University of Southern California Dissertations and Theses

Disrupted ER-to-Golgi trafficking underlies anti-HIV drugs and alcohol induced cellular stress and hepatic injury

(USC Thesis Other)

Disrupted ER-to-Golgi trafficking underlies anti-HIV drugs and alcohol induced cellular stress and hepatic injury

PDF

Download

Open document

Flip pages

Download a page range

Download transcript

Copy asset link

Request this asset

Transcript (if available)

Content

Disrupted ER-To-Golgi Trafficking Underlines Anti-HIV Drugs and
Alcohol Induced Cellular Stress and Hepatic Injury

By

HUI HAN

A Dissertation Submitted to Faculty of the USC Graduate School
in Fulfillment of Requirements for the Degree of
Doctor of Philosophy in Genetic, Molecular and Cellular Biology
University of Southern California

Degree Conferral Date: May 2017

Table of Contents
List of Chapters ........................................................................................................................................................................ 1
List of Abbreviations ............................................................................................................................................................... 2
Acknowledgement ................................................................................................................................................................... 4
Chapter 1 Background Reviews .............................................................................................................................................. 5
1. Human Immunodeficiency Virus (HIV) and HIV Pandemics Worldwide .................................................................... 5
2. Highly Active Anti-Retroviral Therapy (HAART), a Treatment for AIDS but a Problem for the Liver ...................... 6
3. Alcohol Drinking Raises Risks of Liver Disease in Patients under HAART ................................................................ 7
4. Cellular Stress Response as an Important Mechanism in Drug Induced Liver Injuries. ............................................... 8
5. Connection of Golgi Apparatus to the ER stress and Liver Diseases ........................................................................... 8
6. Statement of Purpose and Rationale ............................................................................................................................. 9
Chapter 2 Liver Cell Death Mechanisms under Treatment of Anti-HIV PI Cocktail Ritonavir and Lopinavir ..................... 11
Literature Review ............................................................................................................................................................... 11
1. Ritonavir (RTV) Boosted Lopinavir (LPV) Induces Liver Injuries ............................................................................ 11
2. Insight into Definition and Mode of Cell Death ......................................................................................................... 11
3. Biochemical Features of Major Forms of Regulated Cell Death ................................................................................ 12
4. Mechanisms of Drug Induced Liver Injury – Apoptosis vs Necroptosis .................................................................... 13
Materials & Methods .......................................................................................................................................................... 14
Results ................................................................................................................................................................................ 16
1. R L Induces Cell Death Mediated by Apoptosis in HepG2 Cells ............................................................................... 16
2. RL Induces Cell Death Mediated by Necroptosis and Apoptosis in Cultured Primary Mouse Hepatocytes .............. 17
3. Autophagic Cell Death or Autosis is Not Involved in RL Induced Cell Death ........................................................... 17
Discussion .......................................................................................................................................................................... 18
Figure Legends ................................................................................................................................................................... 19
Figures and Tables .............................................................................................................................................................. 22
Chapter 3 Contribution of the ER stress to Ritonavir and Lopinavir Induced Cell Death in Vitro. ....................................... 29
Literature Review ............................................................................................................................................................... 29
1. The Endoplasmic Reticulum Stress and the Unfolded Protein Response ................................................................... 29
2. Three Canonical Branches of the UPR ....................................................................................................................... 30
3. Role of the ER Stress in HIV Protease Inhibitors Related Injuries ............................................................................. 32
Materials and Methods ....................................................................................................................................................... 33
Results ................................................................................................................................................................................ 34
1. Selective Activation of Two Canonical UPR Branches but not ATF6 Branch in RL-Treated Cells ........................... 34
2. Interference of ATF6 Processing and Localization in HepG2 Cells Treated with RL ................................................ 35
3. Activation of ER Stress Mediated Apoptosis in RL-Treated HepG2 Cells ................................................................. 36
4. Selective Activation of the UPR in the Primary Mouse Hepatocytes under RL Treatment ........................................ 37
Discussion .......................................................................................................................................................................... 37
Figure Legends ................................................................................................................................................................... 40
Figures and Tables .............................................................................................................................................................. 43
Chapter 4 Mechanistic studies on Ritonavir and Lopinavir Induced ER stress – Proteasome Inhibition, Calcium Depletion
and Production of Reactive Oxygen Species ......................................................................................................................... 59
Literature Review ............................................................................................................................................................... 59
1. ER Stress Induced by Pharmacological Compounds .................................................................................................. 59
2. Three Possible Mechanisms for Anti-HIV PIs Induced ER Stress ............................................................................. 59
Materials and Methods ....................................................................................................................................................... 61
Results ................................................................................................................................................................................ 63
1. Weak Association between Proteasome Inhibition and the ER Stress under RL Treatment ....................................... 63
2. Prolonged RL Treatment Leads to Calcium Loss Possibly due to SERCA Degradation ............................................ 63
3. Oxidative Stress Was Evoked in RL-Treated PMH but Did Not Contribute to the ER Stress. ................................... 65
Discussion .......................................................................................................................................................................... 65
Figure Legends ................................................................................................................................................................... 67
Figures and Tables .............................................................................................................................................................. 68
Chapter 5: A Novel Potential Mechanism for the Drug and Alcohol Induced ER Stress and Injuries - Involvement of
Abnormal Golgi Morphology, Function and ER-to-Golgi Transportation ............................................................................ 76
Literature Review ............................................................................................................................................................... 76
1. Mechanisms and Machineries in ER-Golgi Trafficking.............................................................................................. 76
2. Induction of ER Stress under Impaired ER-Golgi Trafficking .................................................................................... 77
Materials & Methods .......................................................................................................................................................... 78
Results ................................................................................................................................................................................ 80
1. RL Affects Golgi Morphology and Function in Liver Cells ....................................................................................... 80
2. Effects of RL on the Golgi Were Correlated with the ER Stress Response ................................................................ 81
3. RL Induced Golgi Fragmentation Possibly by Affecting ER-to-Golgi Trafficking .................................................... 82
4. Alcohol Deteriorated RL Induced Golgi Fragmentation and Injuries in PMH and Mouse Liver ............................... 83
Discussion .......................................................................................................................................................................... 84
Figure Legends ................................................................................................................................................................... 87
Figures and Tables .............................................................................................................................................................. 91
Chapter 6 Conclusions ......................................................................................................................................................... 111
References............................................................................................................................................................................ 114
1

List of Chapters
Chapter 1: Background Information
Chapter 2: Liver Cell Death Mechanisms under Treatment of Anti-HIV PI Ritonavir and Lopinavir
Chapter 3: Contribution of ER stress to Ritonavir and Lopinavir Induced Cell Death in Vitro.
Chapter 4: Mechanistic Studies on Ritonavir and Lopinavir Induced ER Stress – Proteasome Inhibition,
Calcium Depletion and ROS Production
Chapter 5: A Novel Potential Mechanism for the Drug and Alcohol Induced ER Stress and Injuries –
Involvement of Abnormal Golgi Morphology and Function
Chapter 6: Conclusions

2

List of Abbreviations

3-MA 3-methyladenine
ACD accidental cell death
AIDS acquired immune deficiency syndrome
ALT alanine aminotransferase
APV amprenavir
ART anti-retroviral therapy
AST aspartate aminotransferase
ATF4 activating transcription factor 4
ATF6 activating transcription factor 6
AZV atazanavir
BFA brefeldin A
BLA baflomycin A
CASP caspase
CHOP CCAAT/enhancer-binding protein homologous protein
c-JUN jun proto-oncogene
CYP3A4 cytochrome P450 3A4
DAMPs damage associated molecular patterns
DAV darunavir
DDT dithiothreitol
DILI drug induced liver injury
DR5 death receptor 5
ER endoplasmic reticulum
ERGIC ER-Golgi intermediate structure
ERO1 endoplasmic reticulum oxidoreductase alpha
ESLD end-stage liver disease
EtOH ethanol
GADD34 growth arrest and DNA damage-inducible protein 34
GADD45 growth arrest and DNA damage-inducible protein 45
GCP60 Golgi complex-associated protein of 60 kDa
GM130 Golgi matrix protein 130kDa
GRP78 glucose regulated protein 78
GSR Golgi stress response
HAART highly active anti-retroviral therapy
HBV hepatitis B virus
HCC hepatocellular carcinoma
HCV hepatitis C virus
HIV human immunodeficiency virus
HSP47 heat shock protein 47kDa
InSTI(s) integrase inhibitor(s)
IRE1a inositol-requiring enzyme 1a
JNK jun N-terminal kinase
LPV lopinavir
3

MAN2A1 Golgi alpha mannosidase II
MLKL mixed lineage kinase domain like protein
NAS necrosulfamide
NEC1 necrostatin-1
NFV nelfinavir
NNRTI(s) non-nucleoside reverse transcriptase inhibitor(s)
NRTI(s) nucleoside reverse transcriptase inhibitor(s)
p58IPK protein kinase inhibitor of 58 KDa
PBA 4-phenylbutyrate
PBS phosphate buffered saline
PDIA4 protein disulfide isomerase family A member 4
PERK protein kinase R (PKR)-like endoplasmic reticulum kinase
PI(s) protease inhibitor(s)
PMH primary mouse hepatocytes
RAB member RAS oncogene family
RCD regulated cell death
RIP1K receptor interacting serine/threonine kinase 1
RIP3K receptor interacting serine/threonine kinase 3
RL RTV+LPV
ROS reactive oxygen species
RTV ritonavir
SERCA sacroplastic reticulum calcium ATPase
SLB salubrinal
SM sodium monesin
SO superoxide
sXBP1 spliced x-box binding protein 1
TFE3 transcription factor binding to IGHM enhancer 3
THG thapsigargin
TMG TUM+THG
TNFα tumor necrosis factor a
TOYO toyocamyxin
TPV tipranavir
TRIB3 tribbles pseudokinase 3
TUDCA tauroursodeoxycholic acid
TUM tunicamycin
UPR unfolded protein response
uXBP1 un-spliced x-box binding protein 1
VCL vinculin
XBP1 x-box binding protein 1

4

Acknowledgement
Foremost, I would like to express my sincere gratitude to my Ph.D. advisor Prof. Cheng Ji for the continuous
support of my Ph.D. study and research, for his patience, suggestion and immense knowledge. He is a great
mentor who guided me all the time for my study and completion of this dissertation. He is the most talent person
I have never met.
Besides my advisor, I would like to thank the rest of my dissertation committee members. Prof. Jame Ou who
is the chair of the committee greatly helped me to organize annual research appraisal each year to allow me to
review and take suggestions from professionals in the field. I also want to give me deep gratitude to Prof. Laura
D. Deleve who is profession in clinical research and gave me many useful suggestions on animal studies. I
especially want to thank Prof. Ebrahim Zandi who helped me with molecular biology and biochemistry and
provided many suggestions that helped me to increase the depth of my research.
My sincere thank also goes to Prof. Wei Li who helped me with my studies in USC and Prof. Yunfeng Lu from
UCLA who supported another exciting project completed by me. I would like to appreciate the working
experience with Dr, Farzana Choudhury with whom I learned how to teach students and also thank all the faculty
and teachers who generously passed their knowledge to me and assist me to be proficient in field of genetic,
molecular and cellular biology. Except for the mentors, I appreciate all the lab members in Dr. Cheng Ji’s lab and
members research center of liver disease especially for Rhema Lau and Jay Hu for simulative discussion and
training on lab technics.
Last but not the least, I want to greatly thank my wife Yibo Cai who selflessly supported my career and my
daughter Amelia HAN who brought me lots of happiness. I also want to thank my parents Hua HAN and Lili
Wang who are great scientist and gave unconditionally to my growth and education.

5

Chapter 1 Background Reviews
1. Human Immunodeficiency Virus (HIV) and HIV Pandemics Worldwide
Human immunodeficiency virus or HIV , is a lentivirus in the Retroviridae family leading to acquired
immunodeficiency syndrome (AIDS)
1
. HIV was originated from West Central Africa approximately in the early
20
th
century and transmitted from other primates
2
.

In the past decades, HIV has been huge health and wealth
burdens worldwide especially in African countries. Globally, 1.1 million deaths were caused by HIV infection
and 36.9 million of people are still living with HIV
3
.

In United States, approximately 0.4%~0.9% of the adults
are infected with HIV and approximately $30 billion dollars were spent for treatments and cares for HIV patient
in each year
4
.
In contrast to most of the animal viruses, infection with HIV results in prolonged and continuous viral
replication in the infected host
5
. This is not only due to immune-suppression but also effective evasion from the
host immune surveillance
3,5
. When infected with HIV , an individual starts to experience a flu-like syndrome due
to rapid replication of the virus with a dramatic depletion in CD4+ T cells in peripheral blood
6
. Later,
neutralization antibodies are developed by B cells to eradicate the virus. However, a certain population of the
virus could interfere with the antibody binding by varying their surface glycosylation patterns preventing their
detection by the host immune system
6, 7
.
This escaping mechanism relies on an extensive genetic variability and high mutation rate. Hence, HIV has
many groups and subtypes. HIV can be classified into two major groups, HIV-1 and HIV-2 based on genetic
differences. HIV-1 which is the dominant type that causes HIV pandemics could be further divided into 4 groups
and several subtypes
8,9
. The genetic variabilities are 50% between HIV-1 and HIV-2, 37% between groups and
14.7% between subtypes
9
. When an individual is infected with a single viral particle, at the end stage of the
diseases, the genetic diversity of the viral population is close to the diversity of all influenza virus worldwide
10
.
Hence, early attempts in treatment of HIV with single-agent were all failed due to development of drug resistance.
6

Similarly, due to variability in surface epitope, many HIV vaccines have failed during clinical trials
11
.
2. Highly Active Anti-Retroviral Therapy (HAART), a Treatment for AIDS but a Problem for the Liver
To prevent HIV global catastrophe, anti-retroviral therapy (ART) or highly active anti-retroviral therapy
(HAART) was introduced at the end of 20
th
century and remains to be the standard management for HIV patients
up to date
3
. The goal of HAART is to suppress virus load, restore CD4+ T cell count, prevent HIV transmission
and drug resistance, and eventually improve quality of life. Upon introduction of HAART in 1995, death incidence
due to HIV was reduced by ten folds in less than ten years indicating its great efficiency in reducing AIDS related
motalities
12
. HAART combines different small inhibitors targeting multiple conserved structures across all groups
of HIV to achieve rapid reduction of viral load to a limited level
13
. In 2016 US guideline for HIV treatment, the
primary care for all HIV infected individuals is to start HAART as soon as possible because of its effectiveness
in preventing disease progression and transmission
3, 12, 13
. Currently, 39 medications are approved by FDA for
HIV treatment with HARRT including both single substituted and combined medications
14
. The major drug
classes are nucleoside reverse transcriptase inhibitors (NRTIs), non-nucleoside reverse transcriptase inhibitors
(NNRTIs), protease inhibitors (PIs) and integrase inhibitors (InSTIs)
1, 14.
In clinical practice, the optimal regimens
for newly diagnosed patients are either two NRTIs plus an InSTI or a NNRTI plus a PI
1
. However, the regiments
are often switched from one to another during the treatment by considering plasma viral titers, drug resistance,
adverse effects, drug-drug interactions, pregnancy and food restrictions.
Although HAART efficiently reduces AIDS-related mortality, non-AIDS-related mortalities have been
relatively increased contributing to 70% of the death in HIV patients especially due to liver diseases (15%),
cardiovascular diseases (11%) and non-AIDS cancers (15%)
15
. The possible cause includes but not limited to
social status and behavior of HIV patients, dysregulation of immune system and off-target effects under long-
term exposure
16, 17
. In clinical practice, most of the studies agree that acute hepatotoxicity of HAART is tolerable
and clinically manageable. However, clinical trials and data collection on adverse events of anti-HIV drugs study
7

(D:A:D) indicated that HAART does promote liver disease especially in case of life-long exposure
15, 18
. For
instance, cumulative exposure with HIV drugs that induce biomarker defined hepatotoxicity showed increased
risk in end-stage liver disease (ESLD) and hepatocellular carcinoma (HCC)
18,

19
. And the patients under treatment
of HAART with chronic elevation of plasma aminotransferase (ALT, AST) had higher chance to develop
advanced liver diseases
3
.
Anti-HIV PIs account for the largest group of medications in anti-HIV therapy
1
. Since the year 1995, 12
anti-HIV PIs have been approved and 90% of them were based on the principle by which peptidomimetic
structures in PIs inhibit the proteolytic cleavage of HIV Gag and Pol polyproteins that are essential for maturation
of the virus
20
. The most common anti-HIV PIs include ritonavir (RTV), nelfinavir (NFV), amprenavir (APV),
lopinavir (LPV), atazanavir (AZV), darunavir (DAV) and tipranavir (TPV)
14
. Their potencies in inhibiting HIV
protease are similar and the choice in clinical practice depends on pharmacokinetic, drug resistance and tolerance
3,
20
. The liver is the major site for metabolism of anti-HIV PIs. Acute hepatotoxicity induced by anti-HIV PIs is
self-limited with moderately elevated plasma aminotransferase level in patients under PI-based HAARTs
21
. For
long-term exposure, incidence of hepatotoxicity in HAART containing PIs could run up to 40%
22
.
3. Alcohol Drinking Raises Risks of Liver Disease in Patients under HAART
To be worsen, HBV/HCV co-infection and alcohol adherence are often observed among HIV population.
HBV/HCV co-infection which affects 10 million HIV/AIDS patients requires additional anti-viral treatment
increasing the severity of liver toxicities and injuries
23-25
. HIV patients have increased tendency to consume
alcohol and nearly 50% of the HIV-infected patients abuse alcohol, which not only impairs patients’ adherence
but also deteriorates anti-HIV drug-induced hepatotoxicity leading to greater morbidity and mortality
26, 27
.
Association between excessive alcohol drinking with liver diseases among HIV patients was demonstrated by
independent clinical researches
25-31
. Both hazardous drinking and non-hazardous drinking have been shown to
increase the risk of advanced liver fibrosis in both HIV and HIV/HCV co-infected patents in a stepwise manner
27
.
8

Anti-HIV PI-based monotherapies have been shown to increase the prevalence of liver fibrosis by 10% and 25%
in presence or absence of binge drinking, respectively
26
.
4. Cellular Stress Response as an Important Mechanism in Drug Induced Liver Injuries.
Major mechanisms underlying the hepatotoxicity are idiosyncratic hypersensitive reactions mediated by the
adaptive immune system, and cellular stress responses that either induce cell death directly or generate danger
signals co-stimulating the immune system
33, 34
. The cellular stress responses are most relevant because both
alcohol and anti-HIV drugs are metabolized in the liver, which are bound to stress the liver. Previously we and
others reported that alcohol and/or anti-HIV drugs induced the endoplasmic reticulum (ER) stress, lipid
accumulation and hepatic cell death
34-37
. The ER stress initially triggers a protective unfolded protein response
(UPR), which involves three canonical ER stress sensors, IRE1a, PERK and ATF6, to restore the ER
homeostasis
40
. However, prolonged activation of the UPR resulted from chronic alcohol and/or long-term anti-
HIV therapies induces hepatosteatosis, inflammation, cell death and development of fibrosis and cirrhosis
34, 38-40
.
The ER stress/UPR has thus been an important therapeutic target in liver diseases. Xenobiotics including
molecular chaperones (e.g. PBA and TUDCA) and UPR enhancers (e.g. valproate) to ensure proper ER
homeostasis have been developed and tested in a variety of in vitro and in vivo models
41
. However, these
compounds for restoration of ER homeostasis only yield limited protective effects
42
. The incomplete protections
via targeting the ER suggest that although the ER stress is involved in the alcohol and drug-induced hepatotoxicity
and many other forms of liver injury, there are other drug and alcohol-targeted cellular components that either
enhance or are upstream of the ER stress response
42
.
5. Connection of Golgi Apparatus to the ER stress and Liver Diseases
The organelle that is closely associated with the ER is the Golgi apparatus. There is bidirectional ER-Golgi
trafficking transporting proteins and lipids between the two organelles
45
. Anterograde route of ER-to-Golgi is
mediated by COPII complexes and retrograde route of Golgi-to-ER is mediated by COPI complexes. Like the ER,
9

the structure and capacity of the Golgi fluctuate according to physiological demands or pathological stress
conditions. Abnormal Golgi morphologies occur when the protein load and modifications exceed its capacity
44-
46
. For instances, enlargement or hypertrophy of the Golgi has been observed in human liver due to amyloidosis,
toxic hepatitis and other conditions including drug-induced cholestasis and malnutrition
46
. Atrophy or
degeneration occurs during starvation, viral hepatitis and choline or protein deficiency
47
. Under viral infection,
the Golgi is partially fragmented
48
. In man and animal, chronic alcohol feeding with nutritionally adequate diets
induced ultrastructural abnormalities of hepatocytes
27
. There is also an adaptive Golgi stress response (GSR) that
involves GCP60, HSP47 and TFE3 to cope with disrupted homeostasis of the Golgi
42, 44
. However, persistent
stress on the Golgi disturbs metabolisms in the liver. Dysfunction of the Golgi apparatus is well associated with
various stress-induced liver injuries
42, 44-50
. What role of the Golgi plays in the drug and alcohol-induced ER stress
and what underlying molecular mechanism links Golgi dysfunction with the ER stress response and injury are
not known.
6. Statement of Purpose and Rationale
There is no consensus on how to manage AIDS patients suffering from liver diseases because mechanisms
underlying the hepatotoxicity of alcohol abuse and the drugs are still under investigations
18
. Anti-HIV PIs and
non-nucleoside reverse transcriptase inhibitors are the two major groups of HIV drugs causing serum
aminotransferase (ALT, AST) elevation which is the primary indicator for liver diseases in laboratory test
26
.
Currently, one commonly favored mechanism for anti-HIV PIs induced cell death/injury was apoptosis mediated
by the ER stress
20
. For events upstream of the ER stress, three hypotheses have been proposed including
proteasome inhibition, ER calcium perturbation and oxidative stress mediated ER stress
34-37, 43
Indeed, those
events are well-associated with the ER stress and apoptosis in different in vitro models
20, 34, 43
. However, limited
data have been published for liver cells.
In this study, we raised several interesting questions including: (1) whether the ER stress mediated apoptosis
10

is the only mechanism for anti-HIV PIs induced liver cell death; (2) whether proteasome inhibition, ER calcium
perturbation and oxidative stress lead to the ER stress in liver cells due to what mechanism (s); (3) what are the
roles of the Golgi in the anti-HIV PIs induced ER stress; (4) what are the roles of alcohol in promoting anti-HIV
PIs induced liver injuries in vitro and in vivo. These questions are challenging, but the answers will significantly
enhance our understanding of the potential mechanisms of anti-HIV PIs induced liver injury. Trying to address
those questions, I conducted a preliminary in vitro and in vivo studies by examining and charactering some of
biological effects of anti-HIV PIs, specifically using RTV+LPV (RL) cocktail as a drug model, in HepG2 cells,
the primary mouse hepatocytes and mouse liver.

11

Chapter 2 Liver Cell Death Mechanisms under Treatment of Anti-HIV PI
Cocktail Ritonavir and Lopinavir
Literature Review
1. Ritonavir (RTV) Boosted Lopinavir (LPV) Induces Liver Injuries
Ritonavir (RTV) boosted lopinavir (LPV) (RL) has been one of the recommended medications for HIV post-
exposure prophylaxis and popular components of HARRT worldwide due to its good efficiency in treatment of
naï ve or treatment-experienced HIV patients
51
. For instance, 90% of patients with RL have reduced the viral load
to <400 copies/ml after one year and 70% of them could maintain <50 copies/ml after 4 years
52, 53
. Both RTV and
LPV have inhibitory effects on HIV-1 protease with LPV being more active. RTV is usually used as effective
booster due to its inhibitory effects on multi-drug resistant transporter 1 (MDR1) and cytochrome P450 3A4
(CYP3A4) enhancing bioavailability of LPV
54
. Hence, LPV or RTV single therapy has been abandoned and
RTV+LPV (RL) has been dedicated as a constitute of HARRT since 2000
53
. Similar to most of the antiviral drugs,
high proportion of patients with RL encounter elevation of the serum aminotransferase (ALT and AST)
55
, which
increases risk in development of liver diseases
56
. Moreover, life-threatening hepatotoxicity has been reported in
RL-treated patients
57
. In mouse models, RL treatment elevated serum ALT and induced hepatic steatosis
34
. When
co-administrated with alcohol, ALT level reached 15 times of control indicating severe hepatotoxity
34
. In in vitro
model, RL-induced cell death in immortalized cell lines (e.g. Hela cells), the primary macrophage/adipocytes and
cancer cells
58-61
. The cell death mechanism demonstrated by most of the studies was apoptosis mediated by the
ER stress
58-61
. However, limited information was published about mode of cell death induced by RL in liver cells.
2. Insight into Definition and Mode of Cell Death
Studying mechanism of cell death has been an important part to demonstrate potential adverse effects of
chemicals and drugs. According to Nomenclature Committee on Cell Death (NCCD), the dead cells are defined
12

as cells that lose plasma membrane barrier, break into fragments or are engulfed and digested by professional
phagocytes
62
. Instances of cell death could be classified as accidental and regulated cell death (ACD and RCD).
ACD is caused by extreme physical (e.g. high temperature or mechanical damage) and/or chemical (e.g. strong
acid or bases) insults resulting in disintegration of cells without involvement of specific molecular pathways
62
. In
contrast, RCD is regulated by molecular machineries that are genetically encoded
62
. RCD is evoked when cells
fail to restore cellular homeostasis under stress conditions that commonly involve in organelle stress and the
response, e.g. the ER stress and unfolded protein response
63, 64
. Therefore, RCD can often be delayed or modulated
by means of pharmacologic and/or genetic interventions whereas ACD is generally untreatable except for
removing the source of damages
62, 65
. More importantly, when occurred in physiological context, RCD is also
referred as programmed cell death (PCD)
66
.
3. Biochemical Features of Major Forms of Regulated Cell Death
Modes of ACD depend on sources of damage. In contrast, modes of RCD could be classified based on
biochemical standpoints. Apoptosis is a well-defined mode of RCD strictly depends on caspase (CASP)
activities
67
. Apoptotic stimuli can be initiated through intrinsic and extrinsic pathways. Intrinsic apoptotic
pathway is mediated by permeabilization of mitochondrial outer membrane and activation of initiator CASP9
whereas extrinsic apoptotic pathway is mediated by ligand binding of Fas-associated protein with death domain
and activation of initiator CASP8
62, 64, 67
. Both of the initiator CASPs activate executor CASPs (e.g. CASP3,
CASP7) degrading hundreds of cellular components to induce cell death
68
. The second major type of RCD is
necroptosis that critically depends on mixed lineage kinase domain like protein (MLKL) or receptor interacting
serine/threonine kinase 1 (RIP1K) activity
69
. Necroptosis is CASP independent and shares morphological features
with necrosis (e.g. loss of plasma membrane integrity)
62, 69
. Necroptosis is well studied in tumor necrosis factor
α (TNFα) induced cell death in cells lack of CASP8 activation, but it could also be activated by drug treatment
and viral infection
62, 70
. Initiation of necroptosis triggers the formation of RIP1K–RIP3K–MLKL complex or so-
13

called necrosome
69
. RIP1K is responsible for receptor interacting serine/threonine kinase 3 (RIP3K)
phosphorylation, and subsequently, RIP3K phosphorylates MLKL allowing formation of MLKL homotrimers
62,
69
. Then the homotrimers translocate to plasma membrane to causes necrotic membrane permeabilization
accompanied by production of reactive oxygen species (ROS) and reduction of cytosolic ATP
69-71
. Other major
types of RCD include autophagic cell death or autosis that depends on excessive autophagy, parthanatosa that
depends on poly ADP-ribose polymerase-1 activation and ferroptosis that depends on iron molecule
62
.
4. Mechanisms of Drug Induced Liver Injury – Apoptosis vs Necroptosis
Both ACD and RCD mediate drug induced liver injury (DILI). ACD-mediated DILI is often observed in
patients with drug overdose
73, 74
. RCD especially apoptosis and necroptosis of hepatocytes is associated with
chronic DILI
75
. Necroptosis leads to disintegration of plasma membrane and releasing of damage associated
molecular patterns (DAMPs)
69
. DMAPs including some cytoplasmic, nuclear proteins as well as DNA and RNA
could send out danger signals to immune system to indicate local injuries and stimulate a noninfectious
inflammatory response (e.g. neutrophil infiltration and activation of hepatic immune cells)
69
. In contrast,
apoptosis leads to formation of apoptotic bodies which consist of tightly packed organelles and condensed
chromatin without loss of plasma membrane integrity
75
. Apoptotic bodies can be recognized and removed by
Kupffer cells and hepatic stellate cells and degraded by phagolysosomes
62, 75
. Theoretically, apoptotic cells do
not lead to release of DAMPs and stimulate immune system. However, on one hand, excessive apoptosis can
exceed engulfing capacity of liver macrophages while the remaining apoptotic cells could undergo secondary
necrosis releasing DAMPs to extracellular environment
76
On the other hand, intake of apoptotic bodies by Kupffer
cells stimulates their excretion of inflammatory cytokines, e.g. TNFa which induces cell death in hepatocytes
through necroptosis. Therefore, chronic apoptosis has been one of the key features in many types of liver
diseases
76
.
In this chapter, we initiated the study to identify mode of cell death involved by RL treatment in HepG2 cells
14

and the primary mouse hepatocytes (PMH). As indicated in the previous context, apoptosis has been considered
as the primary mechanism for RL-induced cell death in various in vitro models
20
. However, with deepened
understanding in the field of RCD in recent years, we raised a question about whether some other modes of cell
death can also be involved in RL-induced cell death in liver cells. In specific, we focused on examining the
contribution of apoptosis, necroptosis and autosis during RL-induced cell death

Materials & Methods
1. Cell Culture and Drug Treatment
HepG2 cells were purchased from ATCC, cultured on Collagen I (Thermo-Fisher) coated dishes and
maintained in DMEM with 4.5g/ml glucose, 10% FBS, non-essential amino acids and 100U/ml of
penicilin+streptomycin (P/S). The cells were treated with ritonavir (RTV)+lopinavir (LPV) or RL at
concentrations ranging from 0 to 20µg/ml. DMSO (0.05%, v/v) plus ethanol (0.05%, v/v) in DMEM was used as
vehicle control. Primary mouse hepatocytes were isolated by USC Liver Cell Culture Core
34
.

The isolated cells
were allowed for attachment by 4 hours and the medium was switched to William’s E medium supplemented with
dexamethasone, insulin, transferrin, sodium selenium, reduced FBS, GlutMax and P/S. The primary cells were
allowed to stay at 37
o
C/5% CO2 for overnight. On the next day, the cells were treated. After the treatments, the
cells were washed with ice-cold PBS and subjected to fixing or protein and RNA extractions. All in vitro assays
were repeated at least three times for each treatment.
2. Assay of Apoptosis by Flow Cytometry
Apoptosis was indicated by Annexin V (ANXA5)-FITC/propidium iodide (PI) double staining.
Approximately 500,000 cells after treatment were detached with 0.05% trypsin/EDTA+Accutase cocktail and
washed once with ANXA5 binding buffer with 1% bovine serum albumin (BSA). Then the cells were stained
15

with 5ul of ANXA5-FITC (Biolegend) and 20µ g/ml PI (Invitrogen) for 15 minutes. The stained cells were
analyzed by FACS Caliber
®
FACS system (BD Science). Cells without staining served as background control and
drug treated cells with single staining were used to adjust signal compensation. Before the analysis, signals from
cell debris were removed and FSC (forward scatter), SSC (side scatter), FL-1 (green/FITC fluorescent), FL-
2(red/Tritc fluorescent) signals were collected in 50,000 cells. The collected data was analyzed and plotted in
FlowJo
®
7.6.
3. Caspase 3 Activity Assay and Cytotoxicity Assay
Caspase 3 activity was measured with Caspase-Glo® 3/7 assay system (Promega) by detecting luminescent
signals. To assess cytotoxicity, lactate dehydrogenase (LDH) release into the medium was measured by Pierce
TM

LDH Cytotoxicity Assay Kit (Thermo-Fisher).
4. Immunoblotting
Cells after treatment were lysed in RIPA buffer with protease/phosphotase inhibitor cocktails (Sigma) on ice
for 15 minutes. The homogenized mixture was centrifuged at 14000g for 20 minutes at 4
o
C. The whole proteins
in supernatant were differentiated by SDS-PAGE. Then the proteins were transferred to PVDF membrane with
tank blotting. The transferred membrane was subsequently labeled with primary, and secondary antibodies
conjugated with HRP. Protein expression was measured by enhanced chemiluminescence (ECL) that was
recorded by Fuji films. The films were scanned and processed by ImageJ.
5. Immunofluorescent Staining of MLKL
Cells were attached to 18 mm round-glass cover slides coated with Collagen I and cultured in 12-well plates.
After the drug treatments, cells were washed with ice-cold PBS, fixed in 4% parafolmaldehyde (PFA) solution
for 15 minutes, permeabilized in 2% Triton-X 100 in PBS (TPBS) for 15 minutes, and blocked with 10% goat
serum in PBS with 0.05% Tween 20. The fixed cells were probed with monoclonal anti-pMLKL
antibodies(Abcam) which were diluted in the blocking buffer for overnight. The cells on the slides were then
16

labeled with secondary antibodies conjugated with fluorescein isothiocyanate (FITC) and nuclear counter-stained
with 10 ng/ml Hochest blue in PBS. After sequentially washed with PBST, PBS and ddH2O, the cover slides
were mounted on glass slides with Prolonged
®
Gold mounting media (Thermo-Fisher).

Results
1. R L Induces Cell Death Mediated by Apoptosis in HepG2 Cells
Morphological changes in RTV+LPV (RL)-treated HepG2 were observed under light microscope. Upon
treatment with RL for 24 hours, morphological signs of apoptosis appeared including membrane blebbing,
shrinking in cell size and reduced cellular attachment (Fig. 1-1A, right) whereas DMSO-treated cells formed
confluent cell monolayers with epithelial like structure (Fig. 1-1A left). To confirm co-existence of apoptosis with
cell death, cells were stained with Annexin V (ANXA5)-fluorescein isothiocyanate (FITC)+propidium iodide(PI)
and analyzed by flow cytometry. In Figure 1-1B, RL treatment at 10µg/ml increased ANXA5 signal by 6 folds
and resulted in 10% of cell death among total cell population. While, RL treatment at 20 µg/ml increased ANXA5
signal by 20 folds and resulted in 30% of cell death (Fig. 1-1B). For RL treatment at 20 µg/ml, late apoptotic cells
(Fig. 1-1B, population in Q2) accounted for almost 90% of total dead cells (Fig. 1-1B, population in Q1+Q2)
suggesting occurrence of apoptosis with cell death. Next, activities of effector caspase (CASP) and LDH release,
the latter of which is proportional to rate of cell death, were measured at the same time. Upon RL treatment for 8
hours, effector CASP activity but not LDH release was significantly increased indicating that CASP activation
was earlier than cell death (Fig. 1-1C). Moreover, treatment with pan-caspase inhibitor QVD-OPh dose-
dependently inhibited RL-induced cell death (Fig. 1-1D). When the concentration was higher than 6µM, QVD-
OPh could completely rescue the cell death (Fig. 1-1D). Compared with potent ER stress inducer tunicamycin
(TUM), RL treatment resulted in higher CASP activity and cell death (Fig. 1-1C).
17

To identify the presence of necroptosis, phosphorylation of MLKL was detected with immunoblotting. RL
treatment by 24 to 48 hours did not induce any pMLKL (Fig. 1-1E). Meanwhile, cells treated with RL were co-
treated with two structurally unrelated inhibitors for necroptosis including Necrosulfonamide (NAS) and
Necrostatin-1 (NEC1). After treatment for 24 hours, NAS and NEC1 failed to rescue the RL-induced cell death
(Fig. 1-1E). Therefore, necroptosis does not involve in RL-induced cell death in HepG2 cells.
2. RL Induces Cell Death Mediated by Necroptosis and Apoptosis in Cultured Primary Mouse
Hepatocytes
In primary mouse hepatocytes (PMH), treatment with RL also significantly changed cell morphology.
Compared with HepG2 cells, PMH treated with RL also showed membrane blebbing and shrinkage in size
possibly indicating apoptosis (Fig. 1-2A). However, unlike in HepG2 cells, co-treatment with QVD-OPh which
completely abolished effector CASP (Fig. 1-2B) only reduced cytotoxicity by 20% relative to RL treatment in the
PMH (Fig. 1-2C). Compared to TUM-treated cells, effector CASP activity was 47% lower whereas the cell death
was 91% higher in RL-treated cells (Fig. 1-2B&C). Hence, apoptosis only partially contributed to RL-induced
cell death in the PMH model. Next, cells were tried to be rescued with both NEC1 and NAS to confirm the
presence of necroptosis. Under co-treatment with RL for 24 hours, NAS treatment successfully reduced LDH
release by almost 70% (Fig. 1-2D). In contrast, NEC1 had no effect on LDH release induced by RL treatment in
the PMH model (Fig. 1-2D). Moreover, by immunofluorescent staining, MLKL in RL-treated PMH was relocated
to the cell surface and nucleus whereas the distribution was random and even in DMSO or TUM-treated cells
(Fig. 1-2F). Hence, unlike in HepG2 cells, RL caused intermingled mode of cell death mediated majorly by
MLKL dependent necroptosis and partially by apoptosis.
3. Autophagic Cell Death or Autosis is Not Involved in RL Induced Cell Death
In addition to apoptosis and necroptosis, presence of autosis was investigated by blocking of autophagy with
baflomycin A (BLA) and 3-methyladenine (3-MA). Under co-treatment with RL, BLA and 3-MA significantly
18

worsened the cell death (Figure 1-3A) indicated by increased in LDH release by 75% and 100%, respectively.
Consistent with the results in HepG2 cells, co-treatment with BLA or 3-MA significantly increase LDH release
by 20% and 26%, respectively in the PMH (Figure 1-3B). Therefore, instead of contributing to cell death,
autophagy protected the liver cells from RL-induced cell death.

Discussion
Studying mode of liver cell death is essential to illustrate possible mechanism of liver injury
75
. For
RTV+LPV (RL) induced cell death, possibility of ACD could be eliminated since specific molecular mechanisms
were involved in both HepG2 and the PMH models. In this study, apoptotic morphologies were observed in both
of the cell types. In HepG2 cells, presence of apoptosis with cell death was determined by flow cytometry. Earlier
activation of effector CASP at 8 hours and protective effect of QVD-OPh on RL-induced cell death suggested a
cause-effect relationship between apoptosis and cell death. These data were consistent with previous studies in
other immortalized cell lines, e.g. Hela cells and sarcoma cells
58-61
. In both of the cell models, necroptosis was
inhibited by either NEC1 which blocks RIP1K or NAS which blocks MLKL. RIP1K initiates necroptosis while
MLKL executes cells under necroptosis. However, neither NEC1 nor NAS yielded any protective effects on RL-
induced cell death in HepG2 cells indicating that apoptosis was majorly responsible for RL-induced cell death in
HepG2 cells. In contrast, in the PMH model, apoptosis incompletely or slightly contributed to RL-induced cell
death as QVD-OPh exerted limited rescue effects. Rather, the results indicated that necroptosis was majorly
involved in RL-induced cell death in the PMH model since the treatment of NAS remarkably rescued RL-induced
cell death. Consistent with other necroptotic models, RL treatment induced MLKL cell surface translocation,
which is the most critical steps leading to loss of plasma membrane integrity. Meanwhile, RL treatment also
induced translocation of MLKL to nucleus, which was a reported feature during necroptosis in Hela cell model
19

with unknown consequence
77
. However, NEC1 hardly exerted any protective effects on RL-induced cell death
indicating the necroptosis may be independent of RIP1K in the PMH model. Even though MLKL activation
requires RIPK1 in classical TNFa-induced necroptotic model, a growing body of evidences strongly demonstrated
that necroptosis may not be evoked through conventional RIPK1-RIPK3-MLKL pathway
78
. Therefore, it is
possible that RL activated necroptosis which depends on MLKL was mediated by unidentified necroptotic
signaling in the PMH model.
Anti-HIV PIs, RTV and LPV have been shown to interfere with autophagy flux and result in accumulation
of autophagic markers LC3B and p62
36
. Blocking of autophagy could promote cell survival or cell death
depending on the scenario. In this study, treatment with inhibitors blocking autophagic signaling or flux with 3-
MA and BLA worsened apoptosis and cell death. Therefore, under RL-induced cell death, autophagy was possibly
evoked as a protective mechanism in cells.
Cellular stress responses are often relevant to RCD induced cell death
45, 63
. Association of the ER stress with
cell death especially apoptosis in RL-treated cells has been characterized before. In this study, RL-induced
apoptosis and cell death were compared with tunicamycin (TUM) in parallel. In HepG2 cells, RL treatment
induced stronger apoptosis along with more cell death than TUM treatment. In contrast, TUM induced stronger
apoptosis but not necroptosis in the PMH model. Therefore, the ER stress seems to be responsible for apoptosis
but not necroptosis in vitro. Therefore, necroptosis was probably evoked by other organelle stress, e.g. Golgi
stress.

Figure Legends
Figure 1-1 Apoptosis Potentiates Anti-HIV Drugs Induced Cell Death in HepG2 model. (A) Morphological
changes in RTV+LPV (RL)-treated HepG2 cells. Cells were treated with either DMSO or RL for 24 hours and
20

observed under light microscope (40X); DMSO (0.1%) as vehicle control. (B) Assay for detecting cellular
apoptosis. Cells were treated and stained with Annexin V (ANXA5)-FITC and propidium iodide (PI). Data
obtained were analyzed and plotted with FlowJo 7.6; population in each region: Q1, cell death without apoptosis;
Q2, cell death with apoptosis; Q3, viable cells under apoptosis; Q4, viable cells. (C) Activation of effector Caspase
3 under RL treatment. Caspase 3 activity was indicated by relative luminescent reading (RLU); RL 8h/24h, treated
with RL at 20µg/ml for 8/24 hours; QVD, pan-caspase inhibitor QVD-OPh; TUM, tunicamycin (40µg/ml). (D)
Rescue RL-induced cell death by a pan-Caspase inhibitor. Cell death was measured by quantification of LDH
release and indicated by relative absorbance unit (RAU). (E) Expression of pMLKL under RL treatment in HepG2
cells. Cells were treated for indicated time and expression of pMLKL was measured by immunoblotting. MLKL,
MLKL mixed lineage kinase domain like protein. (F) Small inhibitors for necroptosis fail to protect cells from
RL-induced cell death. Cell death was quantitatively measured by flow cytometry with propidium iodide (PI)
single staining. Cells were treated by RL with/without inhibitors for 24 hours; NEC1, necrostatin-; NAS,
necrosulfonamide. *p<0.05; ***p<0.005; ****p<0.0001 compared to control or indicated otherwise; n=4-8.
Figure 1-2 Apoptosis and Necroptosis Are Involved in Anti-HIV Drug Induced Cell Death in the Primary
Mouse Hepatocytes. (A) Morphological changes in RL-treated primary cells. The primary mouse hepatocytes
(PMH) were treated with either DMSO or RL for 24 hours and observed under light microscope (40X); DMSO
(0.1%) as vehicle control. (B) Activation of effector Caspase 3 in the PMH under RL treatment. Caspase 3 activity
was indicated by relative luminescent reading (RLU) after treatment for 24 hours; RL, 20µg/ml; QVD, QVD-
OPh (20µM); TUM, tunicamycin (40µg/ml). (C) Pan-Caspase inhibitor mildly rescued RL-induced cell death in
the PMH. Cell death was measured by LDH release. Cell death was measured by quantification of LDH release
and indicated by relative absorbance unit (RAU). (D) Necroptosis inhibitor dramatically reduced RL-induced cell
death in the PMH. Cells were pre-treated with inhibitors for 4 hours followed by treatment of RL for 24 hours.
(E) Relocation of MLKL to cell surface in RL-treated PMH. Cells were treated with DMSO, RL or TUM by 24
21

hours followed by immunofluorescent staining. The images (63X) were collected and processed by ImageJ.
**p<0.01; ***p<0.005; ****p<0.0001 compared to control or indicated otherwise; n=4-8.
Figure 1-3 Effect of Autophagy Inhibitors on RL Induced Cell Death. After co-treatment of RL with
autophagy inhibitors, LDH release was measured after 24 hours for HepG2 (A) and the PMH (B); BLA,
baflomycin A; 3-MA, 3-methyladenine. ****p<0.0001 compared to control or indicated otherwise; n=4-8.

22

Figures and Tables
Figure 1-1
(A)

(B)

23

Figure 1-1
(C)

(D)

C C 3 L u m in e s c e n c e (R L U )
M e d iu m L D H L e v e l (A U )
D M S O R L 8 h R L 2 4 h T U M
0
1 0 0 0
2 0 0 0
3 0 0 0
4 0 0 0
5 0 0 0
0 .0
0 .1
0 .2
0 .3
0 .4
0 .5
C C 3 : C a s p a s e 3 A c tiv ity
L D H A c tiv ity
* * * *
* * * *
* * * *
* *
* *
M e d iu m L D H L e v e l (A U )
D M S O 0 2 4 6 8 1 0
0 .0
0 .1
0 .2
0 .3
0 .4
0 .5
R L 2 0  g /m l+ Q V D -O P h (  M )
****
***
****
24

Figure 1-1
(E)

(F)

P e rc e n t o f D e a d C e lls (% )
R L R L + N E C 1 R L + N A S R L + Q V D
0
2 0
4 0
6 0
*
***
DMSO RL24h RL48h
25

Figure 1-2
(A)

(B)

C C 3 L u m in e s c e n c e (R L U )
D M S O R L R L + Q V D T U M
0
1 0 0 0
2 0 0 0
3 0 0 0
4 0 0 0
5 0 0 0
****
****
26

Figure 1-2
(C)

(D)

M e d iu m L D H L e v e l (R A U )
D M S O R L R L + Q V D T U M
0 .0
0 .2
0 .4
0 .6
0 .8
****
***
****
27

Figure 1-2
(E)

28

Figure 1-3
(A)

(B)

29

Chapter 3 Contribution of the ER stress to Ritonavir and Lopinavir Induced
Cell Death in Vitro.
Literature Review
1. The Endoplasmic Reticulum Stress and the Unfolded Protein Response
The Endoplasmic reticulum (ER) is an organelle consisting of the largest intracellular membrane system. The
ER is organized as flat sac with branched tubules and its content is physically separated from the cytoplasm by a
lipid bilayer or the ER membrane
79
. The space surrounded by the ER membrane is called the ER lumen where it
is the major site for protein folding
79
. To facilitate the folding, the ER contains high concentration of calcium and
molecular chaperones, the latter of which are a group of proteins presenting widely across the cellular
compartments and organelles to assist proteins folding
80
. Molecular chaperones are conserved across the evolution
and are not necessarily similar in structures and amino acid sequences but they all help polypeptide chains to
overcome the rate-limiting steps during the folding and therefore largely accelerate this process
81
. Representative
examples of the ER located chaperones are glucose-regulated protein (GRP) 78, GRP94 and protein disulfide
isomerase (PDI) family proteins. Although facilitated by various factors, protein folding is thermodynamically
unfavorable in the ER lumen where both protein concentration and synthesis rate are surprisingly high
80, 81
.
Therefore, ER protein folding is easily disturbed by changes in intracellular environment leading to accumulation
of misfolded proteins in the ER or so-called the ER stress
82
. For example, the ER stress can be caused by
chemical/physical insults, nutrient deprivation, hypoxia, impaired protein trafficking and loss of ER calcium
82-84
.
To match the protein folding capacity to the demands, eukaryotic cells have a conserved machinery in response
to the ER stress. This adaptive response is called the ER stress response or the unfolded protein response (UPR)
83
.
When misfolded proteins are accumulated above detectable threshold, three ER transmembrane proteins or so-
called the UPR sensors become activated. The characterized UPR sensors include inositol requiring enzyme 1 α
30

(IRE1α), PRKR-Like Endoplasmic Reticulum Kinase (PERK) and activating transcription factor 6 (ATF6).
Under rest state, luminal domains of the UPR sensors bind to ER resident chaperones, e.g. glucose-regulated
proteins 78 (GRP78). Under the ER stress, misfolded proteins serve as “active ligands” to compete the binding
of UPR sensors to ER resident chaperones leading to conformational changes in each UPR sensor and activation
of their downstream signaling to transduce the UPR signals from the ER to the nucleus
84
.
2. Three Canonical Branches of the UPR
Three UPR sensors govern three canonical braches of the UPR. IRE1a initiates the IRE1a-XBP1 branch of
the UPR. Mammalian IRE1a consists of luminal domains and two enzymatic cytoplasmic domains including a
serine/threonine kinase domain and an endoribonuclease (RNase) domain
83
. Upon activation, the luminal domain
of IRE1a forms multimers to trigger IRE1a auto-phosphorylation which subsequently activates its RNase
domain
85, 86
. IRE1a RNase domain alternatively splices X-box protein 1 (XBP1) by removing a 26nt intron from
its ORF to produce sXBP1 which contains trans-activating domains. Upon translated, sXBP1 is a transcription
factor regulating genes with the unfolded protein response element (UPRE)
87
. The UPRE governs expression of
ER chaperones, ER structural proteins and many members that participate in the ER associated degradation
(ERAD). Hence, IRE1a-XBP1 branch is activated to expand protein folding capacity and promote degradation of
unfolded proteins
42, 83, 87
.
PERK initiates PERK-eIF2a branch of the UPR. Like IRE1a, PERK contains similar luminal domains
turning on PERK kinase activity by formation of multimers followed by autophoshorylation. Activated PERK
phosphorylates the Ser51 of eIF2a which blocks CAP-dependent protein translation upon phosphorylated
89
. This
process causes global protein translational attenuation, and consequently reduces proteins production saving
protein folding capacity for addressing existing unfolded proteins in the ER
88, 89
. However, translation of active
transcription factor 4 (ATF4) and its downstream is facilitated due to unique alternative 5’-proximal elements in
the mRNA reading frame allowing to escapes from peIF2a initiated translational attenuation
90
. As a result, ATF4
31

accumulates upon PERK activation and regulates the expression of stress response genes with cAMP-response
elements (CRE) to accelerate amino acid metabolism, protein synthesis and inhibit cell growth
87
.
ATF6 initiates ATF6 branch of the UPR. Compared to IRE1a and PERK, ATF6 has a unique luminal domain
sensing changes in ER calcium concentration, redox condition and N-glycosylation to directly modulate its
binding to the ER resident chaperones
91-93
. Consequently, ATF6 is particularly inducible by pharmacological ER
stress inducers including thapsigargin (THG) tunicamycin (TUM) and dithiothreitol (DTT)
93
. Unlike the other
two branch, ATF6 requires a trans-organelle activation in the Golgi apparatus. This is initiated by packing the
dissociated ATF6 to COPII cargo followed by anterograde transportation of COPII from the ER to the Golgi
apparatus where ATF6 is subsequently processed by site-1-protease (S1P) and site-2-protease (S2P) to release its
cytoplasmic domain, which is a 55-60kDa cleavage product (ATF6p60)
94
. ATF6p60 or so-called nuclear ATF6
(nATF6) contains bZIP domain to acts as a transcription factor
93
. Genes with ER stress response elements (ESRE)
are the major targets of ATF6 including the most of ER chaperones. Therefore, activation of ATF6 dramatically
expands ER protein folding capacity and benefits restoration of protein homeostasis.
87

UPR acts as “first aid” to reduce protein de novo synthesis and enhance protein folding and degrading
capacities. If adaptive response fails to restore ER homeostasis, the UPR switches on terminal signals to engage
cells death. This usually occurs under prolonged UPR activation. Persistent activation of IRE1a results in
transition of IRE1a multimers to IRE1a high-order oligomers which promote apoptosis mediated by three
pathways. Firstly, oligomerized IRE1a tends to phosphorylate apoptosis signal-regulating kinase 1 (ASK1) and
subsequently activates c-Jun NH2-terminal kinase 1 and 2 (JNK1/2) by phosphorylation
85, 86
. pJNK1/2 leads to
transcriptional up-regulation of pro-apoptotic and inflammatory genes to promote apoptosis
96
. Secondly, activated
JNKs could elevate mitochondria (MITO) calcium loading through modulating ER-MITO calcium flow and cause
fragmentation of MITO to release the pro-apoptotic factors
97
. Thirdly, activity of IRE1a RNase domain could be
amplified during IRE1a oligomerization leading to degradation of anti-apoptotic non-coding RNA unlocking
32

restrain for apoptosis, e.g. microRNA that transcriptionally blocks caspase 2
98
. Prolonged PERK-eIF2a activation
results in accumulation of CHOP, a transcription factor regulated by ATF4 and stimulate transcription of death
receptor 5 (DR5) and BH-3-only protein
99
. Death receptor 5 (DR5) or so-called TRAIL2 is essential for activation
of necroptosis whereas BH-3-only inhibits Bcl-2 to induce apoptosis
99
. Another important pro-apoptotic factor
regulated by CHOP is growth arrest and DNA damage-inducible protein 34 (GADD34)
100, 101
. GADD34 is a
selective phosphatase for eIF2a
100
. Therefore, accumulation of GADD34 releases global translational arrest.
Consequently, the ER would be more severely overloaded by newly synthesized protein causing ATP depletion,
oxidative stress and apoptosis
101
. Moreover, prolonged ER stress reduces ER-to-Golgi trafficking by
downregulating COPII associated components, may reduce ATF6 dependent protection and promote further
cellular injuries
102
,
103
.
3. Role of the ER Stress in HIV Protease Inhibitors Related Injuries
Existence of the ER stress has been reported in HIV protease inhibitors (HIV PIs)-treated in vitro and in vivo
models. In cultured intestinal epithelial cells, treatment with RTV and LPV reduced SEAP secretion while
significantly increased mRNA expression of CHOP and sXBP1 associating to disruption of intestinal permeability.
In cultured macrophage and adipocytes, RTV , indinavar (IDV), atazavair (ATV) and LPV induced lipid
dysregulation correlated with elevated CHOP, ATF4, and XBP1 protein expression suggesting possible
contribution of anti-HIV PIs induced ER stress to atherosclerosis and cardiovascular diseases
59, 60
. In Hela cell
model, RTV , Saquinavir (SQV) and LPV increased CHOP expression while LPV and RTV significantly increased
ATF4 and sXBP1 transcriptional activities with increase in cell death
58
. Nelfinavir (NFV) which induces ER stress
but blocks ATF6 signaling, induces apoptosis in castration-resistant prostate cancer, NSCLC and multiple
myeloma cells
104
. In the liver, treatment with RTV and LPV especially with alcohol increased CHOP and sXBP1
expression and cell death in the primary mouse and human hepatocytes
34, 36
. In mouse model, prolonged injection
of RTV+LPV (RL) resulted in liver steatosis, increased serum ALT, cholesterol and fatty acid concentration
34
.
33

CHOP-/- partially rescued RTV , LPV and RL-induced liver cell apoptosis in mouse model with reduced liver
steatosis and improved liver histology
35
.

Materials and Methods
1. Plasmid and HepG2 Transfection
p3xFLAG-ATF6 and pCGN-ATF6 (1-373) were gifts from Ron Prywes (Addgene plasmid #11975 and #27173).
pcDNA3.1(+)-GRP78/BiP was a gift from Richard C. Austin (Addgene plasmid # 32701). HepG2 cells were
plated on Collagen I coated 12 well plate for 24 hours. Transfection was performed with Lipofectamine® 3000
following the manufacturer’s instructions. In specific, for each well, 1ug of plasmid was mixed with 2ul of P3000
solution in 50ul of OptiMEM medium. In another tube, 1.5ul of Lipofectamine® 3000 transfection reagent was
also mixed with 50ul of OptiMEM medium. After thorough mixing, two mixtures were mixed together and
incubated at room temperature for 20 minutes. Then the 100ul mixture was added to plated cells cultured in 1ml
medium without antibiotics. Overexpression was often achieved after 24 hours and confirmed with western blot.
2. mRNA Extraction and qPCR Assay
After the treatment, the cells were washed with PBS and lysed with Trizol solution. The RNA extraction was
performed with DirectZol total RNA extraction kit (Zymo) following the instructions and RNA quality was
confirmed with 260/280nm absorbance ratio. Then 500ug mRNA was converted to cDNA via High Capacity
cDNA synthesis kit (Thermo) following the instructions. qPCR assay was conducted with PowerSYBR RT-PCR
kit (Life Technology) in ABI 7600HT system. The deltaCT values were obtained and the relative expression was
calculated by RQ manager 1.2 (ABI system). All the primer pairs were synthesized by USC genomic core via
Integrated DNA Technologies Co.

34

Results
1. Selective Activation of Two Canonical UPR Branches but not ATF6 Branch in RL-Treated Cells
Activation of each UPR branch was examined by qPCR and immunoblotting. Phosphorylation of IRE1a was
remarkably detectable on immunoblotting (Fig. 2-1A). In a semi-quantitative analysis, upregulation of pIRE1a to
IRE1a ratio (pIRE1a/IRE1a) by RL treatment at 20µg/ml was comparable with thapsigargin (THG) treatment at
4µM (Fig. 2-1A). At 8 hours, pIRE1a/IRE1a ratio was even higher under RL treatment by 40% (Fig. 2-1A).
Quantitation of sXBP1 transcripts was performed by qPCR assay. Compared to 10µg/ml, RL at 20µg/ml resulted
in significantly faster increases in sXBP1 expression (p<0.0001) while the increases by tunicamycin (TUM) were
relatively equal at 10µg/ml and 20 µg/ml (Fig. 2-1B). In a dose-response analysis with a concentration range 5 to
20µg/ml, RL resulted in 7 to 20-fold increases in sXBP1 mRNA expression compared to control (Fig. 2-1C). In
contrast, smoother sXBP1 increases from 7 to 10 folds were observed in response to TUM at doses from 5 to
20µg/ml (Fig. 2-1C). As one of the downstream targets of IRE1a-XBP1 branch, ER degradation enhancer,
mannosidase alpha-like proteins (EDEM) was also upregulated by 2.3 folds under RL treatment after 16 hours
while the upregulation was more than 4 folds under THG and TUM treatment (Fig. 2-1D). To check activation of
PERK branch, phosphorylation of eIF2a was detected by immunoblotting. Dramatic increase in peIF2a/eIF2a
ratio was observed for cells treated with RL and TUM for 4 hours but contrarily reduced at 8 hours (Fig. 2-1E).
The peIF2a/eIF2a was 24% and 56% stronger under RL compared to TUM treatment at 4 hours and 8 hours,
respectively (Fig. 2-1E). Consistent with peIF2a, accumulation of nuclear ATF4 was detectable in RL and
similarly in TUM-treated cells by immunoblotting (Fig. 2-1E). mRNA expression of CHOP was also
quantitatively studied with qPCR. In time course experiment, RL at 20µg/ml resulted in significantly faster
increases in CHOP expression (p<0.0001) compared to 10µg/ml and the increases by TUM (Fig. 2-1F). In the
dose-response tests with concentrations of 5 to 20 µg/ml, RL resulted in 11 to 123-fold increases in CHOP
expression compared to control (Fig. 2-1G). The ER stress indicated by CHOP expression was dramatically
35

increased when the concentrations of RL were higher than 12.5 µg/ml (Fig. 2-1G). Like sXBP1, smoother CHOP
increases from 22 to 31 folds were observed in response to TUM (Fig. 2-1G). Unlike markers in the other two
canonical UPR branches, upregulation of GRP78 which is the authentic marker for ATF6 branch was undetectable
under RL in either time-course or dose-response experiments demonstrated by immunoblotting and qPCR (Fig.
2-1H&J). In immunoblotting assays, RL abolished upregulation of GRP78 by TUM treatment in a dose-
dependent manner (Fig. 2-1I). In mammalian cells, upregulation of GRP78 relies mainly on activation of the
ATF6 branch. To know whether RL specifically suppressed GRP78 due to ATF6 inhibition, mRNA expression
other downstream markers of ATF6

were investigated in cells under RL treatment. As shown in Fig. 2-1K, GRP78,
p58Ki and PDIA4 were suppressed by 14, 5 and 2 folds, respectively in presence of RL in TUM-treated cells.
However, the suppression was not observable in GADD34, GADD45 and Trib3 that are downstream of the other
two canonical UPR branches indicating that RL branch selectively activated the UPR in HepG2 cells (Fig. 2-1K).
2. Interference of ATF6 Processing and Localization in HepG2 Cells Treated with RL
Activation of ATF6 requires its translocation from the ER to Golgi as well as proteolytic cleavage
95
. To
further confirm an interruption on ATF6 branch, effect of RL on GRP78 expression in cells overexpressing
uncleaved (Flag-ATF6) and fully cleaved (ATF6p60) was studies. In cells overexpressing Flag-ATF6, RL blocked
GRP78 protein expression whereas the expression was unaffected in ATF6p60 overexpressing cells (Fig. 2-2A).
On immunoblotting, intensity of band indicating uncleaved ATF6 was decreased after 4 hours in TUM-treated
cells whereas the band remained unchanged in RL-treated cells indicating an inability in processing ATF6 under
the ER stress (Fig. 2-2B). ATF6 enters nucleus upon cleaved
102
. To assess if ATF6 translocation was also
suppressed by RL, HepG2 cells were transfected with Flag-ATF6 and treated with DMSO, RL or 1,4-dithiothreitol
(DTT) which is known to induce trans-organelle activation of ATF6 (Fig. 2-2C). In transfected cells, nuclear
localizations of ATF6 were observed in more than 90% of the cells treated with DTT but in less than 20% of the
cells treated with RL (Fig. 2-2D). To assess localization of ATF6 in cytoplasmic area, treated cells was stained
36

with ATF6 monoclonal antibody. ATF6 was less co-localized with the Golgi structural protein GM130 in RL-
treated cells than in TUM or THG-treated cells (Fig. 2-2E). RL reduced the Golgi localization of ATF6 by 40%
compared to THG and by 80% compared to TUM, confirming that a suppression of ATF6 Golgi localization and
processing (Fig. 2-2F).
3. Activation of ER Stress Mediated Apoptosis in RL-Treated HepG2 Cells
The UPR contains signaling that modulates survival and death of a cell. Markers downstream of the UPR
that participate in cell death especially apoptosis were examined after treatment with RL for 16 to 24 hours.
Expression of GADD34 and DR5 under PERK branch were significantly up-regulated in RL-treated cell by 28
and 4 folds, respectively (Fig. 2-3A). Compared to RL treatment, TUM upregulated DR5 and GADD34 by 1.5
and 10 folds, respectively (Fig. 2-3A). In consistent with PERK branch, on immunoblotting, pJNK1/2 was
detected after 16 hours in cells under RL treatment whereas it was undetectable in TUM or THG-treated cells
(Fig. 2-3B). To assess contribution from each UPR branch to RL-induced cell death, pharmacological
interventions were conducted by co-administration of UPR branch inhibitors with RL. IRE1a signaling was
inhibited with Toyocamycin (TOYO) and SP600125 (JNKi) that block the formation of sXBP1 and pJNK1/2,
respectively. PERK branch was inhibited with GSK2656157 (PERKi) which blocks PERK kinase activity. ATF6
branch was blocked with PF-429242 dihydrochloride (PF) which is a S1P inhibitor. Treatment with RL increased
rate of cell death while the rate remained unchanged in cells treated with different UPR branch inhibitors (Fig. 2-
3C). In RL-treated cells, co-treatment with TOYO, PERKi and PF further increased the levels of LDH release by
88%, 51% and 20%, respectively suggesting worsened cell death (Fig. 2-3C). In contrast, co-treatment of JNKi
with RL significantly reduced LDH release by 20% along with Caspase 3 (CC3) activity lowered by 17% (Fig.
2-3D&E). Co-treatment of Salubrinal (SLB) which inhibits eIF2a phosphatase further rescued the cells from
death. The level of LDH release in SLB+RL-treated cells was insignificantly different from control indicating a
dramatic protection from RL-induced cell death (Fig. 2-3D). This was also reflected by lowered caspase activities
37

to a larger extent, suggesting a suppression of apoptosis after inhibiting the ER stress (Fig. 2-3E). To assess
whether loss of ATF6 contributed to the cell death during RL treatment, cells overexpressing ATF6p60 were
treated with RL and their death were measured. In ATF6p60-overexpressing cells, LDH release and caspase
activities were significantly reduced by 25% and 32%, respectively, suggesting that the loss of ATF6 promoted
ER stress-induced cellular injury. (Fig. 2-3F& Fig. 2-3G).
4. Selective Activation of the UPR in the Primary Mouse Hepatocytes under RL Treatment
Activation of the UPR was also demonstrated in the primary mouse hepatocytes (PMH). Expression of
sXBP1 and CHOP was upregulated in the PMH in response to RL at concentrations from 12.5 to 20 µg/ml (Fig.
2-4A&B). Treatment with RL at 20µg/ml for 8 hours upregulated CHOP by 30 folds (Fig. 2-4A). Splicing of
XBP1 was detected under RL treatment. The band intensity indicating spliced XBP1 (sXBP1) was increased
whereas the band intensity for un-spliced XBP1 (uXBP1) was decreased (Fig. 2-4B). Compared to RL, TUM
treatment upregulated CHOP by 300 folds. Reduction of uXBP1 was also more robust in THG and TUM-treated
cells (Fig. 2-4A&B). In accordance with HepG2 model, expression of GRP78 at either mRNA or protein level
was unchanged in RL-treated PMH (Fig. 2-4C&D). Co-treatment of RL with TUM suppressed GRP78 mRNA
expression compared to TUM treatment alone indicating an interference with the ATF6 branch (Fig. 2-4E).
Whether JNKi and SLB could reduce RL-induced cell death or not was also assessed in the PMH. Compared to
HepG2 cells, treatment with SLB but not JNKi reduced the level of LDH release by 15% (Fig. 2-3F).

Discussion
In this study, a parallel comparison was made to compare activation of each UPR branch by RL and by potent
pharmacological ER stress inducers, tunicamycin (TUM) and thapsigargin (THG). TUM inhibits N-glycosylation
which is a required process for protein folding and quality control in the ER whereas THG blocks sarcoplasmic
38

reticulum calcium ATPase (SERCA) to reduce ER calcium concentration. In both HepG2 and the primary mouse
hepatocytes (PMH), RL treatment activated IRE1a and PERK branches indicated by corresponding downstream
markers. However, both time-course and dose-dependent tests revealed for the first time that ATF6 was
differentially expressed compared to the other two canonical UPR branches in HepG2 or the primary hepatocytes
in response to RL treatment. This was discovered by deviation in the expression pattern of GRP78 under the ATF6
branch compared to markers under IRE1a and PERK branches. It was further supported by the evidence that RL
treatment inhibited the TUM induced upregulation of GRP78, PDIA4 and p58IPK, which are chaperones
exclusively regulated by ATF6. Also, expression of GRP78 was unaffected in cells expressing ATF6p60 indicating
that RL unlikely blocks transcriptional activity of cleaved ATF6. Therefore, RL possibly affected the proteolytic
processing of ATF6 and thus inhibited the ATF6 branch activation. This was indirectly confirmed with
immunoblotting as uncleaved ATF6 remained unchanged in RL-treated cells whereas TUM which activates the
ATF6 branch apparently reduced the uncleaved ATF6. Translocation of fully cleaved ATF6 to nucleus has been
demonstrated by previous studies using Flag-ATF6 as a marker and DTT as an ER-stress inducer
95
. Hence, we
adopted this system and observed nuclear vs cytoplasmic localization of Flag-ATF6 in RL-treated cells. In this
ATF6 overexpressing system, DTT treatment resulted in more than 90% of the cells with nuclear localization of
Flag-ATF6. In contrast, ATF6 localized in the cytoplasm and failed to translocate to the nucleus in RL- or DMSO-
treated cells. To further specify ATF6 localization in cytoplasm, ATF6 was labeled with ATF6 monoclonal
antibodies. By co-localization analysis with cis-Golgi marker GM130, we found that less ATF6 was localized in
the Golgi apparatus. This suggested the possibility that transportation of ATF6 was affected upon suppression of
the proteolytic cleavage of ATF6 by RL. In brief, delocalization of ATF6 from the Golgi apparatus may interferes
with ATF6 cleavage and subsequent nuclear translocation leading to inactivation of ATF6 branch under RL
treatment.
As described earlier, apoptosis can be induced under prolonged ER stress through accumulation of pro-
39

apoptotic factors GADD34, ERO1a, DR5 and JNK1/2 downstream
82-86
. In HepG2 model, GADD34 and DR5
expression were more remarkably upregulated while phosphorylated JNK1/2 was fairly more detectable in RL-
treated cells than in TUM-treated cells. Expression levels of PERK and IRE1a downstream markers were stronger
under RL treatment. Apoptotic markers, Caspase 3 and Annexin V , expressed more under RL treatment. Therefore,
expression of those UPR pro-apoptotic factor was correlated to apoptotic markers and cell death suggesting that
the ER stress mediated apoptosis could be fully responsible for RL-induced cell death in HepG2 cells. In
supporting this argument, pharmacological treatments that targeting each branch and downstream pro-apoptotic
markers were administered along with RL treatment. RL-induced cell death was markedly worsened when XBP1,
PERK and ATF6 were inhibited and the role of UPR protection was lost. However, specific inhibition of JNK1/2
and eIF2a phosphatases reduce apoptosis and cell death. SLB which inhibited eIF2a phosphatase GADD34
dramatically reduced the cell death rate indicating that premature release of translational arrest under ER stress
was the major mechanism of cell death in RL-treated HepG2 cells.
Studies have shown that GRP78 is central in supporting ER homeostasis and acts more than as a chaperone.
Under ER stress, feedback expression of GRP78 promotes folding capacity of the ER. Meanwhile, initiators of
UPR could restore their binding with GRP78 to become quiescent and prevent prolonged activation of the UPR
that eventually induces apoptosis
105
. In addition, upregulation of GRP78 reduces leakage of calcium from the ER
to cytoplasm consolidating ER calcium homeostasis and chaperone activity
106
. In a previous study, liver specific
deletion of GRP78 in a murine model resulted in dramatic amplification of the ER stress caused by chronic high
fat diet and/or alcohol feeding worsening hepatic apoptosis and liver injuries
39
. Therefore, inhibitions of ATF6
and GRP78 by RL should also worsen the ER stress-induced cell death. As expected, overexpression of either
ATF6p60 or GRP78 that raised GRP78 protein levels reduced caspase activities and dramatically suppressed RL-
induced cell death.
In the PMH, selective activation of the UPR was similarly observed. Compared to HepG2 cells, RL less
40

effectively induced activation of PERK and IRE1a branches than TUM consisting with less apoptosis in RL-
treated PMH. Therefore, treatment with neither SLB nor JNKi significantly rescued RL-induced cell death in the
PMH. This also confirmed that in RL-treated HepG2 cells, cell death was majorly caused by apoptosis mediated
by the ER stress and the situation was different for the PMH in which necroptosis might be the major mechanism
to drive cell death.

Figure Legends
Figure 2-1 Activation of the UPR in RTV+LPV (RL), Tunicamycin (TUM) and/or Thapathigargin (THG)-
Treated HepG2 Cells. (A) pIRE1a and IRE1a expression in RL and THG-treated cells indicated by
immunoblotting. Cells were treated with TUM (20µg/ml), THG 4µM or RL (20µg/ml); quantification was
conducted by ImageJ. (B) Time course of mRNA expression of sXBP1 in RL and/or TUM-treated cells. (C) Dose
response of mRNA expression of sXBP1 in RL and TUM-treated cells. (D) EDEM1 mRNA expression in RL,
THG or TUM-treated cells. (E) peIF2a, eIF2a and ATF4 expression in RL and/or TUM-treated cells indicated by
immunoblotting. (F) Time course of mRNA expression of CHOP in RL and/or TUM-treated cells. (G) Dose
response of mRNA expression of CHOP in RL and/or TUM-treated cells. (H) Immunoblotting indicating GRP78
expression in RL, THG or TUM-treated cells. (I) RL inhibited TUM-induced GRP78 upregulation. (J) Time
course and dose response of mRNA expression of GRP78 in RL and/or TUM-treated cells. (K) mRNA expression
of ATF6-regulated and unregulated factors. mRNA levels were measured 8 hours after each treatment.
****p<0.0001 compared to control or indicated otherwise; n=4-6.
Figure 2-2 RL Affected ATF6 Processing in HepG2 Cells (A) Effects of RL on protein expression of GRP78
in cells transfected with Flag-ATF6 and ATF6p60. Cells were treated with RL for 1 hour followed by transfection
for 16 hours. (B) Immunoblotting of uncleaved ATF6 in HepG2 cells under RL and TUM treatment. (C) Confocal
41

images (63x) showing nuclear localization of ATF6 in the HepG2 cells expressing Flag-ATF6 and under treatment.
DTT is positive control for nuclear localization; DMSO is vehicle control. (D) Quantitation of nuclear localization
of ATF6 in transfected and treated cells. (E) Golgi localization of ATF6. Confocal images (100x) showing changes
of co-localization of ATF6 (green) and the Golgi matrix protein GM130 (red) in the cells exposed to the drugs.
(F) Quantitation of colozalization. **p<0.01; ****p<0.0001 compared to control or indicated otherwise; n=4-7.
Figure 2-3 Contribution of the UPR to RL Induced Cell Death in HepG2 Cells (A) mRNA expression of pro-
apoptotic factors downstream of PERK signaling. Cells were treated for 24 hours. DR5, death receptor 5; ERO1,
ER oxidoreductin 1; GADD34, growth arrest and DNA-damage-inducible 34. (B) pJNK1/2 and JNK1/2 protein
expression in TUM, THG or RL-treated cells. VCL, vinculin was loading control. (C) Inhibition of ATF6, PERK
and IRE1a activity worsened RL-induced cell death. The inhibitors were co-administered with RL for 24 hours
and cytotoxicity was quantified by measuring LDH release that was indicated by relative absorbance unit (RAU).
PERKi, GSK2656157 or PERK inhibitor; TOYO, toyocamycin or IRE1a endonuclease inhibitor; PF, PF-429242
dihydrochloride or S1P inhibitor. (D) Inhibition of RL-induced cell death with JNK inhibitor and eIF2a
phosphatase inhibitor. JNKi, JNK phosphorylation inhibitor; SLB, salubrinal or GADD34 inhibitor. (E) Effects
of JNK inhibitor and eIF2a phosphatase inhibitor on Caspase 3 (CC3) activities in RL-treated HepG2. (F) Effects
of GRP78 or ATF6p60 on RL-induced cell death. (G) Effects of overexpression of GRP78 or ATF6p60 on Caspase
3 (CC3) activities in RL-treated HepG2. *p<0.05; ****p<0.0001 compared to control or indicated otherwise;
n=5-8.
Figure 2-4 Activation of the UPR in the Primary Mouse Hepatocytes under RL Treatment. (A) mRNA
expression of CHOP in the primary mouse hepatocytes (PMH) treated with RL and tunicamycin (TUM). Cells
were treated for 8 hours. RL10/20. RL at 10 or 20µg/ml; TUM at 20ug/ml. (B) Upregulation of XBP1 alternative
splicing in the PMH treated with RL, TUM and thapathigargin (THG); uXBP1, un-spliced XBP1; sXBP1, spliced
XBP1. (C) Immunoblotting of GRP78 protein expression in the PMH treated with RL, TUM or THG. VCL,
42

vinculin was a loading control. (D) GRP78 mRNA expression in the PMH treated with RL, TUM or THG. (E)
Inhibition of GRP78 mRNA expression by RL in TUM-treated PMH. (F) Effects of JNK inhibitor and eIF2a
phosphatase inhibitor on RL-induced cell death. *p<0.05; ****p<0.0001 compared to control or indicated
otherwise; n=5-8.

43

Figures and Tables
Figure 2-1
(A)

DMSO RL4h RL8h THG4h THG8h
pIREa/IREa 1.00 1.79 2.51 1.89 1.98

(B)

T im e (h o u rs )
R e la tiv e s X B P 1 m R N A le v e ls
(fo ld o f c h a n g e )
0 2 4 6 8
0
5
1 0
1 5
R L a t 1 0 u g /m l
R L a t 2 0 u g /m l
T U M a t 1 0 u g /m l
T U M a t 1 0 u g /m l
****
44

Figure 2-1
(C)

(D)

C o n c e n tra tio n (  g /m l)
R e la tiv e s X B P 1 m R N A le v e ls
(fo ld o f c h a n g e )
5 .0 1 0 .0 1 2 .5 1 5 .0 1 7 .5 2 0 .0
0
5
1 0
1 5
2 0
2 5
R L
T U M
*
***
**
45

Figure 2-1
(E)

DMSO RL4h RL8h TUM4h TUM8h
pIREa/IREa 1.00 2.35 1.86 1.39 0.81

(F)

T im e (h o u rs )
R e la tiv e C H O P m R N A le v e ls
(fo ld o f c h a n g e )
0 2 4 6 8
0
5 0
1 0 0
1 5 0
R L a t 2 0  g /m l
R L a t 1 0  g /m l
T U M a t 1 0  g /m l
T U M a t 2 0  g /m l
****
46

Figure 2-1
(G)

(H)

(I)

C o n c e n tra tio n (  g /m l)
R e la tiv e C H O P m R N A le v e ls
(fo ld o f c h a n g e )
5 .0 1 0 .0 1 2 .5 1 5 .0 1 7 .5 2 0 .0
0
5 0
1 0 0
1 5 0
R L
T U M
**
****
****
**
TUM
47

Figure 2-1
(J)

T im e (h o u rs )
R e la tiv e G R P 7 8 m R N A le v e ls
(fo ld o f c h a n g e )
0 2 4 6 8
0
5
1 0
1 5
R L a t 1 0 u g /m l
R L a t 2 0 u g /m l
T U M a t 1 0 u g /m l
T U M a t 1 0 u g /m l
C o n c e n tra tio n (  g /m l)
R e la tiv e G R P 7 8 m R N A le v e ls
(fo ld o f c h a n g e )
5 .0 1 0 .0 1 2 .5 1 5 .0 1 7 .5 2 0 .0
0
5
1 0
1 5
R L
T U M
**
*
48

Figure 2-1
(K)

49

Figure 2-2
(A)

(B)

50

Figure 2-2
(C)

(D)

51

Figure 2-2
(E)

(F)

A re a o f A T F 6 O v e rla p w ith
G o lg i A p p a ra tu s (% )
D M S O
T U M
T H G
R L
R L + T U M
0
2 0
4 0
6 0
8 0
1 0 0
****
****
**
DMSO THG TUM RL
52

Figure 2-3
(A)

(B)

53

Figure 2-3
(C)

(D)

M e d iu m L D H L e v e l (R A U )
D M S O
R L
P E R K i
T O Y O
P F
R L + P E R K i
R L + T O Y O
R L + P F
0 .0
0 .2
0 .4
0 .6
0 .8
1 .0 ****
*
****
M e d iu m L D H L e v e l (R A U )
D M S O
R L
J N K i
S L B
R L + J N K i
R L + S L B
0 .0
0 .1
0 .2
0 .3
0 .4
0 .5
****
*
54

Figure 2-3
(E)

(F)

55

Figure 2-3
(G)

56

Figure 2-4
(A)

(B)

THG TUM
57

Figure 2-4
(C)

(D)

R e la tiv e G R P 7 8 m R N A le v e ls
(fo ld o f c h a n g e )
C tr l R L 1 0 R L 2 0 T g T m
0
1 0
2 0
3 0
****
****
58

Figure 2-4

(E)

(F)

M e d iu m L D H L e v e l (R A U )
D M S O R L R L + J N K i R L + S L B
0 .0
0 .2
0 .4
0 .6
0 .8
*
59

Chapter 4 Mechanistic studies on Ritonavir and Lopinavir Induced ER stress –
Proteasome Inhibition, Calcium Depletion and Production of Reactive Oxygen
Species
Literature Review
1. ER Stress Induced by Pharmacological Compounds
Many drugs in the modern pharmacopeia have been shown to induce the ER stress. In general, any chemicals
that interfere with protein folding could induce the ER stress including but not limited to agents that cause ER
calcium disturbance, oxidative stress, energy depletion, glucose starvation and lipid accumulation
40, 41
. Some
typical ER stress inducers include, thapsigargin (THG), tunicamycin (TUM), dithiothreitol (DTT), brefeldin A
(BFA) as well as various peptidomimetic anti-HIV PIs. Mechanisms for some of those ER stress inducers have
been established previously. For example, THG is a non-competitive inhibitor of sarcoplasmic reticulum calcium
ATPases (SERCAs) which exclusively maintains high concentration of calcium ion in the ER lumen
34
. TUM
inhibits Glc-NAc phosphotransferase (GPT) to block N-glycosaltion and could disrupt the protein quality control
system
39
. DTT is a small-molecule redox reagent containing reduced sulfate group which could break
intramolecular and intermolecular disulfide bonds within proteins
92
. BFA impairs Golgi-to-ER trafficking leading
to Golgi fragmentation and fusion of Golgi materials with the ER to induce the ER stress
82
. However, the precise
mechanism (s) for anti-HIV PIs induced ER stress have been controversial as anti-HIV PIs vary in structures and
off-targets. In published literatures, three mechanisms have been frequently described to explain anti-HIV PIs
induced ER stress including proteasome inhibition, ER calcium perturbation and oxidative stress.
2. Three Possible Mechanisms for Anti-HIV PIs Induced ER Stress
First of all, proteasome inhibition by HIV PIs has been widely proposed as a mechanism for the ER stress as
proteasome is one of the common off-targets for many anti-HIV PIs. In different in vitro models, inhibition of
60

proteasome by anti-HIV PIs has been demonstrated. In myocardial cells, exposure of RL significantly lowered
chymotrypsin-like and caspase-like proteasomal activities. In hepatocyte, RTV treatment induced accumulation
of viral components which are degraded by proteasome machinery
110
. NFV and RTV block degradation of Akt
which requires ubiquitin-proteasome system (UPS) for degradation
111
. In adipose tissue, RTV and SQV interfere
UPS degradation of apolipoprotein B and SREBP possibly enhancing lipid production
112
. Under proteasome
inhibition, misfolded proteins formed are failed to be removed and accumulated possibly leading to the ER
stress
107
. Indeed, potent proteasome inhibitors MG132 and bortezomib kill cancer cells by evoking apoptosis
partially mediated by this mechanism
108
. Hence, proteasome inhibition by anti-HIV PIs may contribute to the ER
stress with the same mechanism.
Secondly, some anti-HIV PIs especially RTV and LPV possibly alters ER calcium to induce ER stress. RTV ,
LPV and RL exposures were highly associated with loss of ER calcium measured by small ligand-evoked transient
rise of cytoplasmic calcium
34, 59, 60
. Treatment with RTV and LPV for 24 hours resulted in reduction of glutamate
or THG-evoked transient rise of cytoplasmic calcium in the primary neuron cells
59
. In non-differentiated 3T3-L1
cells, RTV plus LPV synergistically promotes loss of ER calcium compared to single treatment of RTV or LPV
60
.
In the primary mouse hepatocytes, RL caused calcium perturbation and ER stress likely through inhibiting
expression of SERCAs
34
. However, how SERCAs interact with RL has been difficult to be demonstrated.
The third mechanism for the ER stress induced by anti-HIV PIs is oxidative stress. In the liver, anti-HIV PIs
are metabolized by cytochrome c P450 (CYP) system which produces reactive oxygen species (ROS). Protein
folding is a redox process, therefore, ROS and superoxide (SO) could interfere protein folding especially with
formation of disulfide bonds
92, 114
. In Hela cell model, treatment with LPV at 40µg/ml resulted in ROS production
and JNK1/2 activation leading to apoptosis
58
. In the primary hepatocytes, RL treatment blocked nuclear
translocation of NRF2, which is an essential process for activating anti-oxidant response
36
. In the primary human
skeletal cells, natural anti-oxidant resveratrol down-regulated CHOP which is the potent ER stress marker under
61

RL and THG treatment
115
.
RTV and LPV may also have multiple off-targets especially when they are co-treated as RL. Therefore, the ER
stress induced by RL could through one or multiple mechanisms. In this chapter, we initiated a study to assess
whether proteasome inhibition, ER calcium perturbation and oxidative stress are correlated to the ER stress in
RL-treated liver cells.

Materials and Methods
1. Assay on Intracellular Calcium Level
A graphic explanation is indicated in Fig. 3-1. Calcium distribution before dye loading was indicated in Fig.
3-1 A. Labeling of intracellular calcium was accomplished by Fluo-4AM Direct Calcium Assay kit (Thermo).
Fluo-4 is a fluorescent compound that increases its fluorescent emission intensity by more than 100 folds upon
binding to calcium ions. Fluo-4 included in the kit was synthesized as Fluo-4 chemically conjugated with
acetoxymethyl (AM) to form Fluo-4AM. This conjugation with AM not only prevents binding of calcium to Fluo-
4 but also increase its permeability to lipid bilayer. When the dye molecules were loaded and diffused into cells,
Fluo-4AM molecules were digested by esterase to form Fluo-4. Fluo-4 molecules were membrane impermeable
and locked in cellular compartments that they reached (Fig. 3-1B). The compartments mainly include cytoplasm
but also the ER. In this study, treated or untreated cells seeded in 96 well plate were loaded with Fluo-4AM
reagent dissolved in HBSS for 30 minutes at 37
o
C. Before reading fluoresce, 10mM EGTA was added to block
fluorescence due to binding of extracellular calcium to leaked Fluo-4 (Fig. 3-1C). Then fluorescence at
Ex/EM=490/520 was measured. The fluorescent readout was proportional to total calcium levels in all
intracellular compartment
116
.
2. Thapsigargin (THG) Induced Calcium Discharge Assay
As calcium is a secondary messenger, its concentration in cytoplasm is maintained at nanomolar level at all
62

time
117
. This is accomplished by intake of calcium from cytoplasm to the ER. This process is conducted by active
transportation of SERCAs which collect calcium that passively diffused from the ER to cytoplasm
118
., SERCAs
are blocked under THG treatment, therefore, large amount of calcium is passive transported through translocon
pores resulting rise of cytoplasmic calcium
116-118
. In Fluo-4 loaded cells, as most of the dye molecules were located
in cytoplasm, calcium ions diffused to cytoplasm could bind to Fluo-4 giving rapid increase in fluorescent
intensity relative to background (Fig. 3-1D)
116
. In response to rise in calcium concentration, fast acting sodium
calcium exchangers (NCXs) start to work immediately to remove extra cytoplasmic calcium
119
. Meanwhile, slow
acting plasma membrane calcium ATPases (PMCAs) also remove calcium by active transportation
119
. Hence,
cytoplasmic calcium rise is counteracted leading to decrease in fluorescent intensity (Fig. 3-1E). As a result, in
Fluo-4 loaded cells, THG treatment evokes a transient rise of cytoplasmic calcium indicated by a peak (or pulse)
in the plot of fluorescent intensity with time. The amplitude of the peak is proportional to concentration
differences between cytoplasm and ER calcium concentration at the particular time point
116
. Hence, THG induced
calcium discharge assay has been a useful measure for ER calcium perturbation in a quantitative way. In this study,
cells treated or untreated were loaded with Fluo-4AM reagent dissolved in HBSS for 30 minutes at 37
o
C. Then
EGTA was added to block extracellular background calcium fluorescent. Meanwhile, measurement of fluorescent
intensity was started. For each well, 10 different areas were measured and averaged. The fluorescent levels at this
stage was the total intracellular calcium level indicated by previous background and used as baseline for each
samples. After 10 measurements, 10µM THG was injected within 5s and the fluorescent intensity was recorded
till the intensity was decreased close to the baselines for each sample.
3. Cyclohexamide (CHX) Chasing Assay
CHX blocks overall protein translation. Cells plated on 12-well plate were treated with 20µg/ml CHX for 2
hours and whole proteins were collected as control or 0 hour. For some cells, following CHX treatment for 2
hours, RL or DMSO was treated and proteins were collect after 4, 8, 16 and 24 hours. The collected proteins were
63

analyzed by Western blot. The band intensity was measured by ImageJ and normalized with GAPDH.
4. ROS/SO Quantification Assay and ROS Staining
ROS/SO quantification was performed with ROS/SO detection kits (Enzo) according to manufacturer’s
instruction. Cells plated on 96 well plate were treated for indicated time. ROS and SO substrates were added,
respectively and cells were incubated 37
o
C for 1 hour. Fluorescence at EX/EM=490/520nm and 550/620 nm for
ROS and SO, respectively. In the primary mouse hepatocytes (PMH), treated cells were loaded with CellROX
TM
staining kit (Thermo) according to manufacturer’s instruction. Cells were washed with DPBS and observed under
fluorescent microscope with FITC filter. The fluorescent images were processed by ImageJ.

Results
1. Weak Association between Proteasome Inhibition and the ER Stress under RL Treatment
Inhibition of proteasome activity by RTV+LPV (RL) and/or MG132 was indicated by levels of protein
ubiquitination assessed by immunoblotting. MG132 is a potent proteasome inhibitor. In HepG2 cells, treatment
with MG132 for 8 hours apparently resulted in stronger expression of ubiquitin tagged proteins. (Fig. 3-2A). In
contrast, RL treatment neither resulted in changes in protein ubiquitination levels nor affected MG132 induced
accumulation of ubiqutinated proteins (Fig. 3-2A). UPR marker CHOP was limitedly upregulated by 7.9 folds
under MG132 treatment whereas it was strongly upregulated by more than 100 folds under RL treatment (Fig. 3-
2B). Therefore, protein ubiquitination level which reflexes proteasome activity is not correlated to RL-induced
ER stress.
2. Prolonged RL Treatment Leads to Calcium Loss Possibly due to SERCA Degradation
Loss of calcium from cells to extracellular space reflects ER calcium perturbation. To identify whether RL
resulted in loss of intracellular calcium, intracellular calcium fluorescence was measured after Fluo-4 loaded
64

HepG2 cells. By THG treatment at 50nM and 500nM, total intracellular calcium fluorescence was reduced by 45
and 50%, respectively (Fig. 3-3A). When treated with RL, calcium fluorescence was unchanged unless the
concentration was elevated to 40µg/ml at which RL resulted in reductions of the calcium fluorescent by 11%~15%
relative to control. (Fig. 3-3A). Next, the acute effect of RL on ER calcium was measured by its ability in evoking
transient cytoplasmic calcium rise in Fluo-4 loaded cells. In positive control cells, treatment with 500nM THG
initiated fluorescent peaks for which the average amplitudes were 10 times higher than the background within
100s indicating a temporal movement of calcium from the ER to cytoplasm (Fig. 3-3B). However, fluorescent
peaks were not observed in RL-treated or negative control cells within 15 minutes indicating no acute effects of
RL on ER calcium level. (Fig. 3-3B). In contrast to acute effects, chronic effects RL on ER calcium was
determined by measuring amplitudes of fluorescent peaks in Fluo-4 loaded cells under THG treatment. THG
induced calcium fluorescent peaks are proportional to ER calcium concentration. In time course experiments, RL
treatment at 20µg/ml for 2-4 hours resulted in 20% taller peaks compared to control. However, 12% shorter peaks
were observed when the treatment time was extended to 8 hours indicating a mild loss of ER calcium. In dose
response experiments, RL treatment at 40µg/ml and 80µg/ml resulted in reduced fluorescent peaks by 46% and
66%, respectively, indicating a significant loss of ER calcium (Fig. 3-3D). Therefore, RL did have chronic effects
to reduce ER calcium especially at high concentration. Chronic/gradual perturbation of calcium is strongly
connected to expression of SERCAs because SERCAs exclusively manage the ER-cytoplasm calcium
exchange
118, 119
. Downregulation of SERCAs proteins under RL treatment has been confirmed in the liver cell and
mouse models during our previous studies
34
. To investigate the mechanism of SERCA proteins downregulation,
mRNA expression of SERCA2, the major SERCA isoform in liver cells, was measured in RL-treated liver cells.
Relative to control, both THG and tunicamycin (TUM) treatments upregulated the expression of SERCA2 by 2.5
and 4 folds, respectively (Fig. 3-3E). In contrast, expression of SERCA was unaffected by RL treatment. In this
case, SERCA2 protein downregulation may due to changes in protein turnover or protein degradation rate. To
65

prove this point, SERCA2 protein degradation rate was assessed by cyclohexamide (CHX) chasing assay. By
quantification of SERCA2 bands on immunoblotting, in presence of CHX, SERCA2 proteins degraded constantly
at 0.5% per hour within 24 hours in HepG2 cells (Fig. 3-3F). When treated with RL at 20µg/ml in presence of
CHX, SERCA2 protein degradation rate was elevated to 1.8% per hour after 8 hours in HepG2 cells indicating
increase in protein turnover rate (Fig. 3-3F). Therefore, prolonged RL treatment resulted in downregulation of
SERCA2 by increasing its protein turnover in HepG2 cells.
3. Oxidative Stress Was Evoked in RL-Treated PMH but Did Not Contribute to the ER Stress.
Oxidative stress produces reactive oxygen species (ROS) and superoxide (SO) species. To confirm when
oxidative stress was evoked, production of ROS and SO was measured by ROS/SO assay with microplate assay
or ROS staining in RL-treated liver cells. In HepG2 cells, microplate assay indicated that conversion of
fluorescent substrate by ROS was increased by 2 folds at 16 hours and 1.5 folds at 24 hours in RL-treated cells
(Fig. 3-4A). In contrast, limited changes were observed for SO fluorescence (Fig. 3-4B). Therefore, in HepG2
cells, ROS production must be a later event compared to the UPR which was activated before 2 hours (Fig. 2-1).
In the PMH, ROS production was demonstrated by staining with CellROX green fluorescent dye described in
Materials and Methods. In Fig. 4-4C, a considerable increase in green ROS fluorescence was observed in the
PMH after RL treatment for 4 hours possibly contributing to the ER stress. To confirm cause effect relationship
between ROS production and the ER stress, free radical scarvaging by N-acetyl-cysteine (NAC) was performed
and CHOP expression was measured. Unfortunately, CHOP expression was unaffected under RL and NAC co-
treatment (Fig. 4-4D). Hence, oxidative stress was not a major contributor to RL-induced ER stress in liver cells.

Discussion
There are many ways to induce the ER stress and anti-HIV PIs may target many biomolecules. Many anti-HIV
66

PIs targeted the proteasome. In the liver cells, immunoblotting of global protein ubiquitination indicated that RL
treatment did not affect protein ubiquitination while MG132 induced remarkable accumulation of ubiquitinated
proteins. However, CHOP mRNA expression was more robustly upregulated by RL treatment than by MG132.
Therefore, proteasome inhibition was not a primary contributor for the ER stress.
In this study, we reconfirmed that RL could lead to calcium perturbation. However, RL did not acutely
affected ER calcium upon treatment, like what was caused by thapsigargin (THG). Instead, RL-induced calcium
perturbation was less noticeable unless treatment concentration was elevated 40µg/ml or the time was extended
to 8 hours. In this case, RL was unlikely to target SERCA activity like THG. Instead, RL probably affected SERCA
expression. On transcriptional level, SERCA mRNA expression was unaffected because RL blocked ATF6 branch
which upregulated SERCA under the ER stress. At protein level, SERCA protein expression was significantly
downregulated with extended treatment time. SERCA is a stable and long live protein with half-life of 2-3 days
120
.
However, protein oxidation and subsequent ubiquitination could result in accelerated SERCA degradation
121
. By
CHX chasing assay, it was indicated that the rate of SERCA degradation was increased after 8 hours of RL
treatment at 20µg/ml. This was chronologically correlated to loss of ER calcium indicating that RL caused ER
calcium perturbation through accelerated SERCA degradation.
ROS production was detectable in both HepG2 and the PMH under RL treatment. In HepG2 cells, ROS was
detectable after 16 hours while the time was earlier in RL-treated PMH. The difference was possibly because
Cytochrome C P450 (CYP) system was more active in the PMH
34, 122
. To study the impact of ROS production
on the ER stress, ROS was scavenged by cell-permiable anti-oxidant NAC. However, co-administration of NAC
did not affect RL-induced ER stress indicating that ROS production was not associated with the ER stress
induction.
In sum, both proteasome inhibition and oxidative stress were not well correlated to induction of ER stress in
liver cell under RL treatment. ER calcium perturbation requires prolonged RL treatment. Therefore, all the
67

currently proposed mechanisms may not be the major contributor for RL-induced ER stress in the liver cells.

Figure Legends
Figure 3-1 Graphic Demonstration for Utilizing Fluo-4AM to Monitor Thapsigargin (THG)-evoked
Transient Cytoplasmic Calcium Rise. (A) Calcium distribution before dye loading. (B) Fluo-4 AM (F4-AM) is
transported into cytoplasm and converted to Fluo-4 (F4) which could bind to intracellular calcium to form Fluo-
4-Ca
2+
fluorescent complex (F4-Ca+). (C) Use EGTA to block extracellular calcium level; EGTA, ethylene
glycol-bis (β-aminoethyl ether) -N,N,N',N'-tetraacetic acid. (D) THG evokes transient calcium release from ER
to cytoplasm; leaked calcium binds to F4 in cytoplasm where the dye concentrates leading to increase in
fluorescent signal. (E) Cytoplasmic calcium recovers to rest state due to transportation of calcium to extracellular
space.
Figure 3-2 Proteasome Inhibition and ER stress Induction by RL or MG132. (A) Immunoblotting of
ubiquitinated proteins in RL and MG132-treated cells; HepG2 cells were treated for 8 hours and whole proteins
were extracted; GAPDH, Glyceraldehyde 3-phosphate dehydrogenase was used as loading control. (B) CHOP
mRNA expression in RL and MG132-treated cells. HepG2 cells were treated for 8 hours. Expression data was
normalized with 18s rRNA. ****, p<0.0001, n=6.
Figure 3-3 Calcium Perturbation in RL or Thapsigargin (THG)-Treated Cells. (A) Total intracellular calcium
level in cells under treatment of RL or THG. HepG2 cells plated in 96 well plates were treated and loaded with
Fluo-4 followed by EGTA treatment. Calcium fluorescent was recorded by microplate reader. Percent of reduction
was calculated relative to DMSO-treated samples. (B) Acute effect of RL on ER calcium level. Cells were loaded
with Fluo-4AM followed by injecting with DMSO, RL and THG to final concentration of 0.6%, 20µg/ml and
500nM. Time course fluorescence was measured by microplate reader every 10s (left). Amplitude of peaks in
68

relative cytoplasmic calcium signal was summarized in the right figure. (C) Amplitude of peaks in THG evoked
cytoplasmic calcium rise in cells under RL treatment for different time period. Cells were treated with 20µg/ml
and loaded with Fluo-4AM for 30 minutes. Transient rise in cytoplasmic calcium was evoked by injection of THG
to final concentration of 10µM. The amplitude of peaks was recorded and plotted. (D) Amplitude of peaks in
THG evoked cytoplasmic calcium rise in cells under RL treatment at different concentrations. Cells were treated
with indicated concentration for 4 hours; RL20, 40 80 stand for RL at 20, 40 and 80µg/ml, respectively. (E)
SERCA2 mRNA expression in RL, THG and tunicamycin (TUM)-treated cells. Cells were treated for 8 hours;
SERCA2, sarcoplasmic reticulum calcium ATPase isoform 2. (F) SERCA protein expression under DMSO and
RL treatment in cyclohexamide (CHX) chasing assay. Cells were treated in presence CHX for indicated time.
Protein expression was measured by immunoblotting. *p<0.05; **p<0.01; ***p<0.005; ****p<0.0001 compared
to control (DMSO treatment) or indicated otherwise; n=6-10.
Figure 3-4 Oxidative Stress in RL-treated Liver Cells and its Role in RL Induced ER Stress
(A) Measurement of reactive oxygen species (ROS) production in RTV+LPV (RL)-treated cells. Cells were
treated with RL at 20µg/ml for indicated time and ROS production was measured by fluorescence at
ex/em=490/520; TBHP, tert-Butyl hydroperoxide, positive control for ROS production. (B) Measurement of
superoxide (SO) production in RL-treated cells. Cells were treated with RL at 20µg/ml for indicated time and SO
production was measured by fluorescence at ex/em=555/590; AMA, Antimycin A, positive control for SO
production. (C) ROS staining in the primary mouse hepatocytes treated with RL. Cells were treated with RL at
20µg/ml for indicated time and loaded with CellROX dye. Pictures were taken with fluorescent microscope at
40X. (D) CHOP mRNA expression in cells under co-treatment of RL and N-acetyl cysteine (NAC). NAC
neutralizes ROS production. Cells were treated for 4 hours. ****, p<0.0001, n=6-8.

69

Figures and Tables
Figure 3-1

(A) (B)
( C )
( D )
( E )
70

Figure 3-2
(A)

(B)

R e la tiv e C H O P m R N A le v e ls
(fo ld o f c h a n g e )
D M S O
R L
M G 1 3 2
R L + M G 1 3 2
0
5 0
1 0 0
1 5 0
2 0 0
****
71

Figure 3-3
(A)

(B)

C e llu la r C a lc iu m L e v e l
R e la tiv e to D M S O (% )
T H G 2 h rs R L 2 h rs R L 4 h rs R L 1 6 h rs
0
5 0
1 0 0
1 5 0
0 5 5 0 5 0 0
(n M )
0 1 0 2 0 4 0
(u g /m l)
0 1 0 2 0 4 0
(u g /m l)
0 1 0 2 0 4 0
(u g /m l)
T re a tm e n t
C o n c e n tra tio n
** * *
T im e (s )
R e la tiv e C y to p la s m ic
C a lc iu m S ig n a l (R F U )
2 4 0 4 8 0 7 2 0 9 6 0
-0 .2
0 .0
0 .2
0 .4
0 .6
D M S O
T H G
R L
R e la tiv e C y to p la s m ic
C a lc iu m P e a k (R F U )
D M S O T H G R L
0 .0
0 .2
0 .4
0 .6
****
72

Figure 3-3
(C)

(D)

R e la tiv e C y to p la s m ic
C a lc iu m P e a k (R F U )
D M S O 2 h r s 4 h r s 8 h r s 1 6 h r s
0 .0
0 .5
1 .0
1 .5
****
**
**
*
R e la tiv e C y to p la s m ic
C a lc iu m P e a k (R F U )
D M S O R L 2 0 R L 4 0 R L 8 0
0 .0
0 .5
1 .0
1 .5
****
****
*
73

Figure 3-3
(E)

(F)

R e la tiv e S E R C A 2 m R N A le v e ls
(fo ld o f c h a n g e )
D M S O T H G T U M R L
0
1
2
3
4
5
****
****
T im e (h rs )
S E R C A p ro te in e x p re s s io n re la tiv e
to n o n e -tre a te d c e lls (% )
0 4 8 1 2 1 6 2 0 2 4
0
2 0
4 0
6 0
8 0
1 0 0
R L
D M S O
74

Figure 3-4
(A)

(B)

R O S F lu o re s c e n c e (K U )
N E G
D M S O
R L 2 h
R L 4 h
R L 8 h
R L 1 6 h
R L 2 4 h
T B H P
0
1 0 0
2 0 0
3 0 0
8 5 0
9 0 0
9 5 0
****
****
****
S u p e ro x id e F lu o re s c e n c e (K U )
N E G
D M S O
R L 2 h
R L 4 h
R L 8 h
R L 1 6 h
R L 2 4 h
A M A
0
2
4
6
8
1 0
5 0
5 5
6 0
6 5
7 0
****
75

Figure 3-4
(C)

(D)

DMSO RL 2hrs
RL 4hrs RL 8hrs
76

Chapter 5: A Novel Potential Mechanism for the Drug and Alcohol Induced ER
Stress and Injuries - Involvement of Abnormal Golgi Morphology, Function and
ER-to-Golgi Transportation
Literature Review
1. Mechanisms and Machineries in ER-Golgi Trafficking
One of the popular mechanisms for drug induced ER stress is disruption of ER-Golgi trafficking which is
responsible for intracellular distribution and excretion of biomolecules
82-86
. The Golgi apparatus is a dynamic
organelle that is closely related to ER and its structure is maintained by ER-Golgi trafficking. ER-to-Golgi
trafficking or anterograde transportation moves newly synthesized proteins and lipids to Golgi for processing,
sorting and redistribution
44
. Meanwhile, Golgi-to-ER trafficking or retrograde transportation ensures the recycle
of lipids, fluids, and escaped ER proteins
44
. Coat protein complex II (COPII) vesicle is the only identified
machinery for anterograde transportation
123
. COPII-dependent trafficking or COPII trafficking is initiated by
interaction of GDP bound SAR1 with exchange factors (e.g. Sec12) to exchange its bound GDP for GTP, or so-
called GDP-GTP exchange
124
. This process highly depends on the activities of protein kinase A (PKA), therefore,
H89 which is a potent PKA inhibitor has been used to block ER cargo exportation to Golgi or ER export
125, 126
.
Upon activated, SAR1 which locates at the ER membrane recruit coatomers (e.g. Se24, Sec31) which are to form
cage-like oligomeric vesicles on surface of the ER
127
. Sec24 contains export signals coupling the vesicles to ER
exist sites (ERES) which are long-lived subdomains of the ER and the COPII are enriched at
124
. In contrast to
anterograde transportation, retrograde transportation can be mediated by either coat protein complex I (COPI)
dependent vesicular or COPI independent tubular trafficking
44, 123
. Similar to COPII, COPI is also initiated by
the process of GDP-GTP exchange. The Golgi membrane resident protein ARF1 recruits a machinery different
from COPII and selectively include cargo proteins with KKXX motif which is contained in the protein sequence
77

of ER chaperones124. With a completely different machineries and sorting processes, fidelity and directionality
of ER-Golgi bidirectional trafficking are well maintained44. COPI also involves in inter-compartmental
transportation and maturation of the Golgi
129
. Therefore, inhibition of COPI assembly by brefeldin A (BFA)
results in rapid collapse of rigid Golgi cisternae and induces Golgi fragmentation
129
. Golgi tubular trafficking,
which is an alternative form of retrograde transportation, has been characterized for less extent. However, it has
been implicated that tubular carrier mediates bulk flow of lipids, fluid and some Golgi resident enzymes (e.g. α-
mannosidase II)
130
. Unlike COPI and COPII, Golgi tubular trafficking does not require GDP-GTP exchange but
depends on PKA activities and RAB6A family proteins which coordinate Golgi membrane structure
131-133
. Since
neither COPI nor COPII was involved, inhibition of the Golgi tubular trafficking by means of genetic knockdown
minimally altered Golgi appearance
140
.
2. Induction of ER Stress under Impaired ER-Golgi Trafficking
ER-Golgi trafficking is closely linked to protein homeostasis. Inhibition of COPII overloads the ER and results
in insufficient protein glycosylation which contributes to the tertiary structure of proteins
44
. Inhibition of COPI
leads to Golgi fragmentation and loss of primary Golgi functions
44
. To maintain protein homeostasis between the
ER and the Golgi, kinetics of COPI and COPII are well balanced to maintain the essential components required
for vesicle docking and cargo recognition (e.g. SNARE proteins) to be properly recycled and evenly distributed
across the two organelles
123
. As a result, inhibition of either COPI or COPII collapses the bidirectional trafficking
contributing to the ER stress activating the UPR
44, 123
. For example, brefeldin A (BFA) which inhibits COPI also
abolishes ER-to-Golgi trafficking by collapsing ER export possibly due to depletion of COPII tethering factors
44,
82
. Therefore BFA has been a modeled drug to induce ER stress in vitro
82
. In another example, overexpression of
dominant negative form of SAR1 which inhibits SAR1 GDP-GTP exchange and COPII assembly induces the ER
stress contributing to apoptosis in pancreatic beta cells in diabetic model
134
. Moreover, in amyotrophic lateral
sclerosis patients, mutations in TDP-43, FUS and SOD1 affects RAB1 function leading to impaired ER-to-Golgi
78

trafficking associated to the ER stress and degeneration in neuron
135
.
There was no clue on how RL affects the Golgi apparatus. Based on the previous results on ATF6 dislocation
from the Golgi under the ER stress in presence of RL, we hypothesized that disrupting ER-Golgi trafficking
underlies alcohol and RL-induced ER stress and liver injuries. Therefore, in this chapter, we prove this hypothesis
by carefully demonstrating the effects of RL on Golgi morphology, function as well as ER-Golgi transportation
by using HepG2, the primary mouse hepatocyte and mouse models. Alcohol induces Golgi fragmentation in the
primary rat hepatocytes. Therefore, synergic effects of alcohol and RL on Golgi fragmentation, the ER stress and
liver injuries were investigated in addition.

Materials & Methods
1. siRNA Transfection
TFE3 siRNA and scrambled siRNA were purchased from ThermoFisher Scientific. Briefly, siRNA was diluted
at 30 pmol/200ul OptiMEM and mixed with 3ul of Lipofectamine® 3000 at room temperature for 15 min, to
which 0.1 million of HepG2 cells in 1 ml of antibiotics-free complete medium was added. The resulting cell and
mixture was incubated in 12-well plates coated with Collagen I for 40 hours. The gene knockdown was confirmed
by RT-PCR and immunoblotting.
2. Immunofluorescent Staining, Confocal Microscopy and Quantification
For in vitro experiments, cells were attached to 18 mm round-glass cover slides coated with Collagen I and
cultured in 12-well plates. After the drug and/or alcohol treatments, cells were washed with ice-cold PBS, fixed
in 4% parafolmaldehyde (PFA) solution for 15 minutes, permeabilized in 2% Triton-X 100 in PBS for 15 minutes,
and then blocked with 10% goat serum in PBS with 0.05% Tween 20 (PBST). The fixed cells were probed with
rabbit polyclonal anti-GM130 antibodies (Genetex, 1:250) and/or mouse monoclonal anti-MAN2A1 antibodies
79

(Santa Cruz, 1:50), which were diluted in the blocking buffer overnight. The cells on the slides were then labeled
with secondary antibodies conjugated with fluorescence and/or nuclear counter-stained with 10 ng/ml Hochest
blue in PBS. After sequentially washed with PBST, PBS and ddH2O, the cover slides were mounted on glass
slides with prolonged gold mounting media. For in vivo experiments, liver tissues were embedded in O.C.T., snap
frozen, sectioned at 5μm, and mounted on glass slides, which were fixed and permeabilized in ice-cold acetone
at -20
o
C for 10 min, air dried at room temperature for 30 min, and then processed with the same staining method
as in vitro experiments. The cis-Golgi structure of liver cells was indicated by the immunostaining of GM130
whereas all Golgi regions (cis, medial and trans) indicated by the immunostaining of MAN2A1. The color images
were taken with Zeiss 510 LS confocal microscope and processed with Zen 2.1 using the same parameters. In
order to quantify changes in Golgi morphology in vitro or in vivo, some of the original images were analyzed
with the Particle Analysis function of ImageJ to obtain average size of GM130 positively stained structures or
regions, which were usually decreased during Golgi fragmentation. To quantify COPII accumulation at ERES,
similar strategy was employed by measuring average size of SEC31A positive foci. During quantitation, 10
random microscopic fields including approximately 50 primary mouse hepatocytes or 100 HepG2 cells were
taken. The cells with disseminated Golgi structural components were considered as cells with Golgi fragmentation.
3. Experimental Animals
Male C57BL/6 mice were purchased from Jackson’s Lab. Mice were given with 5% alcohol liquid diet (Dyets,
Inc., Bethlehem, PA) or with isocaloric control diet for 10 days. Mice were I.P. injected with RTV , LPV , or other
anti-HIV drugs from the 5th day of feeding at concentrations of 0 to 50 µg/ml. On the 10th day, mice were gavaged
with 30% alcohol in PBS (5mg/g body weight) or same volume of isocaloric maltose solution. The mice were
sacrificed for serum and liver sampling 8 hours after the alcohol gavage. All animals were treated in accordance
with the Guide for Care and Use of Laboratory Animals and the study was approved by the local animal care
committee.
80

4. Liver Pathological Parameters
Serum alanine aminotransferase (ALT) analysis were described previously
17,18
. For hematoxylin and eosin stain
(H&E), liver tissues were fixed in 10% Formalin overnight at 4
o
C, washed with and stored in 80% ethanol. The
fixed tissues were embedded in paraffin, sectioned at 5μm and proceeded to H&E. For Oil Red O staining, liver
tissues were embedded in O.C.T., snap frozen, sectioned at 5μm, and mounted on glass slides. The tissues on the
slides were fixed in 10% formalin and stained with an Oil Red O isopropanol solution (Electron Microscopy
Sciences, Hatfield, PA).

Results
1. RL Affects Golgi Morphology and Function in Liver Cells
Interruptions of Golgi function are commonly associated with changes in Golgi morphology
44
. In HepG2 cells,
immunofluorescent staining with GM130 and Golgi-resident enzyme -mannosidase II (MAN2A1) demonstrated
dramatic effects of the anti-HIV drugs on the Golgi. RL at 20µg/ml caused complete dispersal of the Golgi
structural marker GM130 (Fig. 4-1A) and redistribution of MAN2A1 (Fig. 4-1B) to ER-like density that were
co-stained with the ER marker protein disulfide isomerase (PDI) (Fig. 4-1C) whereas the Golgi structures
remained as perinuclear located and tightly organized ribbons in the DMSO-treated cells. The Golgi matrix
fragmentation and enzyme redistribution were comparable to those caused by brefeldin A (BFA) which is known
to induce those phenomena by blocking ER-Golgi trafficking (Fig. 4-1A&B). In dose-dependent experiments,
the minimal dose to induce observable Golgi fragmentation was 15µg/ml (Fig. 4-1E), under which 81% of cells
were with Golgi fragmentation, and the size of the GM130 stained structures that was reversely related to severity
of the Golgi fragmentation was 50% of that of control (Fig. 4-1D). In addition, individual treatment with either
RTV or LPV also induced Golgi fragmentation (Fig. 4-1G). Average size of GM130 stained structure was
81

decreased by 60% in the cells treated with LPV at 20µg/ml for 4 hours (Fig. 4-1H). To demonstrate whether Golgi
function besides the morphological changes was affected by RL, three Golgi stress response (GSR molecular)
markers, GCP60, HSP47 and TFE3 were examined. In qPCR analysis of gene expression at transcriptional level,
treatment with RL at 20µg/ml upregulated GCP60 and HSP47 by 3 and 10 folds, respectively (Fig. 4-2A). These
levels were comparable with those induced by modeled Golgi stress inducers which impairs Golgi function,
sodium monesin (SM) and benzyl-GalNAc (GBN). TFE3 coordinate GSR which counteracts Golgi stress. Effects
of TFE3 knockdown (TFE3
KD
) by siRNA on ER stress and cell death injury were investigated. In the TFE3
KD

cells, the RL-induced cell death was increased from 20% to 30% (Fig. 4-2B) whereas expression of CHOP
remained unchanged, indicating the presence of Golgi stress which induces injuries but not ER stress (Fig. 4-2B).
The presence of Golgi fragmentation was also observable in RL-treated primary mouse hepatocytes (PMH)
(Fig. 4-3A). In a similar way, Golgi fragmentation became more severe along with increase in treatment
concentration of RL and exposure time (Fig. 4-3B). Compared to HepG2, the PMH appeared more sensitive to
RL in terms of Golgi fragmentation. Under the treatment at 10µg/ml for 4 hours, half of the PMH were under
Golgi fragmentation whereas less than 5% of HepG2 were under the fragmentation (Fig. 4-3B & 4-1E). At
20µg/ml for 2 hours, Golgi fragmentation occurred in 50% of HepG2 and in greater than 80% of the PMH (Fig.
4-3B & 4-1F).
2. Effects of RL on the Golgi Were Correlated with the ER Stress Response
In time course measurement, the RL-induced Golgi enzyme redistribution occurred earlier than the ER stress.
The Golgi-resident MAN2A1 redistribution known to be a consequence of loss of the Golgi integrity was
observed in HepG2 at 30 minutes after RL treatment (Fig. 4-4A). CHOP was not significantly increased until 1
hour after the drug treatment and no change of GRP78 was detected in the first hour after RL treatment (Fig. 4-
4B). H89 has been known to alleviate BFA-induced Golgi enzyme redistribution and the ER stress. Co-treatments
with H89 reduced RL-induced MAN2A1 redistribution and ER stress response (Fig. 4-4C). Expression of CHOP
82

and sXBP1 was reduced by 20 and 2 folds, respectively in the RL-treated cells whereas in the TUM-treated cells
the CHOP and sXBP1 reductions were not significant (Fig. 4-4D).
Furthermore, there was a correlation between the drug-induced Golgi fragmentation and the ER stress (Fig. 4-
4E). Different degrees of Golgi fragmentation could be induced by different anti-HIV PIs and combinations
including amprenavir (APV), darunavir (DA V), LPV , nelfinavir (NFV), RTV , and RTV-boosted LPV . At 20µg/ml,
treatment with RTV-boosted LPV for 4 hours resulted in fragmented Golgi in 100% of the cells as well as the
strongest ER stress response as indicated by the CHOP expression whereas treatment with DA V , which did not
cause Golgi fragmentation, induced minimal ER stress response. Golgi fragmentation and the ER stress response
by APV , RTV , LPV , and NFV fell in between those by DA V and RTV-boosted LPV . Golgi fragmentations were
correlated well with increases of CHOP expression by APV , RTV , LPV , NFV and RTV boosted LPV (r2=0.972,
p<0.0001).
3. RL Induced Golgi Fragmentation Possibly by Affecting ER-to-Golgi Trafficking
To know which of the two routes (COPI or COPII) of ER-Golgi trafficking was affected by RL, Golgi
fragmentation and Golgi enzyme redistribution were compared between RL and BFA. In HepG2, low
concentration of BFA at 10nM induced apparent Golgi fragmentation indicated by the GM130 staining whereas
RL caused only limited Golgi fragmentation within 2 hours (Fig. 4-5A, red). In contrast, RL caused dramatic
Golgi enzyme redistribution indicated by the MAN2A1 staining at 1 hour after the drug exposure whereas BFA
induced less noticeable MAN2A1 redistribution (Fig. 4-5A, green). Alterations on COPI vesicle assembly or
transportation induce immediate Golgi fragmentation without any exception
136-139
. This was not the case for RL
treatment, therefore, RL was deduced to target COPII route. To support this hypothesis, Sec31A which is a part
of COPII vesicle was labeled with immunofluorescence. Under confocal microscope, COPII vesicles in RL-
treated HepG2 cells were more co-localized to form a larger Sec31A positive foci than those in BFA or DMSO-
treated cells in which COPII vesicles were more evenly distributed (Fig. 4-5B). The average size of Sec31A
83

positive foci was increased by 3 folds in RL-treated than in DMSO-treated cells (Fig. 4-5C). When assembly of
COPII was inhibited by H89, Sec31A positive foci in all the treatment groups were uniformly eliminated
suggesting that the Sec31A positive foci probed should stand for completely assembled COPII which is
accumulated at ERES (Fig. 4-5C). Therefore, it suggested that COPII transportation but not assembly was
affected by RL. To assess this possibility, an assay of Golgi fragmentation and re-assembly was further conducted
based on the fact that Golgi reassembly requires returning of Golgi materials through COPII dependent ER-to-
Golgi trafficking
136, 137
. Treatment of BFA at a high concentration resulted in rapid dispersion of Golgi matrix and
removal of BFA resulted in re-assembly of the Golgi (Fig. 4-5D). Expectedly, RL blocked the reassembly (Fig.
4-5D), indicating that the ER-to-Golgi trafficking was affected. In addition, in the PMH, larger Sec31A positive
foci were observed under RL treatment (Fig. 4-5E). Sec31A staining was highly overlapped with intermediate
structures between ER and Golgi or ERGIC reconfirming that the Sec31A positive foci truly indicated ERES
which is proximally localized with ERGIC in the most of mammalian cells (Fig. 4-5E).
4. Alcohol Deteriorated RL Induced Golgi Fragmentation and Injuries in PMH and Mouse Liver
In HepG2 cells, pre-treatment of alcohol at 85mM up to 96 hours did not affect RL-induced un-regulation of
CHOP indicating limited effect of alcohol on RL-induced ER stress (Fig. 4-6A). In consistent, pre-treatment of
alcohol also did not affect RL-induced Golgi fragmentation (Fig. 4-6B). RL-induced cell death was synergistically
promoted by alcohol only at a high concentration of 350mM (Fig. 4-6C). However, in the context of the PMH,
pre-treatment of alcohol at 85mM for 72 hours significantly increased the LDH release induced by 16-hour RL
treatment for 27% (Fig. 4-6D). The elevated LDH was accompanied by increase in upregulation of CHOP by 4-
hour RL treatment for 21% (Fig. 4-6E). Meanwhile, in cells pre-treated with alcohol followed by RL treatment
for 2 hours, average size of GM130 stained structures which is inversely related to degree of Golgi fragmentation
was reduced by 65% comparing to control whereas the reduction was only 20% in absence of alcohol (Fig. 4-6F
& G). In the presence of alcohol, Golgi fragmentation induced by 2-hour RL treatment was comparable to that
84

was induced by single 4-hour RL treatment which reduced average size of GM130 stained structures by 68%
indicating that alcohol deteriorated RL-induced Golgi fragmentation (Fig. 4-6G).
In accordance with the results from the PMH, Golgi fragmentation was also observed in liver sections from
mice treated with RL or RL plus alcohol (RLA). Representative views were indicated in Fig. 4-7A. Compared to
in vitro model in which GM130 stained structure was robustly reduced for more than 50%, the size of GM130
stained structure was reduced by 20% in the liver section in the liver of mice which was injected with RL at
20µg/ml for 5 days (Fig. 4-6G & 4-7B). However, in the mice pre-feeding with alcohol for 5 days, 45% of
reduction in average size of GM130 stained structure was observed. Along with Golgi fragmentation, ALT levels
were elevated by 4 folds in RL-treated mice and by 6 folds in RL plus alcohol (RLA)-treated mice, respectively
(Fig. 4-7C). CHOP expression was increased significantly by 10 folds in mice with RLA treatment (Fig. 4-7D).
In H&E and Oil Red O staining, accumulation of fatty droplets in the mouse livers was observed under RL
treatment, which was more severe under RLA treatment (Fig. 4-7E & F). The fatty droplets surrounded the central
vein indicating the presence of macrovascular steatosis (Fig. 4 -7F).

Discussion
In this chapter, the results demonstrated a strong association of onset of RL-induced ER stress with Golgi
fragmentation which was possibly cause by disrupted ER-to-Golgi trafficking. As ATF6 requires COPII for ER-
to-Golgi translocation, these could explain why co-localization of ATF6 and Golgi in cells under RL treatment
was significantly lower than the co-localization under the treatment of ER stress inducers (Fig 2-2D). The specific
Golgi fragmentation by RL started to occur within half of an hour, which was earlier than the ER stress response
indicated by CHOP expression and GRP78 expression. Treatment with H89, which is known to inhibit Golgi
fragmentation and the ER stress caused by brefeldin A or Golgicide A, also reduced expression of the UPR
85

markers CHOP and sXBP1 in RL-treated cells. The data strongly indicated a cause-effect relationship between
Golgi fragmentation and the ER stress. In dose response experiments, expression of the UPR markers and degree
of Golgi fragmentation had similar trends and degree of changes (Fig. 2-1 & 4-1). For instance, in HepG2 cells
under RL treatment, treatment at 10µ g/ml and 20µ g/ml resulted in huge differences in both CHOP expression
and average size of GM130 stained structures which reflect the ER stress and Golgi fragmentation, respectively
(Fig. 2-1G & 4-1D). In addition to RL treatment, variations of severity of Golgi fragmentation were observed in
response to treatment with other anti-HIV drugs including amprenavir, darunavir, LPV, nelfinavir, RTV, and
RTV-boosted LPV that are currently in use or recommended by the Department Health and Human Services
(DHHS) guidelines, which were well-correlated with the ER stress.
The Golgi apparatus is a part of the cellular endomembrane system closely associated with the ER
44, 134
. The
structure and function of Golgi are intimately linked. There is a bi-directional ER-Golgi trafficking responsible
for biogenesis and intracellular distribution of biomolecules
44, 134
. Impairment of either anterograde or retrograde
eventually collapses the whole ER-Golgi trafficking and causes cellular stresses and injury
44, 45
. Thus, integrity
of the ER-Golgi trafficking is essential for maintaining rigid Golgi morphology. Golgi fragmentation observed
by us indicates that the anti-HIV PIs stress Golgi and disrupt ER-Golgi trafficking. Indeed, Golgi stress response
markers of GCP60 and HSP47 were increased in the drug-treated liver cells and knockdown of TFE3 (known to
regulate GSR) by siRNA worsened the drug-induced cell death. We also believe that it was the anterograde ER-
to-Golgi trafficking that was mostly affected by the drugs for three reasons. Firstly, brefeldin A is known to
specifically inhibit the assembly of COPI and ARF1 complexes
44
, which disrupts the retrograde route. If RL
inhibit the retrograde route, RL treatment should result in Golgi fragmentation with kinetics similar to or the same
as that by BFA. However, the kinetics for the fragmentation induction were different between RL and BFA. It
took 2 hours for RL to induce the fragmentation in 50% of the cells whereas within 1hour after BFA treatment,
almost all the cells were with Golgi fragmentation. Secondly, the ATF6, known to traffic to the Golgi through the
86

ER-to-Golgi route, was significantly reduced in the Golgi of cells treated with RL. Thirdly, the BFA induced
Golgi fragmentation was primarily associated with abnormal distributions of the Golgi structural protein GM130
whereas the RL-induced fragmentation was primarily associated with abnormal distributions of the Golgi-resident
MAN2A1 to the ER
44, 134, 138
. The observed enzyme redistribution might be due to blocking of ER-to-Golgi
transportation by RL. This was only reported in rat kidney cells overexpressing dominant negative form of SAR1
(SAR1
DN
), a key component of the COPII complexes that mediate the ER-to-Golgi trafficking
139
. In the SAR1
DN

cells, the anterograde route was specifically blocked leading to relocation of the Golgi-resident enzymes before
disrupting Golgi morphology. Our observations support strongly that the drugs induce the Golgi fragmentation
through blocking the COPII trafficking, which has also been linked to the ER stress and apoptosis
134, 135
. In
addition, we further addressed what stage of the ER-to-Golgi trafficking was affected by RL. Initiation and
assembly of the anterograde route requires SAR1, PKA, COPII, Sec proteins (e.g. Sec31A), and GTPases
44
, which
can be blocked by H89. Since H89 did not exert the same effects as RL did on the accumulation of COPII
complexes, the possibility of preventing COPII complexes assembly by RL was ruled out. Thus we have narrowed
the drug targeted molecular sites down to COPII-mediated vesicle transportation. Translocation of COPII vesicle
from ERES to ERGIC involves Rab proteins and Rab effectors controlling vesicle docking and paired sets of
SNARE proteins mediating fusion of vesicles with target membranes
44
. Molecular details on the COPII vesicle
transportation in relation to RL-induced Golgi stress, UPR and liver injury will be our future research goals.
In the liver cell model, alcohol pre-treatment for 72 hours deteriorated the ER stress and cell death in the
primary mouse hepatocyte (PMH) but not in HepG2 cells. This might due to an incomplete alcohol metabolic
system in HepG2 cells which express limited alcohol dehydrogenase and CYP2E1
141
. In the PMH, pre-treatment
of alcohol for 72 hours significantly enhanced RL-induced Golgi fragmentation whereas single administration of
alcohol did not impact Golgi morphology even after 72 hours. Enhancing of Golgi fragmentation by alcohol was
possibly due to a synergistic effect of alcohol and RL in disrupting ER-to-Golgi trafficking since alcohol treatment
87

at 35mM downregulated SAR1A in the primary rat hepatocytes
49
. Furthermore, in mouse model, alcohol feeding
also deteriorated the RL-induced Golgi fragmentation, which was associated with the onset of ER stress, elevated
ALT and liver steatosis. Hence, all the above pieces of evidence from this study suggest for the first time that the
Golgi dysfunction is a robust mechanism underlying or at least contributing to the anti-HIV drugs and/or alcohol-
induced ER stress/UPR and liver injury.

Figure Legends
Figure 4-1 Golgi Fragmentation in HepG2 Cells Treated with RL. (A) Golgi fragmentation in HepG2 cells.
The cells were treated with RL (20µ g/ml), DMSO (0.1%) as vehicle control, or brefeldin A (BFA; 20nM) as
positive control for 4 hours. Golgi matrix was labeled with anti-GM130 antibodies (red) and nuclei were stained
with Hochest blue (blue). (B) Cellular distribution of MAN2A1 (Golgi α-Mannosidase II). The cells were treated
with RL, DMSO and BFA for 2 hours. MAN2A1 was labeled by immunofluorescence (red). (C) Confocal image
indicating ER like distribution and MAN2A1 after RL treatment. Cells were treated for DMSO or RL for 30
minutes. ER was labeled by immunofluorescent staining of PDI. (D) Quantification of Golgi fragmentation under
RL treatment. Golgi was stained with GM130 and size of GM130 positive particles were analyzed by ImageJ;
RL10, 15, 20 and 40, treated with RL at 10, 15, 20 and 40 µ g/ml. (E) Dose response analysis of percent of cells
with Golgi fragmentation under RL treatment. Cells were treated for 4 hours and number of cells with/without
fragmentation was counted at 63X; RL10, 15, 20 and 40, treated with RL at 10, 15, 20 and 40 µ g/ml. (F) Time
course analysis of percent of cells with Golgi fragmentation under RL treatment. Cells were treated at 20µ g/ml
and number of cells with/without fragmentation was counted. (G) Golgi fragmentation induced by single drug
alone or combination of drugs. (H) Quantification of Golgi fragmentation under single drug alone or combination
of drugs treatment. LPV, lopinavir; RTV, ritonavir; ****, p<0.0001 compared to control or indicated otherwise;
88

n=6.
Figure 4-2 Induction of Golgi Stress in HepG2 cells Treated with RL. (A) Expression of Golgi stress marker
GCP60 and HSP47 at 8 hours. RL10, RL at 10µ g/ml; RL20, RL at 20µ g/ml; GCP60, Golgi resident protein 60;
HSP47, heat shock protein 49; SM, sodium monesin. (B) Cell death in the cells deficient in Golgi stress response.
Cell death was revealed by Syntox green staining (left panel) and quantified by ImageJ (right panel) 16 hours
after the treatments. GBN, benzyl-GalNAc; (C) mRNA expression of CHOP at 4 hours in cells deficient in Golgi
stress response. TFE3, transcription factor for immunoglobulin heavy-chain enhancer 3; RL+C
KD
, RL treatment
in cells transfected with control scramble siRNA; RL+TFE3
KD
, RL treatment in cells transfected with TFE3
siRNA;*, p<0.05; **, p<0.01; ***, p<0.005; ****, p<0.0001 compared to control or indicated otherwise; n=5-6.
Figure 4-3 RL Induces Golgi fragmentation in the Primary Mouse Hepatocytes. (A) Golgi fragmentation in
the PMH treated with RL. Confocal images showing Golgi fragmentation in PMH treated with RL at 20µ g/ml for
2 hours; cells were stained with GM130 (Red) and counterstained with Hochest blue (Blue) indicating nucleus.
(B) Quantitation of RL-induced Golgi fragmentation at different dosage and for different treatment time. RL10
and 20, RL treatment at 10 and 20µ g/ml. n=5
Figure 4-4 Association of RL-induced Golgi Fragmentation with Onset of the ER Stress Response. (A)
Confocal images showing the time point of Golgi enzyme redistribution revealed by immunofluorescent staining
with anti-MAN2A1 antibodies. (B) Onset of the ER stress response indicated by expression changes of CHOP
and GRP78 measured with qPCR; (C) Inhibition of RL-induced MAN2A1 redistribution by H89; (D) Inhibition
of RL-induced ER stress by H89. Onset of ER stress was indicated by expression of CHOP and sXBP1 for which
the expression was quantified by qPCR. (E) Correlation of the ER stress response with degree of Golgi
fragmentation in cells treated with different anti-HIV PIs. The cells were treated with RTV (ritonavir), LPV
(lopinavir), APV (amprenavir), NFV (nelfinavir), or RTV boosted LPV at 20µg/ml for 4 hours; CHOP expression
was analyzed by qPCR. Degree of Golgi fragmentation was revealed by quantification of GM130 positive
89

structures in cells under different treatment. ****, p<0.0001 compared to control or indicated otherwise; n=4-5.
Figure 4-5 Effects of Anti-HIV Drugs on ER-Golgi Trafficking. (A) Differences between the RL and brefeldin
A in inducing Golgi fragmentation and enzyme redistribution. Golgi matrix fragmentation was indicated by
GM130 staining (red) while Golgi resident enzyme redistribution was indicated by MAN2A1 staining (green).
(B) COPII staining in cells treated with RL, brefeldin A and H89. COPII at ER exit sites (ERES) was indicated
by immunofluorescent staining with anti-Sec31A antibodies as showed by confocal images. (C) Quantitative
difference in COPII positive foci between RL and BFA treated cells. The plot quantified the size of COPII positive
foci proportional to amount of COPII locates at ERES. (D) Recovery of Golgi morphology after removal of BFA,
which was inhibited by RL. The cells were treated first with BFA (100nM) for 1 hour and then treated with RL
for another 1 hour after BFA had been removed from the media. Quantitation of Golgi fragmentation is on lower
panel. (E) Accumulation of co-localized COPII complex and ERGIC in the PMH treated with RL versus BFA.
Coat protein complex II (COPII) was probed with immunofluorescence (red) using anti-Sec31A antibodies and
ERGIC (the ER-Golgi intermediate compartment) was probed with immunofluorescence (green) using anti-
ERGIC53 antibodies.
Figure 4-6 Effects of Alcohol on RL-induced ER Stress, Golgi Fragmentation and Cell death in HepG2 and
the Primary Mouse Hepatocytes. (A) CHOP mRNA expression in RL and alcohol (ethanol or EtOH)-treated
HepG2 cells. Cells were pre-treated with PBS or alcohol at 85mM for 48 or 96 hours followed by DMSO or RL
treatment at 20µ g/ml for 8 hours. (B) Confocal image comparing RL-induced Golgi fragmentation in presence
and absence of alcohol in HepG2 cells. Cells were pre-treated with alcohol at 85mM for 48 hours followed by
RL treatment at 20µ g/ml for 4 hours; Cells were stained with GM130 (red) to indicate Golgi fragmentation. (C)
Alcohol affected RL-induced cell death in HepG2 cells at high concentration. Cells were pre-treated with alcohol
at indicated concentrations for 72 hours followed by RL treatment at 20µ g/ml for 16 hours; Cell death was
quantified by LDH release. (D) Prolonged alcohol exposure deteriorated RL-induced cell death in the primary
90

mouse hepatocytes (PMH). Cells were pre-treated with alcohol at 85mM for indicated time followed by RL
treatment at 20µ g/ml for 16 hours; Cell death was quantified by LDH release. (E) Prolonged alcohol exposure
also promoted CHOP mRNA expression in RL-treated PMH. Cells were pre-treated with alcohol at 85mM for 72
hours followed by DMSO or RL treatment for 4 hours; CHOP expression was quantified by qPCR. (F) Confocal
image comparing RL-induced Golgi fragmentation in presence and absence of alcohol in the PMH. Cells were
pre-treated with alcohol at 85mM for indicated time followed by RL exposure at 20µ g/ml for different time
periods; Golgi fragmentation was indicated by GM130 staining (green) and counterstained with Hochest blue
(blue) for nucleus. (G) Quantification of Golgi fragmentation. Golgi fragmentation was quantified for 4-6G. *,
p<0.05; **, p<0.01; ***, p<0.005; ****, p<0.0001 compared to control or otherwise indicated; n=5-6.
Figure 4-7 Association of Golgi fragmentation with liver injuries in mice injected anti-HIV drugs and/or
fed with alcohol. (A) Confocal images showing Golgi fragmentation of hepatocytes of liver sections. Golgi
morphology was observed under 100X and revealed by probing with anti-GM130 antibodies through
immunofluorescence (green); the nuclei of hepatocytes were counterstained with Hochest blue staining (blue).
Ctrl, sample from control mice; RL, sample from mice with ritonavir and lopinavir injection. (B) Quantitation of
the Golgi fragmentation; the views from the samples indicated in 4-7A were analyzed by ImageJ to quantify the
average size of GM130 positive structures; three views were included for each sample. (C) Levels of plasma
alanine transaminase (ALT) in mice with RL injection and/or alcohol feeding. (D) H&E of the liver tissues
showing moderate fatty liver injury; (E) Oil Red O staining of the liver tissues showing fat accumulation in the
liver. *, p<0.05; ***, p<0.005 compared to control; n=4-6.

91

Figures and Tables
Figure 4-1
(A)

(B)

92

Figure 4-1
(C)

(D)

A v e ra g e a iz e o f G M 1 3 0
p o s itiv e s tru c tu re (% o f D M S O )
D M S O
R L 1 0
R L 1 5
R L 2 0
R L 4 0
0
2 0
4 0
6 0
8 0
1 0 0
* * * *
* * * *
* * * *
93

Figure 4-1
(E)

(F)

% o f C e ll P o p u la tio n
D M S O R L 1 0 R L 1 5 R L 2 0 R L 4 0
0
2 0
4 0
6 0
8 0
1 0 0
1 2 0
C e lls w ith G o lg i F r a g m e n ta tio n
C e lls w ith o u t G o lg i F ra g m e n ta tio n
T im e (h o u rs )
% o f C e ll P o p u la tio n
0 h 1 h 2 h 4 h
0
2 0
4 0
6 0
8 0
1 0 0
1 2 0
C e lls w ith G o lg i F r a g m e n ta tio n
C e lls w ith o u t G o lg i F ra g m e n ta tio n
94

Figure 4-1
(G)

(H)

A v e ra g e s iz e o f G M 1 3 0 p o s itiv e
s tru c tu re s (% o f C trl)
D M S O R T V L P V R L
0
2 0
4 0
6 0
8 0
1 0 0
1 2 0
* * * *
* * * *
95

Figure 4-2
(A)

(B) (C)

96

Figure 4-3
(A)

(B)

C e ll P o p u la tio n (% )
D M S O
R L 1 0 2 h
R L 1 0 4 h
R L 2 0 1 h
R L 2 0 2 h
R L 2 0 4 h
0
2 0
4 0
6 0
8 0
1 0 0
1 2 0
F ra g m e n te d C e lls
U n fra g m e n te d C e lls
97

Figure 4-4
(A)

(B)

98

Figure 4-4
(C)

(D)

99

Figure 4-4
(E)

(A)

C e lls u n d e r G o lg i fra g m e n ta tio n (% )
A v e ra g e C H O P e x p re s s io n (fo ld )
0 5 0 1 0 0
0
2 0
4 0
6 0
8 0
D A V
R T V
A P V
L P V
N F V
R T V B o o s te d L P V
r
2
= 0 .9 7 2 , p < 0 .0 0 0 1
100

Figure 4-5

(B)

(C)

101

Figure 4-5
(D)

102

Figure 4-5
(E)

(A)

R e la tiv e C H O P m R N A le v e ls
(fo ld o f c h a n g e )
N o E tO H 4 8 h r s 9 6 h r s N o E tO H 4 8 h r s 9 6 h r s
5 0
1 0 0
1 5 0
E tO H
(8 5 m M )
D M S O 0 .6 % R L a t 2 0 u g /m l
103

Figure 4-6
(B)

(C)

M e d iu m L D H L e v e l (R A U )
0 m M 8 5 m M 3 5 0 m M 0 m M 8 5 m M 3 5 0 m M
0 .0
0 .5
1 .0
1 .5
2 .0
2 .5
D M S O 0 .6 % R L a t 2 0 u g /m l
****
104

Figure 4-6
(D)

(E)

M e d iu m L D H L e v e l (R A U )
N o E tO H 2 4 h r s 7 2 h r s N o E tO H 2 4 h r s 7 2 h r s
0 .2
0 .4
0 .6
0 .8
1 .0
*
**
D M S O 0 .6 % R L a t 2 0 u g /m l
E tO H
(8 5 m M )
R e la tiv e C H O P m R N A le v e ls
(fo ld o f c h a n g e )
D M S O R L E tO H R L + E tO H
0
1 0
2 0
3 0
4 0
5 0
**
105

Figure 4-6
(F)

(G)

R e la tiv e s iz e o f G M 1 3 0
s ta in e d s tru c tu re s
0 .2
0 .4
0 .6
0 .8
1 .0
1 .2
R L (2 0 u g /m l) 0 2 h rs 4 h rs 0 2 h rs 2 h rs
E tO H (8 5 m M ) 0 0 0 7 2 h rs 2 4 h rs 7 2 h rs
106

Figure 4-7
(A)

107

Figure 4-7
(B)

(C)

S iz e o f fra g m e n te d G o lg i (% )
C tr l R L R L + A lc o h o l
0
5 0
1 0 0
1 5 0
**
***
A L T (IU /L )
C tr l E tO H R L R L + E tO H
0
1 0 0
2 0 0
3 0 0
4 0 0
5 0 0
***
*
108

Figure 4-7
(D)

R e la tiv e C H O P m R N A le v e ls
(fo ld o f c h a n g e )
D M S O E tO H R L R L + E tO H
0
5
1 0
1 5
2 0
*
109

Figure 4-7
(E)

110

Figure 4-7
(F)

111

Chapter 6 Conclusions
This dissertation systemically studied mechanisms of anti-HIV PIs induced ER stress and cell death in liver
cell models and mouse liver. Two representative anti-HIV PIs, RTV and LPV have been shown to disrupt ER-to-
Golgi trafficking contributing to cellular stress and hepatic injury. In sum, this study resulted in several novel
findings listed below.
In the second chapter, we studied the downstream events of RL-induced liver injury – what is/are the mode(s)
of cell death in RL-treated cells. The study was conducted in HepG2 and PMH models. In both of the cell models,
RL-induced apoptosis consisting with previous publications. However, this was the first time to show that, RL-
induced necroptosis in the PMH model providing a novel possible mechanism for RL-induced liver injury.
Interestingly, pre-treatment necrostatin-1 (NEC1) which worked efficiently in inhibiting TNFa induced
necroptosis hardly inhibited cell death in RL-treated PMH. As necroptotic pathway has not been fully uncovered,
it was difficult to identify the upstream events inducing MLKL activation. Whether MLKL inhibitor,
necrosulfonamide could be an effective antidote for RL-induced liver injury requires further investigation.
In the third chapter, the UPR activation was carefully studied in comparison to potent ER stress inducers in
RL-treated liver cell models. Accidentally, we discovered for the first time that, only two canonical UPR branches
including PERK and IRE1a were strongly activated whereas ATF6 remained inactivated under RL treatment in
liver cell models. During investigation of ATF6 cleavage and localization, we found that Golgi localization of
ATF6 was reduced in presence of RL indicating the possibility that ER to Golgi trafficking was affected.
In the fourth chapter, we indicated that although proteasome inhibition and oxidative stress could affect protein
folding homeostasis, they are not the major contributors for RL-induced ER stress. The ER calcium perturbation
by RL was more systemically studied in HepG2 model. From the results, we determined that the reduction of
SERCA was due to an accelerated protein degradation.
Last but not the least, in the fifth chapter, we are the first to find that Golgi morphology, function, and ER-
112

to-Golgi trafficking are affected by RL. Golgi fragmentation has often been resulted from impair ER-to-Golgi
trafficking which leads to ER stress. In HepG2, PMH and mouse models, Golgi fragmentation was well correlated
with expression of UPR markers and corresponding injuries. The Golgi fragmentation was also associated well
with ER stress induced by different anti-HIV PIs. The reduction of upregulated UPR markers by H89 in RL-
treated cells indirectly but strongly suggested that ER stress is the consequence of impaired ER-Golgi trafficking.
By comparing kinetics of Golgi fragmentation with BFA, staining of COPII component Sec31 and BFA washout
assay, we found that Golgi fragmentation was more likely to be the consequence of impaired COPII trafficking.
More importantly, alcohol promoted RL-induced Golgi fragmentation in PMH and mouse liver, which was
associated with higher levels of expression of UPR markers and liver injuries. Thus we have narrowed the drug
targeted molecular sites down to COPII-mediated vesicle secretion. The vesicle secretion involves different Rab
proteins and Rab effectors controlling vesicle docking and paired sets of SNARE proteins mediating fusion of
vesicles with target membranes. Molecular details on the COPII vesicle secretion in relation to RL-induced Golgi
stress, UPR and liver injury will be the future research goals.
In summary, we have elucidated a ER-to-Golgi trafficking mechanism (Fig. 5-1) by which the anti-HIV drug
and/or alcohol induce ER stress and subsequent hepatic injury, which provides potential therapeutic targets for
HIV-infected patients who are under anti-HIV therapies and suffer from liver injury.

113

Figure 5-1

COPI trafficking
(Lipids/ER chaperones)
COPII trafficking
(ATF6, SREBP, CREB3/lipids)
ARF1
COPI
complex
PKA
SAR1
Sec31
COPII
complex
Anti-HIV PIs/Alcohol
H89
BFA
Flow of Golgi materials
e.g. MAN2A1 redistribution
Endoplasmic
Reticulum
Golgi
Apparatus
MAN2A1
GM130
Golgi Stress
ER Stress
Golgi fragmentation
114

References
1. Clercq ED, Li G. Approved Antiviral Drugs over the Past 50 Years. Clin. Microbiol. Rev. Clinical Microbiology Reviews.
2016;29:695–747.
2. Faria NR, Rambaut A, Suchard MA, Baele G, Bedford T, Ward MJ, et al. The early spread and epidemic ignition of HIV-1 in
human populations. Science. 2014;346:56–61.
3. Günthard HF, Saag MS, Benson CA, Rio CD, Eron JJ, Gallant JE, et al. Antiretroviral Drugs for Treatment and Prevention
of HIV Infection in Adults. JAMA. 2016;316:191.
4. HIV and AIDS in the United States of America (USA) [Internet]. A VERT. Available from:
http://www.avert.org/professionals/hiv-around-world/western-central-europe-north-america/usa
5. Kassutto S, Rosenberg ES. Primary HIV Type 1 Infection. Clinical Infectious Diseases. 2004;38:1447–1453.
6. Acute HIV Infection and Cells Susceptible to HIV Infection. HIV and the Pathogenesis of AIDS, Third Edition. :79–108.
7. Moir S, Chun T-W, Fauci AS. Pathogenic Mechanisms of HIV Disease*. Annu. Rev. Pathol. Mech. Dis. Annual Review of
Pathology: Mechanisms of Disease. 2011;6:223–248.
8. D'angelo LJ, Peele YL, Chandwani S, Peralta L. Determinants of Disease Classification in HIV Infected Adolescents: Is the
Current Classification System Still Relevant? Journal of Adolescent Health. 2009;44.
9. Li G, Piampongsant S, Faria NR, V oet A, Pineda-Peña A-C, Khouri R, et al. An integrated map of HIV genome-wide variation
from a population perspective. Retrovirology. 2015;12.
10. Coffin J, Swanstrom R. HIV Pathogenesis: Dynamics and Genetics of Viral Populations and Infected Cells. Cold Spring
Harbor Perspectives in Medicine. 2013;3.
11. Safrit JT, Koff WC. Novel approaches in preclinical HIV vaccine research. Current Opinion in HIV and AIDS. 2016;:1.
12. Mocroft A. Starting highly active antiretroviral therapy: why, when and response to HAART. Journal of Antimicrobial
Chemotherapy. 2004;54:10–13.
13. Granich R, Crowley S, Vitoria M, Smyth C, Kahn JG, Bennett R, et al. Highly active antiretroviral treatment as prevention
of HIV transmission: review of scientific evidence and update. Current Opinion in HIV and AIDS. 2010;5:298–304.
14. FDA-Approved HIV Medicines | HIV/AIDS Fact Sheets | Education Materials | AIDSinfo [Internet]. U.S National Library
of Medicine. ; Available from: https://aidsinfo.nih.gov
15. Smith CJ, Ryom L, Weber R, Morlat P, Pradier C, Reiss P, et al. Trends in underlying causes of death in people with HIV
from 1999 to 2011 (D:A:D): a multicohort collaboration. The Lancet. 2014;384:241–248.
16. Jia Z, Ruan Y, Lu Z. HIV incidence and mortality in China. The Lancet. 2015;385:1510.
17. Krentz H, Kliewer G, Gill M. Changing mortality rates and causes of death for HIV-infected individuals living in Southern
Alberta, Canada from 1984 to 2003. HIV Medicine HIV Med. 2005;6:99–106.
18. Joshi D, O'grady J, Dieterich D, Gazzard B, Agarwal K. Increasing burden of liver disease in patients with HIV infection.
The Lancet. 2011;377:1198–1209.
19. Wittkop L. End-Stage Liver Disease in HIV Infection: An Avoidable Burden? Clin Infect Dis. Clinical Infectious Diseases.
2016;
20. Wang Y, Lv Z, Chu Y. HIV protease inhibitors: a review of molecular selectivity and toxicity. HIV HIV/AIDS - Research
and Palliative Care. 2015;:95.
21. NIDDK, Livertox, Antiviral Agents. Available from: https://livertox.nlm.nih.gov/proteaseinhibitors.htm
22. Sulkowski MS. Drug-Induced Liver Injury Associated with Antiretroviral Therapy that Includes HIV-1 Protease Inhibitors.
Clinical Infectious Diseases. 2004;38.
23. Sherman KE, Rockstroh J, Thomas D. Human immunodeficiency virus and liver disease: An update. Hepatology.
2015;62(6):1871-82.
24. Esposito I, Labarga P, Barreiro P, Fernandez-Montero JV, de Mendoza C, Bení tez-Gutié rrez L, et al. Dual antiviral therapy
for HIV and hepatitis C-drug interactions and side effects. Expert Opin Drug Saf. 2015;14(9):1421-34.
115

25. Muga R, Sanvisens A, Fuster D, Tor J, Martí nez E, Pé rez-Hoyos S, Muñ oz A. Unhealthy alcohol use, HIV infection and
risk of liver fibrosis in drug users with hepatitis C. PLoS One. 2012; 7(10):e46810.
26. Bilal U, Lau B, Lazo M, Mccaul ME, Hutton HE, Sulkowski MS, et al. Interaction Between Alcohol Consumption Patterns,
Antiretroviral Therapy Type, and Liver Fibrosis in Persons Living with HIV . AIDS Patient Care and STDs. 2016; 30:200–
207.
27. Kahler CW, Liu T, Cioe PA, Bryant V, Pinkston MM, Kojic EM, et al. Direct and Indirect Effects of Heavy Alcohol Use
on Clinical Outcomes in a Longitudinal Study of HIV Patients on ART. AIDS Behav. 2016. PMID: 27392417.
28. Neuman MG, Schneider M, Nanau RM, Parry C. HIV-Antiretroviral Therapy Induced Liver, Gastrointestinal, and Pancreatic
Injury. International Journal of Hepatology. 2012;2012:1–23.
29. A. A. Chaudhry, M. S. Sulkowski, G. Chander, and R. D. Moore, “Hazardous drinking is associated with an elevated aspartate
aminotransferase to platelet ratio index in an urban HIV-infected clinical cohort,” HIV Medicine, vol. 10, no. 3, pp. 133–142,
2009.
30. E. Rosenthal, D. Salmon-Céron, C. Lewden et al., “Liver-related deaths in HIV-infected patients between 1995 and 2005 in
the French GERMIVIC Joint Study Group Network (Mortavic 2005 Study in collaboration with the Mortalité 2005 survey,
ANRS EN19),” HIV Medicine, vol. 10, no. 5, pp. 282–289, 2009.
31. R. Lana, M. Núñez, J. L. Mendoza, and V . Soriano, “Rate and risk factors of liver toxicity in patients receiving antiretroviral
therapy,” Medicina Clinica, vol. 117, no. 16, pp. 607–610, 2001.
32. Uetrecht, Jack. "Idiosyncratic Drug Reactions: Current Understanding." Annual Review of Pharmacology and Toxicology
47.1 (2007): 513-39. Web.
33. D Pirmohamed, Munir. "Pharmacogenetics of Idiosyncratic Adverse Drug Reactions." Handbook of Experimental
Pharmacology Adverse Drug Reactions (2009): 477-91.
34. Kao E, Shinohara M, Feng M, Lau MY , Ji C. Human immunodeficiency virus protease inhibitors modulate Ca2 homeostasis
and potentiate alcoholic stress and injury in mice and primary mouse and human hepatocytes. Hepatology. 2012; 56:594–604.
35. Wang Y , Zhang L, Wu X, Gurley EC, Kennedy E, Hylemon PB, et al. The role of CCAAT enhancer-binding protein
homologous protein in human immunodeficiency virus protease-inhibitor-induced hepatic lipotoxicity in mice. Hepatology.
2013; 57:1005–1016.
36. Hu J, Han H, Lau MY , Lee H, Macveigh-Aloni M, Ji C. Effects of Combined Alcohol and Anti-HIV Drugs on Cellular Stress
Responses in Primary Hepatocytes and Hepatic Stellate and Kupffer Cells. Alcohol Clin Exp Res. 2015; 39:11–20.
37. De Gassart A, Bujisic B, Zaffalon L, Decosterd LA, Di Micco A, Frera G, Tallant R, Martinon F. An inhibitor of HIV-1
protease modulates constitutive eIF2α dephosphorylation to trigger a specific integrated stress response. Proc Natl Acad Sci
USA. 2016;113(2): E117-26.
38. Ji C, Kaplowitz N. Betaine decreases hyperhomocysteinemia, endoplasmic reticulum stress, and liver injury in alcohol-fed
mice. Gastroenterology. 2003;124(5):1488-99.
39. Ji C, Kaplowitz N, Lau MY, Kao E, Petrovic LM, Lee AS. Liver-specific loss of glucose-regulated protein 78 perturbs the
unfolded protein response andexacerbates a spectrum of liver diseases in mice. Hepatology. 2011;54(1): 229-39.
40. Wang M, Kaufman RJ. Protein misfolding in the endoplasmic reticulum as a conduit to human disease. Nature.
2016;529:326–335.
41. Lafleur MA, Stevens JL, Lawrence JW. Xenobiotic perturbation of ER stress and the unfolded protein response. Toxicol
Pathol. 2013; 41(2):235-62.
42. Ji C. Advances and New Concepts in Alcohol-Induced Organelle Stress, Unfolded Protein Responses and Organ Damage.
Biomolecules. 2015; 5(2):1099-121.
43. Resveratrol protects against protease inhibitor-induced reactive oxygen species production, reticulum stress and lipid raft
perturbation.
44. Brandizzi F, Barlowe C. Organization of the ER–Golgi interface for membrane traffic control. Nat Rev Mol Cell Biol. 2013;
14:382–392.
45. Sasaki K, Yoshida H. Organelle autoregulation-stress responses in the ER, Golgi, mitochondria and lysosome. J Biochem.
116

2015;157(4):185-95.
46. Rubin E, Rybak BJ, Lindenbaum J, Gerson CD, Walker G, Lieber CS. Ultrastructural changes in the small intestine induced
by ethanol. Gastroenterology. 1972; 63(5):801-14.
47. Chang PL, Sharma RN, Sturgess JM, Moscarello MA. Characterization of intracisternal and membrane subfractions from
sonically disrupted rough microsomal and Golgi-rich fractions of the rat liver. Exp Cell Res. 1978;112(1):187-97.
48. Avitabile E, Di Gaeta S, Torrisi MR, Ward PL, Roizman B, Campadelli-Fiume G. Redistribution of microtubules and Golgi
apparatus in herpes simplex virus-infected cells and their role in viral exocytosis. J Virol. 1995;69(12):7472-82.
49. Petrosyan A, Cheng PW, Clemens DL, Casey CA. Downregulation of the small GTPase SAR1A: a key event underlying
alcohol-induced Golgi fragmentation in hepatocytes. Sci Rep. 2015; 5:17127.
50. Gang H, Lieber CS, Rubin E. Ethanol increases glycosyl transferase activity in the hepatic Golgi apparatus. Nat New Biol.
1973; 243(125):123-5.
51. Consolidated guidelines on the use of antiretroviral drugs for treating and preventing HIV infection: recommendations for a
public health approach [Internet]. Geneva: World Health Organization; 2016. Available from:
http://www.who.int/hiv/pub/arv/arv-2016/en/
52. Chandwani A, Shuter J. Lopinavir/ritonavir in the treatment of HIV-1 infection: a review. Therapeutics and Clinical Risk
Management. 2008;4(5):1023-1033.
53. Huang X, Xu Y, Yang Q, et al. Efficacy and biological safety of lopinavir/ritonavir based anti-retroviral therapy in HIV-1-
infected patients: a meta-analysis of randomized controlled trials. Scientific Reports. 2015;5:8528. doi:10.1038/srep08528.
54. Ernest CS. Mechanism-Based Inactivation of CYP3A by HIV Protease Inhibitors. Journal of Pharmacology and Experimental
Therapeutics. 2004;312:583–591.
55. NIDDK, Livertox, Antiviral Agents. Available from: https://livertox.nlm.nih.gov/proteaseinhibitors.htm
56. Wang Y, Lv Z, Chu Y. HIV protease inhibitors: a review of molecular selectivity and toxicity. HIV HIV/AIDS - Research
and Palliative Care. 2015;:95.
57. Cohen J, Opal SM, Powderly WG, Calandra Thierry Franc ̧ ois. Infectious diseases. Edinburgh: Mosby; 2010.
58. Taura M, Kariya R, Kudo E, Goto H, Iwawaki T, Amano M, et al. Comparative analysis of ER stress response into HIV
protease inhibitors: Lopinavir but not darunavir induces potent ER stress response via ROS/JNK pathway. Free Radical
Biology and Medicine. 2013;65:778–788.
59. Zhou H. HIV Protease Inhibitors Activate the Unfolded Protein Response in Macrophages: Implication for Atherosclerosis
and Cardiovascular Disease. Molecular Pharmacology. 2005;
60. Zha BS, Wan X, Zhang X, Zha W, Zhou J, Wabitsch M, et al. HIV Protease Inhibitors Disrupt Lipid Metabolism by Activating
Endoplasmic Reticulum Stress and Inhibiting Autophagy Activity in Adipocytes. PLoS ONE. 2013;8.
61. Bernstein WB, Dennis PA. Repositioning HIV protease inhibitors as cancer therapeutics. Current Opinion in HIV and AIDS.
2008;3:666–675.
62. Galluzzi L, Pedro JMB-S, Vitale I, Aaronson SA, Abrams JM, Adam D, et al. Essential versus accessory aspects of cell death:
recommendations of the NCCD 2015. Cell Death Differ Cell Death and Differentiation. 2014;22:58–73.
63. Lamech LT, Haynes CM. The unpredictability of prolonged activation of stress response pathways. J Cell Biol The Journal
of Cell Biology. 2015;209:781–787.
64. Pinkoski MJ. Life and Death Decisions in Response to Stress. Apoptosis and Cancer Therapy. :757–776.
65. Green DR. Pharmacological manipulation of cell death: clinical applications in sight? Journal of Clinical Investigation.
2005;115:2610–2617.
66. Fuchs Y, Steller H. Live to die another way: modes of programmed cell death and the signals emanating from dying cells.
Nature Reviews Molecular Cell Biology Nat Rev Mol Cell Biol. 2015;16:329–344.
67. Apoptosis: controlled demolition at the cellular level :
68. Mcilwain DR, Berger T, Mak TW. Caspase Functions in Cell Death and Disease: Figure 1. Cold Spring Harbor Perspectives
in Biology Cold Spring Harb Perspect Biol. 2015;7.
69. Pasparakis M, Vandenabeele P. s and its role in inflammation. Nature. 2015;517:311–320.
117

70. Yang X, Chao X, Wang Z-T, Ding W-X. The end of RIPK1-RIPK3-MLKL-mediated necroptosis in acetaminophen-induced
hepatotoxicity? Hepatology. 2015;64:311–312.
71. Cai Z, Jitkaew S, Zhao J, Chiang H-C, Choksi S, Liu J, et al. Plasma membrane translocation of trimerized MLKL protein is
required for TNF-induced necroptosis. Nature Cell Biology Nat Cell Biol. 2013;16:55–65.
72. Ying Y, Padanilam BJ. Regulation of necrotic cell death: p53, PARP1 and cyclophilin D-overlapping pathways of regulated
necrosis? Cell. Mol. Life Sci. Cellular and Molecular Life Sciences. 2016;73:2309–2324.
73. Yuan L, Kaplowitz N. Mechanisms of Drug-induced Liver Injury. Clinics in Liver Disease. 2013;17:507–518.
74. Karch SB. Karch's pathology of drug abuse. Boca Raton: CRC Press; 2002.
75. Luedde T, Kaplowitz N, Schwabe RF. Cell Death and Cell Death Responses in Liver Disease: Mechanisms and Clinical
Relevance. Gastroenterology. 2014;147.
76. Guicciardi ME, Malhi H, Mott JL, Gores GJ. Apoptosis and Necrosis in the Liver. Comprehensive Physiology. 2013;
77. Yoon S, Bogdanov K, Kovalenko A, Wallach D. Necroptosis is preceded by nuclear translocation of the signaling proteins
that induce it. Cell Death Differ Cell Death and Differentiation. 2015;23:253–260.
78. Silke J, Rickard JA, Gerlic M. The diverse role of RIP kinases in necroptosis and inflammation. Nature Immunology Nat
Immunol. 2015;16:689–697.
79. Baumann O, Walz B. Endoplasmic reticulum of animal cells and its organization into structural and functional domains.
International Review of Cytology. 2001;:149–214.
80. Braakman I, Hebert DN. Protein Folding in the Endoplasmic Reticulum. Cold Spring Harbor Perspectives in Biology. 2013;5.
81. Hartl FU, Bracher A, Hayer-Hartl M. Molecular chaperones in protein folding and proteostasis. Nature. 2011;475:324–332.
82. Hetz C, Chevet E, Oakes SA. Erratum: Proteostasis control by the unfolded protein response. Nature Cell Biology Nat Cell
Biol. 2015;17:1088–1088.
83. Hetz C. The unfolded protein response: controlling cell fate decisions under ER stress and beyond. Nature Reviews Molecular
Cell Biology Nat Rev Mol Cell Biol. 2012;
84. Maly DJ, Papa FR. Druggable sensors of the unfolded protein response. Nature Chemical Biology Nat Chem Biol.
2014;10:892–901.
85. Chen Y, Brandizzi F. IRE1: ER stress sensor and cell fate executor. Trends in Cell Biology. 2013;23:547–555.
86. Li H, Korennykh AV, Behrman SL, Walter P. Mammalian endoplasmic reticulum stress sensor IRE1 signals by dynamic
clustering. Proceedings of the National Academy of Sciences. 2010;107:16113–16118.
87. Carrara M, Prischi F, Ali M. UPR Signal Activation by Luminal Sensor Domains. IJMS International Journal of Molecular
Sciences. 2013;14:6454–6466.
88. Cui W, Li J, Ron D, Sha B. The structure of the PERK kinase domain suggests the mechanism for its activation. Acta
Crystallogr D. 2011;67:423–428.
89. Li Y, Guo Y, Tang J, Jiang J, Chen Z. New insights into the roles of CHOP-induced apoptosis in ER stress. Acta Biochimica
et Biophysica Sinica. 2015;47:146–147.
90. Vattem KM, Wek RC. Reinitiation involving upstream ORFs regulates ATF4 mRNA translation in mammalian cells.
Proceedings of the National Academy of Sciences. 2004;101:11269–11274.
91. Hong M. Underglycosylation of ATF6 as a Novel Sensing Mechanism for Activation of the Unfolded Protein Response.
Journal of Biological Chemistry. 2004;279:11354–11363.
92. Nadanaka S, Okada T, Yoshida H, Mori K. Role of Disulfide Bridges Formed in the Luminal Domain of ATF6 in Sensing
Endoplasmic Reticulum Stress. Molecular and Cellular Biology. 2006;27:1027–1043.
93. Li B, Yi P, Zhang B, Xu C, Liu Q, Pi Z, et al. Differences in endoplasmic reticulum stress signalling kinetics determine cell
survival outcome through activation of MKP-1. Cellular Signalling. 2011;23:35–45.
94. Schindler AJ, Schekman R. In vitro reconstitution of ER-stress induced ATF6 transport in COPII vesicles. Proceedings of
the National Academy of Sciences. 2009;106:17775–17780.
95. Shen J, Prywes R. Dependence of Site-2 Protease Cleavage of ATF6 on Prior Site-1 Protease Digestion Is Determined by
the Size of the Luminal Domain of ATF6. Journal of Biological Chemistry. 2004;279:43046–43051.
118

96. Chen F. JNK-Induced Apoptosis, Compensatory Growth, and Cancer Stem Cells. Cancer Research. 2012;72:379–386.
97. Verma G, Bhatia H, Datta M. JNK1/2 regulates ER-mitochondrial Ca2 cross-talk during IL-1 -mediated cell death in
RINm5F and human primary -cells. Molecular Biology of the Cell. 2013;24:2058–2071.
98. Hassler J, Cao SS, Kaufman RJ. IRE1, a Double-Edged Sword in Pre-miRNA Slicing and Cell Death. Developmental Cell.
2012;23:921–923.
99. Tabas I, Ron D. Integrating the mechanisms of apoptosis induced by endoplasmic reticulum stress. Nature Cell Biology.
2011;13:184–190.
100. Teng Y, Gao M, Wang J, Kong Q, Hua H, Luo T, et al. Inhibition of eIF2α dephosphorylation enhances TRAIL-induced
apoptosis in hepatoma cells. Cell Death Dis Cell Death and Disease. 2014;5.
101. Han J, Back SH, Hur J, Lin Y-H, Gildersleeve R, Shan J, et al. ER-stress-induced transcriptional regulation increases protein
synthesis leading to cell death. Nature Cell Biology Nat Cell Biol. 2013;15:481–490.
102. Morishima N, Nakanishi K, Nakano A. Activating Transcription Factor-6 (ATF6) Mediates Apoptosis with Reduction of
Myeloid Cell Leukemia Sequence 1 (Mcl-1) Protein via Induction of WW Domain Binding Protein 1. Journal of Biological
Chemistry. 2011;286:35227–35235.
103. Amodio G, Venditti R, Matteis MAD, Moltedo O, Pignataro P, Remondelli P. Endoplasmic Reticulum stress reduces COPII
vesicle formation and modifies Sec23a cycling at ERESs. FEBS Letters. 2013;587:3261–3266.
104. Guan M, Su L, Yuan Y-C, Li H, Chow WA. Nelfinavir and Nelfinavir Analogs Block Site-2 Protease Cleavage to Inhibit
Castration-Resistant Prostate Cancer. Sci. Rep. Scientific Reports. 2015;5:9698.
105. Liu M, Dudley SC. Targeting the unfolded protein response in heart diseases. Expert Opinion on Therapeutic Targets.
2014;18:719–723.
106. Hammadi M, Oulidi A, Gackiere F, Katsogiannou M, Slomianny C, Roudbaraki M, et al. Modulation of ER stress and
apoptosis by endoplasmic reticulum calcium leak via translocon during unfolded protein response: involvement of GRP78.
The FASEB Journal. 2013;27:1600–1609.
107. Reyskens KM, Essop MF. HIV protease inhibitors and onset of cardiovascular diseases: A central role for oxidative stress
and dysregulation of the ubiquitin–proteasome system. Biochimica et Biophysica Acta (BBA) - Molecular Basis of Disease.
2014;1842:256–268.
108. Ustundag Y. Proteasome inhibition-induces endoplasmic reticulum dysfunction and cell death of human cholangiocarcinoma
cells. World Journal of Gastroenterology WJG. 2007;13:851.
109. Reyskens KMSE, Fisher T-L, Schisler JC, O'connor WG, Rogers AB, Willis MS, et al. Correction: Cardio-Metabolic Effects
of HIV Protease Inhibitors (Lopinavir/Ritonavir). PLoS ONE. 2013;8.
110. Schmidtke G, Holzhutter H-G, Bogyo M, Kairies N, Groll M, Giuli RD, et al. How an Inhibitor of the HIV-I Protease
Modulates Proteasome Activity. Journal of Biological Chemistry. 1999;274:35734–35740.
111. Bernstein WB, Dennis PA. Repositioning HIV protease inhibitors as cancer therapeutics. Current Opinion in HIV and AIDS.
2008;3:666–675.
112. Domingo P, Sambeat MA, Pé rez A, Ordoñ ez J. Effect of Protease Inhibitors on Apolipoprotein B Levels and Plasma Lipid
Profile in HIV-1–Infected Patients on Highly Active Antiretroviral Therapy. JAIDS Journal of Acquired Immune Deficiency
Syndromes. 2003;33:114–116.
113. Vivithanaporn P, Asahchop EL, Acharjee S, Baker GB, Power C. HIV protease inhibitors disrupt astrocytic glutamate
transporter function and neurobehavioral performance. Aids. 2016;30:543–552.
114. Malhotra JD, Kaufman RJ. Endoplasmic Reticulum Stress and Oxidative Stress: A Vicious Cycle or a Double-Edged Sword?
Antioxidants & Redox Signaling. 2007;9:2277–2294.
115. Touzet O, Philips A. Resveratrol protects against protease inhibitor-induced reactive oxygen species production, reticulum
stress and lipid raft perturbation. Aids. 2010;24:1437–1447.
116. Poenie M. Measurement of Intracellular Calcium with Fluorescent Calcium Indicators. Intracellular Messengers. :129–174.
117. Michelangeli F, East JM. A diversity of SERCA Ca 2 pump inhibitors. Biochm. Soc. Trans. Biochemical Society
Transactions. 2011;39:789–797.
119

118. Periasamy M, Kalyanasundaram A. SERCA pump isoforms: Their role in calcium transport and disease. Muscle & Nerve
Muscle Nerve. 2007;35:430–442.
119. Peacock M. Calcium Metabolism in Health and Disease. Clinical Journal of the American Society of Nephrology. 2010;5.
120. Louch WE, Hougen K, Mø rk HK, Swift F, Aronsen JM, Sjaastad I, et al. Sodium accumulation promotes diastolic
dysfunction in end-stage heart failure following Serca2 knockout. The Journal of Physiology. 2010;588:465–478.
121. Ying J, Sharov V, Xu S, Jiang B, Gerrity R, Schö neich C, et al. Cysteine-674 oxidation and degradation of sarcoplasmic
reticulum Ca2 ATPase in diabetic pig aorta. Free Radical Biology and Medicine. 2008;45:756–762.
122. Westerink WM, Schoonen WG. Cytochrome P450 enzyme levels in HepG2 cells and cryopreserved primary human
hepatocytes and their induction in HepG2 cells. Toxicology in Vitro. 2007;21:1581–1591.
123. Borgese N. Getting membrane proteins on and off the shuttle bus between the endoplasmic reticulum and the Golgi complex.
Journal of Cell Science J Cell Sci. 2016;129:1537–1545.
124. Jensen D, Schekman R. COPII-mediated vesicle formation at a glance. Journal of Cell Science. 2010;124:1–4.
125. Hughes H, Stephens DJ. Assembly, organization, and function of the COPII coat. Histochemistry and Cell Biology.
2007;129:129–151.
126. Aridor M, Balch WE. Kinase Signaling Initiates Coat Complex II (COPII) Recruitment and Export from the Mammalian
Endoplasmic Reticulum. Journal of Biological Chemistry. 2000;275:35673–35676.
127. Gü rkan C, Stagg SM, Lapointe P, Balch WE. The COPII cage: unifying principles of vesicle coat assembly. Nature Reviews
Molecular Cell Biology. 2006;7:727–738.
128. Majoul I, Straub M, Hell SW, Duden R, Soeling H-D. KDEL-Cargo Regulates Interactions between Proteins Involved in
COPI Vesicle Traffic. Developmental Cell. 2001;1:139–153.
129. Pellett PA, Dietrich F, Bewersdorf J, Rothman JE, Lavieu G. Inter-Golgi transport mediated by COPI-containing vesicles
carrying small cargoes. eLife. 2013;2.
130. Spang A. Retrograde Traffic from the Golgi to the Endoplasmic Reticulum. Cold Spring Harbor Perspectives in Biology.
2013;5.
131. Cancino J, Capalbo A, Di Campli A, Giannotta M, Rizzo R, Jung JE, et al. Control Systems of Membrane Transport at the
Interface between the Endoplasmic Reticulum and the Golgi. Developmental Cell. 2014;30:280–294.
132. Lee TH, Linstedt AD. Potential Role for Protein Kinases in Regulation of Bidirectional Endoplasmic Reticulum-to-Golgi
Transport Revealed by Protein Kinase Inhibitor H89. Molecular Biology of the Cell. 2000;11:2577–2590.
133. Matsuto M, Kano F, Murata M. Reconstitution of the targeting of Rab6A to the Golgi apparatus in semi-intact HeLa cells:
A role of BICD2 in stabilizing Rab6A on Golgi membranes and a concerted role of Rab6A/BICD2 interactions in Golgi-to-
ER retrograde transport. Biochimica et Biophysica Acta (BBA) - Molecular Cell Research. 2015;1853:2592–2609.
134. Tian L, Dai LL, Yin ZJ, Fukuda M, Kumamaru T, Dong XB, et al. Small GTPase Sar1 is crucial for proglutelin and -
globulin export from the endoplasmic reticulum in rice endosperm. Journal of Experimental Botany. 2013;64:2831–2845.
135. Soo KY, Halloran M, Sundaramoorthy V, Parakh S, Toth RP, Southam KA, et al. Rab1-dependent ER–Golgi transport
dysfunction is a common pathogenic mechanism in SOD1, TDP-43 and FUS-associated ALS. Acta Neuropathologica Acta
Neuropathol. 2015;130:679–697.
136. Langhans M, Hawes C, Hillmer S, Hummel E, Robinson DG. Golgi Regeneration after Brefeldin A Treatment in BY-2 Cells
Entails Stack Enlargement and Cisternal Growth followed by Division. Plant Physiology. 2007;145:527–538.
137. Bannykh SI, Plutner H, Matteson J, Balch WE. The Role of ARF1 and Rab GTPases in Polarization of the Golgi Stack.
Traffic. 2005;6:803–819.
138. Sengupta P, Satpute-Krishnan P, Seo AY, Burnette DT, Patterson GH, Lippincott-Schwartz J. ER trapping reveals Golgi
enzymes continually revisit the ER through a recycling pathway that controls Golgi organization. Proceedings of the National
Academy of Sciences Proc Natl Acad Sci USA. 2015;112.
139. Seemann J, Jokitalo E, Pypaert M, Warren G. Matrix proteins can generate the higher order architecture of the Golgi
apparatus. Nature. 2000;407(6807):1022-6.
140. Ma R, Zhang J, Liu X, Li L, Liu H, Rui R, et al. Involvement of Rab6a in organelle rearrangement and cytoskeletal
120

organization during mouse oocyte maturation. Scientific Reports. 2016;6:23560.
141. Cederbaum AI. Ethanol Cytotoxicity to a Transfected HepG2 Cell Line Expressing Human Cytochrome P4502E1. Journal
of Biological Chemistry. 1996;271:23914–23919.

Abstract (if available)

Abstract Endoplasmic reticulum (ER) stress and unfolded protein response (UPR) is involved in anti-HIV drugs and alcohol-induced liver disease in a significant number of HIV-infected patients. However, the precise mechanism by which the drugs and alcohol cause the ER stress remains elusive. We reported that ritonavir-boosted lopinavir (RL) activated two canonical UPR branches without activation of the third canonical ATF6 branch in either HepG2 cells or primary mouse hepatocytes (PMH). In the RL-treated cells, ATF6 localization in the Golgi apparatus required for its activation was markedly reduced, which was proceeded by dramatic fragmentation of the Golgi structure and ER-like redistribution of Golgi-resident enzymes. Severities of the Golgi fragmentation induced by other anti-HIV drugs varied and were correlated with ER stress response. In the liver of mice fed RL, alcohol feeding deteriorated the Golgi fragmentation, which was correlating with ER stress, elevated ALT and liver steatosis. Golgi stress response (GSR) markers of GCP60 and HSP47 were increased in RL-treated liver cells and knockdown of TFE3 of GSR by siRNA worsened RL-induced cell death. Co-treatment of pharmacological agent H89 with RL inhibited the RL-induced ER-like redistribution of Golgi enzymes and ER stress. Moreover, the COPII complexes mediating ER-to-Golgi trafficking were accumulated in the RL-treated liver cells and the accumulation was not due to the interference of RL with the assembly of the COPII complexes. In addition, RL inhibited the Golgi fragmentation and reassembly induced by short treatment and removal of brefeldin A. Our study elucidates a novel ER-to-Golgi trafficking mechanism by which the anti-HIV drug and/or alcohol induce ER stress and hepatic injury, which provides more precise therapeutic targets for HIV-infected patients suffering from drug hepatotoxicity.

Linked assets

University of Southern California Dissertations and Theses

Conceptually similar

PDF

Cell surface translocation mechanism of stress-inducible GRP78 in human cancer

PDF

Pathways of cell death in response to HIV protease cocktail Ritonavir/Lopinavir in primary hepatocytes

PDF

NAP-2100 and analogues have improved anti-cancer effect in TNBC through SERCA2 depletion induced ER stress

PDF

PRAS40 connects microenvironmental stress signaling to exosome-mediated secretion

PDF

The roles of endoplasmic reticulum chaperones in regulating liver homeostasis and tumorigenesis

PDF

Examining perillyl alcochol in glioblastoma treatment and relating its effect to endoplasmic reticulum stress response to further advance its usage in combination treatment with dimethyl-celecoxib

Asset Metadata

Creator Han, Hui (author)

Core Title Disrupted ER-to-Golgi trafficking underlies anti-HIV drugs and alcohol induced cellular stress and hepatic injury

School Keck School of Medicine

Degree Doctor of Philosophy

Degree Program Genetic, Molecular and Cellular Biology

Publication Date 02/13/2019

Defense Date 10/24/2016

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

Tag ER stress,ER-Golgi trafficking,golgi stress,hepatotoxicity of drugs/alcohol,OAI-PMH Harvest,UPR

Language English

Contributor Electronically uploaded by the author (provenance)

Advisor Ou, James (committee chair), DeLeve, Laurie (committee member), Ji, Cheng (committee member), Zandi, Ebrahim (committee member)

Creator Email bzhhwll@gmail.com,huihan@usc.edu

Permanent Link (DOI) https://doi.org/10.25549/usctheses-c40-337019

Unique identifier UC11258131

Identifier etd-HanHui-5056.pdf (filename),usctheses-c40-337019 (legacy record id)

Legacy Identifier etd-HanHui-5056.pdf

Dmrecord 337019

Document Type Dissertation

Rights Han, Hui

Type texts

Source University of Southern California (contributing entity), University of Southern California Dissertations and Theses (collection)

Access Conditions The author retains rights to his/her dissertation, thesis or other graduate work according to U.S. copyright law. Electronic access is being provided by the USC Libraries in agreement with the a...

Repository Name University of Southern California Digital Library

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