replication (Fig. 2-S1C).
Key metabolites, including pathway intermediates, checkpoint metabolites, of carbon
metabolism are increased, such as Acetyl-CoA, Ribulose, and 6PG (Fig. 2-1F,G,H). Acetyl-CoA,
the metabolite connecting glycolysis with TCA cycle, increased more than 60-fold at 24 hpi,
suggesting the high glycolytic flux and the need of acetyl-CoA for energy production or lipid
synthesis. Ribulose, an intermediate connecting carbon metabolism and nucleotide synthesis,
increased more than ten fold, indicating the sufficient supply for virus DNA replication. An
indicator of the PPP, 6-phosphogluconate (6PG), increased more than 30-fold in KSHV-infected
cells. The PPP is the branch pathway of glycolysis and produces ribose-5-phosphate (R5P) for
de novo nucleotide synthesis and antioxidant. These findings support the conclusion that KSHV
40

activates anabolic pathways in central carbon metabolism with the KSHV lytic replication.

41

Figure 2- 1 Metabolomics analysis showed KSHV
(A) Summary of metabolic profiling of KSHV lytic replication from a total of 560 named
biochemicals. One-way ANOVA analysis was conducted to calculate p-value.
(B) Venn plot shows overlapping metabolites between 24hpi/Mock and 48hpi/Mock of both
consistently increased and decreased metabolite that has p-value < 0.05.
(C) Principal component analysis (PCA) reduces the dimension and complex of
metabolomics showing the segregation between different time point samples.
(D) Compounds identified from metabolomics were analyzed using ANOVA to determine
42

statistical significance to select metabolites with FDR lower than 0.05. By comparing
metabolites expression between 24 hpi/Mock, and 48hpi/Mock, we selected up-
regulated metabolites and subjected to KEGG pathway enrichment analysis.
(E) Heatmap showing top30 significantly changed metabolites between hpi 24 and mock
samples and labelled with their pathway annotation.
(F) Metabolite levels in KSHV lytic replicated cells.
43

Supplementary Figure 2- 1
(A) Schematic view of r219.KSHV virus
(B) Representative images of KSHV lytic replication on HOK16B cells. RFP is associated
E
44

with lytic gene promoter, RTA, and marking lytic replication, while GFP is an indicator of
infection.
(C) Metabolites of central carbon metabolism pathway are shown in a network layout.
Colored circles are detected metabolites, the darker color indicated higher fold-changes.
Gray circles indicate non-detected metabolites.
(D) Metabolites with more than 2 times fold changes, both increased and decreased,
comparing between 24 hpi/Mock, and 48hpi/Mock were analyzed by their overlapping as
shown in the Venn plot.
(E) Heatmap showing all metabolites detected in metabolomics with their pathway
annotation.

45

Flux analysis identifies Glycolysis and Pentose Phosphate Pathway Increase
During KSHV lytic Infection
To understand the role of central carbon metabolism in KSHV infection, we measured
the glucose utilization during KSHV infection. Both the glucose consumption and lactate
secretion increased upon KSHV lytic replication, indicating the increased glycolysis that
corresponds with metabolomic results (Fig. 2-2 A, B). To profile the transient metabolic flux
changes, especially for pathways with high turnover rates, we performed [U-
13
C] glucose
isotope tracing on KSHV infected HOK16B cells (Fig. 2-2C). Indicators of sub-pathways in
central carbon metabolism have significantly increased, including glycolysis indicators,
Glycerate-3-phosphate (G3P); PPP pathway indicator, 6-phosphogluconate (6PG); TCA cycle
indicator, malate; and nucleotide synthesis precursor, ribose-5-phosphate (R5P), although R5P
and ribulose-5-phosphate (Ru5P) are undistinguished from MS (Fig. 2-2D,E; Fig. 2-S2A, B).
However, both traced serine and glycine, belonging to one-carbon amino acid metabolism, have
minor changes (Fig. 2-S2C, D). Those results collectively show that central carbon metabolism
is up-regulated by KSHV infection, especially for PPP and glycolysis. To gain insights into the
balance of glycolysis and PPP, we performed [1,2-
13
C] glucose tracing to quantify the ratio of
glycolysis flux to PPP flux (Fig. 2-2F). Glycolysis produces pyruvate M+2, while the PPP
produces pyruvate M+1, and pyruvate will be converted into lactate by lactate dehydrogenase
(LDH). Therefore, the ratio of lactate M1 to M2 reflect the flux of PPP/glycolysis rate, which has
no significant change upon KSHV infection (Fig. 2-2G, J)

FLIM Free NAD(P)H Increases as Glycolytic Rate is Higher Than OXPHOS
Fluorescence lifetime imaging (FLIM) measures the lifetime of the autofluorescence
molecule and can be applied to intracellular metabolites, such as NAD(P)H and FAD(P)H, in live
cells without treatment. FLIM has been used in cancer research to detect the pro-glycolytic
states of cancer cells by measuring the bound/unbound NAD(P)H ratio (Pate et al., 2014).
46

Bound/unbound NAD(P)H ratio is an indicator of glycolysis/OXPHOS ratio, which represents
LDH – bound and – unbound NAD(P)H. Here, we applied the FLIM on KSHV infected HOK16B
cells to measure the glycolytic rate. KSHV-infected cells are more glycolytic than mock-infected
cells and reach the peak of the glycolysis at 12 hpi. Interestingly, the 12 hpi sample shows a
highly glycolytic core near the nucleus, indicated by the red arrow (Fig. 2-2H, I), suggesting a
metabolic core formed during KSHV infection. As the location of the metabolic core is close to
nucleus, it can provide energy and intermediates for viral protein synthesis and genome
assembly. The FLIM results contain spatial information and correspond with the metabolomic
analysis where the upper central carbon metabolism, including glycolysis and PPP, are more
active than TCA cycle. Also, the early activation post infection suggests the virus reprogram the
host cell metabolism based on the transient modification, such as post-translational modification
and protein interaction, instead of the transcriptional regulation.
47

Figure 2- 2 Central carbon metabolism is up regulated during KSHV lytic replication
(A-B) Supernatant of mock- and KSHV- infected HOK16B cells are harvested as indicated time
points and measured for glucose concentration and lactate concentration. The original glucose
concentration in HOK medium is 1 g/L.
(C) [U -
13
C]-glucose metabolite flux overview.
(D) Glycolysis indicator, G3P, in [U-
13
C]-glucose labelled KSHV lytic replicated cells
(E) Pentose phosphate pathway indicator, 6PG, in [U-
13
C]-glucose labelled KSHV lytic
replicated cells.
48

(F) [1, 2 -
13
C]-glucose metabolite flux overview.
(G) [1, 2 -
13
C]-glucose medium replaced the original medium of KSHV infected HOK16B cells
24 hpi or 48 hpi for 30 minutes before harvested for metabolite extraction. Lactate M1 is
produced though PPP pathway, while lactate M2 is produced from glycolysis. So, the PPP/
glycolysis flux is measured by lactate M1/M2 ratio.
(H) Cells were seeding in dishes with a glass bottom and imaging under the Leica multi-
spectrum to detect the fluorescence life-time images. Images showing the detected
autofluorescence NAD(P) signals. Color scale of FLIM image is shown in Fig. S2-2.
(I) The ratio of glycolysis and OXPHOS of each Z-stack figure is shown as one dot.

49

Supplementary Figure 2- 2
(A) TCA cycle intermediate, malate, levels in [U-
13
C]-glucose labelled KSHV lytic replicated
cells indicate TCA cycle metabolism increased
(B) Ribose-5-phosphate and ribulose-5-phosphate levels in [U-
13
C]-glucose labelled KSHV
50

lytic replicated cells indicate PPP and nucleotide metabolism increased
(C-E) Serine, glycine, and glutamine levels in [U-
13
C]-glucose labelled KSHV lytic replicated
cells indicate amino acid synthesis are not increased
(F-H) Malate, citrate/iso-citrate, and ribose-5-phosphate/ ribulose-5-phosphate levels in [1,2-
13
C]-glucose labelled KSHV lytic replicated cells.
(I) Schematic view of FLIM detected NAD(P)H as indicator of OXPHO and Glycolysis.
(J) FLIM phasor plot showing bound/free NAD(P)H ratio color scale corresponding to Fig.2-
2 H.

51

Multi-omics identify deamidation as a regulatory mechanism for central carbon
metabolism
To probe the molecular mechanism of KSHV-mediated metabolic reprogramming, we
performed RNA-sequencing analysis of KSHV and Mock infection profile in two lytic replication
cell lines, HOK16B and lymphatic endothelial cells (LEC). According to the gene profiling, we
didn’t observe an increased expression pattern in central carbon metabolism genes during the
KSHV infection, suggesting the metabolic reprogramming is not regulated though transcription
(Fig. 2-S3A). To verify this result, we used the previous reported (Purushothaman et al., 2015).
RNA profiling conducted on TIVE, CD4+ T cells. The heatmap of glycolysis and PPP pathway
gene expression showed no significant change (Fig. 2-S3B), while the pathway enrichment
analysis recapitulates the previous results on HOK16B and LEC cells (Fig.S3C). Besides the
transcriptional regulation, post-translational regulation also regulates the metabolism
reprogramming, such as the glycosylation of PFKL in regulating cancer growth. Therefore, we
hypothesize that PTM might participate the metabolic reprogramming. To test this idea, we
screened the global PTM between Mock-infected and KSHV-infected cell using tandem MS/MS
where we observed that deamidation is enriched in metabolism enzymes. MS identified 950
deamidated peptides from 383 proteins, in which 515 peptides with increased deamidation and
385 peptides with decreased deamidation (Fig. 2-S3D). To determine the role of protein
deamidation in KSHV lytic replication, we analyzed the proteome-wide post-translational
modification of Mock-infected and KSHV-infected HOK16B cells at 48hpi. We conducted KEGG
pathway enrichment analysis on 268 increased deamidation proteins, which includes carbon
metabolism in the top ten enriched pathways, suggesting carbon metabolism is a deamidation-
regulated process (Fig. 2-3B, C). Other modifications, such as acetylation and phosphorylation,
were observed in MS/MS but shows no regulation on metabolism enzymes. Other pathways,
such as the ribosome pathway and aging-related diseases (prion diseases, Parkinson’s
disease), are enriched correlated with another research (Hains and Truscott, 2010). KEGG
52

pathway enrichment analysis of 199 decreased deamidated proteins showed that protein
associated with protein aging and processing, such as prion disease, ER processing, might be
downregulated during KSHV infection (Fig. 2-2-S3E). Scatter plot shows the detected peptides
by its’s deamidation change and p-value with highlight of some metabolic enzymes (Fig.S3F).
The, by integrating protein interaction database, we conducted protein-protein interaction
network analysis on proteins with increased deamidation. We found six clusters including
carbon metabolism enzymes and amino acid synthesis (Fig.3C) Although the enrichment of
keratinocyte cornification in PPI might be specific in HOK16B cells due to keratinocyte
programmed cell death.

2D screening of carbohydrate metabolism
As the detection sensitivity of deamidations depends on the protein abundance and MS
performance, we also used two-dimensional gel electrophoresis (2DGE) to corroborate with the
MS analysis. The gain of negative charges by deamidation can be detected using 2DGE by
showing shift toward the positive pole of the strip. Therefore, we constructed a stable cell line
library that each cell line expressing an exogenous v5-tagged metabolic gene in 293T cells. To
ensure the lytic replication, we transfected KSHV lytic promoter, kRTA, prior to KSHV infection
and harvested 293T cells at 48 hpi. 2DGE of stable cell lines produced results supporting those
deamidations in GPI and LDHA identified using endogenous proteins (Fig. 2-3D) and revealed
new deamidations in proteins not detected in endogenous proteins, including G6PD (PPP),
PDHA1 (glycolysis), malate dehydrogenase 2 (MDH2) (TCA cycle) and glycogen phosphorylase
(PYGB, glycogen) (Fig. 2-3E). These results suggest that deamidation is a common mechanism
in regulating central carbon metabolic enzymes. Other 2D screening experiments results are
summarized as well. (Fig. 2-S4, S5)

53

Figure 2- 3 Proteomics analysis of KSHV Lytic Replication
(A) KEGG pathway enrichment analysis of proteins with increased deamidated peptides
54

(B) Volcano plot showing deamidated peptides differentially modified between Mock and KSHV
infected HOK16B cells. Peptides in red are upregulated metabolic enzymes’ peptides, peptides
in blue are downregulated.
(C) Network showing the high-confidence interaction between deamidated proteins. In total 249
deamidated proteins with 10532 interactions were extracted from the STRING database.
(D-E) 293T cells were lentivirus that contains the V5-tagged metabolism enzymes and selected
with puromycin. Then, 293T cells were transfected with RTA and infected with Mock and KSHV
24 hours after transfection. Whole cell lysate of 293T/V5-enzyme cells were harvested at 24hpi
of Mock- or KSHV-infection and analyzed by 2D gel electrophoresis and immunoblotting by V5
antibody. (D) 2DGE of GPI, LDHA, and MDH2, validating globally tandem MS/MS result. (E)
2DGE of PYGB, PDHA1, and G6PD, result of 2D screening using KSHV lytic replication stable
cell line
55

Supplementary Figure 2- 3
(A) Heatmap representing the Z-scores of PPP metabolism gene expression (FPKM) of
Mock- and KSHV-infected HOK16B, HUVEC, and LEC cells at 48 hpi.
(B) Heatmap representing the Z-scores of PPP metabolism gene expression (FPKM) of
Mock- and KSHV-infected PBMCs, TIVE and CD14+ cells at indicated time points
generated from public RNA-sequencing data (GSE62344).
(C) Pathway enrichment analysis of significantly changed genes (adjust.P value > 0.01) from
56

time-series analysis in Mock- and KSHV-infected PBMCs, TIVE and CD14+ cells
(D) Summary of deamidated peptide profiling of KSHV lytic replication from a total of 950
identified peptides.
(E) KEGG pathway enrichment analysis of proteins with increased deamidated peptides.
(F) Schematic view of glycolysis protein and their deamidation sites in KEGG protein
network. Red dots indicate proteins with identified deamidation sites in KSHV infected
cells, gray dots indicated peptides or deamidation not detected.

57

Supplementary Figure 2- 4
(A-B) 293T cells were lentivirus that contains the V5-tagged metabolism enzymes and selected
with puromycin. Then, 293T cells were transfected with RTA and infected with Mock and KSHV
24 hours after transfection. Whole cell lysate of 293T/V5-enzyme cells were harvested at 24hpi
of Mock- or KSHV-infection and analyzed by 2D gel electrophoresis and immunoblotting by V5
antibody.
58

Supplementary Figure 2- 5
(A-D) 293T cells were lentivirus that contains the V5-tagged metabolism enzymes and selected
with puromycin. Then, 293T cells were transfected with RTA and infected with Mock and KSHV
24 hours after transfection. Whole cell lysate of 293T/V5-enzyme cells were harvested at 24hpi
of Mock- or KSHV-infection and analyzed by 2D gel electrophoresis and immunoblotting by V5
antibody.

59

Pentose Phosphate Pathway is required for KSHV Lytic Replication
To probe the role of central carbon metabolism sub pathways in KSHV lytic replication,
we used metabolism inhibitors to block sub pathways during KSHV de novo infection and tested
virus replication. Pharmacological inhibitors include PFK15, glycolysis inhibitor; sodium
oxamate, lactate dehydrogenase inhibitor; 6-AN, PPP pathway inhibitor; DCA, pyruvate
dehydrogenase kinase (PDK) inhibitor; and SB204990, the ATP citrate lyase (ACLY) inhibitor
(Fig. 2-4A). IC50 of inhibitors were measured in HOK16B cells and determined by XTT assay at
48 hours post-treatment (Fig. 2-S6A). To ensure homogenized infection, we first infected
r219.KSHV with spin inoculation at MOI 10, then we treated infected cells with chemical
inhibitors at two hpi. Fluorescence images, capturing GFP and RFP expression, were taken at
48 and 72 hpi and quantified using ImageJ, where the GFP and RFP fluorescence area were
quantified separately as an indicator of infected cells and lytic replication cells. Among those
inhibitors, 6-AN and DCA treatment significantly reduced RFP/GFP ratio, the lytic replication
indicator, suggesting that PPP and pyruvate utilization are essential for KSHV lytic replication
(Fig. 2-4B,C, Fig. 2-2-S6B, C, D). Virus infection causes a strong oxidative response, but the
GSH/GSSG ratio has not significantly increase during KSHV infection, indicating an increased
anti-oxidative metabolism (Fig. 2-S7A). Since PPP is critical not only for its role in producing
nucleotide intermediates, R5P, but also as a major resource for antioxidant, NADPH, we
hypothesize that PPP is critical for KSHV lytic replication. To further investigate PPP, we
quantified the virus titer in supernatant and KSHV DNA genome in cell pellets after 6AN and
PFK15 treatment, where 6AN treatment efficiently reduced KSHV genome copies and infectious
virus production (Fig, 4D, E), as well as the KSHV lytic protein expression (Fig. 2-4F). Besides
the energy utilization, PPP produces antioxidant reagents, NADPH, as shown by GSSH/GSG
ratio, the anti-oxidative reaction is up-regulated upon KSHV infection (Fig.S7A). To test if the
result is mammalian gamma-herpesvirus specific, we treated mouse gamma-herpesvirus,
mHV68, infected mouse MEF cells with metabolism inhibitors, and measured the virus titer in
60

cell supernatant. 6AN treated mHV68-infected cell showed an intense decrease in infectious
mHV68 numbers, suggesting PPP pathway is also critical for mHV68 replication (Fig. 2-S7B).
We also quantified the interferon-stimulated genes (ISGs) expressions in 6-AN treated cell,
where ISGs have no significant expression changes, implying that 6-AN regulate KSHV
replication in a metabolism-dependent manner (Fig. 2-S7C). [1,2-
13
C] glucose tracing of 6-AN
treated cells validated that oxidative PPP is blocked, as well as the PPP-produced M2 lactate
(Fig. 2-S7D, E, F). Considering the chemical inhibitors might have side effects, we established
the shRNA stable cell lines that targeting the metabolic enzymes, including PFKL, G6PD, TKT,
GPI, and DLAT, and tested the KSHV replication (Fig. 2-S7G). Both PFKL and GPI are
glycolytic enzymes, while G6PD and TKT are PPP enzymes, DLAT is part of the pyruvate
dehydrogenases enzymes that connect the pyruvate to the TCA cycle. ShG6PD stable cell lines
produced less infectious KSHV in supernatant and had lower DNA genome copy in cell pellets
compared to other stable cell lines, which validated the chemical inhibitor treatment. ISG
expression in shG6PD was stable comparing to shGIPZ control cells (Fig. 2-S7H), while the
6PG, G6PD catalyzed product, was diminished (Fig. 2-S7I). Taken together, PPP is required for
KSHV lytic replication as PPP supports the anti-oxidative response and ribose for the nucleotide
synthesis.
61

Figure 2- 4 PPP is required for KSHV Lytic Replication
(A) Model of the metabolism inhibitor treatment experiment and schematic view of glycolysis
inhibitor treatment experiment.
(B) Cells were infected with KSHV at MOI 10 and treated with metabolism inhibitors two
hours after spin inoculation. Images were taken at 48 hpi and 72 hpi with at least three
independent areas for replicates. The RFP and GFP fluorescence areas were measured
using ImageJ. Inhibitor treatment concnetration: PFK15 (1uM), 6AN (100 μM), SB-
204990 (5μM), DCA (10μM), Sodium Oxamate (10mM).
(C) Fluorescence Image showing virus replication changes after chemical treatment (6AN,
PFK15)
62

(D - E) Cell supernatants were harvested at 48 hpi and 72 hpi and subjected to plaque
assay in 293T cells. (D) shows KSHV virus titers after chemical treatment in HOK16B cells,
and (E) shows KSHV virus titers after chemical treatment in LEC cells.
(F) Whole cell lysate was harvested at 72 hpi and 96 hpi and immunoblotted with KSHV lytic
proteins, kRTA and kTK.
(G) Cell supernatants were harvested at 48 hpi from shRNA (shHK1, shPFKP, shG6PD,
shTKTL1, shDLAT) stable cell lines and subjected to plaque assay in 293T cells.
(H) Cell pellets were harvested at 48 hpi from shRNA (shHK1, shPFKP, shG6PD, shTKTL1,
shDLAT) stable HOK16B cell lines and quantified virion genome copy using qPCR.

63

Supplementary Figure 2- 6
(A) Chemical inhibitors’ IC50 measured in HOK16B cells using XTT cell viability assay.
(B - C) Cells were infected with KSHV at MOI 10 and treated with metabolism inhibitors two
hours after spin inoculation. Images were taken at 48 hpi and 72 hpi with at least three
independent areas for replicates. The RFP and GFP fluorescence areas were measured using
ImageJ. (B) GFP quantification of fluorescence image of KSHV infected cells. (C) RFP
quantification of fluorescence Image of KSHV infected cells
(D) Representative fluorescence images after chemical treatment (6AN, PFK15, SO, SB, DCA)
64

Supplementary Figure 2- 7
(A) Intensities of metabolites, including oxidized glutathione (GSSG) and reduced glutathione
(GSH), were measured in metabolomics data as indicated in Fig. 1. GSH is the major
antioxidant that reduce the lipid hydroperoxides by GPx, while GSSG is product of the GPx
catalyzed antioxidant reaction. GSSG/GSH ratio indicates the oxidative stress, the higher
GSSG/GSH ratio, the higher oxidative stress.
(B) Metabolic inhibitors were treated two hours post mHV68 infection (MOI =1). Cell supernatant
of infected MEF cells were harvested at 96 hpi and measured by plaque assay for its virus titer.
(C) RNA was extracted from cell pellets harvested after 6-AN or DMSO treated for 24 hours.
After reverse-transcription, cDNA samples are subjected to real-time quantitative PCR to
determine the ISG gene expression.
65

(D-F) [1, 2 -
13
C]-glucose medium replaced the original medium of KSHV infected HOK16B cells
at 24 hpi or 48 hpi for 45 minutes before harvested for metabolite extraction.
(G-H) RNA was extracted from stable cell lines and quantified by real-time quantitative PCR
after reverse-transcription to determine the knockdown efficiency (G) or ISG gene expression

66

G6PD Activity and Deamidation Increased during KSHV Lytic
G6PD is the rate-limiting enzyme of PPP and one of the enzymes with high allele
mutations among humans. Mutation of G6PD can lead to enzyme deficiency which will cause
severe hemolytic diseases. G6PD deficiency is common in Africa and the Mediterranean, where
KSHV has higher seroprevalence, although the intrinsic correlation was unknown. PPP pathway
is required for the KSHV infection during which G6PD is deamidated, so we measured both
deamidation and enzymatic activity of G6PD during KSHV infection in HOK16B, LEC cells, and
HCT116 cells, which all undergo lytic replication spontaneously. We found G6PD activity
increased in a time dependent manner, where G6PD has a parallel increase in deamidation,
suggesting that the deamidation might regulate G6PD activity (Fig. 2-5A, 5B, Fig. 2-S8A, B). To
test this, we then purified the G6PD from mock-infected and KSHV-infected cells and analyzed
G6PD’s post-translational modification by the tandem MS/MS (Fig. 2-S8 C, D). We identified
three increased deamidated sites, including asparagine (N) 126, N301, and N302, which N126
has increased most (Fig. 2-5C). Since G6PD is well studied and has 3D dimerized structure
(PDB: 2BHL), we mapped the deamidation sites on its 3D structure. The 3D structure of G6PD
suggests that the deamidation of N126 might form a salt bridge with Arginine nearby for a more
stable conformation, while the deamidation of N301&302D might promote the stabilization of
hydrophobic pocket near the second NADP+ binding site (Fig. 2-5D).

Deamidation of G6PD Increased Activity and Promoted KSHV Replication
To probe the role of the deamidation specificity, we constructed three aspartic acid
deamidation mutants (N > D): N122D, N301&302D, and triple mutant (TM, N122D, N301D, and
N302D), validating by 2DGE of G6PD wild type and mutants from E. coli (Fig. 2-S8E).
Deamidation mutants and wild type G6PD are purified from E. Coli. and subjected to the G6PD
activity assay. N126D and N301&302D mutants have higher G6PD activities compared to the
wild type, while the TM mutant has similar activity. (Fig. 2-5E) Then we constructed the G6PD
67

CRISPR/Cas9 knock-out HOK16B cells with reconstituted Flag labeled G6PD wild type and
mutants and subjected to [1,2-
13
C] glucose tracing. We found that the metabolic flux to produce
6PG in N126D and N301&302D stable cell lines were faster than in wild type cells, with the
increase of the R5P flux suggest that N126D G6PD produced more nucleotide metabolism
intermediates (Fig. 2-5F, G) In addition, mammalian cells have different 2DGE species
compared to the G6PD in E. coli, suggesting mammalian G6PD might have other PTM besides
deamidation. The phosphorylation of G6PD can regulate its activity, so we conducted
radioactive P
32
-labeled kinase assay that treated purified G6PD with Mock and KSHV-infected
cell lysate. We found that N126D and TM mutant has less phosphorylation in KSHV- infected
cell lysate, indicating the evasion of the canonical regulatory mechanism. We measured the
lactate secretion and glucose consumption in both the wild type and mutant cells and found that
the higher activity mutants, N126D and N301&302D mutants intake less glucose than wild type
and produce less lactate, suggesting an unknown mechanism in regulating glucose intake and
G6PD activity (Fig. 2-S9C,D). In addition, we measure if the higher activity mutants lead to the
changes of the KSHV replication by measuring the infectious virus in the supernatant and virus
genome copy in cell pellet. We found that cells expressing deamidation mutants have higher
KSHV replication compared to wild type cells suggesting that deamidation of G6PD promote
virus replication which is correlated with G6PD activity. Worth to mentioning, although G6PD
mutant cells grow slower compared to the wild type cell lines, the mutant G6PD is more stable
compared to the wild type (Fig. 2-S9B, E, G).

68

Figure 2- 5 The Functional Role of G6PD Deamidation
(A) Whole cell lysate of HOK/V5-G6PD cells were analyzed by 2D gel electrophoresis
and immuno- blotting from HOK16B cells and LEC cells.
69

(B) G6PD enzyme activity during KSHV infection. G6PD was harvested from HOK16B
cells and LEC cells with or without KSHV infection at hpi 24, 48 and 72 and analyzed by
biochemistry enzymatic activity assay.
(C) G6PD was purified from HOK cells with Mock or KSHV infection at hpi 48 and
analyzed by tandem mass spectrometry for deamidation. Three increased deamidation residues
were identified, in which N126 was the most significantly increased one. MS/MS representation
figures showing N126D deamidation peptides.
(D) Ribbon structure of two subunits from human G6PD (PDB:2bhl), highlighting the
asparagine (N) residues was converted into D (in red) due to deamidation. D126 forms a
possible salt chain with Asparagine nearby, while D301&302 participate in NADP
+
binding.
(E) 6xHis tag G6PD wild type and mutants are purified from E. Coli. and measured by
the concentration by comparing to BSA concentration. Sample amounts of 6xHis-G6PD wild
type and mutants were assayed for G6PD activity by measuring the absorbance 340nm for
NADPH production.
(F-G) [1, 2 -
13
C]-glucose medium replaced the original medium of Flag-G6PD WT and
mutants reconstituted HOK16B cells for 45 minutes before harvested for metabolite extraction.
(F) 6PG is the product of G6PD, (G) R5P-Ru5P is the product of oxidative PPP.
(H) Supernatant were harvested at indicated time points and serial diluted, then diluted
medium infected KSHV titers in infected G6PD mutant stable cell lines
(I) Cell pellets were harvested at 48 hpi from Flag-G6PD reconstituted stable HOK16B
cell lines and quantified virion genome copy using qPCR.
(J) Whole cell lysate of Flag-G6PD stable cell lines were analyzed by native gel for
dimerization assay and immunoblot by anti-flag.
70

Supplementary Figure 2- 8
(A) Whole cell lysate of HOK/V5-G6PD cells were analyzed by 2D gel electrophoresis
and immuno- blotting from HCT116 cells.
(B) G6PD enzyme activity during KSHV infection. G6PD was harvested from HCT116
cells with or without KSHV infection at hpi 24, 48 and 72 and analyzed by biochemistry
enzymatic activity assay.
71

(C) HOK/V5-G6PD cells were infected with mock and KSHV (moi = 30) for 24 hrs. Whole
cell lysates (WCLs) were precipitated with anti-V5 agarose gel and analyzed by Coomassie
staining. Gel pieces were then cut and in-gel digest and subject to MS analysis.
(D) MS post-translational modification analysis.
(E-F) Whole cell lysate of G6PD were analyzed by 2DGE. GST-G6PD wild type and
mutants were purified from E. coli (E), Flag-G6PD wild type and mutants were purified from
HOK stable cell lines.

72

Supplementary Figure 2- 9
(A) Cell growth curves in shGPIZ and shG6PD
(B) Cell growth curves of G6PD mutants
(C) Supernatant of mock- and KSHV- infected HOK16B cells are harvested as indicated time
points and measured for glucose concentration and lactate concentration. The original glucose
concentration in HOK medium is 1 g/L.
73

(D-E) 293T cells were transfected with wt and mutant Flag-G6PD overnight before treatment of
translational inhibitor, cycloheximide (CHX) at 50ug/ml. Whole cell lysates were harvested at
indicated time points and analyzed by SDS-PADE and immunoblotting.
(F) 6xHis-G6PD wt and mutants were incubated with Mock/KSHV whole cell lysate or purified
Flag-P53 in the presence of [γ-32P]-ATP at 30 °C for 30 min. The 6xHis-G6PD was purified
from E. coli. Kinase assay samples were boiled with SDS/PAGE sample buffer before separated
and transferred in NC membranes. Membrane were exposed overnight before imaging.
(G) Flag-G6PD wt and mutants were purified from reconstituted stable cell lines. Bead-
associated G6PD were treated alkaline phosphatase, Calf Intestinal (CIP) to remove the
phosphate group at 37 °C for 30 mins. Treated/untreated samples were boiled with SDS/PAGE
sample buffer and analyzed by SDS-PADE and immunoblotting.
(H-J) [1, 2 -
13
C]-glucose medium replaced the original medium of Flag-G6PD wt and mutants
reconstituted stable HOK16B cells for 45 minutes before harvested for metabolite extraction.

74

CAD catalyze G6PD deamidation during KSHV infection
Our published and unpublished data demonstrate that nucleotide synthesis enzymes,
PPAT, PFAS, and CAD are GATs and regulate fundamental biological processes. Combining
affinity purification with MS analysis, we discovered that GATs interacted with multiple central
carbon metabolic enzymes, such as LDHA, ACLY, transketolase (TKT), and G6PD (Fig. 2-
S10A). Cellular PFAS and CAD catalyze the de novo synthesis of purine and pyrimidine,
respectively, and are activated during the KSHV infection (Fig. 2-S10). To identify how the
deamidation of G6PD is regulated during the KSHV process, we hypothesize that highly
activated nucleotide enzymes catalyzed the G6PD deamidation. To test this idea, we used the
size-exclusion column filtration to identify the co-elution of CAD and G6PD, no other GATs (Fig.
2-6A). Additionally, we screened both the host GATs for G6PD-binding ability by co-expressing
G6PD and GATs in 293T cells and using co-immunoprecipitation to validate (Fig. 2-6B). The
result demonstrated that CAD interacts with G6PD better than other enzymes, suggesting CAD
might be the GAT that catalyzed the G6PD deamidation during KSHV infection. To test this
idea, we constructed the stable cell line with knockdown of CAD expression by shRNA in
HOK16B cell lines and measured the intracellular G6PD activity. We found shCAD has reduced
G6PD enzymatic activities compared to shGIPZ (Fig. 2-6C), while G6PD is less deamidated in
shCAD cell lines as shown by 2DGE (Fig. 2-6D). To eliminate the contamination by other
cellular modification, we performed in vitro deamidation assay used purified G6PD form E. coli
and purified CAD wild type and CAD enzymatic-dead (ED) mutants. We found that wild type
CAD, not CAD-ED, can deamidate G6PD in vitro and increase enzymatic activity (Fig. 2-6E,F),
confirming that CAD is the bone fide deamidase of G6PD. As previously reported, knockdown
of CAD might reduce KSHV replication, so we infected the shCAD stable cell lines with KSHV
(MOI= 30) and tested the infectious virus in supernatant. We found that shCAD can reduce
KSHV replication (Fig. 2-6I). Treatment of the general deamidation inhibitor, DON, and the
CAD-specific inhibitor, 2TCPA, can block the G6PD deamidation and reduce its enzymatic
75

activity. DON, the glutamine-analog, can inhibit all glutamine-aminotransferase activity and
inhibit glutamine metabolism, showed more profound effect compare to specific inhibitor,
suggesting other unknown regulatory mechanism of G6PD. In addition, treatment of DON and
2TCPA inhibit the KSHV replication in cells, might as a clinical inhibitor for those infection
diseases that activate PPP pathway by G6PD deamidation.
76

Figure 2- 6 CAD deamidates G6PD to promote KSHV replication
77

(A) Whole cell lysate of 293T/Flag-CAD CRISPR-Cas9 Knock-in cell line was analyzed by gel
filtration and immunoblotting with anti-Flag antibody, anti-G6PD antibody, anti-PPAT
antibody, and anti-PFAS antibody.
(B) G6PD-V5 and flag-tagged mammalian GATs proteins, including CAD, PFAS, and PPAT, are
expressed in 293T cell by transfection for 48 hours. Whole cell lysates were precipitated with
anti-Flag (GATs). Precipitated proteins and whole cell lysate were analyzed by
immunoblotting.
(C) Whole cell lysate of HOK/shRNA-GIPZ, HOK/shRNA-CAD1, and HOK/shRNA-CAD2 was
analyzed by G6PD activity assay.
(D) Whole cell lysate of HOK/shRNA-GIPZ, HOK/shRNA-CAD1, and HOK/shRNA-CAD2 was
analyzed by 2D gel electrophoresis and immunoblotted using G6PD antibody to examine
G6PD deamidation.
(E) In vitro deamidation of G6PD was analyzed by 2D gel electrophoresis and immunoblotting
(right). GST-G6PD was purified from E. Coli., CAD-WT and CAD-ED were purified from
293T cells to homogeneity and analyzed by silver staining (left). WT and ED, wild-type and
the enzyme-dead mutant of CAD.
(F) In vitro deamidation was carried out as in (E), sample were diluted 1:100 and subjected to
G6PD enzymatic assay. Error bar denotes SD (n =3).
(G) HOK cells were treated with DMSO, DON and 2TCPA for 24 hours before KSHV infection.
Whole cell lysate was harvested at 48hpi and analyzed by 2DGE.
(H) HOK cells were treated with DMSO, DON and 2TCPA for 24 hours. Whole cell lysate
sample were diluted 1:100 and subjected to G6PD enzymatic assay. Error bar denotes SD
(n =3).
(I) Supernatant of KSHV infected HOK/shRNA-GIPZ, HOK/shRNA-CAD1, and HOK/shRNA-
CAD2 stable cell lines were harvested at 48 hpi and 72hpi and analyzed by plaque assay to
determine the KSHV virus titer.
78

(J) HOK cells were treated with DMSO, DON and 2TCPA for 24 hours before KSHV infection.
Supernatant was harvest at 48hpi and 72 hpi to measure the virus titer.
(K) ) [1, 2 -
13
C]-glucose medium replaced the original medium of DON/2TCPA treated HOK16B
cells for 45 minutes before harvested for metabolite extraction.

Supplementary Figure 2- 10
79

(A) Summary of PFAS and CAD affinity purification – MS results.
(B) 293T cells were transfected with Flag-tag GATs for 24 hours. Whole cell lysate was
harvested and analyzed by 2DGE.
(C) Whole cell lysate of Mock and KSHV infected HOK16B cells was analyzed with CAD
phosphorylation antibody (Ser1859).
(D) Cell pellets were harvested at 48 hpi from HOK/shRNA-GIPZ, HOK/shRNA-CAD1, and
HOK/shRNA-CAD2 stable cell lines and quantified virion genome copy using qPCR
(E) LEC cells were treated with DMSO, DON and 2TCPA for 24 hours. Whole cell lysate sample
were diluted 1:100 and subjected to G6PD enzymatic assay. Error bar denotes SD (n =3).
(F) HOK16B cells were treated with DMSO, DON and 2TCPA for 24 hours. Whole cell lysate of
HOK16B cells with Mock infection and analyzed by 2DGE for isoelectric shift.
(G) [1, 2 -
13
C]-glucose medium replaced the original medium of DON/2TCPA treated HOK16B
cells for 45 minutes before harvested for metabolite extraction.

80

Chapter 3 Method
Data and Code Availability
Datasets supporting this research are available on request, including imaging data,
metabolomics, proteomics, and RNA-sequencing data. Original image data of microscopy, FLIM
imaging, and immunoblots in this paper are available at Mendeley Data sets:
https://data.mendeley.com/datasets/myfxss42jh/draft?preview=1
The accession numbers for RNA-sequencing datasets reported in this dissertation are
available at the NCBI GEO database.
The access of code is available at GitHub: https://github.com/maotian06/KSHV_Metabolism

Experimental Model and Subject Details
Cell Culture
HEK293T, HCT116, mouse embryonic fibroblasts+ (MEF) and 3T3 cells were
maintained in Dulbecco’s modified Eagle’s medium (DMEM, HyClone) with 10% fetal bovine
serum (FBS; Gibco), and penicillin/streptomycin (10,000 U/mL, Corning®). HOK16B cells were
cultured in KBM Keratinocyte Growth Medium supplemented with KGM
TM
-2 SingleQuots
TM

Supplements (Lonza). LEC cells were cultured in EBM
TM
-2 Endothelial Cell Growth Basal
Medium with EGM
TM
-2 Endothelial SingleQuots
TM
Kit (Lonza). All other cell lines were cultured
in their corresponding medium according to the ATCC recommendation.
Plasmids
The non-silencing (control) shRNA and shRNA for CAD were purchased from Thermo
Scientific. The shRNA (GIPZ lentiviral shRNA) for G6PD, HK1, HK2, PFKL, PFKP were kindly
provided by Dr. Chengyu Liang. Mammalian expression plasmid (pLX317 with C-terminal V5-
tag sequence) to construct the metabolism enzyme stable cell lines were purchased from
MISSION® TRC3 Open Reading Frame (ORF) clones. Mammalian expression plasmid
(pcDNA5/frt/to with N-terminal Flag-tag sequence) and lentiviral expression plasmids (pCDH-
81

CMV-EF1-PURO) for G6PD was generated based on its cDNA construct in TRC3 ORF clones.
All point mutants were generated in pENTR clones by site-directed mutagenesis with
sequencing confirmation, then pENTR clones were transferred to pCDH vectors/pDEST vector
using Gateway® Cloning kit (Invitrogen), including G6PD-N122D, N122A, N126D, N126A,
N301D&N302D, N301A&N302A, Q471E, Q471A, double mutant (N126D, Q471E), and triple
mutant (N126D, N301D&N302D).

Experiment Method
RNA Extraction and Quantitative Real-time PCR (qRT-PCR)
Cellular RNA was extracted from whole cell lysate using TRIzol reagent (Invitrogen) and
dissolved in RNase-free DI water with concentration measured by nanodrop. PrimeScript
Reverse Transcriptase (Clontech) synthesized complementary cDNA from 1 microgram of total
RNA. cDNA was diluted 1:20~50 and analyzed by qRT-PCR using SYBR® Green Master Mix
(BIO-RAD) with CFX ConnectReal-Time PCR (BIO-RAD). Target genes’ mRNA expression was
calculated using the 2
-ΔΔCt
method and β-actin as an internal control for cellular genes and orf50
as an internal control for viral genes.
KSHV Production and Infection
Recombinant KSHV, r219.KSHV, was produced from the iSLK.219 cell lines. iSLK.219
cells are maintained in DMEM medium with 10 µg/ml puromycin, 800 µg/ml G418, and 1200
µg/ml hygromycin. As for virus production, iSLK.219 cells are induced with doxycycline (1 µg/ml)
and sodium butyrate (900 µM) for 48 hours or 72 hours. Then the supernatant of iSLK.219
containing r219.KSHV will be harvested and concentrated using ultracentrifuge (23,000 rpm, 2
hours, 4°C). After concentration, virus pellets will be dissolved in cell culture medium that will be
used in the following cells and quantified based on serial dilution and plaque assay.
As for KSHV infection, cells are seed into 6-well or 12-well one day before and infected
with KSHV. The infection MOI is based on the virus concentration and cell confluency. After
82

adding the virus to the cells, cells will be spin-inoculated at 1,800 rpm for 45 mins at 37 °C and
then incubated at 37 °C incubator for two hours before changing medium.
Chemical Inhibitor Treatment
Chemical inhibitors includes: PFK158 (SelleckChem), 6-Aminonicotinamide(6AN,
CaymanChem), SB204990 (CaymanChem), sodium oxamate(Sigma), sodium dichloroacetate
(DCA, Sigma). Chemical inhibitor IC50 were measured using the XTT Cell Viability Assay Kits
(Biotium #30007) as the manufacturer’s protocols. Each data point was performed in triplicate.
The IC50 value of chemical inhibitors were listed in Supplementary Figure 2-6A. Chemical
inhibitors were added into the medium two hours post KSHV spin-inoculation. Medium were
changed every 24 hours into fresh medium containing the chemical inhibitors.
Whole Cell Lysate Preparation
The whole cell lysate for western blot assay was lysed with Triton X-100 buffer or NP-40
lysis buffer (20 mM Tris, pH 7.5, 150 mM NaCl, 20 mM β-glycerophosphate, 1 mM sodium
orthovanadate, 10% glycerol, 2 mM EDTA, 1% NP40) supplemented with a protease inhibitor
cocktail (Roche) for 15 mins at 4°C before sonification (20%Amp, 10 seconds, 3 times) and
centrifugation (12,000 rpm for 15 min at 4°C). Supernatant of whole cell lysate with 6x loading
buffer (375 mM Tris.HCl, 9% SDS, 50% Glycerol, 0.03% Bromophenol blue) was boiled at 95 °C
before analyzed by SDS gel electrophoresis and immunoblotting.
Mammalian Protein Purification
HOK16 cells that stably expressed G6PD-V5 were infected with Mock or KSHV. Cells
were harvested at 48 hpi and lysed with Triton X-100 buffer (20 mM Tris, pH 7.5, 150 mM NaCl,
1.5 mM MgCl 2
, 20 mM β-glycerophosphate, 1 mM sodium orthovanadate, 10% glycerol, 0.5 mM
EGTA, 0.5% Triton X-100) supplemented with a protease inhibitor cocktail (Roche). Whole cell
lysates were incubated for 15 mins at 4°C in lysis buffer before sonicated and centrifuged at
12,000 rpm for 15 mins at 4°C. Supernatant was harvested, filtered, pre-cleared with protein
83

A/G agarose beads at 4°C for 1 h and then incubated with anti-V5 agarose beads (Sigma,
A7345) at 4°C for 3 h. The agarose beads were washed with lysis buffer for 4 times and
analyzed by SDS gel electrophoresis and silver staining. Then the purified proteins were cut
from the gel and subjected to tandem MS/MS.
E. coli Expression Protein Purification
6xHis-G6PD wild type and mutants are constructed in pET45b plasmids or cloned using
Gateway system in pDEST destination plasmids and transformed in BL21 (DE3) Competent E.
coli. BL21 E. coli was cultured at 37°C for around 4 hours and induced with 0.5mM IPTG when
the OD is 0.6, then the E. coli was continuing cultured at 20°C overnight. For GST-tag proteins,
samples were harvested and lysed using the lysis buffer (1% Triton X-100, 0.1% Sarcosyl PBS)
supplement with lysosome (0.5 mg/ml) for 30 mins before sonification (30%Amp, 10 seconds, 3
times) and centrifugation (12,000 rpm for 20 min at 4°C). Cell lysate were incubated with Pierce
Glutathione Agarose Resin (ThermoFisher) for one hour in the rotator. Then, GST-beads bound
proteins were washed in lysis buffer for 4 times before elution or other experiments.
For His-tag proteins, samples were harvested and lysed using the His-binding buffer
(50mM Tris-Cl pH8.0, 0.5 mM NaCl, 0.1 mM EDTA) supplement with 0.5 mg/ml lysosome for 30
mins at 4°C before sonification and centrifugation. Then cell lysate were incubated with Ni-
agarose beads for 1 hour at 4°C before washed using His-Wash buffer(50mM Tris-Cl pH8.0, 0.5
mM NaCl, 0.1 mM EDTA, and 10 mM Imidazole).
Co-immunoprecipitation (Co-IP) and Immunoblotting
Expression plasmids were transfected into HEK293T cells for 48 hours before
harvesting, while KSHV infected cells were harvested at 48 hpi. Cell pellets were lysed with
NP40 for 15 mins at 4°C, pipet up and down for 20 times, centrifuged and harvested for
supernatant. Cell lysates were pre-cleared with Sepharose® CL-4B beads, incubated with
antibodies overnight, and conjugated with protein A/G agarose for 1.5 h at 4°C. The agarose
84

beads were washed 4~6 times extensively and boiled at 95°C for 10 min. 5% of IP sample were
analyzed with precipitated proteins by SDS gel electrophoresis and immunoblotting.
All immunoblotting was performed using the indicated primary antibodies (1:500 ~ 2000
dilution) and IRDye®800CW Streptavidin secondary antibodies (1:10,000 dilution, Li-Cor 926-
32230). Proteins were visualized by Odyssey® CLx imaging system (Li-Cor) and Image
Studio™.
RNA Library Preparation and Sequencing
RNA was harvested using TRIzol™ Reagent (Invitrogen™) and QIAGEN RNeasy Mini
Kit (Cat#74104). According to the manufacturer’s protocol, 1 ug of total RNA was used for the
construction of sequencing libraries. Libraries were prepared by KAPA RNA-Seq library
preparation kit (KAPA Biosystems, Roche). Then prepared libraries were checked for quality
control and sequenced at the USC Molecular Genomics Core using Illumina NextSeq 500. The
quality of raw files of RNA sequencing was examined by FASTQC for quality control before
processing using the Tuxedo protocol. We used the UCSC hg38 genome to build Homo sapiens
genome index and aligned raw files to sequencing reads using HISAT (v2.1.0)(Kim et al., 2015).
Sequencing reads were assembled and quantified using Samtools and StringTie (v1.3.6)(Pertea
et al., 2016), then processed for differential gene expression by DESeq2 (v1.16.1)(Love et al.,
2014) and ballgown (Ref). Pathway enrichment analysis were conducted using R package
clusterProfiler(Yu et al., 2012) Analysis code could be found at Github. Sequencing data are
available at: https://www.ncbi.nlm.nih.gov/geo/.
Lactate Colorimetric Assay
Cells (2 × 10
5
) were seeded in 6-well over night and infected with KSHV with spin
inoculation at 1,800 rpm for 45 mins. The cell supernatant was collected at 24 hpi, 48 hpi, and
72 hpi. The medium was harvested and filtered by a 10kD MW spin filter (Thermo) and diluted
at the range of 1:20~40 with lactate assay buffer. Then samples were measured using
spectrometer by the Lactate Assay Kit (Sigma).
85

Metabolic Profiling and Isotope Tracing
HOK16B cells or LEC cells were seed in 6-well one day before infection. KSHV infected
and mock infected cells (2 × 10
7
) were harvested at indicated time points, and each sample has
at least triplicates or four replicates. The global metabolomic samples were treated with trypsin
for ten minutes before neutralizing and washing. Cell pellets were frozen at -80 °C and send to
Metabolon Inc. for metabolomics analysis.
Isotope tracing experiments were performed as previously described (Delfarah et al.,
2019) (Sauer, 2006). For glycolytic rate analysis, KSHV/Mock infected HOK16B cells were
cultured with medium containing 1g/ml [U-
13
C]-labeled glucose or [1,2-
13
C]-labeled glucose
(Cambridge Isotope Laboratories) for 30 min. For nucleotide synthesis flux, KSHV/Mock
infected HOK16B cells were cultured with medium containing 1g/mL [1,2-
13
C]-labeled for 2 hrs.
Before harvesting, 80% MeOH (MS grade, Sigma) was pre-cooled at −80°C for at least
one hour. To harvest the intracellular metabolites, medium was vacuumed, and cells were
washed on ice with 1 mL pre-cooled 150 mM ammonium acetate (NH
4
AcO, pH 7.3). After
vacuuming 150 mM NH
4
AcO washing buffer, 1 mL of −80°C pre-cooled 80% MeOH was added
to each well and wrapped by parafilm to prevent MeOH evaporation. Samples were incubated at
−80°C overnight to release metabolites. The next day, cells were scraped off from the 6-well
using clean cell scrapers, and supernatants were transferred to Eppendorf tube and centrifuged
at 4°C for 5 min at 12,000 rpm. The supernatants were harvested into a new Eppendorf tube.
The residual cell pellets were re-suspended using 200 μl ice-cold 80% MeOH and collected
after similar centrifugation. The resuspend supernatant then merged with the initial supernatant
and vacuum-dried at room temperature for 3~4 hours. The final sample is a gel-like pellets at
the bottom of the Eppendorf tube.
Samples were analyzed on a Q Exactive Plus hybrid quadrupole-Orbitrap mass
spectrometer coupled to an UltiMate 3000 UHPLC system (Thermo Scientific) with three or four
biological replicates. Metabolites were detected and quantified as area intensity and normalized
86

using median intensity under the curve based on retention time and accurate mass (≤10 ppm)
using the TraceFinder 3.3 (Thermo Scientific) software. Raw data was corrected for naturally
occurring
13
C abundance using IsoCorrectoR (Heinrich et al., 2018).
Two-dimensional Gel Electrophoresis
Cells (0.5~1 × 10
6
) were lysed in Thiourea rehydration buffer (7 M urea, 2 M thiourea,
2% CHAPS, 0.5% IPG Buffer, 0.002% bromophenol blue) for 30 minutes and sonicated three
pulses. The whole cell lysates were centrifuged at 12,000 rpm for 15 min at 4°C. 120ul of
supernatants were loaded to IEF strips (GE) and running in the isoelectric focusing system with
the same program as indicated in (Zhao et al., 2016b). After IEF, strips were incubated with
SDS equilibration buffer with 10 mg/mL DTT for 20 min and SDS equilibration buffer with 2-
iodoacetamide for 20 min. The recipe of SDS equilibration buffer was described in (Zhao et al.,
2020) . Strips were washed with SDS-PAGE buffer before loading to the 2
nd
dimension SDS
gels and analyzing by SDS-PAGE and immunoblotting.
In Vitro Deamidation Assay
GST-G6PD was purified from E. coli., CAD and CAD-C/S enzymatic-dead mutant were
purified from transfected 293T cells and purified by Flag-conjugated agarose beads before silver
staining quantification. Purified proteins or bead-conjugated proteins were added with 50%
glycerol and kept at -80°C. In vitro on-column deamidation of G6PD was performed as
previously reported (Zhao et al., 2016b). Briefly, ~ 0.2 μg of CAD or CAD enzymatic-dead C/S
mutant, and 0.4 μg of GST-G6PD (bound to glutathione-conjugated agarose) were incubated at
30 μl in vitro deamidation buffer as described in (Zhao et al., 2020) at 30°C for 45 min. Then,
deamidation buffer were vacuumed because the left of any residual salt in reaction can casue
the 2DGE failure. GST-G6PD-conjugated GST beads was eluted with 2DGE rehydration buffer
(7 M urea, 2 M thiourea, 2% CHAPS, 0.5% IPG Buffer, 0.002% bromophenol blue) at room
temperature for 30 minutes. Samples were then analyzed by two-dimensional gel
electrophoresis and immunoblotting.
87

Metabolomics Analysis
Metabolomics result from MS in Mock/KSHV infected cell extracts was normalized using
MetaboAnalystR package in R, where all empty metabolite abundance was replaced with the
least value. Original metabolite abundance (MS intensity area) was normalized based on sum of
all metabolite abundance of a sample. After normalized, multivariate ANOVA analysis was
performed to identify the metabolite significance and pathway enrichment analysis on
normalized data using MetaboAnalystR. Flux data were corrected and normalized using
IsoCorrectoR (Heinrich et al., 2018). The graph of flux data were presented using ggplot2
(Wickham, 2009) and ggpubr package in R, and the code could be found at Github.
Protein Mass Spectrometry Analysis
For identification of deamidation sites, HOK/G6PD-V5 stable cell line was infected with
Mock and KSHV. HOK/G6PD-V5 cells was harvested 48 hours post infection and purified by
anti-V5-conjugated agarose beads for 4 hours at 4°C. Beads were then washed with Triton X-
100 lysis buffer for 5 times and boiled with 1x SDS loading buffer at 95 °C for ten minutes.
Purified G6PD was subjected to SDS-PAGE electrophoresis and gel slices were cut and frozen
at -80°C before in-gel digestion and MS analysis (Poochon Scientific).
Protein-protein Interaction Network Analysis
Deamidation proteomics data were normalized and calculated for P-value. Significantly
increased proteins were analyzed by their protein-protein interaction using STRING database
(Szklarczyk et al., 2021). Then, PPI was constructed in R using igraph package and converted
into Cytoscape using RCys3 package. Clustering analysis was analyzed by degree
centrality and betweenness centrality using Cytoscape (Shannon et al., 2003).
Native Gel Dimerization Assay
The cell pellet of one 6-well Flag-G6PD stable cells was lysed using Triton-X100 lysis
buffer and pipet up and down for full lysis. No sonification needed. After 15 mins incubation on
ice, the cell lysate was centrifuged similar to protein sample purification to remove the
88

undissolved proteins. Then half of the samples were load into the native gel. The recipe of
native gels and running buffer was adapted from previous research (Tanaka and Chen, 2012).
After running gels for 2 hours at constant voltage 80v, the native gels were incubated at SDS-
PAGE running buffer for 30 minutes before transferred to membranes for immunoblotting.
Lentivirus-Mediated Stable Cell Line Construction
Lentiviruses were produced as previously described (Dong and Feng, 2011, Feng et al.,
2008). As for the lentivirus production, HEK293T cells were transfected with the 3
rd
generation
lentiviral packaging plasmids (VSV-G and DR8.9) and the targeted lentiviral expression vectors
(pCDH expression vector, pLxTRC3 expression vector, GIPZ puromycin selection shRNA
vector, and pLKO-Hygro shRNA hygromycin selection vector). At 48 hours post transfection, the
supernatant was harvested and centrifuged at 4,000 rpm for 20 mins to remove the cell pellets
before filtered using 22 um filter. If necessary, the lentivirus supernatant can be concentrated by
ultracentrifugation at 32,000 rpm for 2 hours at 4 °C.
As for the lentivirus infection, 8 ug/ml polybrene were added into the lentivirus medium
before the spin-inoculation at 1,800 rpm for 45 mins at 37 °C. After 4 hours post spin
inoculation, lentivirus medium was changed into original medium. Infected cells were split and
selected at 48 h post infection and maintained in original medium with supplementation of
puromycin (1~2 μg/mL) or hygromycin (200 μg/mL). To establish G6PD-knockout cell lines,
HOK and LEC cells were transduced with lentivirus expressing sgRNA for G6PD (pL-
CRISPR.EFS.PAC-sgRNA-Targeting-G6PD, 3’-UTR) and selected with 1 μg/mL puromycin.
Single colonies were isolated and screened by immunoblotting with G6PD antibody.
G6PD Enzymatic Assay
G6PD enzymatic assay was adopted from previous research (Löhr and Waller, 1974).
6xHis- The concentration G6PD wild type and mutants are quantified with BSA standard using
SDS-PAGE and Coomassie staining. For general activity assay, 0.1ug of G6PD were added
89

into 100ul G6PD enzymatic assay buffer () with 10ul of 3mM NADP+ in each 96-wells. After
incubating for 5 minutes, add 10 ul 3mM Glucose-6-phosphate (Sigma, 10127647001) to the
reaction and quantify 340nm observance under the plate reader in a time-series plate mode.
Fluorescence Image Processing and Quantification
For quantification of KSHV infected cells with GFP or RFP fluorescence marker, images
of live cells in 12-well were taken by confocal microscopy (Leica M, Germany). Fluorescence
intensity of the images was quantified using ImageJ (NIH). For each picture, original RGB image
was transformed to an 8-bit gray image and threshold by a fixed value to remove the
background fluorescence. Areas and integrated densities of images were measured that
indicates the fluorescence cells and their fluorescence intensities. The image processing was
recorded by the ImageJ Macro and coded to store all the measurement at the same csv file.
GFP/RFP fluorescence cell ratio = RFP fluorescence areas/ RFP fluorescence areas.
All image analyses were performed in a blinded fashion and independently by three
individuals. The image processing code and description can be found at:
https://github.com/maotian06/KSHV_Metabolism/FluorescenceImage
Fluorescence Lifetime Imaging Microscopy of KSHV-infected Cells
For imaging KSHV/Mock-infected HOK16B cells were plated onto 35 mm round #1.5
glass coverslips dishes (MatTek, P35GCOL-1.5-14-C) in the growth medium and grown to 70–
80% confluence before KSHV/Mock infection. No treatment needed; dishes were imaged using
Leica SP8 confocal/multiphoton microscopy system. Two spectral was collected: Ch2: 460–
500 nm (NAD(P)H) and Ch3: 520–560 nm (FAD). Leica LAS X software collected and analyzed
the time-resolved fluorescence imaging.
Quantification of the FLIM Optical Redox Ratio
For the glycolytic rate calculation, we mapped the phasor signal to the to linear
regression line between bound NAD(P)H, 400 ps, and unbound NAD(P)H, 2,500 ps. The
90

position of the phasor center was quantified as glycolytic rate. Code can be found at:
https://github.com/peiyuwan/TIC_FLIM_Collaborators.git
Quantification and Statistical Analysis
Data are presented as mean ± standard deviation (SD). For two groups comparison,
statistical analyses were performed by unpaired, two-tailed Student’s t test. For more than two
groups comparison, data were analyzed by ANOVA followed by the unpaired Student’s t test.
Statistical analyses were performed using Graphpad Prism for qPCR, biochemistry assay, and
fluorescence imaging assay; MATLAB for FLIM imaging statistical analysis, R for genomics,
metabolomics, and proteomics analysis. P-value less than 0.05 is considered statistically
significant.
∗
, p < 0.05;
∗∗
, p < 0.01;
∗∗∗
, p < 0.001; ****, p < 0.0001.

91

Chapter 4 Discussion and Perspective
Discussion
In this research, we have made several major findings of the KSHV-mediated metabolic
reprogramming process and provided the multi-omics approach to uncover the molecular
mechanism. We found that KSHV infection upregulates central carbon metabolism, including
glycolysis, TCA cycle, and PPP, in which PPP is necessary for KSHV lytic replication. Central
carbon metabolism enzymes are the putative deamidation targets, including G6PD, LDHA, and
PDHA1. Deamidated G6PD has higher enzymatic activities and promotes KSHV lytic
replication. De novo pyrimidine synthesis enzymes, CAD, deamidates G6PD during KSHV
infection and increases its enzymatic activity. Both the CAD knockdown and CAD inhibitor
treatment decreased G6PD deamidation and its enzymatic activity, thus reducing KSHV
replication.
Metabolism reprogramming is an emerging disease maker since the metabolism
changes are correlated with disease progress and diagnosis. The most well-known example is
cancer’s Warburg effect, where anaerobic glycolysis increased along with the higher local
concentration of lactate in tumor microenvironment. Following the finding of Warburg effect,
researchers also find the increased lactate can prime the immune cells, such as tumor-
infiltrating T cells and NK cells, to benefit cancer proliferation (Brand et al., 2016). However,
based on the observation Figure. 2-5, we found that PPP inhibitor, 6AN, can significantly
decrease the lactate production of KSHV infected cells comparing to the glycolysis inhibitor,
PFK15. We also observed that G6PD deamidation mutant reconstituted cells secreted less
lactate compared to wild type cells, which correlates to the slower proliferation rate. These two
results show that the G6PD might have a regulatory role in lactate production or cell growth,
possibly correlate with G6PD’s interaction with LDHA, which needs further validation. Besides
the potential regulatory role of G6PD in lactate production, there is an irreversible metabolic
cycle between glycolysis and PPP, in which the oxidative PPP product, R5P, can be converted
92

into G3P and G6P by gluconeogenesis. This cycle maximizes the production of NADPH with
limited glucose, thus, can be a fast-responsive mechanism to producing antioxidants. During
KSHV infection, this process is upregulated at 24 hpi based on the [1,2-
13
C]-glucose labeling
result, as shown that the M+5 6PG has higher concentration at 24 hpi compared to Mock or 48
hpi. To validate this, the pyruvate labeling of KSHV infected cells can measure the
gluconeogenesis flux. In general, the critical role of PPP might be underestimated in virus
infection and carcinogenesis, suggesting PPP is a potential drug target for diseases.
Oxidative stimulus reactivates KSHV from latent infected cells, while G6PD response as
a major antioxidant enzyme during oxidative stress. Therefore, the balance of oxidative
response mediated by G6PD can regulate KSHV reactivation. Both G6PD deficiency and KSHV
have high prevalence in Africa, where malaria is one of the leading health concerns. As a blood
transmitted pathogen, malaria will transfer to the liver and infect blood cells for proliferation and
transmission. However, in G6PD-deficient patient, the blood cell has less antioxidant response
and cannot tolerate malaria infection, which might reduce the production and transmission of
malaria in patients’ blood. As for the KSHV infection, the malaria and KSHV co-infected blood
cells cannot tolerate strong oxidative stress and will undergo apoptosis, showing a reduced
malaria transmission. However, those two hypotheses are hard to exam in animal models for
human parasites’ co-infection. Using epidemiologic methods to study the co-infection will be
expensive. KSHV also has high prevalence in Mediterranean area, where prevalence of
inheriting G6PD-deficiency is high, although intrinsic relevance is not clear yet. With the
development of system biology, more omics data will be collected and analyzed with advanced
analysis models and tools, so the physiology of disease and hypothesis will be tested in a more
effective way.
A clinical report published recently mentioned that G6PD-deficient patients has more
severe symptoms, such as blood oxygen level and hospitalization time, comparing to normal
patients(Youssef et al., 2021). This correlated with the critical role of G6PD in blood cells which
93

transport oxygen from lung to different organs, suggesting testing G6PD deficiency before
treatment is important for COVID19 patients.
Most post-translational modification is a transient process at a certain time, making it
hard to recapture. PTMs that requires high energy to reverse, such as deamidation, are stable
and can be identified by MS. But rare and unstable PTMs like glycosylation and phosphor-
ribosylation are hard to detect. Therefore, methods to capture the transient interaction will be a
breakthrough for future PTM research.
Recently, spatial genomics have been developed to provide both spatial information of
gene expression. As for the protein level spatial information, FLIM could provide spatial
information as FLIM detect the fluorescent-labeled protein in live cells. As shown in Fig. 2-2F,
the glycolytic core aggregates near the nucleus at 12 hpi were detected by the autofluorescence
product, NAD(P)H. With GFP-labeled metabolic enzymes, we can get the localization of
metabolic enzymes and metabolic core.
In the deamidation proteomics results, we also found that other biological processes,
such as the ribosomal process, protein processing in ER, are regulated by deamidation. A
cluster of protein also highlighted the cornification protein, KRT, which is keratinocyte specific,
suggesting the cell death process might be regulated by deamidation as well. In general, more
mechanism research of deamidation needs to be done as deamidation might be an indicator of
regulation. In addition, the irreversible deamidation could impact other modification due to the
structural changes of the protein, making the mechanism more complicated.

Perspective
As for multi-omics research, developing system methods to identify targets will be in high
needs. Current multi-omics data includes genomics (RNA-seq, scRNA-seq, CRISPR-sgRNA
screening), proteomics (quantitative & PTM), protein-interaction database, and metabolomics.
Patients’ data are more complicated than molecular level data, including blood testing, hormone
94

levels, age, weights, genders, and disease phases. To integrate the multi-omics research for
biomarker identification, here I propose a possible solution includes three steps. The first one is
the interpretation of molecular level results by constructing a multi-layer network including multi-
omics data. The second step is calculating the probability of gene/protein as a hub gene/protein.
The third step is mathematically computing the association between protein and disease.
With the experience of studying KSHV, finding new oncogenic viruses is critical for
cancer prevention and therapy. Current oncogenic virus includes hepatitis B virus (HBV),
hepatitis C virus (HCV), human papillomavirus (HPV), EBV, KSHV, Merkel cell polyomavirus
(MCPyV), and HTLV-1.(Luo and Ou, 2015) The number of oncogenic virus would be
underestimated. With the development of personalized medicine, TCGA database has over 2.5
petabytes of genomic, epigenomic, transcriptomic, and proteomic data, including 20,000 primary
cancer and matched normal samples spanning 33 cancer types. The tremendous amount of the
data could help identify new oncogenic virus. My potential solution is using the unmapped
mRNA reads from the raw RNA-sequencing and DNA-sequencing file, which is 1~5 % of the
whole mRNA reads. Then use the unmapped read to assemble the de novo virus genome and
calculate the virus occurrence and disease relationship epidemiology analysis.
In general, identifying novel pathogens and developing multi-omics methods are critical
for future medical biology research. With the COVID19 pandemic, the public has realized how a
small pathogen impact the whole society. Methods to foresee and prevent such global health
crisis is needed.

95

References
Antman, K., and Chang, Y. (2009). Kaposi’s Sarcoma.
Aon, M.A., Bernier, M., Mitchell, S.J., Germanio, C.D., Mattison, J.A., Ehrlich, M.R., Colman,
R.J., Anderson, R.M., and Cabo, R. de (2020). Untangling Determinants of Enhanced Health
and Lifespan through a Multi-omics Approach in Mice. Cell Metab. 32, 100-116.e4.
AuCoin, D.P., Colletti, K.S., Xu, Y., Cei, S.A., and Pari, G.S. (2002). Kaposi’s Sarcoma-
Associated Herpesvirus (Human Herpesvirus 8) Contains Two Functional Lytic Origins of DNA
Replication. J. Virol. 76, 7890–7896.
AuCoin, D.P., Colletti, K.S., Cei, S.A., Papousková, I., Tarrant, M., and Pari, G.S. (2004).
Amplification of the Kaposi’s sarcoma-associated herpesvirus/human herpesvirus 8 lytic origin
of DNA replication is dependent upon a cis-acting AT-rich region and an ORF50 response
element and the trans-acting factors ORF50 (K-Rta) and K8 (K-bZIP). Virology 318, 542–555.
Avey, D., Brewers, B., and Zhu, F. (2015). Recent advances in the study of Kaposi’s sarcoma-
associated herpesvirus replication and pathogenesis. Virol. Sin. 30, 130–145.
Baslan, T., and Hicks, J. (2017). Unravelling biology and shifting paradigms in cancer with
single-cell sequencing. Nat. Rev. Cancer 17, 557–569.
Bechtel, J.T., Liang, Y., Hvidding, J., and Ganem, D. (2003). Host Range of Kaposi’s Sarcoma-
Associated Herpesvirus in Cultured Cells. J. Virol. 77, 6474–6481.
Bellare, P., and Ganem, D. (2009). Regulation of KSHV Lytic Switch Protein Expression by a
Virus-Encoded MicroRNA: An Evolutionary Adaptation that Fine-Tunes Lytic Reactivation. Cell
Host Microbe 6, 570–575.
Bhatt, A.P., and Damania, B. (2012). AKTivation of PI3K/AKT/mTOR signaling pathway by
KSHV. Front. Immunol. 3, 401.
Bolaños, J.P., and Almeida, A. (2010). The pentose-phosphate pathway in neuronal survival
against nitrosative stress. IUBMB Life 62, 14–18.
Bolaños, J.P., Almeida, A., and Moncada, S. (2010). Glycolysis: a bioenergetic or a survival
pathway? Trends Biochem. Sci. 35, 145–149.
Brand, A., Singer, K., Koehl, G.E., Kolitzus, M., Schoenhammer, G., Thiel, A., Matos, C., Bruss,
C., Klobuch, S., Peter, K., et al. (2016). LDHA-Associated Lactic Acid Production Blunts Tumor
Immunosurveillance by T and NK Cells. Cell Metab. 24, 657–671.
Brown, H.J., Song, M.J., Deng, H., Wu, T.-T., Cheng, G., and Sun, R. (2003). NF-κB Inhibits
Gammaherpesvirus Lytic Replication. J. Virol. 77, 8532–8540.
Brulois, K., Wong, L.-Y., Lee, H.-R., Sivadas, P., Ensser, A., Feng, P., Gao, S.-J., Toth, Z., and
Jung, J.U. (2015). Association of Kaposi’s Sarcoma-Associated Herpesvirus ORF31 with
ORF34 and ORF24 Is Critical for Late Gene Expression. J. Virol. 89, 6148–6154.
96

Bu, W., Carroll, K.D., Palmeri, D., and Lukac, D.M. (2007). Kaposi’s sarcoma-associated
herpesvirus/human herpesvirus 8 ORF50/Rta lytic switch protein functions as a tetramer. J.
Virol. 81, 5788–5806.
Bu, W., Palmeri, D., Krishnan, R., Marin, R., Aris, V.M., Soteropoulos, P., and Lukac, D.M.
(2008). Identification of Direct Transcriptional Targets of the Kaposi’s Sarcoma-Associated
Herpesvirus Rta Lytic Switch Protein by Conditional Nuclear Localization. J. Virol. 82, 10709–
10723.
Burns, J.S., and Manda, G. (2017). Metabolic Pathways of the Warburg Effect in Health and
Disease: Perspectives of Choice, Chain or Chance. Int. J. Mol. Sci. 18, 2755.
Butler, A., Hoffman, P., Smibert, P., Papalexi, E., and Satija, R. (2018). Integrating single-cell
transcriptomic data across different conditions, technologies, and species. Nat. Biotechnol. 36,
411–420.
Cai, Q., Lan, K., Verma, S.C., Si, H., Lin, D., and Robertson, E.S. (2006). Kaposi’s sarcoma-
associated herpesvirus latent protein LANA interacts with HIF-1 alpha to upregulate RTA
expression during hypoxia: Latency control under low oxygen conditions. J. Virol. 80, 7965–
7975.
Cai, Q., Cai, S., Zhu, C., Verma, S.C., Choi, J.-Y., and Robertson, E.S. (2013). A unique
SUMO-2-interacting motif within LANA is essential for KSHV latency. PLoS Pathog. 9,
e1003750.
Cambiaghi, A., Ferrario, M., and Masseroli, M. (2017). Analysis of metabolomic data: tools,
current strategies and future challenges for omics data integration. Brief. Bioinform. 18, 498–
510.
Candido, E.P., Reeves, R., and Davie, J.R. (1978). Sodium butyrate inhibits histone
deacetylation in cultured cells. Cell 14, 105–113.
Cannon, M.L., and Cesarman, E. (2004). The KSHV G protein-coupled receptor signals via
multiple pathways to induce transcription factor activation in primary effusion lymphoma cells.
Oncogene 23, 514–523.
Carroll, K.D., Khadim, F., Spadavecchia, S., Palmeri, D., and Lukac, D.M. (2007). Direct
interactions of Kaposi’s sarcoma-associated herpesvirus/human herpesvirus 8 ORF50/Rta
protein with the cellular protein octamer-1 and DNA are critical for specifying transactivation of a
delayed-early promoter and stimulating viral reactivation. J. Virol. 81, 8451–8467.
Cattelan, A.M., Calabrò, M.L., Gasperini, P., Aversa, S.M., Zanchetta, M., Meneghetti, F., De
Rossi, A., and Chieco-Bianchi, L. (2001). Acquired immunodeficiency syndrome-related
Kaposi’s sarcoma regression after highly active antiretroviral therapy: biologic correlates of
clinical outcome. J. Natl. Cancer Inst. Monogr. 44–49.
Cesarman, E. (2014). Gammaherpesviruses and lymphoproliferative disorders. Annu. Rev.
Pathol. 9, 349–372.
97

Chandriani, S., and Ganem, D. (2010). Array-based transcript profiling and limiting-dilution
reverse transcription-PCR analysis identify additional latent genes in Kaposi’s sarcoma-
associated herpesvirus. J. Virol. 84, 5565–5573.
Chang, P.-J., and Miller, G. (2004). Autoregulation of DNA binding and protein stability of
Kaposi’s sarcoma-associated herpesvirus ORF50 protein. J. Virol. 78, 10657–10673.
Chang, H., Dittmer, D.P., Chul, S.-Y., Hong, Y., and Jung, J.U. (2005a). Role of Notch Signal
Transduction in Kaposi’s Sarcoma-Associated Herpesvirus Gene Expression. J. Virol. 79,
14371–14382.
Chang, J., Renne, R., Dittmer, D., and Ganem, D. (2000). Inflammatory cytokines and the
reactivation of Kaposi’s sarcoma-associated herpesvirus lytic replication. Virology 266, 17–25.
Chang, M., Brown, H.J., Collado-Hidalgo, A., Arevalo, J.M., Galic, Z., Symensma, T.L., Tanaka,
L., Deng, H., Zack, J.A., Sun, R., et al. (2005b). beta-Adrenoreceptors reactivate Kaposi’s
sarcoma-associated herpesvirus lytic replication via PKA-dependent control of viral RTA. J.
Virol. 79, 13538–13547.
Chang, P.-C., Fitzgerald, L.D., Van Geelen, A., Izumiya, Y., Ellison, T.J., Wang, D.-H., Ann,
D.K., Luciw, P.A., and Kung, H.-J. (2009). Kruppel-associated box domain-associated protein-1
as a latency regulator for Kaposi’s sarcoma-associated herpesvirus and its modulation by the
viral protein kinase. Cancer Res. 69, 5681–5689.
Chang, P.-J., Shedd, D., and Miller, G. (2008). A mobile functional region of Kaposi’s sarcoma-
associated herpesvirus ORF50 protein independently regulates DNA binding and protein
abundance. J. Virol. 82, 9700–9716.
Chang, Y., Cesarman, E., Pessin, M.S., Lee, F., Culpepper, J., Knowles, D.M., and Moore, P.S.
(1994). Identification of herpesvirus-like DNA sequences in AIDS-associated Kaposi’s sarcoma.
Science 266, 1865–1869.
Chen, J., Ueda, K., Sakakibara, S., Okuno, T., Parravicini, C., Corbellino, M., and Yamanishi, K.
(2001). Activation of latent Kaposi’s sarcoma-associated herpesvirus by demethylation of the
promoter of the lytic transactivator. Proc. Natl. Acad. Sci. U. S. A. 98, 4119–4124.
Chen, M., Sun, F., Han, L., and Qu, Z. (2016). Kaposi’s sarcoma herpesvirus (KSHV) microRNA
K12-1 functions as an oncogene by activating NF-κB/IL-6/STAT3 signaling. Oncotarget 7,
33363–33373.
Cheong, W.-C., Park, J.-H., Kang, H.-R., and Song, M.J. (2015). Downregulation of Poly(ADP-
Ribose) Polymerase 1 by a Viral Processivity Factor Facilitates Lytic Replication of
Gammaherpesvirus. J. Virol. 89, 9676–9682.
Cohen, A., Brodie, C., and Sarid, R. (2006). An essential role of ERK signalling in TPA-induced
reactivation of Kaposi’s sarcoma-associated herpesvirus. J. Gen. Virol. 87, 795–802.
Cosentino, C., Grieco, D., and Costanzo, V. (2011). ATM activates the pentose phosphate
pathway promoting anti-oxidant defence and DNA repair. EMBO J. 30, 546–555.
Couzin-Frankel, J. (2013). Cancer Immunotherapy. Science 342, 1432–1433.
98

Cui, J., Yao, Q., Li, S., Ding, X., Lu, Q., Mao, H., Liu, L., Zheng, N., Chen, S., and Shao, F.
(2010). Glutamine Deamidation and Dysfunction of Ubiquitin/NEDD8 Induced by a Bacterial
Effector Family. Science 329, 1215–1218.
Dai, W., Jia, Q., Bortz, E., Shah, S., Liu, J., Atanasov, I., Li, X., Taylor, K.A., Sun, R., and Zhou,
Z.H. (2008). Unique structures in a tumor herpesvirus revealed by cryo-electron tomography
and microscopy. J. Struct. Biol. 161, 428–438.
Dai, X., Gong, D., Lim, H., Jih, J., Wu, T.-T., Sun, R., and Zhou, Z.H. (2018). Structure and
mutagenesis reveal essential capsid protein interactions for KSHV replication. Nature 553, 521–
525.
Dalton-Griffin, L., Wilson, S.J., and Kellam, P. (2009). X-box binding protein 1 contributes to
induction of the Kaposi’s sarcoma-associated herpesvirus lytic cycle under hypoxic conditions.
J. Virol. 83, 7202–7209.
Davis, D.A., Rinderknecht, A.S., Zoeteweij, J.P., Aoki, Y., Read-Connole, E.L., Tosato, G.,
Blauvelt, A., and Yarchoan, R. (2001). Hypoxia induces lytic replication of Kaposi sarcoma–
associated herpesvirus. Blood 97, 3244–3250.
Delfarah, A., Parrish, S., Junge, J.A., Yang, J., Seo, F., Li, S., Mac, J., Wang, P., Fraser, S.E.,
and Graham, N.A. (2019). Inhibition of nucleotide synthesis promotes replicative senescence of
human mammary epithelial cells. J. Biol. Chem. 294, 10564–10578.
Di Bartolo, D.L., Hyjek, E., Keller, S., Guasparri, I., Deng, H., Sun, R., Chadburn, A., Knowles,
D.M., and Cesarman, E. (2009). Role of defective Oct-2 and OCA-B expression in
immunoglobulin production and Kaposi’s sarcoma-associated herpesvirus lytic reactivation in
primary effusion lymphoma. J. Virol. 83, 4308–4315.
Ehrlich, E.S., Chmura, J.C., Smith, J.C., Kalu, N.N., and Hayward, G.S. (2014). KSHV RTA
abolishes NFκB responsive gene expression during lytic reactivation by targeting vFLIP for
degradation via the proteasome. PloS One 9, e91359.
Fang, Z., Jiang, C., Feng, Y., Chen, R., Lin, X., Zhang, Z., Han, L., Chen, X., Li, H., Guo, Y., et
al. (2016). Effects of G6PD activity inhibition on the viability, ROS generation and mechanical
properties of cervical cancer cells. Biochim. Biophys. Acta BBA - Mol. Cell Res. 1863, 2245–
2254.
Feng, P., Moses, A., and Früh, K. (2013). Evasion of adaptive and innate immune response
mechanisms by γ-herpesviruses. Curr. Opin. Virol. 3, 285–295.
FILOSA, S., FICO, A., PAGLIALUNGA, F., BALESTRIERI, M., CROOKE, A., VERDE, P.,
ABRESCIA, P., BAUTISTA, J.M., and MARTINI, G. (2003). Failure to increase glucose
consumption through the pentose-phosphate pathway results in the death of glucose-6-
phosphate dehydrogenase gene-deleted mouse embryonic stem cells subjected to oxidative
stress. Biochem. J. 370, 935–943.
Fleming, J.C., Woo, J., Moutasim, K., Mellone, M., Frampton, S.J., Mead, A., Ahmed, W.,
Wood, O., Robinson, H., Ward, M., et al. (2019). HPV, tumour metabolism and novel target
identification in head and neck squamous cell carcinoma. Br. J. Cancer 120, 356–367.
99

Földi, M., Stickeler, E., Bau, L., Kretz, O., Watermann, D., Gitsch, G., Kayser, G., zur Hausen,
A., and Coy, J.F. (2007). Transketolase protein TKTL1 overexpression: A potential biomarker
and therapeutic target in breast cancer. Oncol. Rep. 17, 841–845.
Fujii, T., Taguchi, H., Katano, H., Mori, S., Nakamura, T., Nojiri, N., Nakajima, K., Tadokoro, K.,
Juji, T., and Iwamoto, A. (1999). Seroprevalence of human herpesvirus 8 in human
immunodeficiency virus 1-positive and human immunodeficiency virus 1-negative populations in
Japan. J. Med. Virol. 57, 159–162.
Gao, S.-J., Kingsley, L., Li, M., Zheng, W., Parravicini, C., Ziegler, J., Newton, R., Rinaldo, C.R.,
Saah, A., Phair, J., et al. (1996). KSHV antibodies among Americans, Italians and Ugandans
with and without Kaposi’s sarcoma. Nat. Med. 2, 925–928.
Geschwind, J.-F.H., Ko, Y.H., Torbenson, M.S., Magee, C., and Pedersen, P.L. (2002). Novel
Therapy for Liver Cancer: Direct Intraarterial Injection of a Potent Inhibitor of ATP Production.
Cancer Res. 62, 3909–3913.
González-Molleda, L., Wang, Y., and Yuan, Y. (2012). Potent antiviral activity of topoisomerase
I and II inhibitors against Kaposi’s sarcoma-associated herpesvirus. Antimicrob. Agents
Chemother. 56, 893–902.
Gottwein, E., Mukherjee, N., Sachse, C., Frenzel, C., Majoros, W.H., Chi, J.-T.A., Braich, R.,
Manoharan, M., Soutschek, J., Ohler, U., et al. (2007). A viral microRNA functions as an
orthologue of cellular miR-155. Nature 450, 1096–1099.
Gottwein, E., Corcoran, D.L., Mukherjee, N., Skalsky, R.L., Hafner, M., Nusbaum, J.D.,
Shamulailatpam, P., Love, C.L., Dave, S.S., Tuschl, T., et al. (2011). Viral microRNA targetome
of KSHV-infected primary effusion lymphoma cell lines. Cell Host Microbe 10, 515–526.
Gould, F., Harrison, S.M., Hewitt, E.W., and Whitehouse, A. (2009). Kaposi’s sarcoma-
associated herpesvirus RTA promotes degradation of the Hey1 repressor protein through the
ubiquitin proteasome pathway. J. Virol. 83, 6727–6738.
Gradoville, L., Gerlach, J., Grogan, E., Shedd, D., Nikiforow, S., Metroka, C., and Miller, G.
(2000). Kaposi’s sarcoma-associated herpesvirus open reading frame 50/Rta protein activates
the entire viral lytic cycle in the HH-B2 primary effusion lymphoma cell line. J. Virol. 74, 6207–
6212.
Gregory, S.M., West, J.A., Dillon, P.J., Hilscher, C., Dittmer, D.P., and Damania, B. (2009). Toll-
like receptor signaling controls reactivation of KSHV from latency. Proc. Natl. Acad. Sci. U. S. A.
106, 11725–11730.
Grundhoff, A., and Ganem, D. (2004). Inefficient establishment of KSHV latency suggests an
additional role for continued lytic replication in Kaposi sarcoma pathogenesis. J. Clin. Invest.
113, 124–136.
Guasparri, I., Keller, S.A., and Cesarman, E. (2004). KSHV vFLIP is essential for the survival of
infected lymphoma cells. J. Exp. Med. 199, 993–1003.
100

Gui, S., Yang, L., Li, J., Luo, J., Xu, X., Yuan, J., Chen, L., Li, W., Yang, X., Wu, S., et al.
(2020). ZEAMAP, a Comprehensive Database Adapted to the Maize Multi-Omics Era. IScience
23.
Guihot, A., Dupin, N., Marcelin, A.-G., Gorin, I., Bedin, A.-S., Bossi, P., Galicier, L.,
Oksenhendler, E., Autran, B., and Carcelain, G. (2006). Low T Cell Responses to Human
Herpesvirus 8 in Patients with AIDS-Related and Classic Kaposi Sarcoma. J. Infect. Dis. 194,
1078–1088.
Guito, J., Gavina, A., Palmeri, D., and Lukac, D.M. (2014). The cellular peptidyl-prolyl cis/trans
isomerase Pin1 regulates reactivation of Kaposi’s sarcoma-associated herpesvirus from latency.
J. Virol. 88, 547–558.
Günther, T., and Grundhoff, A. (2010). The epigenetic landscape of latent Kaposi sarcoma-
associated herpesvirus genomes. PLoS Pathog. 6, e1000935.
Guo, H.-G., Sadowska, M., Reid, W., Tschachler, E., Hayward, G., and Reitz, M. (2003).
Kaposi’s sarcoma-like tumors in a human herpesvirus 8 ORF74 transgenic mouse. J. Virol. 77,
2631–2639.
Guo, H.-G., Pati, S., Sadowska, M., Charurat, M., and Reitz, M. (2004). Tumorigenesis by
human herpesvirus 8 vGPCR is accelerated by human immunodeficiency virus type 1 Tat. J.
Virol. 78, 9336–9342.
Gwack, Y., Nakamura, H., Lee, S.H., Souvlis, J., Yustein, J.T., Gygi, S., Kung, H.-J., and Jung,
J.U. (2003a). Poly(ADP-Ribose) Polymerase 1 and Ste20-Like Kinase hKFC Act as
Transcriptional Repressors for Gamma-2 Herpesvirus Lytic Replication. Mol. Cell. Biol. 23,
8282–8294.
Gwack, Y., Baek, H.J., Nakamura, H., Lee, S.H., Meisterernst, M., Roeder, R.G., and Jung, J.U.
(2003b). Principal role of TRAP/mediator and SWI/SNF complexes in Kaposi’s sarcoma-
associated herpesvirus RTA-mediated lytic reactivation. Mol. Cell. Biol. 23, 2055–2067.
Hains, P.G., and Truscott, R.J.W. (2010). Age-Dependent Deamidation of Lifelong Proteins in
the Human Lens. Invest. Ophthalmol. Vis. Sci. 51, 3107–3114.
Haque, M., Wang, V., Davis, D.A., Zheng, Z.-M., and Yarchoan, R. (2006). Genetic organization
and hypoxic activation of the Kaposi’s sarcoma-associated herpesvirus ORF34-37 gene cluster.
J. Virol. 80, 7037–7051.
Harris, S.A., and Harris, E.A. (2018). Molecular Mechanisms for Herpes Simplex Virus Type 1
Pathogenesis in Alzheimer’s Disease. Front. Aging Neurosci. 10.
Harrison, S.M., and Whitehouse, A. (2008). Kaposi’s sarcoma-associated herpesvirus (KSHV)
Rta and cellular HMGB1 proteins synergistically transactivate the KSHV ORF50 promoter.
FEBS Lett. 582, 3080–3084.
He, S., Zhao, J., Song, S., He, X., Minassian, A., Zhou, Y., Zhang, J., Brulois, K., Wang, Y.,
Cabo, J., et al. (2015). Viral Pseudo Enzymes Activate RIG-I via Deamidation to Evade
Cytokine Production. Mol. Cell 58, 134–146.
101

Heinrich, P., Kohler, C., Ellmann, L., Kuerner, P., Spang, R., Oefner, P.J., and Dettmer, K.
(2018). Correcting for natural isotope abundance and tracer impurity in MS-, MS/MS- and high-
resolution-multiple-tracer-data from stable isotope labeling experiments with IsoCorrectoR. Sci.
Rep. 8, 17910.
Heinrich, P.C., Morris, H.P., and Weber, G. (1976). Behavior of Transaldolase (EC 2.2.1.2) and
Transketolase (EC 2.2.1.1) Activities in Normal, Neoplastic, Differentiating, and Regenerating
Liver. Cancer Res. 36, 3189–3197.
Hoffman, N.J., Parker, B.L., Chaudhuri, R., Fisher-Wellman, K.H., Kleinert, M., Humphrey, S.J.,
Yang, P., Holliday, M., Trefely, S., Fazakerley, D.J., et al. (2015). Global Phosphoproteomic
Analysis of Human Skeletal Muscle Reveals a Network of Exercise-Regulated Kinases and
AMPK Substrates. Cell Metab. 22, 922–935.
Hu, T., Zhang, C., Tang, Q., Su, Y., Li, B., Chen, L., Zhang, Z., Cai, T., and Zhu, Y. (2013).
Variant G6PD levels promote tumor cell proliferation or apoptosis via the STAT3/5 pathway in
the human melanoma xenograft mouse model. BMC Cancer 13, 251.
Huang, L.-M., Chao, M.-F., Chen, M.-Y., Shih, H., Chiang, Y.-P., Chuang, C.-Y., and Lee, C.-Y.
(2001). Reciprocal Regulatory Interaction between Human Herpesvirus 8 and Human
Immunodeficiency Virus Type 1 *. J. Biol. Chem. 276, 13427–13432.
Huck, J.H.J., Verhoeven, N.M., Struys, E.A., Salomons, G.S., Jakobs, C., and van der Knaap,
M.S. (2004). Ribose-5-Phosphate Isomerase Deficiency: New Inborn Error in the Pentose
Phosphate Pathway Associated with a Slowly Progressive Leukoencephalopathy. Am. J. Hum.
Genet. 74, 745–751.
Hyun, T.S., Subramanian, C., Cotter, M.A., Thomas, R.A., and Robertson, E.S. (2001). Latency-
associated nuclear antigen encoded by Kaposi’s sarcoma-associated herpesvirus interacts with
Tat and activates the long terminal repeat of human immunodeficiency virus type 1 in human
cells. J. Virol. 75, 8761–8771.
Izumiya, Y., Izumiya, C., Hsia, D., Ellison, T.J., Luciw, P.A., and Kung, H.-J. (2009). NF-kappaB
serves as a cellular sensor of Kaposi’s sarcoma-associated herpesvirus latency and negatively
regulates K-Rta by antagonizing the RBP-Jkappa coactivator. J. Virol. 83, 4435–4446.
Izumiya, Y., Kobayashi, K., Kim, K.Y., Pochampalli, M., Izumiya, C., Shevchenko, B., Wang, D.-
H., Huerta, S.B., Martinez, A., Campbell, M., et al. (2013). Kaposi’s sarcoma-associated
herpesvirus K-Rta exhibits SUMO-targeting ubiquitin ligase (STUbL) like activity and is essential
for viral reactivation. PLoS Pathog. 9, e1003506.
Jaber, T., and Yuan, Y. (2013). A virally encoded small peptide regulates RTA stability and
facilitates Kaposi’s sarcoma-associated herpesvirus lytic replication. J. Virol. 87, 3461–3470.
Jalloh, I., Carpenter, K.L.H., Grice, P., Howe, D.J., Mason, A., Gallagher, C.N., Helmy, A.,
Murphy, M.P., Menon, D.K., Carpenter, T.A., et al. (2015). Glycolysis and the Pentose
Phosphate Pathway after Human Traumatic Brain Injury: Microdialysis Studies Using 1,2-13C2
Glucose. J. Cereb. Blood Flow Metab. 35, 111–120.
102

Jiang, W., Oikonomou, P., and Tavazoie, S. (2020). Comprehensive Genome-wide
Perturbations via CRISPR Adaptation Reveal Complex Genetics of Antibiotic Sensitivity. Cell
180, 1002-1017.e31.
Jiang, Y., Xu, D., Zhao, Y., and Zhang, L. (2008). Mutual inhibition between Kaposi’s sarcoma-
associated herpesvirus and Epstein-Barr virus lytic replication initiators in dually-infected
primary effusion lymphoma. PloS One 3, e1569.
Jr, W.H., Sieczkowski, L., Sosa, C., Chan-a-Sue, S., Cai, J.P., Cabral, L., and Wood, C. (1997).
Activation of HHV-8 by HIV-1 tat. The Lancet 349, 774–775.
Karijolich, J., Zhao, Y., Peterson, B., Zhou, Q., and Glaunsinger, B. (2014). Kaposi’s sarcoma-
associated herpesvirus ORF45 mediates transcriptional activation of the HIV-1 long terminal
repeat via RSK2. J. Virol. 88, 7024–7035.
Kathagen, A., Schulte, A., Balcke, G., Phillips, H.S., Martens, T., Matschke, J., Günther, H.S.,
Soriano, R., Modrusan, Z., Sandmann, T., et al. (2013). Hypoxia and oxygenation induce a
metabolic switch between pentose phosphate pathway and glycolysis in glioma stem-like cells.
Acta Neuropathol. (Berl.) 126, 763–780.
Kathagen-Buhmann, A., Schulte, A., Weller, J., Holz, M., Herold-Mende, C., Glass, R., and
Lamszus, K. (2016). Glycolysis and the pentose phosphate pathway are differentially associated
with the dichotomous regulation of glioblastoma cell migration versus proliferation. Neuro-Oncol.
18, 1219–1229.
Kaul, R., Purushothaman, P., Uppal, T., and Verma, S.C. (2019). KSHV lytic proteins K-RTA
and K8 bind to cellular and viral chromatin to modulate gene expression. PLOS ONE 14,
e0215394.
Kim, W., Bennett, E.J., Huttlin, E.L., Guo, A., Li, J., Possemato, A., Sowa, M.E., Rad, R., Rush,
J., Comb, M.J., et al. (2011). Systematic and Quantitative Assessment of the Ubiquitin-Modified
Proteome. Mol. Cell 44, 325–340.
Kimball, L.E., Casper, C., Koelle, D.M., Morrow, R., Corey, L., and Vieira, J. (2004). Reduced
Levels of Neutralizing Antibodies to Kaposi Sarcoma–Associated Herpesvirus in Persons with a
History of Kaposi Sarcoma. J. Infect. Dis. 189, 2016–2022.
King, C.A., Li, X., Barbachano-Guerrero, A., and Bhaduri-McIntosh, S. (2015). STAT3
Regulates Lytic Activation of Kaposi’s Sarcoma-Associated Herpesvirus. J. Virol. 89, 11347–
11355.
Konrad, A., Wies, E., Thurau, M., Marquardt, G., Naschberger, E., Hentschel, S., Jochmann, R.,
Schulz, T.F., Erfle, H., Brors, B., et al. (2009). A systems biology approach to identify the
combination effects of human herpesvirus 8 genes on NF-kappaB activation. J. Virol. 83, 2563–
2574.
Kuehne, A., Emmert, H., Soehle, J., Winnefeld, M., Fischer, F., Wenck, H., Gallinat, S.,
Terstegen, L., Lucius, R., Hildebrand, J., et al. (2015). Acute Activation of Oxidative Pentose
Phosphate Pathway as First-Line Response to Oxidative Stress in Human Skin Cells. Mol. Cell
59, 359–371.
103

Kukhanova, M.K., Korovina, A.N., and Kochetkov, S.N. (2014). Human herpes simplex virus: life
cycle and development of inhibitors. Biochem. Biokhimiia 79, 1635–1652.
Lambert, M., Gannagé, M., Karras, A., Abel, M., Legendre, C., Kerob, D., Agbalika, F., Girard,
P.-M., Lebbe, C., and Caillat-Zucman, S. (2006). Differences in the frequency and function of
HHV8-specific CD8 T cells between asymptomatic HHV8 infection and Kaposi sarcoma. Blood
108, 3871–3880.
Lan, K., Kuppers, D.A., Verma, S.C., and Robertson, E.S. (2004). Kaposi’s sarcoma-associated
herpesvirus-encoded latency-associated nuclear antigen inhibits lytic replication by targeting
Rta: a potential mechanism for virus-mediated control of latency. J. Virol. 78, 6585–6594.
Lan, K., Kuppers, D.A., and Robertson, E.S. (2005a). Kaposi’s sarcoma-associated herpesvirus
reactivation is regulated by interaction of latency-associated nuclear antigen with recombination
signal sequence-binding protein Jkappa, the major downstream effector of the Notch signaling
pathway. J. Virol. 79, 3468–3478.
Lan, K., Kuppers, D.A., Verma, S.C., Sharma, N., Murakami, M., and Robertson, E.S. (2005b).
Induction of Kaposi’s sarcoma-associated herpesvirus latency-associated nuclear antigen by
the lytic transactivator RTA: a novel mechanism for establishment of latency. J. Virol. 79, 7453–
7465.
Lander, E.S., Linton, L.M., Birren, B., Nusbaum, C., Zody, M.C., Baldwin, J., Devon, K., Dewar,
K., Doyle, M., FitzHugh, W., et al. (2001). Initial sequencing and analysis of the human genome.
Nature 409, 860–921.
Lee, C.-P., Liu, P.-T., Kung, H.-N., Su, M.-T., Chua, H.-H., Chang, Y.-H., Chang, C.-W., Tsai,
C.-H., Liu, F.-T., and Chen, M.-R. (2012). The ESCRT machinery is recruited by the viral BFRF1
protein to the nucleus-associated membrane for the maturation of Epstein-Barr Virus. PLoS
Pathog. 8, e1002904.
Lei, X., Bai, Z., Ye, F., Xie, J., Kim, C.-G., Huang, Y., and Gao, S.-J. (2010). Regulation of NF-
kappaB inhibitor IkappaBalpha and viral replication by a KSHV microRNA. Nat. Cell Biol. 12,
193–199.
Lei, X., Zhu, Y., Jones, T., Bai, Z., Huang, Y., and Gao, S.-J. (2012). A Kaposi’s sarcoma-
associated herpesvirus microRNA and its variants target the transforming growth factor β
pathway to promote cell survival. J. Virol. 86, 11698–11711.
Lepone, L., Rappocciolo, G., Knowlton, E., Jais, M., Piazza, P., Jenkins, F.J., and Rinaldo, C.R.
(2010). Monofunctional and polyfunctional CD8+ T cell responses to human herpesvirus 8 lytic
and latency proteins. Clin. Vaccine Immunol. CVI 17, 1507–1516.
Li, D.-J., Verma, D., Mosbruger, T., and Swaminathan, S. (2014). CTCF and Rad21 act as host
cell restriction factors for Kaposi’s sarcoma-associated herpesvirus (KSHV) lytic replication by
modulating viral gene transcription. PLoS Pathog. 10, e1003880.
Li, H.-M., Yang, J.-G., Liu, Z.-J., Wang, W.-M., Yu, Z.-L., Ren, J.-G., Chen, G., Zhang, W., and
Jia, J. (2017). Blockage of glycolysis by targeting PFKFB3 suppresses tumor growth and
metastasis in head and neck squamous cell carcinoma. J. Exp. Clin. Cancer Res. 36, 7.
104

Li, Q., Zhou, F., Ye, F., and Gao, S.-J. (2008). Genetic disruption of KSHV major latent nuclear
antigen LANA enhances viral lytic transcriptional program. Virology 379, 234–244.
Li, Q., Qin, T., Bi, Z., Hong, H., Ding, L., Chen, J., Wu, W., Lin, X., Fu, W., Zheng, F., et al.
(2020). Rac1 activates non-oxidative pentose phosphate pathway to induce chemoresistance of
breast cancer. Nat. Commun. 11, 1456.
Li, X., Feng, J., and Sun, R. (2011). Oxidative stress induces reactivation of Kaposi’s sarcoma-
associated herpesvirus and death of primary effusion lymphoma cells. J. Virol. 85, 715–724.
Liang, Y., and Ganem, D. (2003). Lytic but not latent infection by Kaposi’s sarcoma-associated
herpesvirus requires host CSL protein, the mediator of Notch signaling. Proc. Natl. Acad. Sci. U.
S. A. 100, 8490–8495.
Liang, Y., Chang, J., Lynch, S.J., Lukac, D.M., and Ganem, D. (2002). The lytic switch protein of
KSHV activates gene expression via functional interaction with RBP-Jkappa (CSL), the target of
the Notch signaling pathway. Genes Dev. 16, 1977–1989.
Lin, H.-R., and Ganem, D. (2011). Viral microRNA target allows insight into the role of
translation in governing microRNA target accessibility. Proc. Natl. Acad. Sci. U. S. A. 108,
5148–5153.
Lin, C.L., Li, H., Wang, Y., Zhu, F.X., Kudchodkar, S., and Yuan, Y. (2003). Kaposi’s sarcoma-
associated herpesvirus lytic origin (ori-Lyt)-dependent DNA replication: identification of the ori-
Lyt and association of K8 bZip protein with the origin. J. Virol. 77, 5578–5588.
Liu, C., Okruzhnov, Y., Li, H., and Nicholas, J. (2001). Human herpesvirus 8 (HHV-8)-encoded
cytokines induce expression of and autocrine signaling by vascular endothelial growth factor
(VEGF) in HHV-8-infected primary-effusion lymphoma cell lines and mediate VEGF-
independent antiapoptotic effects. J. Virol. 75, 10933–10940.
Liu, X., Romero, I.L., Litchfield, L.M., Lengyel, E., and Locasale, J.W. (2016). Metformin Targets
Central Carbon Metabolism and Reveals Mitochondrial Requirements in Human Cancers. Cell
Metab. 24, 728–739.
Liu, X., Zhao, Y., Zhang, E., Yan, H., Lv, N., and Cai, Z. (2019). The synergistic effect of PFK15
with metformin exerts anti-myeloma activity via PFKFB3. Biochem. Biophys. Res. Commun.
515, 332–338.
Liu, Y., Yang, M., Deng, Y., Su, G., Enninful, A., Guo, C.C., Tebaldi, T., Zhang, D., Kim, D., Bai,
Z., et al. (2020). High-Spatial-Resolution Multi-Omics Sequencing via Deterministic Barcoding in
Tissue. Cell 183, 1665-1681.e18.
Löhr, G.W., and Waller, H.D. (1974). Glucose-6-phosphate Dehydrogenase. In Methods of
Enzymatic Analysis (Second Edition), H.U. Bergmeyer, ed. (Academic Press), pp. 636–643.
Lu, C.-C., Li, Z., Chu, C.-Y., Feng, J., Feng, J., Sun, R., and Rana, T.M. (2010a). MicroRNAs
encoded by Kaposi’s sarcoma-associated herpesvirus regulate viral life cycle. EMBO Rep. 11,
784–790.
105

Lu, F., Zhou, J., Wiedmer, A., Madden, K., Yuan, Y., and Lieberman, P.M. (2003). Chromatin
remodeling of the Kaposi’s sarcoma-associated herpesvirus ORF50 promoter correlates with
reactivation from latency. J. Virol. 77, 11425–11435.
Lu, F., Day, L., Gao, S.-J., and Lieberman, P.M. (2006). Acetylation of the latency-associated
nuclear antigen regulates repression of Kaposi’s sarcoma-associated herpesvirus lytic
transcription. J. Virol. 80, 5273–5282.
Lu, F., Stedman, W., Yousef, M., Renne, R., and Lieberman, P.M. (2010b). Epigenetic
regulation of Kaposi’s sarcoma-associated herpesvirus latency by virus-encoded microRNAs
that target Rta and the cellular Rbl2-DNMT pathway. J. Virol. 84, 2697–2706.
Lucarelli, G., Galleggiante, V., Rutigliano, M., Sanguedolce, F., Cagiano, S., Bufo, P., Lastilla,
G., Maiorano, E., Ribatti, D., Giglio, A., et al. (2015). Metabolomic profile of glycolysis and the
pentose phosphate pathway identifies the central role of glucose-6-phosphate dehydrogenase
in clear cell-renal cell carcinoma. Oncotarget 6, 13371–13386.
Lukac, D.M., Renne, R., Kirshner, J.R., and Ganem, D. (1998). Reactivation of Kaposi’s
sarcoma-associated herpesvirus infection from latency by expression of the ORF 50
transactivator, a homolog of the EBV R protein. Virology 252, 304–312.
Lukac, D.M., Kirshner, J.R., and Ganem, D. (1999). Transcriptional activation by the product of
open reading frame 50 of Kaposi’s sarcoma-associated herpesvirus is required for lytic viral
reactivation in B cells. J. Virol. 73, 9348–9361.
Lukac, D.M., Garibyan, L., Kirshner, J.R., Palmeri, D., and Ganem, D. (2001). DNA binding by
Kaposi’s sarcoma-associated herpesvirus lytic switch protein is necessary for transcriptional
activation of two viral delayed early promoters. J. Virol. 75, 6786–6799.
Luo, G.G., and Ou, J.J. (2015). Oncogenic viruses and cancer. Virol. Sin. 30, 83–84.
Luo, R., and Zhao, H. (2012). Protein quantitation using iTRAQ: Review on the sources of
variations and analysis of nonrandom missingness. Stat. Interface 5, 99–107.
Mallery, S.R., Pei, P., Landwehr, D.J., Clark, C.M., Bradburn, J.E., Ness, G.M., and Robertson,
F.M. (2004). Implications for oxidative and nitrative stress in the pathogenesis of AIDS-related
Kaposi’s sarcoma. Carcinogenesis 25, 597–603.
Manzano, M., Shamulailatpam, P., Raja, A.N., and Gottwein, E. (2013). Kaposi’s sarcoma-
associated herpesvirus encodes a mimic of cellular miR-23. J. Virol. 87, 11821–11830.
Mars, R.A.T., Yang, Y., Ward, T., Houtti, M., Priya, S., Lekatz, H.R., Tang, X., Sun, Z., Kalari,
K.R., Korem, T., et al. (2020). Longitudinal Multi-omics Reveals Subset-Specific Mechanisms
Underlying Irritable Bowel Syndrome. Cell 182, 1460-1473.e17.
Martin, D., Galisteo, R., Ji, Y., Montaner, S., and Gutkind, J.S. (2008). An NF-kappaB gene
expression signature contributes to Kaposi’s sarcoma virus vGPCR-induced direct and
paracrine neoplasia. Oncogene 27, 1844–1852.
Maschek, G., Savaraj, N., Priebe, W., Braunschweiger, P., Hamilton, K., Tidmarsh, G.F., Young,
L.R.D., and Lampidis, T.J. (2004). 2-Deoxy-d-glucose Increases the Efficacy of Adriamycin and
106

Paclitaxel in Human Osteosarcoma and Non-Small Cell Lung Cancers In Vivo. Cancer Res. 64,
31–34.
Massière, F., and Badet-Denisot, M.-A. (1998). The mechanism of glutamine-dependent
amidotransferases. Cell. Mol. Life Sci. 54, 205–222.
McGarry, T., Biniecka, M., Veale, D.J., and Fearon, U. (2018). Hypoxia, oxidative stress and
inflammation. Free Radic. Biol. Med. 125, 15–24.
Mele, L., Paino, F., Papaccio, F., Regad, T., Boocock, D., Stiuso, P., Lombardi, A., Liccardo, D.,
Aquino, G., Barbieri, A., et al. (2018). A new inhibitor of glucose-6-phosphate dehydrogenase
blocks pentose phosphate pathway and suppresses malignant proliferation and metastasis in
vivo. Cell Death Dis. 9, 1–12.
Merat, R., Amara, A., Lebbe, C., de The, H., Morel, P., and Saib, A. (2002). HIV-1 infection of
primary effusion lymphoma cell line triggers Kaposi’s sarcoma-associated herpesvirus (KSHV)
reactivation. Int. J. Cancer 97, 791–795.
Mercader, M., Taddeo, B., Panella, J.R., Chandran, B., Nickoloff, B.J., and Foreman, K.E.
(2000). Induction of HHV-8 lytic cycle replication by inflammatory cytokines produced by HIV-1-
infected T cells. Am. J. Pathol. 156, 1961–1971.
Merelli, I., Tordini, F., Drocco, M., Aldinucci, M., Liò, P., and Milanesi, L. (2015). Integrating
multi-omic features exploiting Chromosome Conformation Capture data. Front. Genet. 6.
Mesri, E.A., Cesarman, E., and Boshoff, C. (2010). Kaposi’s sarcoma herpesvirus/ Human
herpesvirus-8 (KSHV/HHV8), and the oncogenesis of Kaposi’s sarcoma. Nat. Rev. Cancer 10,
707–719.
Mettenleiter, T.C., Klupp, B.G., and Granzow, H. (2009). Herpesvirus assembly: An update.
Virus Res. 143, 222–234.
Metzker, M.L. (2010). Sequencing technologies - the next generation. Nat. Rev. Genet. 11, 31–
46.
Middeldorp, J.M., Brink, A.A.T.P., van den Brule, A.J.C., and Meijer, C.J.L.M. (2003).
Pathogenic roles for Epstein–Barr virus (EBV) gene products in EBV-associated proliferative
disorders. Crit. Rev. Oncol. Hematol. 45, 1–36.
Miller, G., Heston, L., Grogan, E., Gradoville, L., Rigsby, M., Sun, R., Shedd, D., Kushnaryov,
V.M., Grossberg, S., and Chang, Y. (1997). Selective switch between latency and lytic
replication of Kaposi’s sarcoma herpesvirus and Epstein-Barr virus in dually infected body cavity
lymphoma cells. J. Virol. 71, 314–324.
Min, J., and Katzenstein, D.A. (1999). Detection of Kaposi’s sarcoma-associated herpesvirus in
peripheral blood cells in human immunodeficiency virus infection: association with Kaposi’s
sarcoma, CD4 cell count, and HIV RNA levels. AIDS Res. Hum. Retroviruses 15, 51–55.
Mocroft, A., Kirk, O., Clumeck, N., Gargalianos-Kakolyris, P., Trocha, H., Chentsova, N.,
Antunes, F., Stellbrink, H.-J., Phillips, A.N., and Lundgren, J.D. (2004). The changing pattern of
Kaposi sarcoma in patients with HIV, 1994-2003: the EuroSIDA Study. Cancer 100, 2644–2654.
107

Montaner, S., Sodhi, A., Molinolo, A., Bugge, T.H., Sawai, E.T., He, Y., Li, Y., Ray, P.E., and
Gutkind, J.S. (2003). Endothelial infection with KSHV genes in vivo reveals that vGPCR initiates
Kaposi’s sarcomagenesis and can promote the tumorigenic potential of viral latent genes.
Cancer Cell 3, 23–36.
Nyholm, L., Koziol, A., Marcos, S., Botnen, A.B., Aizpurua, O., Gopalakrishnan, S., Limborg,
M.T., Gilbert, M.T.P., and Alberdi, A. (2020). Holo-Omics: Integrated Host-Microbiota Multi-
omics for Basic and Applied Biological Research. IScience 23.
Ozsolak, F., and Milos, P.M. (2011). RNA sequencing: advances, challenges and opportunities.
Nat. Rev. Genet. 12, 87–98.
Papalexi, E., and Satija, R. (2018). Single-cell RNA sequencing to explore immune cell
heterogeneity. Nat. Rev. Immunol. 18, 35–45.
Papp, B., Motlagh, N., Smindak, R.J., Jang, S.J., Sharma, A., Alonso, J.D., and Toth, Z. (2019).
Genome-Wide Identification of Direct RTA Targets Reveals Key Host Factors for Kaposi’s
Sarcoma-Associated Herpesvirus Lytic Reactivation. J. Virol. 93.
Pate, K.T., Stringari, C., Sprowl-Tanio, S., Wang, K., TeSlaa, T., Hoverter, N.P., McQuade,
M.M., Garner, C., Digman, M.A., Teitell, M.A., et al. (2014). Wnt signaling directs a metabolic
program of glycolysis and angiogenesis in colon cancer. EMBO J. 33, 1454–1473.
Pellet, C., Chevret, S., Blum, L., Gauvillé, C., Hurault, M., Blanchard, G., Agbalika, F., Lascoux,
C., Ponscarme, D., Morel, P., et al. (2001). Virologic and immunologic parameters that predict
clinical response of AIDS-associated Kaposi’s sarcoma to highly active antiretroviral therapy. J.
Invest. Dermatol. 117, 858–863.
Perl, A., Hanczko, R., Telarico, T., Oaks, Z., and Landas, S. (2011). Oxidative stress,
inflammation and carcinogenesis are controlled through the pentose phosphate pathway by
transaldolase. Trends Mol. Med. 17, 395–403.
Pertea, M., Kim, D., Pertea, G.M., Leek, J.T., and Salzberg, S.L. (2016). Transcript-level
expression analysis of RNA-seq experiments with HISAT, StringTie and Ballgown. Nat. Protoc.
11, 1650–1667.
Plaisance-Bonstaff, K., Choi, H.S., Beals, T., Krueger, B.J., Boss, I.W., Gay, L.A., Haecker, I.,
Hu, J., and Renne, R. (2014). KSHV miRNAs decrease expression of lytic genes in latently
infected PEL and endothelial cells by targeting host transcription factors. Viruses 6, 4005–4023.
Prakash, O., Tang, Z.-Y., Peng, X., Coleman, R., Gill, J., Farr, G., and Samaniego, F. (2002).
Tumorigenesis and aberrant signaling in transgenic mice expressing the human herpesvirus-8
K1 gene. J. Natl. Cancer Inst. 94, 926–935.
Purushothaman, P., Thakker, S., and Verma, S.C. (2015). Transcriptome Analysis of Kaposi’s
Sarcoma-Associated Herpesvirus during De Novo Primary Infection of Human B and
Endothelial Cells. J. Virol. 89, 3093–3111.
Ralser, M., Wamelink, M.M., Kowald, A., Gerisch, B., Heeren, G., Struys, E.A., Klipp, E.,
Jakobs, C., Breitenbach, M., Lehrach, H., et al. (2007). Dynamic rerouting of the carbohydrate
flux is key to counteracting oxidative stress. J. Biol. 6, 10.
108

Rao, X., Duan, X., Mao, W., Li, X., Li, Z., Li, Q., Zheng, Z., Xu, H., Chen, M., Wang, P.G., et al.
(2015). O-GlcNAcylation of G6PD promotes the pentose phosphate pathway and tumor growth.
Nat. Commun. 6, 8468.
Renne, R., Barry, C., Dittmer, D., Compitello, N., Brown, P.O., and Ganem, D. (2001).
Modulation of cellular and viral gene expression by the latency-associated nuclear antigen of
Kaposi’s sarcoma-associated herpesvirus. J. Virol. 75, 458–468.
Rey, G., Valekunja, U.K., Feeney, K.A., Wulund, L., Milev, N.B., Stangherlin, A., Ansel-
Bollepalli, L., Velagapudi, V., O’Neill, J.S., and Reddy, A.B. (2016). The Pentose Phosphate
Pathway Regulates the Circadian Clock. Cell Metab. 24, 462–473.
Rodriguez-Rodriguez, P., Fernandez, E., and Bolaños, J.P. (2013). Underestimation of the
Pentose–Phosphate Pathway in Intact Primary Neurons as Revealed by Metabolic Flux
Analysis. J. Cereb. Blood Flow Metab. 33, 1843–1845.
Sauer, U. (2006). Metabolic networks in motion: 13C-based flux analysis. Mol. Syst. Biol. 2, 62.
Schieber, M., and Chandel, N.S. (2014). ROS Function in Redox Signaling and Oxidative
Stress. Curr. Biol. CB 24, R453–R462.
Shannon, P., Markiel, A., Ozier, O., Baliga, N.S., Wang, J.T., Ramage, D., Amin, N.,
Schwikowski, B., and Ideker, T. (2003). Cytoscape: A Software Environment for Integrated
Models of Biomolecular Interaction Networks. Genome Res. 13, 2498–2504.
Silveira, W.A. da, Fazelinia, H., Rosenthal, S.B., Laiakis, E.C., Kim, M.S., Meydan, C., Kidane,
Y., Rathi, K.S., Smith, S.M., Stear, B., et al. (2020). Comprehensive Multi-omics Analysis
Reveals Mitochondrial Stress as a Central Biological Hub for Spaceflight Impact. Cell 183,
1185-1201.e20.
Simpson, G.R., Schulz, T.F., Whitby, D., Cook, P.M., Boshoff, C., Rainbow, L., Howard, M.R.,
Gao, S.-J., Bohenzky, R.A., Simmonds, P., et al. (1996). Prevalence of Kaposi’s sarcoma
associated herpesvirus infection measured by antibodies to recombinant capsid protein and
latent immunofluorescence antigen. The Lancet 348, 1133–1138.
Staiger, W.I., Coy, J.F., Grobholz, R., Hofheinz, R.-D., Lukan, N., Post, S., Schwarzbach, M.H.,
and Willeke, F. (2006). Expression of the mutated transketolase TKTL1, a molecular marker in
gastric cancer. Oncol. Rep. 16, 657–661.
Stern-Ginossar, N., Weisburd, B., Michalski, A., Le, V.T.K., Hein, M.Y., Huang, S.-X., Ma, M.,
Shen, B., Qian, S.-B., Hengel, H., et al. (2012). Decoding Human Cytomegalovirus. Science
338, 1088–1093.
Sychev, Z.E., Hu, A., DiMaio, T.A., Gitter, A., Camp, N.D., Noble, W.S., Wolf-Yadlin, A., and
Lagunoff, M. (2017). Integrated systems biology analysis of KSHV latent infection reveals viral
induction and reliance on peroxisome mediated lipid metabolism. PLoS Pathog. 13, e1006256.
Szklarczyk, D., Gable, A.L., Nastou, K.C., Lyon, D., Kirsch, R., Pyysalo, S., Doncheva, N.T.,
Legeay, M., Fang, T., Bork, P., et al. (2021). The STRING database in 2021: customizable
protein–protein networks, and functional characterization of user-uploaded gene/measurement
sets. Nucleic Acids Res. 49, D605–D612.
109

Tanaka, Y., and Chen, Z.J. (2012). STING specifies IRF3 phosphorylation by TBK1 in the
cytosolic DNA signaling pathway. Sci. Signal. 5, ra20.
Ter-Pogossian, M.M. (1983). Positron Emission Tomography (PET). In Diagnostic Imaging in
Medicine, R.C. Reba, D.J. Goodenough, and H.F. Davidson, eds. (Dordrecht: Springer
Netherlands), pp. 273–277.
Tirelli, U., and Bernardi, D. (2001). Impact of HAART on the clinical management of AIDS-
related cancers. Eur. J. Cancer 37, 1320–1324.
Valle-Casuso, J.C., Angin, M., Volant, S., Passaes, C., Monceaux, V., Mikhailova, A., Bourdic,
K., Avettand-Fenoel, V., Boufassa, F., Sitbon, M., et al. (2019). Cellular Metabolism Is a Major
Determinant of HIV-1 Reservoir Seeding in CD4+ T Cells and Offers an Opportunity to Tackle
Infection. Cell Metab. 29, 611-626.e5.
Venter, J.C., Adams, M.D., Myers, E.W., Li, P.W., Mural, R.J., Sutton, G.G., Smith, H.O.,
Yandell, M., Evans, C.A., Holt, R.A., et al. (2001). The sequence of the human genome.
Science 291, 1304–1351.
Vizán, P., Alcarraz‐ Vizán, G., Díaz‐ Moralli, S., Solovjeva, O.N., Frederiks, W.M., and Cascante,
M. (2009). Modulation of pentose phosphate pathway during cell cycle progression in human
colon adenocarcinoma cell line HT29. Int. J. Cancer 124, 2789–2796.
Völker, H.-U., Hagemann, C., Coy, J., Wittig, R., Sommer, S., Stojic, J., Haubitz, I., Vince, G.H.,
Kämmerer, U., and Monoranu, C.-M. (2008). Expression of Transketolase-like 1 and Activation
of Akt in Grade IV Glioblastomas Compared With Grades II and III Astrocytic Gliomas. Am. J.
Clin. Pathol. 130, 50–57.
Wang, C., Qu, J., Yan, S., Gao, Q., Hao, S., and Zhou, D. (2018a). PFK15, a PFKFB3
antagonist, inhibits autophagy and proliferation in rhabdomyosarcoma cells. Int. J. Mol. Med. 42,
359–367.
Wang, M., Li, A., Sekiya, M., Beckmann, N.D., Quan, X., Schrode, N., Fernando, M.B., Yu, A.,
Zhu, L., Cao, J., et al. (2021). Transformative Network Modeling of Multi-omics Data Reveals
Detailed Circuits, Key Regulators, and Potential Therapeutics for Alzheimer’s Disease. Neuron
109, 257-272.e14.
Wang, X., He, Y., Ye, Y., Zhao, X., Deng, S., He, G., Zhu, H., Xu, N., and Liang, S. (2018b).
SILAC–based quantitative MS approach for real-time recording protein-mediated cell-cell
interactions. Sci. Rep. 8, 8441.
Warburg, O. (1925). The Metabolism of Carcinoma Cells. J. Cancer Res. 9, 148–163.
Weiss, R.A., and Vogt, P.K. (2011). 100 years of Rous sarcoma virus. J. Exp. Med. 208, 2351–
2355.
Wickham, H. (2009). ggplot2: Elegant Graphics for Data Analysis (New York: Springer-Verlag).
Xu, I.M.-J., Lai, R.K.-H., Lin, S.-H., Tse, A.P.-W., Chiu, D.K.-C., Koh, H.-Y., Law, C.-T., Wong,
C.-M., Cai, Z., Wong, C.C.-L., et al. (2016). Transketolase counteracts oxidative stress to drive
cancer development. Proc. Natl. Acad. Sci. 113, E725–E734.
110

Xu, P., Duong, D.M., Seyfried, N.T., Cheng, D., Xie, Y., Robert, J., Rush, J., Hochstrasser, M.,
Finley, D., and Peng, J. (2009). Quantitative Proteomics Reveals the Function of
Unconventional Ubiquitin Chains in Proteasomal Degradation. Cell 137, 133–145.
Youssef, J.G., Zahiruddin, F., Youssef, G., Padmanabhan, S., Ensor, J., Pingali, S.R., Zu, Y.,
Sahay, S., and Iyer, S.P. (2021). G6PD deficiency and severity of COVID19 pneumonia and
acute respiratory distress syndrome: tip of the iceberg? Ann. Hematol. 100, 667–673.
Zampella, E.J., Bradley, E.L., and Pretlow, T.G. (1982). Glucose-6-phosphate dehydrogenase:
A possible clinical indicator for prostatic carcinoma. Cancer 49, 384–387.
Zerilli, M., Amato, M.C., Martorana, A., Cabibi, D., Coy, J.F., Cappello, F., Pompei, G., Russo,
A., Giordano, C., and Rodolico, V. (2008). Increased expression of transketolase-like-1 in
papillary thyroid carcinomas smaller than 1.5 cm in diameter is associated with lymph-node
metastases. Cancer 113, 936–944.
Zhang, J., Zhao, J., Xu, S., Li, J., He, S., Zeng, Y., Xie, L., Xie, N., Liu, T., Lee, K., et al. (2018).
Species-Specific Deamidation of cGAS by Herpes Simplex Virus UL37 Protein Facilitates Viral
Replication. Cell Host Microbe 24, 234-248.e5.
Zhang, R., Andersson, C.E., Savchenko, A., Skarina, T., Evdokimova, E., Beasley, S.,
Arrowsmith, C.H., Edwards, A.M., Joachimiak, A., and Mowbray, S.L. (2003). Structure of
Escherichia coli Ribose-5-Phosphate Isomerase: A Ubiquitous Enzyme of the Pentose
Phosphate Pathway and the Calvin Cycle. Structure 11, 31–42.
Zhang, S., Yue, J.-X., Yang, J.-H., Cai, P.-C., and Kong, W.-J. (2008). Overexpression of
transketolase protein TKTL1 is associated with occurrence and progression in nasopharyngeal
carcinoma: A potential therapeutic target in nasopharyngeal carcinoma. Cancer Biol. Ther. 7,
517–522.
Zhang, X.D., Deslandes, E., Villedieu, M., Poulain, L., Duval, M., Gauduchon, P., Schwartz, L.,
and Icard, P. (2006). Effect of 2-deoxy-D-glucose on various malignant cell lines in vitro.
Anticancer Res. 26, 3561–3566.
Zhao, J., Li, J., Xu, S., and Feng, P. (2016a). Emerging Roles of Protein Deamidation in Innate
Immune Signaling. J. Virol. 90, 4262–4268.
Zhao, J., Zeng, Y., Xu, S., Chen, J., Shen, G., Yu, C., Knipe, D., Yuan, W., Peng, J., Xu, W., et
al. (2016b). A Viral Deamidase Targets the Helicase Domain of RIG-I to Block RNA-induced
Activation. Cell Host Microbe 20, 770–784.
Zhao, J., Tian, M., Zhang, S., Delfarah, A., Gao, R., Rao, Y., Savas, A.C., Lu, A., Bubb, L., Lei,
X., et al. (2020). Deamidation Shunts RelA from Mediating Inflammation to Aerobic Glycolysis.
Cell Metab. 31, 937-955.e7.
Zhu, W., Ye, L., Zhang, J., Yu, P., Wang, H., Ye, Z., and Tian, J. (2016). PFK15, a Small
Molecule Inhibitor of PFKFB3, Induces Cell Cycle Arrest, Apoptosis and Inhibits Invasion in
Gastric Cancer. PLOS ONE 11, e0163768.
111

Zhu, Y., Li, T., Ramos da Silva, S., Lee, J.-J., Lu, C., Eoh, H., Jung, J.U., and Gao, S.-J. (2017).
A Critical Role of Glutamine and Asparagine γ-Nitrogen in Nucleotide Biosynthesis in Cancer
Cells Hijacked by an Oncogenic Virus. MBio 8.
(1989). Glucose-6-phosphate dehydrogenase deficiency. WHO Working Group. Bull. World
Health Organ. 67, 601–611.

Abstract (if available)

Abstract Deamidation is a post-translational modification of glutamine and asparagine, which is catalyzed by cellular glutamine aminotransferase (GATs). Deamidation regulates basic biological functions, such as the immune response and transcription activity. Recently, our group identified that deamidated RelA upregulates anaerobic glycolysis in cancers. Kaposi’s sarcoma-associated herpesvirus (KSHV) is an oncogenic virus that causes Kaposi’s Sarcoma (KS) and B cell lymphoma in immune-compromised individuals. Previous studies have shown that latent infection of KSHV reprograms host metabolism. However, the mechanism of how KSHV lytic replication reprograms host cells remains rudimentary. Here, we reported that KSHV activates CAD, a de novo pyrimidine synthesis enzyme, to deamidate G6PD, the pentose phosphate pathway (PPP) rate-limiting enzyme, to upregulate carbon metabolism to favor viral lytic replication. We applied both metabolomics and proteomics as a system approach to study the reprogramming process with biochemical validation for the regulatory mechanism. We found a robust parallel activation in central carbon metabolism and deamidation in key central carbon metabolism enzymes by analyzing proteomics and metabolomics analyses. We applied ¹³C-glucose labeled tracing for metabolic fluxes and chemical inhibitor screening to reveal that PPP is critical for KSHV replication. Then we identified G6PD is deamidated during KSHV infection, and G6PD’s deamidation is coupled with its increased activity during KSHV infection. Reconstructed G6PD deamidation mutants have higher activities than wild type and promote KSHV replication. We screened mammalian GATs and identified CAD deamidated G6PD during KSHV infection, where inhibition of CAD can impair G6PD deamidation and reduce KSHV replication. Our work uncovers the key role of protein deamidation directly regulating metabolic enzymes and metabolic reprogramming in general, expanding the functional repertoire of CAD as a responsive deamidase that regulates PPP to reprograming the central carbon metabolism. The system approach we developed will provide a new pipeline to research the metabolism reprogramming in virus-associated disease.

Linked assets

University of Southern California Dissertations and Theses

Conceptually similar

PDF

CAD-mediated metabolic reprogramming is regulated by innate immune kinases TBK1 and IKKβ

PDF

Study of protein deamidation in innate immune signaling

PDF

FoxO1 suppresses Kaposi's sarcoma-associated herpesvirus lytic replication and controls viral latency

PDF

Lymphatic cell environment promotes sustained KSHV lytic replication and viral maintenance

PDF

Viral and cellular N⁶-methyladenosine and N⁶,2'-O-dimethyladenosine epitranscriptomes in the Kaposi’s sarcoma‐associated herpesvirus life cycle

PDF

Regulation of acute KSHV infection by SIRT1

PDF

Selective innate immune activation by murine gamma-herpesvirus 68 (ɣHV68)

PDF

Mechanism of ethanol-mediated increase of SARS-COV-2 entry through GRP78 induction

PDF

Virus customization of host protein machinery for efficient propagation

PDF

KSHV-mediated modulation of immunoreceptor surface expression

PDF

Kaposi's sarcoma associated herpes-virus induces cellular proxi expression to modulate host gene expression that benefits viral infection and oncogenesis

PDF

Beclin 2-mediated autophagic degradation of the pathogenic proteins in neurodegenerative diseases

PDF

The effects of hepatitis C virus infection on host immune response and signaling pathways

PDF

Kaposi's sarcoma‐associated herpesvirus‐Bcl-2 mediated regulation of NM23-H2 is required for KSHV lytic replication

PDF

Expression of recombinant hepatitis B virus e antigen and analysis of its effect on macrophages

PDF

Adaptive metabolic strategies of Mycobacterium tuberculosis to combat stress from antibiotics and ROS

PDF

The analysis and modeling of signaling pathways induced by the interactions of the SARS-CoV-2 spike protein with cellular receptors

PDF

Characterization of three novel variants of the MAVS adaptor

Asset Metadata

Creator Tian, Mao (author)

Core Title Protein deamidation mediated metabolic reprogramming during KSHV lytic replication

School Keck School of Medicine

Degree Doctor of Philosophy

Degree Program Medical Biology

Degree Conferral Date 2021-05

Publication Date 11/05/2021

Defense Date 03/10/2021

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

Tag deamidation,KSHV,metabolism,OAI-PMH Harvest,post-translational modification

Format application/pdf (imt)

Language English

Contributor Electronically uploaded by the author (provenance)

Advisor Ou, Jing-Hsiung (committee chair), Feng, Pinghui (committee member), Xu, Jian (committee member), Yuan, Weiming (committee member)

Creator Email maot@usc.edu

Permanent Link (DOI) https://doi.org/10.25549/usctheses-oUC112720082

Unique identifier UC112720082

Identifier etd-TianMao-9601.pdf (filename)

Legacy Identifier etd-TianMao-9601

Document Type Dissertation

Format application/pdf (imt)

Rights Tian, Mao

Type texts

Source 20210507-wayne-usctheses-batch-835-shoaf (batch), University of Southern California (contributing entity), University of Southern California Dissertations and Theses (collection)

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

Repository Name University of Southern California Digital Library

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

Repository Email cisadmin@lib.usc.edu