Mucosal and Systemic Immune Phenotype Is Altered

MUCOSAL AND SYSTEMIC IMMUNE PHENOTYPE IS ALTERED DURING HIV-1 INFECTION AND IS PARTIALLY RESTORED AND FURTHER DISRUPTED IN THE ABSENCE OF DETECTABLE VIRAL REPLICATION

MARIE ROSE MCCAUSLAND

Submitted in partial fulfillment of the requirements for the degree of

Doctor of Philosophy

Department of Molecular Biology and Microbiology

CASE WESTERN RESERVE UNIVERSITY

January, 2017

CASE WESTERN RESERVE UNIVERSITY SCHOOL OF GRADUATE STUDIES

We hereby approve the thesis/dissertation of

Marie Rose McCausland

candidate for the Doctor of Philosophy*

Committee Chair John Tilton

Committee Member David McDonald

Committee Member Calvin Cotton

Committee Member Alan Levine

Date of Defense July 25, 2016

*We also certify that written approval has been obtained for any proprietary material contained therein

This dissertation is dedicated to my loving and amazing husband Jeffrey Alexander

McCausland.

iii

Table of Contents List of Tables ...... ix

List of Figures ...... x

Acknowledgements ...... 1

Abbreviations ...... 3

Abstract ...... 5

Chapter 1: Introduction ...... 7

HIV ...... 8

The Epidemic ...... 8

The Virus and Viral Life Cycle ...... 9

The Advent of Combination Antiretroviral Therapy ...... 13

Living with HIV in the Treatment Era- Immune successes and Immune failures ...... 17

Increased plasma indices of inflammation and correlation to increased morbidity and

mortality ...... 18

Monocytes in Disease ...... 18

Intestinal Barrier Function and Intestinal Epithelial Cell Polarity ...... 20

Intestinal Barrier function is decreased in HIV disease ...... 23

Toll-Like Receptors ...... 25

The Microbiome ...... 28

Mucosal Immune cells depleted in HIV disease and our resulting model of systemic

inflammation ...... 29

Chapter 2: Altered Monocyte Phenotype in HIV-1 Infection Tends to Normalize with

Integrase-Inhibitor-Based Antiretroviral Therapy ...... 32

Summary ...... 33

Background ...... 33

Methods ...... 33

Results...... 33

Conclusions ...... 34

Introduction ...... 34

Results ...... 35

Patient and Control Samples ...... 35

Monocyte Phenotypes in Fresh and Cryopreserved PBMCS ...... 36

Altered Proportions of Monocyte Subsets in Untreated HIV-1 Infection ...... 37

Monocyte Phenotypes are Altered in Untreated HIV-1 Infection and Change with ART

...... 38

Greater HLA-DR Expression and Density in Untreated HIV-1 Infection, Tend to

Decrease with ART ...... 38

Greater CD86 Expression and Density on Monocytes in Untreated HIV-1 Infection

decrease with ART ...... 39

Lower CD40 Density on Traditional Monocytes in Untreated HIV-1 Infection ...... 40

Chemokine Receptor (CCR2 and CX3CR1) Expression on Inflammatory and Patrolling

Monocyte Subsets is Lower in Untreated HIV-1 Infection and Normalizes with ART ...... 41

Discussion ...... 43

Methods ...... 46

Ethics Statement ...... 46

Study Design ...... 47

Sample Collection ...... 47

Flow Cytometry ...... 48

Statistics...... 49

Chapter 3- Toll Like Receptors in the Intestine are altered in HIV disease ...... 68

Summary ...... 69

Introduction ...... 71

Results ...... 74

TLR expression is increased and TLR location modified in the epithelium and lamina propria of Viremic HIV+ patients ...... 74

Recovery of TLR 4 and TLR 9 expression in the Immune Success Population; Toxicity of ART for epithelial TLR3 Expression ...... 82

Elevated TLR expression during a Failure to reconstitute immune homeostasis ...... 85

Inflammation and H&E staining ...... 88

Discussion ...... 90

Materials and Methods ...... 94

Patient Data ...... 94

Immunofluorescence ...... 96

Metamorph ...... 97

Statistics...... 98

vii

Chapter 4: Discussion ...... 100

Summary of Findings ...... 101

Discussion ...... 105

Future Directions ...... 108

Confirm TLR expression alterations in Colonic Epithelium of HIV (+) and HIV (-) individuals using Flow cytometry...... 108

TLR ligands and cytokines alter the abundance and localization of TLRs on the colonic epithelium and lamina propria ...... 115

Evaluate the toxicity of an HIV virion and ART compounds on TLR abundance and localization in Colonic Epithelium and Lamina Propria ...... 120

Concluding Remarks ...... 121

viii

List of Tables Table 1.1. Abridged functions of the Human Immunodeficiency Virus proteins. .... 11

Table 1.2. Timeline of HIV drug discovery [36]...... 15

Table 1.4. Toll-Like Receptors and their Foreign Ligands...... 26

Table 1.5. Toll-Like Receptors and their Endogenous Ligands...... 26

Table 2.1. Patient characteristics...... 51

Table 2.2. Alterations in frequencies of surface marker expression on total monocytes and monocyte subsets of HIV-1-infected patients before and after initiation of

ART...... 56

Table 2.3. Alterations in surface marker density on total monocytes and monocyte subsets of HIV-1-infected patients before and after initiation of ART...... 57

Table 2.4. Mean and Standard Error for frequencies of surface marker expression on total monocytes and monocyte subsets of HIV-1-infected patients before and after initiation of ART...... 58

Table 2.5. Mean and Standard Error for density of surface marker expression on total monocytes and monocyte subsets of HIV-1-infected patients before and after initiation of

ART...... 59

Table 3.1 Table of Characteristics for both HIV (-) and HIV (+) samples...... 95

List of Figures Figure 1.1. Diagram of the Human Immunodeficiency Virus...... 12

Figure 1.2. Model of Colonic Epithelium...... 22

Figure 1.3 Model of Intestinal Dysfunction...... 31

Figure 2.1. Gating strategy for flow cytometry and comparison between fresh and cryopreserved monocyte surface marker expression...... 53

Figure 2.2. Monocyte subset proportions at baseline and after ART initiation compared to proportions among controls...... 55

Figure 2.3. Expression and density (MFI) of HLA-DR on patient monocytes at baseline and after ART initiation compared to values among controls...... 61

Figure 2.4. Expression and density (MFI) of CD86 on patient monocytes at baseline and after ART initiation compared to values among controls...... 63

Figure 2.5. Expression and density (MFI) of CCR2 on patient monocytes at baseline and after ART initiation compared to values among controls...... 65

Figure 2.6. Expression and density (MFI) of CX3CR1 on patient monocytes at baseline and after ART initiation compared to values among controls...... 67

Figure 3.1. Thresholding using Metamorph...... 75

Figure 3.2. Quantification of TLR expression in colonic sections using Metamorph software...... 77

Figure 3.3. Immunofluorescence staining of TLR3, TLR4, and TLR9 in Controls,

Viremic Patients, Imuune Successes, and Immune Failures...... 79

Figure 3.4 Toll-Like Receptor Expression Alterations in the Surface Epithelium,

Crypt, and Lamina Propria of Viremic Patients and Uninfected Control Colons...... 81

Figure 3.5 Toll-Like Receptor Expression Alterations in the Surface Epithelium,

Crypt, and Lamina Propria of Immune Successes, Viremic Patients, and Uninfected

Control Colons...... 84

Figure 3.6. Toll-Like Receptor Expression Alterations in the Surface Epithelium,

Crypt, and Lamina Propria of Immune Failures, Immune Successes, Viremic Patients, and Uninfected Control Colons...... 87

Figure 3.7. H&E images at 20x...... 89

Figure 4.1. Phenotypic alterations in the colon of patients with HIV-1...... 104

Acknowledgements I would first like to thank my amazing husband, Jeffrey McCausland. You have made me a better version of myself through your kindness and your ambition and your unwavering support and love. You held my hand when I wanted to quit, and you told me

I was worth it and could do this. You also made me food and coffee, which is pretty important. I love you so much.

I would also like to thank my family and my husband’s family for their love and support throughout my 20+ years of schooling. I’d like to thank my parents, Bill and Susie

Ebner, for fostering my love for books and science and learning, through their example.

I’d like to thank my mom for showing me how to be a strong independent woman, and my dad for sending me handwritten cards, just when I needed them, that told me I was going to be a “great science nerd” one day. Thank you to my brother, Billy, and his family (Carrie, Taylor, Tyler, and Hannah) for giving me a reason to do childish science experiments again. I am also so grateful to have found a second family and home in

Jeff’s family, Elaine, Bill, and Chris McCausland. I thank them for all of their love and emotional support, and for treating me like one of their own.

I would also like to thank the amazing network of friends that I have made during my time here at Case Western Reserve University. I’d like to specifically thank Steve

Juchnowski for being a great friend and fellow lover of cat calendars and monocytes, and

Mark Lucera for always having time for coffee, venting, and troubleshooting (science and otherwise). I’d like to thank Chelsey Judge and Elane Reyes for their support and words

of encouragement and amazing friendship. Thank you to the members of the Lederman and Sieg lab for providing a wonderful and fun environment for the first three years of my graduate degree, and the members of the Levine lab for making the past year so memorable. Thank you also to Olu. You called me “The Dr.” long before this day, and told me during my hardest times that I just needed to “speak it in to existence”. Your friendship has been a blessing.

I would also like to thank my amazing committee. Thank you Dr. Tilton and Dr.

McDonald for helping me to navigate an unbelievably difficult situation with grace. You were my advocates, and for that I am forever grateful. Thank you also to Dr. Cotton for joining my committee late in the game and for giving me a fresh perspective on my project.

Finally, I would like to thank my advisor, Dr. Alan Levine. Words cannot fully express my gratitude for you, Alan. You showed me what a great mentor can be, and I am forever grateful for the time that I spent in your lab under your mentorship. Thank you for reinvigorating my love for academia through your own passion, and helping me to remember my own potential.

Abbreviations

ACS-Acute Coronary Syndrome AIDS - Acquired Immunodeficiency Syndrome AMP-Anti-microbial Peptides APC-Allophycocyanin ART- Antiretroviral Therapy BD-Becton Dickinson BI- Boehringer Ingelheim BMS- Bristol-Myers Squibb CA-capsid CDC- Center for Disease Control CRFs- Circulating Recombinat forms DMSO-Dimethyl Sulfoxide DTT-Dithiothreitol EDTA-Ethylenediaminetetraacetic acid FBS-Fetal Bovine Serum FDA- Food and Drug Administration FTC/TDF-Emtricitiabine/tenofovir disroxil fumarate GEE- Generalized Estimating Equation GRID- Gay-Related Immunodeficiency GSK-GlaxoSmithKline HBSS-Hanks Balanced Salt Solution HIV – Human Immunodeficiency Virus HMGB1- High Mobility Group Box Protein 1 IBD-Inflammatory Bowel Disease IF-Immune Failure IFN-Interferon IL-Interleukin INI-Integrase Inhibitor IRFs- Interferon Regulatory Factors IS-Immune Success J&J-Johnson and Johnson LPS-Lipopolysaccharide MA- Matrix MBN-Morel, Bokossa, and Neerchal MFI-Mean Fluorescence Intensity NC- Nucleocapsid NES- Nuclear Export Signal NIH-National Institutes of Health NRTI-Nucleoside Analog Reverse Transriptase Inhibitor NNRTI-Non-nuceloside Reverse Transcriptase Inhibitors PAMPs-Pathogen Associated Molecular Patterns PBMC-Peripheral Blood Mononuclear Cells PE-Phycoerythrin PerCP-Peridinin Chlorophyll protein PIC- Pre-integration Complexes PCP- Pneumocystis carinii pneumonia PKE-Pharmokinetic enhancer PSA- Polysaccharide A

RAL-Raltegravir RT- Reverse Transcriptase SIV-Simian Immunodeficiency Virus TAR- Transactivation response element TER- Transepithelial Resistance TLR-Toll-Like Receptor TJ- Tight Junction

TNF-Tumor Necrosis Factor

Mucosal and Systemic Immune Phenotype is Altered During HIV-1 Infection and is Partially Restored and Further Disrupted in the Absence of Detectable Viral Replication

Abstract

MARIE ROSE MCCAUSLAND

Human Immunodeficiency Virus (HIV) infection, while now treatable with antiretroviral therapy (ART), remains a problem for millions of patients worldwide. Patients on ART continue to have an increased risk of Acquired Immunodeficiency Syndrome (AIDS)-related and non-

AIDS-related diseases such as cardiovascular disease, cancer, and neurocognitive defects. One potential reason for the increased morbidity and mortality seen in HIV (+) patients is the intestinal barrier defect known to be present in these patients. Decreased barrier function allows for microbial translocation, localized inflammation, partially due to Toll-Like receptor (TLR) ligation, and eventually systemic immune activation. We analyzed the phenotype of monocytes, known players in cardiovascular disease which travel to sites of inflammation, by flow cytometry, as well as TLR phenotype in the colon by immunofluorescence. Our results showed an increase in monocyte activation (HLA-DR, CD86) and decreased chemokine receptor expression (CCR2,

CX3CR1) in viremic HIV (+) patients when compared to uninfected controls, which was only partially normalized with ART. Additionally, CD40 is further disrupted on monocytes with ART.

Similarly, we find increased TLR3, TLR4, and TLR9 expression in the colons of Viremic HIV

(+) patients, whose abundance and location do not fully normalize (TLR4, TLR9), or was further disrupted by ART (TLR3). Together our data suggest that ART, which is successful at controlling viral replication in tissue and HIV RNA levels found in the blood, does not allow for

normalization of the immune activation seen in the HIV (+) patients, and in fact could be contributing to the continued pathology.

Chapter 1: Introduction

HIV

The Epidemic

Human immunodeficiency virus, or HIV, is a retrovirus which was identified as the causative agent in a deadly epidemic that began its rise in the 1980’s. In the early days of the epidemic, clinicians were perplexed as they began seeing patients with enlarged lymph nodes, Kaposi’s sarcoma, and Pneumocystis carinii pneumonia (PCP) which typically occurred in patients with immunodeficiency[1]. These patients, however, were not on immunosuppressive drugs and they had no other clear immunodeficiency, which made their symptoms particularly puzzling. Additionally, the mortality rate was startlingly high (59%) in patients who were being diagnosed with both Kaposi’s sarcoma and PCP by March 1982[1]. One of the earliest reports noted that the majority of patients were homosexual men, and of the men and women who reported to be heterosexual, 56% were reported to be IV drug users. Remarkably, even in this 1982 report, the possibility of a new retrovirus being the causative agent in this disease outbreak was discussed, with transmission similar to that of Hepatitis B[1]. Unfortunately, by the time these symptoms were being tracked by the medical community, an estimated 100,000-200,000 people had already been infected with HIV[2].

Due to the prevalence of this immunodeficiency in the homosexual community, one of the early names for this disease was Gay-Related Immunodeficiency or GRID. It was apparent, though, that this was not a disease strictly constrained to the homosexual community and, eventually, the CDC adopted a more universal naming: Acquired

Immunodeficiency Syndrome or AIDS.

It wasn’t until 1984 that it was determined that a retrovirus, later termed Human

Immunodeficiency Virus (HIV), was the causative agent of AIDS[3].

The Virus and Viral Life Cycle

In 1984, three years after the disease was first described, two independent research groups led by Robert Gallo and Luc Montagnier found the virus HIV to be the causative agent of AIDS[3]. Since then, much has been discovered about the virus, its replication, and its diversity.

There are two types of HIV: HIV-1 and HIV-2. HIV-1 is the predominant type that is responsible for the HIV epidemic we think of in the 1980s. While HIV-2 infections occur, they are predominantly located in West Africa and is less virulent[4]. There are 4 strains of HIV-1: Group M (the ‘major’ group), Group N, Group O, and Group

P[5]. Additionally, Group M can be divided into 9 distinct subtypes (A, B, C, D, F, G, H,

J, K), or clades, and circulating recombinant forms (CRFs), which are hybrid viruses resulting from molecular recombination between viruses of multiple subtypes[5]. When

HIV is discussed in this dissertation, it is in reference to HIV-1.

HIV is a retrovirus with an RNA genome that is reverse transcribed to DNA and then inserted into the host genome. The HIV genome consists of around 10,000 base pairs and is composed of nine main genes: gag, pol, env, vif, vpr, tat, rev, vpu, and nef[6]. The functions of each are described in Table 1.1, and a diagram of their location is provided in Figure 1.1. The most important are the structural proteins, as they provide the basic functions for viral replication.

The first step in viral replication is entry into the cell, mainly CD4+ T cells or macrophages, via GP120 binding to CD4, and one of two accessory proteins (CCR5 or

CXCR4)[7]. CD4+ T cells in the gut have the highest proportion of CCR5, which is likely why there is such a large depletion in the gut[8]. After binding to the cell, GP41 mediates viral fusion and entry into the cell[7]. Once in the cell, the capsid is broken down, the RNA genome is reverse transcribed into DNA via the viral reverse transcriptase, and is transported into the nucleus as pre-integration complexes (PIC)[9].

Once in the nucleus, the PICs are then integrated into the host genome, using a viral integrase[9]. Once integrated into the genome, it will either remain latent (not actively producing virus), or the genome will be transcribed and translated using host machinery[10]. The virus is then assembled (2 copies of the RNA genome, Env, Gag, and Gag polyprotein-integrase, protease and reverse transcriptase) and released (budding) from the host lipid bilayer[11]. At this stage, the virus is not yet infectious, as it needs to undergo maturation where the viral protease cleaves the Gag-Pol poly-protein and allows for the formation of the capsid[11]. It was important to understand the functions of each protein and the role they play in the viral life cycle in order to find more specific and effective treatments for HIV.

Type Protein Name Known Functions REF Recruits two copies of the RNA viral genome [12] p55 Cleaved into four proteins by the viral protease during maturation [13] Located along the inner surface of the virion Matrix (MA) Aids in nuclear transport of the viral DNA, even in non-dividing cells [14] Gag Capsid (CA) Forms the core [15] Nucleocapsid Binds to the packaging signal on the viral RNA [16] (NC) Aids in reverse transcription [17] p6 Mediates p55 and Vpr interaction, aiding in Vpr's incorporation into virions [18] Protease Cleaves Gag and Gag-Pol polyprotein [19] Structural Reverse Transcriptase Makes a double-stranded DNA from RNA template Pol (RT) [20] Aids reverse transcription by removing the RNA template from the Rnase H RNA:DNA hybrid [21] Integrase Mediates integration of the HIV proviral DNA into host DNA [22] gp160 Cleaved by cellular protease to generate gp41 and gp120 [23] gp41 Mediates fusion between viral and cellular lipid bilayers [24] Env Exists in trimers gp120 Binds to CD4 Interacts with co-receptors (CCR5 and CXCR4) [25]

Binds to the transactivation response element (TAR), increasing transcription 1000-fold Transcriptional Tat Promotes elongation of transcripts [26] Transactivator Activate TGFb, TNFb [27, 28] Regulatory Downregulate bcl-2, MIP-1a [29, 30] Binds to Rev response element (RRE) and is required for HIV-1 replication

Rev Required to induce early to late gene expression transition [31] Has 3 domains: RNA binding domain, multimerization domain, and a nuclear export signal (NES)

Downregulate CD4 by trapping it in the Endoplasmic reticulum Vpu Increases viral release [32] Aids in localizing the preintegration complex (PIC) near the nuclear pore Vpr Accessory Can block cell cycle progression [33] Vif Essential for PBMC replication, but still unclear why [34] negative Decreases CD4 expression by increasing its degradation Nef factor Decreases Class I MHC [35]

Table 1.1. Abridged functions of the Human Immunodeficiency Virus proteins.

Figure 1.1. Diagram of the Human Immunodeficiency Virus.

The Advent of Combination Antiretroviral Therapy

In 1987 the first antiretroviral therapy was approved by the FDA, and now there are dozens of treatment options targeting multiple stages of the viruses life cycle[36] (Table

1.2). The first type of drug class was the nucleoside analog reverse transcriptase inhibitor

(NRTI) which incorporates into the virally synthesized c-DNA chain causing chain termination[36]. Unfortunately, NRTI’s also cause chain termination during human

DNA replication and therefore produce toxic side effects in the host[37]. The next class was the protease inhibitor (PI) which targets HIV protease, and therefore HIV maturation, using multiple mechanisms[36]. The non-nucleoside reverse transcriptase inhibitors

(NNRTI) were approved next[36] and these prevent reverse transcriptase from functioning by binding to an area near the active site; unlike NRTI’s, the NNRTI do not have off target host effects[38]. Enfuvirtide was the first fusion inhibitor, which blocks

HIV entry and fusion by blocking the viral protein GP41 and the conformational changes needed for fusion[36, 39]. Raltegravir was the first in the integrase inhibitor class, becoming approved in 2007[36, 40]. Maraviroc was the first CCR5 based entry inhibitor which works by binding to the co-receptor and preventing HIV from binding [36, 41].

Cobicistat is a pharmokinetic enhancer (PKE) that has no direct effect against HIV, but which boosts the drug levels of atazanavir or darunavir [42].

Single treatments, however, led to drug resistance because the HIV reverse transcriptase has an incredibly high error rate leading to mutations which made treatment ineffective [43, 44]. It then became clear that by using multiple drugs which target different stages of the viral life cycle, HIV drug resistance could be greatly reduced

However, this meant taking multiple pills a day with a specific regimen which was hard for patients to adhere to. Today there are several combination drugs on the market that are in a singular pill form, which makes combination therapy much more effective (Table

1.3). One of the most successful is Truvada which allowed delivery of both Tenofovir and Emtricitabine in a single pill (Table 1.3).

Year of FDA Trade Name Generic name Drug Class Company Approval 1987 Retrovir Zidovudine NRTI GSK 1991 Videx Didanosine NRTI BMS 1992 (discont. HIVID Zalcitabine NRTI Roche 2006) 1994 Zerit Stavudine NRTI BMS 1995 Eipivir Lamivudine NRTI GSK 1995 Fortovase/Invirase Saquinavir NRTI Roche 1996 Norvir Ritonavir PI Abbott 1996 Viramune Nevirapine NNRTI BI 1996 Crixiva Indinavir PI Merk 1997 Viracept Nelfinavir PI Pfizer 1997 Rescriptor Delavirdine NNRTI Pfizer 1998 Sustiva Efavirenz PI BMS 1998 Ziagen Abacavir NRTI GSK 1999 (discont. Agenerase Amprenavir PI GSK 2004) 2001 Viread Tenofovir NRTI Gilead 2003 Reyata Atazanavir PI BMS 2003 Emtriva Entricitabine PI Gilead 2003 Fuzeon Enfuvirtide entry/gp41 Roche 2004 Lexiva Fosamprenavir PI GSK 2005 Aptivus Tipranavir PI BI 2006 Prezista Darunavir 2nd-gen. PI Tibotec/ 2007 Isentress Raltegravir INI Merck 2007 Selzentry Maraviroc Entry/CCR5 Pfizer 2011 Edurant Rilpivirine NNRTI Janssen 2013 Tivicay Dolutegravir INI ViiV 2014 Vitekta Elvitegravir INI Gilead 2014 Tybost Cobicistat PKE Gilead

Table 1.2. Timeline of HIV drug discovery [36]. Abbreviations: Nucleoside Analog Reverse Transcriptase Inhibitor (NRTI), Non-nucleoside

Reverse Transcriptase Inhibitor (NNRTI), Protease Inhibitor (PI), Pharmokinetic enhancer (PKE),

Integrase Inhibitor (INI), GlaxoSmithKline (GSK), Bristol-Myers Squibb (BMS), Boehringer

Ingelbeim (BI).

Year of FDA Trade Name Generic name Drug Class Company Approval 1997 Combivir Zidovudine+ lamivudine NNRTIs GSK Abacavir + zidovudine + 2000 Trizivir NRTIs GSK lamivudine 2000 Kaletra Lopinavir + ritonavir PI Abbott 2004 Truvada Tenofovir + emtricitabine NRTIs Gilead 2004 Epzicom Abacavir + lamivudine NRTIs GSK Tenofovir + emtricitabine + NRTIs + Gilead and 2006 Atripla efavirenz NNRTI BMS Emtricitabine, rilpivirine, and PI + NNRTI + 2011 Complera Gilead tenofovir NRTI 2Elvitegravir, cobicistat, INI, PKE, PI, 2012 Stribild Gilead emtricitabine, and tenofovir NRTI 2015 Evotaz Atazanavir and cobicistat PI + PKE BMS 2015 Prezcobix Darunavir and cobicistat INI + PKE Janssen Table 1.3. Combination therapy for HIV [36]. Abbreviations: Nucleoside Analog Reverse

Transcriptase Inhibitor (NRTI), Non-nucleoside Reverse Transcriptase Inhibitor (NNRTI),

Protease Inhibitor (PI), Pharmokinetic enhancer (PKE), Integrase Inhibitor (INI),

GlaxoSmithKline (GSK), Bristol-Myers Squibb (BMS).

Today, while ART is now saving lives for many, HIV remains a huge public health concern. In a study conducted in 2014 it found 36.9 million people in the world were living with HIV, with cases in sub-Saharan Africa accounting for 70% of the global total, many of them children who contracted HIV through mother to child transmission [45].

Transmission occurs through an exchange of bodily fluids from infected individuals via sexual contact (semen or vaginal fluids), breast milk, and blood (sharing contaminated needles, non-sterile piercings or tattoos, and in the very early stages of the epidemic, blood transfusions)[45]. Additionally, 59% of people living with HIV are still unable to

access treatment in 2015[45]. Moreover, treatment is not currently curative and there are still sites which treatment may not reach, including the gut and brain[46]. These pitfalls in therapy leave patients with a lifelong prescription to ART, highlighting the importance of a cure as well as the giving patients with HIV a high quality of life.

Living with HIV in the Treatment Era- Immune successes and Immune failures Patients with HIV infection are placed on ART with the intent of decreasing their viral load below the limit of detection and recovering their CD4+ T cells above 500 cells/uL. Some patients on ART, however, fail to recover their CD4+ T cells and are deemed “immune failures.” Risk factors for immune failure include age, interrupted treatment, low nadir CD4+ T cell count (the lowest CD4+ T cell count for the patient), co-infections (HCV, TB), and incomplete viral suppression[47]. Immune failure patients have an increased risk of morbidity and mortality due to AIDS-related and non-AIDS related diseases, such as cancer and cardiovascular disease[48]. Because of these increased risks, understanding the underlying cause of the pathology of immune failure patients is of utmost importance. It is not yet understood why immune failure patients do not recover their CD4+ T cells, but there are several possible explanations including sequestered latent virus, dysfunctional replenishment from the thymus, decreased IL-7 homeostatic proliferation, lymphoid fibrosis[49] affecting activation and emigration of cells, and, important to our studies, microbial translocation and the resulting immune activation. We propose that perturbations in the gut mucosal barrier persist after ART administration and may contribute to immune failure through systemic immune activation. By looking at monocyte activation in peripheral blood mononuclear cells

(PBMCs) and characteristics of the epithelial cells in colonic tissue we hope to further our understanding of this phenomenon.

Increased plasma indices of inflammation and correlation to increased morbidity and mortality

To reiterate, decreased barrier function will not only cause local inflammation, but it may also cause systemic inflammation, which has very detrimental effects.

Importantly, it has been shown that immune activation is an important predictor of morbid outcomes and disease progression in HIV, with levels of IL-6, d-dimers, sCD14 correlating with increased mortality[50]. As sCD14 is an indicator of monocyte activation, and it is increased in HIV (+) patients compared to healthy controls, we predict that monocytes will be activated in the periphery. Additionally, sCD14 levels decrease with ART, but not to the levels seen in healthy patients; therefore, we would also predict that HIV+ patients on ART will normalize their extent of monocyte activation but not to that of a healthy uninfected control. We propose that perturbations in the gut mucosal barrier persist after ART administration and may contribute to continued systemic immune activation and failure to reconstitute the immune system.

Monocytes in Disease

There is increased risk for cardiovascular events in patients with HIV disease and efforts have begun to treat these complications with statins[51]. In parallel, there are efforts to understand the underlying monocyte phenotypes associated with cardiovascular disease. Monocytes are important mediators of plaque formation, can differentiate into

foam cells, and play a vital role in immune defense. There are three monocytes classifications: traditional (CD14++CD16-), inflammatory (CD14++CD16+), and patrolling (CD14l+, CD16++)[52]. The traditional monocytes produce the most reactive oxygen species, when compared to the other classifications, and in response to LPS, produce more of the anti-inflammatory cytokine IL10, though they also produce IL8, IL6,

CCL3, and CCL2[52]. Both the inflammatory monocytes and the traditional monocytes are efficient at phagocytosis[52], and inflammatory monocytes produce TNFa, IL1b,

IL10, IL8, and IL6 in response to LPS. Inflammatory monocytes are potent producers of

TNFa, IL1b, and IL6 in response to viral infections, specifically TLR7 (3M2) and TLR8

(3M13) agonists[52]. Patrolling monocytes, are less phagocytic, and patrol the endothelium of the blood vessels, responding mostly to viral products and inflammation[52].

In one study, researchers examined a cohort of HIV+ patients that were virologically suppressed, and it was found that monocyte CX3CR1 expression independently predicted carotid artery thickness in HIV-positive individuals[53]. In addition, monocytes of HIV(+) patients have a shared monocyte phenotype with ACS patients (patients which have recently had a heart attack)[54]. These studies highlight the importance of determining an expansive monocyte phenotype, since they focus on only a targeted number of monocyte markers. Using targeted approaches such as this, can allow us to better determine why patients with HIV have increased cardiac risk.

Interestingly, monocytes have also been found to have an altered phenotype in

Inflammatory Bowel Disease, with increased levels of activation (HLA-DR) and CCR9

in active disease[55], and patients with IBD have higher cardiovascular risk[56], creating a possible link between cardiovascular disease and inflammatory diseases stemming from the gut.

Intestinal Barrier Function and Intestinal Epithelial Cell Polarity

The intestinal tract is colonized by commensal microbiota, making an intact intestinal barrier essential. The intestinal tract is made up of columnar epithelial cells which are polar, meaning they have an apical and a basolateral side (Figure 1.2). In order to form an intact intestinal barrier, the epithelial cells require functional tight junctions[57], which prevent paracellular (between cell) trafficking. These tight junctions are located near the luminal surface and consist of several proteins: occludins, zonulins, and claudins[58].

In addition to tight junctions, there are several other junctional complexes between epithelial cells: gap junctions, adherens junctions and desmosomes. The adherens junctions are found apically, just below the tight junctions[58], and the desmos2omes are just under the adherens junctions; both add strength to the barrier[59].

Gap junctions are more basolateral, and are involved in intercellular communication[60].

Figure 1.2. Model of Colonic Epithelium. Colonic epithelial cells are columnar and polar; the apical side is near the lumen of the gut and the basolateral side is nearest the lamina propria. The gut lumen consists of commensal bacteria and sometimes pathoge2nic bacteria and viruses. In addition to the tight junctions, shown between the epithelial cells, a protective layer of mucus (mucin), antimicrobial peptides (AMPs), and

IgA (brought to the surface by pIgR) aid in protecting barrier function. Toll-like receptors can be found mostly on the basolateral side of the epithelium as well as in endosomes. TLR7, TLR8, and TLR9 are located in endosomes in the gut and interact

2with MyD88 to induce inflammatory cytokine production and Interferons. TLR3 is also found in endosomes and interact with TRIF to cause interferon production. TLR2, TLR5, and TLR9 can be found alongside TLR4 on the basolateral side of the epithelium and interact with MyD88. TLR4 can also interact with TRIF to induce inflammatory cytokine production and interferons. NFkB (shown as p50 and p65), AP1, and the

Interferon regulatory factors (IRFs) are transcription factors which increase the expressio2n of interferons and inflammatory cytokines.

The intestinal barrier consists of not only the junctional complexes formed between the intestinal epithelium, but also a thin mucous layer in the small intestine, a thicker mucous layer in the large intestine, antimicrobial peptides, and secreted antibodies IgA and IgM

(Figure 1.2) [61, 62]. When this essential barrier loses integrity, it allows for bacterial products to enter the lamina propria and can cause local and systemic inflammation.

In addition to the physical intestinal barrier, immune cells in the lamina propria play an important role in maintaining gut function. Dendritic cells sample the lumen of the gut with the help of M cells, a highly specialized epithelial cell that aids in presentation of peptides to the cells of the lamina propria[63], and subsequently aid in maintaining tolerance to resident gut flora[64]. Lamina Propria T cells have trafficked there after being activated in Peyers patches, or Gut-associated lymphoid tissue[65]. Th17 cells produce both IL17 and IL22 which aid in maintaining tight junctions between the epithelial cells and, along with T-regulatory cells[66], aid in regulating the mucosal response[67, 68]. B cells produce IgA antibodies which help to coat bacteria in the lumen of the gut, as well as in the instance of a barrier breach[69].

Some clinical outcomes of impaired intestinal barrier function include celiac disease and inflammatory bowel disease[70]. Important to our study, patients with HIV infection have also been found to have decreased or impaired intestinal barrier function and has been to predict mortality[71] through an unknown mechanism.

Intestinal Barrier function is decreased in HIV disease

In HIV disease, CD4+ T cells are not only depleted in the blood, but also in the intestinal tract at a much higher rate[8], and CD4+ T cell depletion is not fully restored after ART intervention[72, 73]. The depletion in CD4+ T cells in the gut coincides with alterations in the intestinal barrier integrity in patients with HIV.

Intestinal barrier integrity in HIV disease has been assessed using several methods, including urinary excretion of saccharides[74, 75]. These studies have highlighted increased lactulose: mannitol ratio in HIV patients when compared to uninfected controls, an indicator of increased small intestine permeability; additionally, sucralose excretion has been shown to be elevated in HIV+ patients when compared to healthy controls, an indicator of increased colonic permeability[76]. It was also recently found that tight junction protein ZO-1 mRNA is progressively decreased proximal to distal in the HIV(+) colon, suggesting decrease in ZO-1 at the entrance to the large intestine that becomes greater as you approach the rectum[77]. Taken together, this alludes to one potential mechanism for increased intestinal permeability in HIV (+) individuals[77].

It has also been found that TLR4 agonist lipopolysaccharide (LPS) is increased in the periphery of HIV+ patients, when compared to levels of LPS in healthy controls, indicating the ability of microbial products to traverse the intestinal barrier has increased in HIV patients[62]. Unfortunately, this decreased intestinal barrier integrity has been shown to continue with antiretroviral therapy, especially if treatment is delayed, and the reasons for this are not understood[78]. We believe a better understanding of barrier function in HIV disease can be uncovered by the

exploration of altered Toll-Like receptor abundance and localization in these tissues.

Toll-Like Receptors Toll-like receptors (TLRs) play an important role in intestinal barrier integrity and homeostasis[61, 79]. These receptors bind to bacterial and viral products (Table 1.4), as well as endogenous ligands (Table 1.5) resulting in a signaling cascade (Figure 1.2), which alerts the cell to invasion and intestinal barrier dysfunction by activating several transcription factors, resulting in the production of inflammatory cytokines and chemokines. IRF3 has several target genes including IFNα and IFNβ[80]; IRF5 targets

IL-6, IL-12, TNF, and NFKB[81]; IRF7 targets IFNα and IFNβ[82, 83]. AP1 has several targets including IL-10[84]. NFkB has an ever growing number of target genes, including IL-8[85, 86], IL-6[87-89], IL-12[90, 91], and TNF [92, 93]. In HIV disease,

IL-6 [94, 95], IL-8[96], and IFNalpha[97] are increased in HIV disease, suggesting TLR ligation as a potential source for the observed inflammation.

Ligand References TLR1 (with TLR2) triacyl lipoproteins [98] TLR2 liproteins, heat shock proteins [98, 99] TLR3 dsRNA, Poly I:C [100] TLR4 LPS [101] TLR5 Flagellin [102] TLR6 (with TLR2) diacyl lipoproteins [103] TLR7 ssRNA [104] TLR8 GU-containing RNA, Imidizaquinolines [104] unmethylated Bacterial DNA, CpG ODN Dna, RNA:DNA TLR9 [105, 106] hybrids

Table 1.4. Toll-Like Receptors and their Foreign Ligands.

Ligand References Biglycan, endoplasmin, HMBG1, HSP60, human cardiac TLR2 [107-118] myosin, hyaluronan and monosodium urate crystals TLR3 mRNA [119, 120] Biglycan, CD138, HMBG1, HSP60, HSP70, HSP72, fibronectin, [107, 111, 118, TLR4 heparin sulfate 121-130] TLR7 RNA and siRNA [131-133] TLR8 Human cardiac myosin and siRNA [112, 133] TLR9 DNA and HMBG1 [134-136]

Table 1.5. Toll-Like Receptors and their Endogenous Ligands.

The work presented here focuses on TLR3, TLR4 and TLR9. TLR3s ligands include double stranded RNA and Poly I:C (Table 1.4), as well as endogenous mRNA (Table

1.5). In the context of HIV infection, secondary structures generated by viral RNA act as

double-stranded RNA pathogen associated molecular patterns (PAMPs) to trigger TLR3.

TLR9s ligands include unmethylated bacterial DNA and RNA:DNA hybrids, as well as endogenous DNA and HMGB1. During the replication cycle of HIV, RNA:DNA hybrids are made, and these can not only bind to TLR9 by themselves[106], act in concert with the alarmin, High mobility Group Box Protein-1 (HMGB1)[136]. TLR4s ligands include

LPS, as well as several endogenous ligands including HMGB1[123].

Importantly, TLR3, TLR4, and TLR9 are located along the surface of the colonic epithelium, whereas in most other cell types TLR3 and TLR9 are located in endosomes.

In the native state, they are located on the basolateral side, versus the apical side, such that they will only see their ligand during infection or a breach in the intestinal barrier.

Additionally, their location along the surface allows for their ligation without active infection. To aid in this process, the alarmin HMBG1, which is released upon cell death, can bind to TLR2, TLR4, TLR5 and TLR9 ligands and present them to the TLR[109,

123, 136]. This protein is thought to play a role in linking microbial translocation to

TLR stimulation and the resultant immune response by binding to CPG (TLR9), LPS

(TLR4), and flagellin (TLR5)[137, 138]. It is possible that this is playing a role in the altered TLR abundance and localization that we are seeing in HIV disease, both prior to and after ART in the IF patients. Given this information, it is unsurprising that HMBG1 has been found to be elevated in the plasma of patients with HIV-1 and correlated with disease progression [139].

Both TLR2 and TLR5 ligation in intestinal epithelial cells in vitro has been shown to increase transepithelial resistance (TER), where increased resistance correlates with increased barrier integrity[140]. TLR3 and TLR4 are known to be altered in

inflammatory bowel disease[140], and TLR9 has been shown to either induce or repress inflammatory responses depending on the polarity of the signal, where apical TLR9 stimulation is anti-inflammatory and basolateral TLR9 stimulation is pro- inflammatory[79]. TLRs have not been studied in colonic epithelium of HIV (+) patients, representing a novel area of study which could yield high impact results. To date, only one study in Rhesus Macaques with Simian Immunodeficiency Virus (SIV), has shown altered TLR RNA in the SIV infected colon when compared to healthy macaque colon[141], lending evidence to our theory. Alterations in the polarity, or the responses to polar stimuli, in HIV disease could be a potential mechanism for the continued inflammation and activation seen in HIV disease and after ART administration.

The Microbiome The microbiome refers to the billions of bacteria resident in the gastrointestinal tract, mainly the colon. These resident bacteria, along with the resident mycobiome-or fungus- in the gastrointestinal tract aid in the processing of food [142]. The intestinal tract is colonized as an infant, and colonization appears important in developing immune tolerance, through mechanisms involving T-regulatory cells and Th17 cells [66].

Segmented filamentous bacteria, has been reported to induce Th17 cells in the small intestine. Th17, as discussed earlier, are important in maintaining gut integrity through the secretion of IL-17. Polysaccharide A (PSA), expressed by B. fragilis, has been shown to mediate the conversion to Tregs via TLR2, a cell type which suppresses Th17 responses[143]. Tregs are important in decreasing inflammation and regulating host response by secreting TGFb and IL10[143].

Perturbations in the resident commensal microenvironment, or dysbiosis, have been found in diseases such as IBD[144], and patients with allergies[145-148]. More recently, dysbiosis has been reported in the HIV-infected population [149, 150]. Coupled with possible TLR alterations which aid in the regulation of T cell differentiation, dysbiosis could lead to further CD4+ T cells depletion in the gut and loss of the tolerance mechanism. As is pointed out though in Li, et al, studies assessing the microbiome are wrought with difficulty, as there are many influencing factors including lifestyle, diet, age, gender, as well as HIV-associated problems such as the degree of immunodeficiency, co-infection, antibiotic use (a major confounder), ART duration/type, and degree of viral suppression. Finding matching cohorts for research studies is crucial when measuring alterations in the microbiota, and in determining downstream mechanisms.

Mucosal Immune cells depleted in HIV disease and our resulting model of systemic inflammation

It has long been known that CD4+ T cells are rapidly depleted in the intestinal mucosa, much more rapidly than that in the periphery. Here we propose our model of gut barrier dysfunction after the initial CD4+ T cell depletion due to HIV infection, leading to increased localized inflammation and systemic indices of inflammation, such as plasma IL6 and monocyte activation (Figure 1.3).

(1) The intestinal epithelium forms a tight barrier between the lumen, which contains commensal microbiota, and the lamina propria, which contains a number of immune

cells. The colonic epithelium is lined with TLRs which are pro-inflammatory when activated basally (or on the side of the lamina propria), while apical (on the side of the lumen) TLRs are anti-inflammatory. This polarizing-tolerizing mechanism allows for the intestinal lumen to be colonized without chronic colitis. Macrophages which reside in the lamina propria are largely tolerant, expressing extremely low levels of HLADR, CD80,

CD86, CD14, and CD16.

In HIV infection (2), CD4+ T cells are infected, including Th17 cells. The Th17 CD4+ T cells normally express the cytokines IL17 and IL22, which maintain gut barrier integrity.

(3) Infected CD4+ T cells perish, causing decreased expression of maintenance cytokines

IL17 and IL22. (4) Without IL17 and IL22, gut barrier integrity is diminished. This diminished gut barrier integrity can cause alterations in TLR abundance and localization.

(5) Due to the decreased barrier integrity, bacterial products are allowed to enter into the lamina propria which activate basolateral TLRs and macrophages. (6) These macrophages will then express higher levels of HLADR, CD80, and CD86 leading to decreased tolerance of intestinal microbiota. Macrophages which have become activated will produce IL6, which is known to be elevated in the peripheral blood of HIV (+) patients. (7) IL6 will then cause CD127, the IL7 receptor, to be down-regulated on the surface of immune cells. (8) Without CD127, immune cells are unable to respond to

IL7, a homeostatic cytokine, and also perish, causing further decreases in IL17 and IL22.

This further decrease in IL17 and IL22 will then cause further decreases in gut barrier integrity.

Figure 1.3 Model of Intestinal Dysfunction. Description is in text.

Our goal is to understand the phenotypic inflammatory alterations present in patients with

HIV disease, both prior to and after ART. We examined immune phenotype both peripherally in Chapter 2 and the intestinal mucosa in Chapter 3, using a combination of techniques including flow cytometry and immunofluorescence.

Chapter 2: Altered Monocyte Phenotype in HIV-1 Infection Tends to Normalize with

Integrase-Inhibitor-Based Antiretroviral Therapy

This chapter can be found published in PlosOne [151].

Summary

Background

Monocytes are increasingly implicated in the inflammatory consequences of HIV-1 disease, yet their phenotype following antiretroviral therapy (ART) initiation is incompletely defined. Here, we define more completely monocyte phenotype both prior to ART initiation and during 48 weeks of ART.

Methods

Cryopreserved peripheral blood mononuclear cells (PBMCs) were obtained at baseline

(prior to ART initiation) and at weeks 12, 24, and 48 of treatment from 29 patients participating in ACTG clinical trial A5248, an open label study of raltegravir/emtricitibine/tenofovir administration. For comparison, cryopreserved PBMCs were obtained from 15 HIV-1 uninfected donors, each of whom had at least two cardiovascular risk factors. Thawed samples were stained for monocyte subset markers

(CD14 and CD16), HLA-DR, CCR2, CX3CR1, CD86, CD83, CD40, CD38, CD36,

CD13, and CD163 and examined using flow cytometry.

Results

In untreated HIV-1 infection there were perturbations in monocyte subset phenotypes, chiefly a higher frequency and density (mean fluorescence intensity–MFI) of HLA-DR

(%-p = 0.004, MFI-p = .0005) and CD86 (%-p = 0.012, MFI-p = 0.005) expression and lower frequency of CCR2 (p = 0.0002) expression on all monocytes, lower CCR2 density on inflammatory monocytes (p = 0.045) when compared to the expression and density of these markers in controls’ monocytes. We also report lower expression of CX3CR1 (p =

0.014) on patrolling monocytes at baseline, compared to levels seen in controls. After

ART, these perturbations tended to improve, with decreasing expression and density of

HLA-DR and CD86, increasing CCR2 density on inflammatory monocytes, and increasing expression and density of CX3CR1 on patrolling monocytes.

Conclusions

In HIV-1 infected patients, ART appears to attenuate the high levels of activation (HLA-

DR, CD86) and to increase expression of the chemokine receptors CCR2 and CX3CR1 on monocyte populations. Circulating monocyte phenotypes are altered in untreated infection and tend to normalize with ART; the role of these cells in the inflammatory environment of HIV-1 infection warrants further study.

Introduction

Monocytes are increasingly recognized as contributors to inflammation and coagulation in HIV-1 infection [53, 54, 152]. These antigen-presenting cells can be segregated into three functionally distinct populations based on CD14 and CD16 expression [153, 154]. “Traditional” monocytes express high levels of CD14, are lacking

CD16 (CD14+CD16-), and produce pro-inflammatory cytokines in response to microbial elements, though to a lesser degree than do “inflammatory” monocytes (CD14+CD16+)

[52]. “Patrolling” monocytes (CD14dimCD16+) produce IL6 and IL8 in response to viral elements, and patrol the vascular endothelium [52]. Increased proportions of both the inflammatory and patrolling monocytes have been reported previously in untreated

HIV-1 infected patients when compared to the proportions in a healthy control population[54]. This nomenclature describes the function of these monocytes; others

have characterized these cells as classical, intermediate and non-classical monocyte subsets respectively[154].

Since monocyte phenotype perturbations in HIV-1 infection and changes in monocyte phenotype with antiretroviral therapy (ART) are incompletely defined, we implemented a flow cytometry panel for cryopreserved cells that explored the expression and density of: activation and maturation markers, HLA-DR, CD38, CD13, and CD83; the co-stimulatory molecules CD40 and CD86; chemokine receptors CCR2 and

CX3CR1; and the scavenger receptors CD36 and CD163. Using this monocyte phenotyping panel, we found that in untreated HIV-1 infection there is lower density of

CCR2 on inflammatory monocytes and lower expression of CX3CR1 on patrolling monocytes. We also found that untreated HIV-1-infected individuals had higher expression of HLA-DR and CD86 on total blood monocytes, and on most subsets, reflective of increased activation. Finally, we reported that many, but not all indices, normalized after ART.

Results

Patient and Control Samples

HIV-1 infected patients were 86% male, with a median age of 46, and fewer than half were White Non-Hispanic (Table 2.1). Before ART initiation, the median plasma

HIV-1 RNA and CD4+ T cell count were 34,469 copies/mL and 283 cells/uL respectively. The control population was 80% male, 67% White non-Hispanic, with a median age of 48, and one-third of the subjects were cigarette smokers. Smoking data were not available for the HIV-1 infected patients.

Monocyte Phenotypes in Fresh and Cryopreserved PBMCS

Multicenter clinical trials designed to monitor immune cell subsets typically utilize cryopreserved samples since multi-parameter flow cytometry using fresh blood samples has not yet been standardized to support on-site performance at multiple centers.

While most T cell phenotypes are stable with cryopreservation [155], the stability of monocyte phenotypes after cryopreservation is not adequately described, and certain monocyte functions have been found to be diminished after cryopreservation [156, 157].

We therefore needed to determine which monocyte surface markers were relatively unaltered with cryopreservation.

We obtained peripheral blood mononuclear cell (PBMC) samples prepared by

Ficoll density sedimentation of EDTA-anti-coagulated whole blood from four healthy controls and four HIV-1-infected subjects and stained a portion of each sample with our monocyte phenotyping panel. The remaining cells were cryopreserved and later thawed and stained using the same phenotyping panel.

Monocyte subset distributions in fresh and cryopreserved monocytes are similar

(Fig. 2.1a). Expression of CD40, CD163, CD86, CD38, HLA-DR, CCR2, CX3CR1, and

CD13 on monocytes was consistent in fresh and cryopreserved PBMCs of healthy (Fig.

2.1B) and virologically suppressed HIV-1-infected subjects (Fig. 2.1C), though the HIV-

1-infected subjects appear to have higher variability in staining. The density (MFI) of

CD36 appeared to be greater in the cryopreserved PBMCs of the healthy subjects, though this was driven by relatively low expression of CD36 in the fresh preparation from one healthy subject. Although the expression of CD83 was relatively diminished in cryopreserved PBMCs of the controls, we elected to retain this marker in the panel as

CD83 expression is a marker of monocyte maturation that increases in response to viral products[158]. In contrast, the expression of LOX1, PDL1 and PDL2 was diminished so dramatically after cryopreservation that these markers were excluded from the panel (data not shown).

Altered Proportions of Monocyte Subsets in Untreated HIV-1 Infection

In earlier work using fresh whole blood samples, we found lower proportions of traditional monocytes, and increased proportions of inflammatory and patrolling monocytes in HIV-1 infected patients with uncontrolled viremia [54] . Our new data evaluating a smaller number of cryopreserved samples from different patient and control populations are similar. The proportion of traditional (CD14+CD16-) monocytes tended to be lower in HIV-1-infectedpatients (median-76.6%) compared to the proportion of traditional monocytes among control subjects (median-82.5%), though not significantly lower (p =.089). The proportion of inflammatory (CD14+CD16+) monocytes also tended to be higher in the setting of HIV-1 infection (median-14.8%) than among controls

(median-13.9%), though not significantly (p=0.19). The proportion of patrolling

(CD14dimCD16+) monocytes was significantly higher in HIV-1 infection (median-6.0%) than among controls (median-2.9%, p=.029, Fig. 2.2A). These subset proportions did not change with ART (Fig. 2.2B-2.2D).

Monocyte Phenotypes are Altered in Untreated HIV-1 Infection and Change with ART

Shown on Tables 2.2 and 2.3 are summaries of the baseline and on treatment data sets presented in this manuscript. Proportions and mean fluorescent intensities that differ significantly at baseline from controls’ values are shown in red (greater in patients) and green (lower in patients). Likewise, significant changes from baseline in these values during ART are shown in red if they rise and in green if they diminish. Selected graphic distributions of these phenotypes are also shown in Fig. 2.3-2.6. The means and standard errors of the means are shown in Table 2.4 and Table 2.5.

Greater HLA-DR Expression and Density in Untreated HIV-1 Infection, Tend to

Decrease with ART

HLA-DR is utilized for presentation of peptide antigen to CD4+ T cells [159, 160] and HLA-DR expression on monocytes typically increases in the setting of activation

[161]. We found that both frequency of HLA-DR expression and the HLA-DR mean fluorescence intensity (MFI) on total monocytes were significantly greater in patient samples (p=0.004, 0.0005 respectively) than they were among controls’ samples (Fig.

2.3A, 2.3B). With initiation of ART, the proportions of HLA-DR+ monocytes did not decrease significantly and remained elevated when compared to proportions among healthy controls. HLA-DR density (MFI) decreased significantly on total monocytes at

weeks 24 and 48. At baseline, the frequency of HLA-DR expression was higher on each monocyte subset in the setting of HIV-1 infection (traditional-p=0.002; inflammatory- p<0.0001; patrolling-p=0.039) compared to the frequency of HLA-DR expression on monocyte subsets from controls. When these monocyte subpopulations were examined individually, both the proportions and density of HLA-DR decreased significantly on both inflammatory and patrolling monocytes at week 48; the frequencies of HLA-DR expression on these subsets was no longer different from these frequencies in subsets from controls’ samples (Fig. 2.3E, 2.3G). At baseline, HLA-DR density on traditional

(p=0.003) and inflammatory (p=0.003) monocytes was significantly higher on samples from HIV-1-infected subjects compared to HLA-DR density on samples from controls

(Fig. 2.3D, 2.3F). Density of HLA-DR tended to be higher on the patrolling monocytes of patients when compared to the density on controls’ samples, but these differences were not significant (p=0.0893) (Fig. 2.3H). Density of HLA-DR decreased on all monocyte subsets after 48 weeks of ART (Fig. 2.3D, 2.3F).

Greater CD86 Expression and Density on Monocytes in Untreated HIV-1 Infection decrease with ART

CD86 is a surface receptor that binds T cell CD28 and aids in co-stimulation [162,

163]. The proportion of total monocytes expressing CD86 was significantly greater in the untreated HIV-1-infected patients (p=0. 012), and CD86 density was significantly higher (p=0.005) among patients’ cells than among controls’ cells (Fig. 2.4A, 2.4B). The proportion of CD86+ monocytes did not change with treatment while CD86 density

decreased significantly (by GEE and signed rank test) at week 24 and by signed rank test at week 48, achieving levels not different from levels seen in controls.

Both the proportion of CD86 expressing traditional monocytes and the density of

CD86 on traditional monocytes were significantly higher in untreated HIV-1 infection compared to the levels seen in the healthy control population (p=0.008, p=0.009 respectively) (Fig. 2.4C, 2.4D). After 48 weeks of ART, both the density and expression of CD86 on traditional monocytes fell to levels no longer different from those seen among controls and the changes from baseline were significant at week 24.

The proportion of inflammatory monocytes expressing CD86 tended to be greater in untreated HIV-1 infection (p=0.050) and the MFI of CD86 on these cells was greater in patients than among controls (p=0.004) (Fig. 2.4E, 2.4F). The proportions of CD86+ patrolling monocytes decreased from baseline at weeks 24 and 48, and at these times neither their frequencies nor MFIs were different from those seen among controls’ cells

(Fig. 2.4G, 2.4H).

Lower CD40 Density on Traditional Monocytes in Untreated HIV-1 Infection

In monocytes, the ligation of CD40 can result in the production of several pro- inflammatory cytokines, such as IL-6 and IL-1β, and also can result in greater costimulatory molecule expression through interaction with CD40 ligand on CD4 T cells

[164]. The proportions of monocytes expressing CD40 were not significantly different

from the proportions in controls (total-p=0.75, traditional-p=0.33; inflammatory-p=0.97; patrolling-p=0.92), however, the density of CD40 was marginally lower on traditional monocytes in HIV-1 infection (total-p=0.71; traditional-p=0.048; inflammatory-p=0.38; patrolling-p=0.89) and the density of CD40 on total monocytes and on patrolling monocytes decreased significantly from baseline after 48 weeks of ART (Table 2.3).

Chemokine Receptor (CCR2 and CX3CR1) Expression on Inflammatory and Patrolling

Monocyte Subsets is Lower in Untreated HIV-1 Infection and Normalizes with ART

We next examined the distribution of the chemokine receptors CCR2 and CX3CR1 on monocyte subsets (Fig. 2.5, 2.6). As expected, traditional monocytes had the highest expression of CCR2 and the lowest expression of CX3CR1, and, conversely, patrolling monocytes had the highest expression of CX3CR1 and the lowest expression of CCR2, with expression of these molecules on inflammatory monocytes falling in between[165].

At baseline, the proportion of total monocytes expressing CCR2 was lower in patients

(p=0.0002) when compared to the proportion of CCR2+ monocytes in controls (Fig.

2.5A) and there was a trend towards lower CCR2 density (MFI, p= 0.053) on patient monocytes (Fig. 2.5B). At baseline, inflammatory monocytes from patients tended to less frequently express CCR2 (p=0.057) and to have lower CCR2 MFIs (p=0.045) compared to inflammatory monocytes from controls (Fig. 2.5E, 2.5F). There were no significant differences in CCR2 expression at baseline among traditional (MFI-p=0.55;

%- p=0.14) (Fig. 2.5C, 2.5D) or patrolling subsets (MFI-p=0.075; %- p=0.097) (Fig.

2.5G, 2.5H) compared to expression levels and frequencies on controls’ monocytes. Both the proportion of CCR2+ monocytes and the CCR2 densities increased from baseline by week 12 of ART on total monocytes and on both traditional and inflammatory populations, and became comparable to levels seen among controls.

The frequencies and densities of the fractalkine receptor (CX3CR1) on total monocytes, traditional monocytes, and inflammatory monocytes, were comparable in samples from controls and patients at baseline (Fig. 2.6A- 2.6F). Though they were unaltered at baseline, the proportions of CX3CR1+ traditional and inflammatory monocytes, and the density of this receptor, rose from baseline at 12 weeks (Fig. 2.6C,

2.6E). Among patrolling monocytes from patient samples, both the proportion of

CX3CR1+ cells and CX3CR1 density were lower than among controls at baseline

(p=0.0453, p=0.0235 respectively) and rose by week 12 of ART to be no longer different from these values among controls’ cells (Fig. 2.6G, 2.6H).

CD13 is a homotypic cell adhesion molecule which is expressed on both monocytes and endothelial cells, and is thought to play a role in monocyte adhesion and migration, and is used as a marker of monocyte maturation [166]. The proportion of CD13+ monocytes was lower (p=0.013) in untreated HIV-1 infection and this was related to low frequencies of CD13 expression in a subset of patients. The proportion of total monocytes expressing CD13 was decreased further after 48 weeks on treatment (Table

2.2). At baseline, the proportion of CD13+ traditional monocytes were significantly

lower in samples from patients than among samples from controls (p=0.019), but this was not seen among inflammatory (p=0.37) or patrolling subsets (p=0.48).

Discussion

Our study examined the phenotype of monocytes and their subsets in untreated

HIV-1 infection, after ART initiation, and in a healthy control population also at risk for coronary artery disease. We were able to confirm previous reports demonstrating an increased proportion of patrolling monocytes in HIV-1 infection [50] and we also provide new insights into the kinetics of ART effects on monocyte phenotype in HIV-1 disease.

We report here the phenotypes of total circulating monocytes, as well as the phenotypes of circulating monocyte subpopulations as defined by CD14 and CD16 expression[54]. While there are a number of ways to distinguish the monocyte subsets

[167, 168], we selected the approach used by Cros et al [52] and by us in our earlier works [54, 169], recognizing that there is not yet a consensus as to how best to phenotype circulating monocytes. In untreated HIV-1 infection, circulating monocytes have greater expression of HLA-DR and CD86 that may reflect in vivo activation and might affect their ability to co-stimulate T cells. Higher expression and density of HLA-DR on monocytes has been reported previously [170]. Interestingly, we found lower CD40 density on traditional monocytes in HIV-1 infection. Earlier work by our group has found diminished induction of CD40L expression on activated (CD38+) CD4 T cells after T cell receptor stimulation [171, 172] . Concurrent lower level CD40 expression on

antigen presenting cells may contribute to an impaired CD4 T cell response to antigen in vivo; a recognized complication of HIV-1 infection[173].

We confirmed our previous finding that patrolling monocyte proportions are increased in untreated HIV disease when compared to the patrolling monocyte proportions found in a control population [54]. An increased proportion of CD16+

(inflammatory and patrolling) monocytes in HIV disease has been reported by several groups [174-176], and was recently found to be predictive of greater coronary artery calcium progression in HIV-1-infected patients[176]. We also confirmed previous reports of unaltered CD163 expression on CD14+CD16+ monocytes after administration of ART[177].

Lower CCR2 expression and higher CX3CR1 expression on cryopreserved monocytes from elite controllers and patients on ART with viral suppression has been reported previously [178] while CX3CR1 expression on classical monocytes has been shown previously to be higher in patients with uncontrolled viremia [54]. The expression of CCR2 on monocytes was previously shown to be unaltered in treated HIV disease, similar to what we found after 48 weeks of therapy [169]. We extend these findings by reporting for the first time, significantly lower CCR2 expression on total circulating monocytes, lower CCR2 density on inflammatory monocytes, and lower CX3CR1 expression and density on patrolling monocytes in patients with untreated HIV-1 disease.

Lower proportions of CCR2- or CX3CR1-expressing cells may reflect an increased systemic exposure to their ligands, CCL2 or fractalkine respectively, causing impaired monocyte migration. The lowered chemokine receptor expression could also indicate that patient monocytes have already egressed into sites of inflammation and are no longer

represented in the blood, leaving behind those monocytes with lower chemokine receptor expression. Importantly, these alterations appear to be abrogated with ART, as shown by the increase in both CCR2 and CX3CR1 expression after ART initiation.

Limitations of this study include a relatively modest sample size and the complexity of dealing with missing samples, due in part to the inconsistent quality and numbers of monocytes present in samples thawed after cryopreservation. We therefore needed to apply statistical methods to deal with missing data (GEE). Also, patients in this study were treated with an integrase-inhibitor based ART regimen. Additionally, as plasma inflammatory profiles have different trajectories after initiation of integrase-based versus non-nucleoside based therapy [179, 180] it is not certain that the findings here will be superimposable after initiation of different ART-based regimens.

It should be noted that our healthy control population is comprised of subjects with defined cardiovascular risk, and may differ from healthy controls utilized in other studies. As our HIV-1-infected patients are also at greater risk for cardiovascular disease

[54, 176, 181-184], these controls gave us the opportunity to link the perturbations we observed in circulating monocyte phenotypes to HIV-1 infection itself. On the other hand, as we do not have access to smoking histories of our patients, this remains a limitation of our study. Also, by including controls with defined cardiovascular risk, we may have limited our ability to identify some monocyte phenotypic changes associated with HIV-1 infection and the cardiovascular risks that are its consequence; for example,

CCR2 expression that did not differ between patients and controls at baseline, but rose with ART.

In summary, we have identified distinct perturbations in circulating monocyte phenotypes in patients with untreated HIV-1 infection: elevated expression of HLA-DR and CD86 that normalized with ART while expression of the chemokine receptors

CX3CR1 and CCR2 rose with ART. In contrast, we found lower expression of CD40 in untreated HIV-1 infection which decreased with ART, rather than normalizing. These data demonstrate that cryopreserved monocytes can be used to examine monocyte phenotypes and in HIV-1 infection, and perturbations of circulating monocyte phenotypes tend to improve with administration of suppressive antiretroviral therapy. The role of circulating monocytes in the sustained inflammatory environment of HIV-1 infection warrants further study.

Methods

Ethics Statement

This study was approved by institutional review boards at all participating sites:

Brigham and Women's Hospital Clinical Research Site (CRS), Johns Hopkins Adult

AIDS CRS, UCSD, AVRC CRS, University of Rochester ACTG CRS, AIDS Care CRS,

Washington University CRS, The Ohio State University AIDS CRS, MetroHealth CRS,

Northwestern University CRS, The Miriam Hospital ACTG CRS, Vanderbilt

Therapeutics CRS, IHV Baltimore Treatment CRS, University of Colorado Hospital

CRS, Houston AIDS Research Team CRS, and the Harlem ACTG CRS. Participants provided their written consent to participate in this study. This trial is registered with

Clinicaltrials.gov # NCT00660972.

Study Design

The study design has been more completely described in [50, 185], but briefly, A5248 was a prospective, open-label, multicenter, pilot study performed in the United States of

America. Recruitment began June 2008 and ended January 2009; follow-up ended April

2010. A5248 was a single arm study of raltegravir (RAL, 400 mg twice daily) and emtricitabine/tenofovir disproxil fumarate (FTC/TDF, 200mg/300mg once daily) in HIV-

1 infected ART-naïve patients who were enrolled if plasma HIV-1 RNA levels were

>10,000 and <300,000 copies/mL. Patients were excluded from this analysis if they experienced virologic failure (plasma HIV-1 RNA level ≥ 1000 copies/mL between week

16 and week 24, or ≥ 200 copies/mL at or after 24 weeks) or clinical rebound (plasma

HIV-1 RNA >0.3 log10 c/mL above the previous measurement).

Sample Collection

Blood samples collected pre-entry and at study entry, week 12, week 24, and week 48 after ART initiation were utilized for analysis. Peripheral blood mononuclear cell

(PBMC) samples were prepared by Ficoll density sedimentation and cryopreserved in

90% Fetal Bovine Serum (FBS) and 10% dimethyl sulfoxide (DMSO) until analyzed in batch. Cryopreserved samples available from patients who experienced virologic response to therapy, as described above, were used for these analyses. Baseline samples were pre-entry or entry samples according to availability. Patient characteristics are described in Table 1. Healthy control cryopreserved samples were obtained from a cohort of HIV uninfected persons without cardiovascular disease, but who had at least two cardiovascular risks including age > 50 years old, male gender, cigarette smoking,

high blood pressure, high blood cholesterol, low HDL, Type II diabetes, or a history of cardiovascular disease in a first degree relative.

Flow Cytometry

Cryopreserved PBMCs were thawed and immediately stained for viability

(LIVE/DEAD fixable Yellow Dead Cell Stain-Life Technologies, Grand Island, NY).

All samples from the same patient were thawed and examined on the same day, and all patient and control samples were examined during a two-month period. PBMCs were then washed twice with complete medium (RPMI- 10% FBS, 1% Pen/Strep, 1% Hepes,

1% L-glutamine) and stained for 30 minutes using a panel that excluded T cells, B cells,

NK cells, and neutrophils (anti-CD3-Phycoerythrin-Cy7 (PE-Cy7) (557581-Becton

Dickinson (BD), San Jose, CA), anti-CD15-PE-Cy7 (323030-Biolegend, San Diego,

CA), anti-CD19-PE-Cy7 (302216-Biolegend), anti-CD56-PE-Cy7 (557747-BD)), and stained for monocyte markers using anti-CD14-Pacific Blue (558121-BD) and anti-

CD16-Phycoerythrin (PE) (555407-BD), as well as for anti-HLA-DR-Peridinin chlorophyll protein (PerCP) (307628-Biolegend), anti-CD13- Allophycocyanin-Cy7

(APC-Cy7) (301710-Biolegend), anti-CD163-PE-CF594(562670-BD), anti-CD38-AF700

(560676-BD), anti-CCR2- PerCP-Cy5.5 (357203-Biolegend), anti-CX3CR1-APC

(341610 Biolegend), anti-CD86-PerCP-Cy5.5 (305420-BD), anti-CD36-APC (550956-

BD), CD40- AF700 (561208-BD). Cells were then fixed with 4% paraformaldehyde and examined using an LSRII flow cytometer (BD), which is calibrated daily by a dedicated technician using standard CS&T set up beads (BD Biosciences) to assure consistency of fluorescence detection.

Gates to identify positive expression of surface markers were determined using isotype control antibodies: MsIgG2a-Pacific Blue (558118-BD), MsIgG1- PE (555749-

BD), MsIgG2a-PerCP (400250-Biolegend), MsIgG1-APC-Cy7 (400128-Biolegend),

MsIgG1-PE-CF594 (562292-BD), MsIgG1-AF700 (557882-BD), MsIgG2a-PerCP-

Cy5.5 (400252-Biolegend), Rat IgG2b-APC (400612-Biolegend), MsIgG2b-PerCP-

Cy5.5 (400338-Biolegend), MsIgM-APC (555585-BD).

Data were analyzed using FACSDiva software (Version 6.2 BD Bioscience, San

Diego CA). Monocytes were identified based on singlets, exclusion of viability dye, forward and side scatter characteristics, exclusion of cells with a high density of CD3,

CD15, CD19, CD56, and were divided into the three subsets based on CD14 and CD16 expression. The CD14 and CD16 monocyte subset gates were based on the staining of their respective isotypes[54]. A “total monocyte” gate was also drawn around all three monocyte subsets. Total monocytes and monocyte subsets were characterized further by expression of HLA-DR, CD13, CD163, CD38, CCR2, CX3CR1, CD86, CD36, and

CD40. Between 100,000 and 300,000 events were collected in the Forward/Side Scatter gate.

Statistics

Comparisons between baseline findings in patients and among controls were performed using two-tailed Mann Whitney U tests. Comparisons among treatment time points were performed using both the generalized estimating equation (GEE) and two-

tailed signed rank tests. Whereas the signed rank tests considered each pre- and post-

ART comparison separately, the GEE allowed all the data to be considered together, and more fully considered repeated measures for each individual. Specifically, expression and density on monocytes prior to beginning ART was compared to the levels seen after ART initiation at week 12, week 24, and week 48 (Tables 2, 3). GEE is best suited for studies with longitudinal data with correlated results [186]. Rather than using a repeated measures ANOVA, which requires a full dataset, we used GEE which does not require this approach. While our total HIV (+) patient population is 29, our final dataset includes

17 patients at baseline, 21 patients at week 12, 18 patients at week 24, and 23 patients at week 48, necessitating the GEE approach. Significance was defined as p ≤ 0.05. All graphs and analyses were performed using RStudio [187, 188] or Graphpad Prism

Software (Version 5.04). Specifically, GEE was performed using the GEE function in

RStudio[189].

HIV-1 (+) HIV-1 (-) n=29 n=15 Female (%) 14 20 Gender Male (%) 86 80

Median 46 48 Age (years) Range 25-58 36-65

White Non-Hispanic 48 67 (%) Black Non-Hispanic 24 16 (%) Demographics Hispanic (%) 24 0 Asian or Pacific 0 7 Islander (%) Other (%) 0 13 Not Reported (%) 1 0

Median: HIV-1 RNA ---- 34469 Range: 6644- (copies/mL) ---- 264210 HIV-1 Status CD4+ T cell Count Median: 283 ---- Range: 10.5- (cells/uL) ---- 547.5

Table 2.1. Patient characteristics. HIV-1-infected patients are from the A5248 cohort and controls were taken from a cohort of uninfected persons with at least two cardiovascular disease risk factors.

A Isotype Fresh PBMC Cryopreserved PBMC

CD16 MsIgG1 MsIgG2a CD14

RatIgG2b-Iso

CX3CR1

B HIV (-) Fresh PBMC Cryopreserved PBMC

C HIV (+) Fresh PBMC Cryopreserved PBMC

Figure 2.1. Gating strategy for flow cytometry and comparison between fresh and cryopreserved monocyte surface marker expression. (A) Shown are isotype control dot plots and CD14 and CD16 expression in freshly obtained and cryopreserved PBMC from the same healthy volunteer and the expression of CX3CR1 on each monocyte subset in the cryopreserved sample. Monocyte subsets were gated using the isotype staining as a guide, as seen in the farthest left panel. Traditional monocytes are in purple, inflammatory monocytes are in pink, patrolling monocytes are in green. Gates are drawn based on negative isotype staining using Rat IgG2b, in the case of CX3CR1. (B) Summary surface marker expression, both proportion and MFI, in fresh and cryopreserved PBMCs on total monocytes of healthy control subjects, with medians, are shown. Each shape (triangle, square, diamond, and circle) represents an individual healthy control subject. (C) Summary surface marker expression, both proportion and MFI, in fresh and cryopreserved PBMCs on total monocytes of virologically suppressed HIV-infected subjects, with medians, are shown. Each shape (triangle, square, diamond, and circle) represents an individual HIV-infected subject.

C Uninfected Control 0 Baseline

12, 24, 48 Week 12, Week 24, and Week 48 after initiation of ART

Significant difference from uninfected controls (Mann Whitney p ≤ # 0.05)

Change from Baseline significantly different (Wilcoxon signed rank p ≤

* 0.05) Change from Baseline significantly different (Generalized estimating ** equation (GEE) p ≤ 0.05)

Change from Baseline significantly different (Both Wilcoxon Signed *** Rank and GEE p ≤0.05)

Figure 2.2. Monocyte subset proportions at baseline and after ART initiation compared to proportions among controls. (A) Jitterplot comparing the subset proportions in HIV-1-infected individuals prior to ART initiation and subset proportions in controls. Medians are shown, and p values were determined using Mann Whitney U tests. Figures B-D display Tukey boxplots of medians and interquartile ranges. Outliers are shown as open circles. Tukey boxplots show the proportions of traditional monocytes

(B), inflammatory monocytes (C) and patrolling monocytes (D) in controls (red) and in

HIV-1-infected subjects at baseline and over the course of 48 weeks of ART.

GEE GEE Signed Rank test Proportion Robust z Robust P p value 0-12 0-24 0-48 0-12 0-24 0-48 0-12 0-24 0-48 HLADR 0.74 0.99 -0.92 0.459 0.322 0.358 0.328 0.839 0.255 CCR2 0.48 0.48 1 0.631 0.631 0.317 0.000 0.001 0.004 CX3CR1 1.8 -0.62 0.56 0.072 0.535 0.575 0.194 0.831 1.000 CD38 -0.15 0.61 -0.04 0.881 0.542 0.968 0.572 0.831 0.003 CD13 0.76 0.51 0.16 0.447 0.610 0.873 0.414 0.898 0.021 Total CD40 -0.53 -1.95 -0.59 0.596 0.051 0.555 0.296 0.014 0.040 CD86 -0.65 -1.93 -0.69 0.516 0.054 0.490 0.952 0.148 0.636 CD36 -0.11 -0.57 -0.85 0.912 0.569 0.395 0.610 0.520 0.109 CD163 0.18 0.48 0.53 0.857 0.631 0.596 0.903 0.898 0.701 CD83 1.22 1.78 1.95 0.222 0.075 0.051 1.000 0.174 0.100 HLADR 1.23 1.08 -1.26 0.219 0.280 0.208 0.117 0.824 0.272 CCR2 2.11 1.45 2.37 0.035 0.147 0.018 0.012 0.102 0.039 CX3CR1 3.21 0.46 0.92 0.001 0.646 0.358 0.009 0.123 0.784 CD38 -0.27 0.05 -0.27 0.787 0.960 0.787 0.294 0.359 0.080 CD13 1.06 1.01 0.99 0.289 0.312 0.322 0.217 0.898 0.305 Traditional CD40 0.12 -0.98 0.25 0.904 0.327 0.803 0.730 0.148 0.045 CD86 -0.31 -2.06 -0.8 0.757 0.039 0.424 0.761 0.123 0.636 CD36 -0.19 -0.7 -0.75 0.849 0.484 0.453 0.754 0.765 0.327 CD163 0.47 0.15 0.36 0.638 0.881 0.719 0.808 0.859 0.497 CD83 1.44 1.73 1.85 0.150 0.084 0.064 0.423 1.000 0.181 HLADR -0.65 -1.82 -2.7 0.516 0.069 0.007 0.826 0.042 0.036 CCR2 2.38 2.59 2.14 0.017 0.010 0.032 0.049 0.067 0.340 CX3CR1 3.21 0.03 0.85 0.001 0.976 0.395 0.013 0.859 0.497 CD38 -0.45 -0.28 -0.64 0.653 0.779 0.522 0.808 0.240 0.004 CD13 0.31 0.48 -0.2 0.757 0.631 0.841 0.780 0.756 0.110 Inflammatory CD40 -0.22 -1.62 -0.55 0.826 0.105 0.582 0.834 0.083 0.094 CD86 -0.46 -2.13 -0.86 0.646 0.033 0.390 0.952 0.067 0.588 CD36 0.4 -0.15 -0.61 0.689 0.881 0.542 0.706 0.520 0.224 CD163 0.21 -0.03 0.15 0.834 0.976 0.881 0.583 0.700 0.946 CD83 1.38 2.02 2.34 0.168 0.043 0.019 0.162 0.363 0.154 HLADR -0.18 -0.13 -2.13 0.857 0.897 0.033 0.761 0.610 0.127 CCR2 0.64 0.6 1.64 0.522 0.549 0.101 0.855 0.520 0.839 CX3CR1 3.34 1.7 2.3 0.001 0.089 0.021 0.001 0.067 0.152 CD38 -1.87 -2.43 -2.53 0.061 0.015 0.011 0.191 0.010 0.001 CD13 0.11 0.45 -0.41 0.912 0.653 0.682 0.414 0.831 0.255 Patrolling CD40 -1.5 -3.18 -2.96 0.134 0.001 0.003 0.153 0.002 0.000 CD86 0.25 -1.77 -0.99 0.803 0.077 0.322 0.761 0.413 0.455 CD36 -1.81 -2.71 -2.71 0.070 0.007 0.007 0.025 0.024 0.008 CD163 -1.02 -0.21 -0.51 0.308 0.834 0.610 0.391 0.240 0.787 CD83 1.31 2.02 1.89 0.190 0.043 0.059 0.208 0.083 0.168

Table 2.2. Alterations in frequencies of surface marker expression on total monocytes and monocyte subsets of HIV-1-infected patients before and after initiation of ART. Baseline proportions are in red if significantly greater than among uninfected controls and in green if significantly lower by Mann Whitney U tests. Comparisons were made between week 12 and baseline (0-12), week 24 and baseline (0- 24), and week 48 and baseline (0-48) using both Wilcoxon signed rank test and the generalized estimating equation. Significant increases (p<0.05) are highlighted in red and significant decreases are in green. Robust Z score for the GEE plot are included to aid in the understanding of the direction of the change; negative Z score indicates decreased values after ART initiation and positive Z scores indicate increased values after ART initiation.

GEE GEE Signed Rank test MFI Robust z Robust P p value 0-12 0-24 0-48 0-12 0-24 0-48 0-12 0-24 0-48 HLADR -1.710 -2.740 -5.730 0.087 0.006 0.000 0.194 0.175 0.001 CCR2 5.060 4.070 4.770 0.000 0.000 0.000 0.000 0.003 0.003 CX3CR1 0.950 -0.610 0.570 0.342 0.542 0.569 0.727 1.000 1.000 CD38 -0.410 -1.350 -2.450 0.682 0.177 0.014 0.715 0.831 0.011 Total CD13 -1.220 -1.610 -3.610 0.222 0.107 0.000 0.593 0.278 0.000 CD40 -2.540 -2.660 -1.130 0.011 0.008 0.258 0.035 0.024 0.056 CD86 -1.940 -2.400 -0.680 0.052 0.016 0.497 0.068 0.024 0.040 CD36 0.680 -0.190 -2.580 0.497 0.849 0.010 0.675 1.000 0.080 HLADR -0.350 -1.490 -3.470 0.726 0.136 0.001 0.855 0.700 0.013 CCR2 3.290 1.970 2.940 0.001 0.049 0.003 0.001 0.032 0.080 CX3CR1 1.440 -0.010 0.880 0.150 0.992 0.379 0.224 0.154 1.000 CD38 -1.150 -2.440 -3.290 0.250 0.015 0.001 1.000 0.278 0.006 Traditional CD13 -0.460 -1.040 -2.840 0.646 0.298 0.005 0.952 0.638 0.000 CD40 -0.590 -1.160 0.120 0.555 0.246 0.904 0.802 0.123 0.083 CD86 -1.430 -1.840 -0.160 0.153 0.066 0.873 0.135 0.042 0.057 CD36 0.240 -0.640 -2.620 0.810 0.522 0.009 0.442 0.839 0.069 HLADR -1.530 -1.880 -4.050 0.126 0.060 0.000 0.626 0.320 0.002 CCR2 3.360 3.310 3.500 0.001 0.001 0.000 0.033 0.005 0.127 CX3CR1 3.230 0.570 1.610 0.001 0.569 0.107 0.012 0.221 0.266 CD38 0.170 -0.580 -2.440 0.865 0.562 0.015 0.808 0.966 0.021 Inflammatory CD13 -2.510 -2.690 -4.610 0.012 0.007 0.000 0.119 0.024 0.000 CD40 -0.720 -1.510 -1.340 0.472 0.131 0.180 0.808 0.230 0.055 CD86 -2.290 -2.980 -1.390 0.022 0.003 0.165 0.049 0.007 0.110 CD36 1.110 0.160 -2.410 0.267 0.873 0.016 1.000 0.959 0.110 HLADR -2.290 -1.830 -4.070 0.022 0.067 0.000 0.013 0.123 0.001 CCR2 -0.430 0.600 0.900 0.667 0.549 0.368 1.000 1.000 1.000 CX3CR1 5.160 2.590 1.920 0.000 0.010 0.055 0.001 0.042 0.147 CD38 -3.750 -3.850 -2.680 0.000 0.000 0.007 0.001 0.004 0.021 Patrolling CD13 -3.380 -3.540 -4.900 0.001 0.000 0.000 0.013 0.032 0.001 CD40 -2.360 -3.450 -4.600 0.018 0.001 0.000 0.049 0.014 0.003 CD86 -1.770 -2.810 -3.560 0.077 0.005 0.000 0.078 0.083 0.021 CD36 -0.510 -1.570 -3.370 0.610 0.116 0.001 0.119 0.206 0.057

Table 2.3. Alterations in surface marker density on total monocytes and monocyte subsets of HIV-1-infected patients before and after initiation of ART.

Baseline MFI values are in red if significantly greater than among uninfected controls and in green if significantly lower by Mann Whitney U tests. Comparisons were made between week 12 and baseline (0-12), week 24 and baseline (0-24), and week 48 and baseline (0-48) using both Wilcoxon signed rank test and the generalized estimating equation. Significant increases (p<0.05) are highlighted in red and significant decreases are in green. Robust Z score for the GEE plot are included to aid in the understanding of the direction of the change; negative Z score indicates decreased values after ART initiation and positive Z scores indicate increased values after ART initiation

Table 2.4. Mean and Standard Error for frequencies of surface marker expression on total monocytes and monocyte subsets of HIV-1-infected patients before and after initiation of ART.

Table 2.5. Mean and Standard Error for density of surface marker expression on total monocytes and monocyte subsets of HIV-1-infected patients before and after initiation of ART.

A B

C D

E F

G H

C Uninfected Control 0 Baseline 12, 24, 48 Week 12, Week 24, and Week 48 after initiation of ART # Significant difference from uninfected controls (Mann Whitney p ≤ 0.05) * Change from Baseline significantly different (Wilcoxon signed rank p ≤ 0.05) ** Change from Baseline significantly different (Generalized estimating equation (GEE) p ≤ 0.05) *** Change from Baseline significantly different (Both Wilcoxon Signed Rank and GEE p ≤0.05)

Figure 2.3. Expression and density (MFI) of HLA-DR on patient monocytes at baseline and after ART initiation compared to values among controls. Values for frequency and density of HLA-DR on monocyte subsets in untreated HIV-1 infection were compared to levels among controls (in red) using Mann Whitney U tests, and baseline values among patients were compared to values on ART using GEE and Signed

Rank test (see boxed legend). Figures show Tukey boxplots of medians and interquartile ranges; outliers are shown as open circles. HLA-DR densities and proportions were increased at baseline when compared to the levels seen in the controls (A-G). Tukey boxplots show the proportions of HLA-DR+ total monocytes (A), traditional monocytes

(C), inflammatory monocytes (E), and patrolling monocytes (G) in controls’ samples

(red) and in patient samples at baseline and after ART initiation. Tukey boxplots show the density of HLA-DR on total monocytes (B), traditional monocytes (D), inflammatory monocytes (F), and patrolling monocytes (H) in controls’ samples (red) and in patient samples at baseline and after ART initiation.

Figure 2.4. Expression and density (MFI) of CD86 on patient monocytes at baseline and after ART initiation compared to values among controls. Values for frequency and density of CD86 on monocyte subsets in untreated HIV-1 infection were compared to levels among controls (in red) using Mann Whitney U tests, and baseline values among patients were compared to values on ART using GEE and Signed Rank test (see boxed legend). Figures show Tukey boxplots of medians and interquartile ranges; outliers are shown as open circles. Tukey boxplots show the proportion of CD86+ total monocytes

(A), traditional monocytes (C), inflammatory monocytes (E), and patrolling monocytes

(G) in control samples (red) and in patient samples at baseline and after ART initiation.

Tukey boxplots show the density of CD86 on total monocytes (B), traditional monocytes

(D), inflammatory monocytes (F), and patrolling monocytes (H) in control samples (red) and in patient samples at baseline and after ART initiation.

A B

C D

E F

H G

Figure 2.5. Expression and density (MFI) of CCR2 on patient monocytes at baseline and after ART initiation compared to values among controls. Values for frequency and density of CCR2 on monocyte subsets in untreated HIV-1 infection were compared to levels among controls using Mann Whitney U tests, and baseline values among patients were compared to values after ART initiation using GEE and Signed

Rank test (see boxed legend). Figures show boxplots of medians and interquartile ranges; outliers are shown as open circles. Tukey boxplots show the proportion of

CCR2+ total monocytes (A), traditional monocytes (C), inflammatory monocytes (E), and patrolling monocytes (G) in control samples (red) and in patient samples at baseline and after ART initiation. Tukey boxplots show the density of CCR2 on total monocytes

(B), traditional monocytes (D), inflammatory monocytes (F), and patrolling monocytes

(H) in control samples (red) and in patient samples at baseline and after ART initiation.

A B

C D

E F

G H

Figure 2.6. Expression and density (MFI) of CX3CR1 on patient monocytes at baseline and after ART initiation compared to values among controls. Values for frequency and density of CX3CR1 on monocytes in untreated HIV-1 infection were compared to levels among controls using the Mann Whitney U test, and baseline values were compared to values after ART initiation using GEE and Signed Rank test. Figures show boxplots of medians and interquartile ranges; outliers are shown as open circles.

Tukey boxplots show the proportion of CX3CR1+ total monocytes (A), traditional monocytes (C), inflammatory monocytes (E), and patrolling monocytes (G) in control samples (red) and in patient samples at baseline and after ART initiation. Tukey boxplots show the density of CX3CR1 on total monocytes (B), traditional monocytes (D), inflammatory monocytes (F), and patrolling monocytes (H) in control samples (red) and in patient samples at baseline and after ART initiation.

Chapter 3- Toll Like Receptors in the Intestine are altered in HIV disease

Summary

Background: Since the intestinal mucosa and microbiome are disrupted in HIV disease, we studied the abundance of innate immune sensors, Toll-Like Receptors (TLR), in the epithelia of patients with Viremia, prior to antiretroviral therapy (ART) initiation,

Immune Success (IS- >500 CD4 T cells/ul after 2 years of ART; suppressed viremia), and Immune Failure (IF- <350 CD4 T cells/ul after 2 years of ART; suppressed viremia).

Viremia and IF show increased systemic immune activation; therefore, modulation of intestinal TLR abundance may offer a mechanism behind persistent inflammation in HIV.

Method: Immunofluorescence for TLR3, TLR4, and TLR9 on sections cut from paraffin embedded colonic biopsies of uninfected, viremic, IS, and IF was imaged with a

DeltaVision microscope and quantified with Metamorph software.

Results: Viremic patients have significantly higher viral TLR9 and TLR3, and significantly lower bacterial TLR4, on the surface epithelium when compared to levels seen on uninfected control. Viremic patients have significantly higher TLR9, TLR3, and

TLR4 expression in the crypts when compared to uninfected control. TLR9 abundance remains elevated in IF, while normalization occurs in IS. TLR3 abundance on surface epithelium and crypt is even more elevated in IF and IS, suggesting an effect of ART.

TLR4 abundance on surface epithelium normalizes in IF and IS; in the crypt, in IS TLR4 decreases further.

TLR4 is equally distributed between the surface epithelium and the crypt in the uninfected colon, while the distribution is shifted to the colon in Viremic patients and to

the surface epithelium after ART. TLR3 is skewed toward the surface in all groups;

TLR9 is also skewed toward the surface in uninfected controls, Viremic, and IF. TLR9 is distributed further toward the crypt in IS.

Conclusion: Epithelial TLR abundance and location is selectively altered in HIV disease; effective ART partially normalizes some TLRs, and further alters others.

Mucosal TLR desensitization or hyper-vigilance, possibly caused by disrupted cytokine levels in the HIV+ colon, may contribute to systemic inflammation.

Introduction

It is widely recognized that intestinal barrier function, a predictor of mortality [190], is decreased in HIV disease, and delayed anti-retroviral therapy (ART) fails to reestablish full barrier integrity [191-197]. Components of the intestinal mucosa disrupted in treated and untreated HIV disease include dysbiosis on the luminal microbiota, paracellular permeability[74], and selective loss of Th17 cell subpopulation[198]. One consequence of colonic barrier loss of integrity is an increase in microbial translocation, where upon microbial products enter the lamina propria and often the blood stream, causing local and systemic inflammation[198-200]. Microbial translocation in HIV disease contributes to increased systemic immune activation[201, 202] and may in turn lead to increased morbidity and mortality, most notably non-AIDS multi-organ complications. Disruption of epithelial and immune function in the GI tract may also affect the establishment or perpetuation of the latent HIV reservoir[200], a reduction in lamina propria T cell repletion during effective ART treatment[203], and poor mucosal vaccine responses[204].

Underlying the critical role for barrier integrity in the GI tract is the presence of Pattern

Recognition Receptors, with specific reference to the Toll-Like Receptors (TLRs) located on the basolateral surface of the colonic epithelium and on immune and stromal cells of the lamina propria[61]. These TLRs are sentinels of the innate immune system, and when the barrier is breached or weakened, they signal the release of a myriad of immune mediators, cytokines and chemokines[205]. These cytokines and chemokines then aid in repairing gut integrity as well as alerting neighboring epithelial and infiltrating immune

cells of the breach[61]. In general, TLRs are expressed on the basolateral surface of the epithelium and on cells of the lamina propria. However, in HIV+ patients, the balance of innate and adaptive immune cells is altered resulting in a large proportion of the CD4+ T cells, especially the Th17 subset [198], depleted in the colon. As a consequence, the profile of cytokines released is similarly altered yielding a highly dysregulated and dysfunctional mucosal immune response. We therefore hypothesize that this resultant immune imbalance leads to and is partially mediated by a change in expression levels of

TLRs, and their subsequent signals, in the intestinal mucosa. Overall, we propose that

HIV infection stimulates a parallel imbalance in the surveillance provided by TLR expression levels and their strategic location within the mucosa. This hypothesis is supported by evidence of perturbations in TLR abundance found in inflammatory bowel disease[140] and TLR RNA levels in the colons of Rhesus Macaques infected with

SIV[206]. In light of this, we sought to investigate if similar perturbations in epithelial cell and lamina propria TLR abundance and localization may contribute to or result from barrier dysfunction in HIV disease.

In addition, while ART is now highly effective in inhibiting viral replication, there is a sizable minority of patients who have received two years of ART but fail to recover their circulating CD4 T cells (<350 CD4 T cells/L) aptly referred to as Immune Failure (or

Immune Non-Responders) [207, 208]. These Immune Failures (IF) have higher risk for morbidity and mortality due to AIDS related, as well as non-AIDS related diseases, when compared to their successful counterparts (Immune Success- >500 CD4 T cells/L after 2 years of ART) and uninfected controls [208-211]. Specifically, IF patients demonstrate.

We therefore also hypothesized that IF patients will experience a mucosal immune imbalance that is distinct from IS patients, as reflected by a loss or degeneration of TLR innate surveillance. To test these two hypotheses we utilized quantitative immunofluorescence to explore TLR abundance and localization on colonic epithelial cells and lamina propria immune and stromal cells to assess the potential for these cells to respond to translocated microbial products during HIV disease and after ART treatment.

Results

TLR expression is increased and TLR location modified in the epithelium and lamina propria of Viremic HIV+ patients Systemic and mucosal immune dysfunction is a hallmark of HIV infection, not only due to the direct viral depletion of CD4+ T cells[212], but also associated with the imbalance in cytokine expression and leukocyte activation in circulation[94, 207, 213], in monocyte activation[151], in lymph nodes[49], and on mucosal surfaces[71-73, 76-78, 149, 191-

199, 202, 206, 212, 214, 215]. This disruption of the immune system is likely to not only lead to, but also result from modifications in TLR expression and anatomical distribution.

To quantify these predicted changes, we focused on two families of TLR: (i) TLR3 and

TLR9, whose ligands are most frequently viral products such as double stranded RNA and RNA:DNA hybrids, respectively, and (ii) TLR4, a sensor of the bacterial cell wall, and thus a sentry for microbial translocation through an HIV-induced leaky epithelium[77]. Examining the expression and location of these TLR by immunohistochemistry we quantified their distribution using imaging software, as described in Figure 3.1 and Figure 3.2.

Figure 3.1. Thresholding using Metamorph. (A) Secondary alone staining of a viremic colon is shown without thresholding. (B) A threshold of 1000 units was placed on the secondary alone staining of the viremic colon. Notice the small cluster of positive orange staining just left of center in the frame. (C) A threshold of 1000 units was placed on the TLR3 staining of the viremic colon. Notice the higher density of staining near the apical/luminal surface versus deeper in the crypt.

Figure 3.2. Quantification of TLR expression in colonic sections using Metamorph software. Colonic sections are stained with a nuclear DAPI stain and FITC-conjugated anti-TLR stain. (A) A region is created around each crypt on the DAPI image. We create regions using the DAPI image rather than the FITC image so that we are unbiased in our circling of the crypts. (B) The regions are transferred onto the FITC image. (C) A threshold is established with the FITC secondary Ab image, and positive events are masked in orange. We collect regional measurements on this image to determine the threshold area % of each crypt. (D) The regions created around each crypt on the DAPI image are turned to black. (E) The regions on the DAPI image are then deleted. (F) The

DAPI image is double dilated. As you can see when comparing (E) to (F), the nuclei are expanded. We do this so as to encompass the cytoplasm around each nuclei. (G) We then use the “Count Nuclei” function to draw a mask around each double dilated nuclei.

(H) This image is the same as G but with each nucleus as a different color. (I) The threshold is established for each image to capture every double dilated nuclei. (J) We then draw a region around every double dilated nuclei. The program is able to recognize every orange dot as a region, so we do not have to draw these regions by hand. (K) We then transfer these regions back onto the FITC image. We collect region measurements on this image to determine the threshold area % of every cell.

Both viral-sensing TLR 3 and 9 abundances are significantly increased on the surface epithelium (Figures 3.3, 3.4A, 3.4B), TLR3 expression is increased in the crypt epithelium (Figures 3.4A), and TLR9 expression is elevated within the lamina propria

(Figures 3.4B). In addition, in the Viremic patient TLR3 and TLR9 expression is preferentially increased on the surface epithelium, suggesting a potential protective response by the host to further viral infection (Figure 3.4D). In contrast the bacterial sensor TLR4 concentrations are either not increased or modestly increased within the surface and crypt epithelium and on immune and non-immune cells of the lamina propria is significantly in the viremic population compared to that of the healthy control (Figures

3.3, 3.4C). TLR4 location along the epithelium is preferentially elevated in the crypt

(Figures 3.4D), in close proximity to and possibly in consort with the lamina propria, possibly reflecting and responding to the increase in microbial translocation evident in the HIV mucosa[74, 77, 198, 199].

Figure 3.3. Immunofluorescence staining of TLR3, TLR4, and TLR9 in Controls, Viremic Patients, Imuune Successes, and Immune Failures.

Surface Epithelium Crypt Lamina Propria A

Figure 3.4 Toll-Like Receptor Expression Alterations in the Surface Epithelium,

Crypt, and Lamina Propria of Viremic Patients and Uninfected Control Colons.

TLR3 (A), TLR9 (B), and TLR4 (C) expression is shown for both Viremic and Uninfected

controls. TLR localization in the crypt and surface epithelium is shown in (D) for TLR3, TLR9,

and TLR4. Significance was determine using Mann Whitney vs Uninfected (*), vs

Viremic (#), or vs IF (**) and a p value <0.05, and if the value was also significant (p

value <0.1) using GEE it is highlighted in red.

Recovery of TLR 4 and TLR 9 expression in the Immune Success Population; Toxicity of ART for epithelial TLR3 Expression After successful ART intervention, CD4+ T cells rebound and inflammation in the periphery decreases, although in most cases not to the levels seen in uninfected controls.

Similarly, mucosal TLR9 and TLR4 abundance appears to normalize after successful

ART (Figures 3.3, 3.5B, 3.5C, 3.5D). In fact, TLR9 expression on the Surface epithelium and in the lamina propria is somewhat less than Uninfected controls, after successful ART intervention (Figures 3.3, 3.5B). In both the crypt and the lamina propria, Indeed, since TLR9 abundance in the crypt in IS patients is significantly low compared to the Uninfected controls, the Viremic Patients, and the Immune Failures

(Figure 3.5B), we propose that the decline in TLR9 in all regions of the mucosa after successful ART regimen and immune recovery will have some immunological consequences for concurrent viral co-infections. Finally, TLR9 expression after successful T cell recovery uniquely predominates in the epithelial crypts, further suggesting an anatomical modulation of innate immunity in the intestine despite repletion of adaptive T cell populations.

In contrast, TLR3 expression after ART and T cell recovery remains elevated in both the surface and crypt epithelium (Figures 3.3, 3.5A), suggesting a possible toxicity of ART and/or a failure to recover from the effects of Viremia, which include increased localized inflammation.

Lamina Propria A Surface Epithelium Crypt

Figure 3.5 Toll-Like Receptor Expression Alterations in the Surface Epithelium,

Crypt, and Lamina Propria of Immune Successes, Viremic Patients, and Uninfected

Control Colons. TLR3 (A), TLR9 (B), and TLR4 (C) expression is shown for Immune

Successes, Viremic, and Uninfected Controls. TLR localization in the crypt and surface

epithelium is shown in (D) for TLR3, TLR9, and TLR4. Significance was determine using

Mann Whitney vs Uninfected (*), vs Viremic (#), or vs IF (**) and a p value <0.05, and

if the value was also significant (p value <0.1) using GEE it is highlighted in red.

Elevated TLR expression during a Failure to reconstitute immune homeostasis Patients within the Immune Failure cohort, in addition to inadequate reconstitution of their circulating CD4+ T cells, have increased levels of serum LPS, IL-6, IL-8 among other immune mediators when compared to their successful counterparts and uninfected controls[207]. We now report that mucosal TLR3, TLR4, and TLR9 expression is also elevated in these patients, and in some circumstances to levels greater than that seen in

Viremic patients. TLR9 trends toward being significantly higher in the surface and crypt epithelium in Immune Failure patients when compared to the uninfected controls and

Viremic patients (Figures 3.3, 3.6B). In Immune Failure patients TLR4 expression normalizes in both the crypt and lamina propria (Figure 3.3, 3.6C), while TLR4 expression on the surface epithelium is increased to levels far greater than all other groups (Figure 3.6C, 3.6D).

TLR3 expression in IF is significantly elevated in the lamina propria when compared to uninfected controls, Viremic patients, and Immune Success (Figures 3.3, 3.6A). Viremic,

Immune Success, and Immune Failure patients maintain a skewed distribution of TLR3 toward the Surface epithelium, with the Immune Failure patients having almost exclusive expression on the surface (Figure 3.6D).

A Surface Epithelium Crypt Lamina Propria

Figure 3.6. Toll-Like Receptor Expression Alterations in the Surface Epithelium,

Crypt, and Lamina Propria of Immune Failures, Immune Successes, Viremic

Patients, and Uninfected Control Colons. TLR3 (A), TLR9 (B), and TLR4 (C) expression

is shown for Immune Failures, Immune Successes, Viremic, and Uninfected Controls. TLR

localization in the crypt and surface epithelium is shown in (D) for TLR3, TLR9, and TLR4.

Significance was determine using Mann Whitney vs Uninfected (*), vs Viremic (#), or vs

IF (**) and a p value <0.05, and if the value was also significant (p value <0.1) using

GEE it is highlighted in red.

Inflammation and H&E staining The altered expression and location of TLRs in the HIV mucosa, even after successful or unsuccessful T cell recovery, may be due to the effects of chronic HIV infection or, in the case of TLR3, ART toxicity. An alternate interpretation of the results may reflect, as seen in IBD, that TLR abundance is modulated by the state of intestinal inflammation.

To eliminate this possibility, samples were stained with Hematoxylin and Eosin to assess inflammation (Figure 3.6), and no consistent pattern of inflammation was observed.

A. Uninfected

B. Viremic

C. Immune Success

D. Immune Failure

Figure 3.7. H&E images at 20x. Uninfected (A), Viremic (B), Immune Success (C), and

Immune Failure (D) images are shown. Each image is a from a different donor.

Discussion

Traditionally, HIV progression is defined as a decrease in CD4+ T cells[212], however, patients have alterations in their innate and adaptive immunity both systemically[54, 94,

207] and at specific sites, such as the mucosa[190]. Systemically, these alterations include increased levels of systemic inflammatory markers (IL6, IFN (alpha), sCD14)[50,

94], monocyte activation[54, 151], T cell activation[50]. Alterations in the intestinal mucosa include decreased barrier integrity[71, 74], increased microbial translocation[191, 202], and decreased tight junction proteins proximally to distally along the colon[77]. Our study extends these observations this by demonstrating that TLR expression in the epithelium and the lamina propria of the colon are selectively altered in various outcomes of HIV disease.

As previously stated, viremic patients have decreased barrier function, so we expected

TLR alterations due to the increased levels of microbial translocation. What we found were increased levels of TLR3 and TLR9 in all three compartments (the surface epithelium, crypt, and lamina propria), increased levels of TLR4 in the crypt and lamina propria, and decreased levels of TLR4 in surface of Viremic patients when compared to the uninfected controls. Similar alterations have been found in IBD where there is decreased TLR3 and increased TLR4 abundance, as well as TLR4 localization shifts toward the surface epithelium[140]. In graft vs host disease reports show that variants in

TLR9 that are associated with reduced TLR9 expression had significantly lower treatment-related mortality[216].

ART has turned HIV infection into a chronic inflammatory disease by blocking viral replication, and, in the majority of cases, allowing for restoration of CD4+ T cells.

However, full immune restoration does not occur, especially in the gut, despite restoration of CD4+ T cells in the periphery. Immune success patients have elevated plasma IL-6, d-Dimer, and sCD14 when compared to healthy controls[207]. In the same study it was found that plasma sCD14, an indicator of microbial translocation, significantly correlated with the levels of IL6, indicating a possible relationship between the activation seen in the periphery and microbial translocation. Additionally, ART does not fully diminish the monocyte[151] or T cell activation seen in HIV+ patients[50].

Our results show that in IS patients TLR4 localization becomes skewed toward the surface epithelium and expression in the lamina propria is diminished; TLR9 abundance is lower in all three compartments, as opposed to the high expression in Viremic and IF patients, and remains skewed toward the surface epithelium; TLR3 abundance is high in all three compartments. These three findings indicate that while TLR9 appears to be normalizing in these IS patients, both TLR4 and TLR3 are distinctly altered by ART.

It is well established that intestinal barrier function, a predictor of mortality[190], is decreased in HIV disease, and delayed ART fails to reestablish intestinal barrier function[191-197]. Additionally, there are patients who have received two years of ART but fail to recover their CD4 T cells (<350 CD4 T cells/uL) aptly deemed Immune

Failures (or Immune Non-Responders)[207, 208]. These Immune Failures (IF) have higher risk for morbidity and mortality due to AIDS related, as well as non-AIDS related diseases, when compared to their successful counterparts (Immune Success- >500 CD4 T cells/uL after 2 years of ART) and uninfected controls [208-211]. IF also have higher levels of IL6, sCD14, and D-Dimer, indicating higher levels of activation, microbial translocation, as well as indices of T cell exhaustion and senescence[207, 213]. Together these indicate an impaired immune response and an activated phenotype. Our results show high TLR3 and TLR9 expression in all three compartments, normalized TLR4 expression in the crypt and surface epithelium, though it was skewed more toward the surface, and lower TLR4 expression in the lamina propria.

One likely reason for these alterations is the increased microbial translocation in the gut and active HIV replication, but our data does not suggest any correlation between local tissue inflammation and TLR expression. There is some evidence that TLR ligation can cause increases in the abundance of TLRs, so increases in the available ligand in the gut

(whether it is from HIV or microbial translocation) could be directly causing the increases in TLR expression that we are seeing. Both the Viremic patients and the IF have high levels of TLR9, which points toward persistent low levels of HIV viral replication in the IF patients, one of many hypotheses as to the cause of IF. Additionally, there is evidence that IFN alpha, which is increased in HIV + patients[97], can increase

TLR3 abundance[217]. We propose that higher levels of these TLR would then increase local immune activation, due to the production of cytokines and chemokines after ligation. With the depletion of T cells in the gut, which is only partially replenished after

ART, the local immune activation is unable to be regulated, causing further dysfunction and disruption of the intestinal barrier and eventually the systemic immune activation and exhaustion seen in HIV (+) patients.

Here we highlighted that TLRs are independently altered in HIV disease both prior to and after ART, implicating a role for both viremia and ART for their disruption. Our findings indicate that there are likely low levels of viral reactivation, though not necessarily replication, in the gut of IF patients, which has been previously speculated[78, 218], and that these TLR are playing a role in a feedback loop which perpetuates the residual peripheral immune activation seen in both IS and IF patients.

Materials and Methods

Patient Data

HIV (-) control biopsies were taken from colonic resections due to either polyps or adenocarcinoma. HIV (+) biopsies were removed during a colonoscopy procedure.

There was a total of 5 HIV (-), 3 HIV (+) Viremic, 4 Immune Failures (<350 CD4+ T cells/uL, at least 2 years of ART), and 2 Immune successes (>500 CD4+ T cells/uL, at least 2 years of ART) which were generously provided Dr. Michael Lederman and

Heather Pilch-Cooper. Samples were placed in 4% Paraformaldehyde the same day as it was removed, and placed in 80% EtOH the next day. The HIV (-) samples were predominantly male (60%), white non-Hispanic (100%), with a median age of 79. HIV

(+) viremic patient’s samples were from black non-Hispanic males, with a median age of

56, CD4 + T cell count of 552, and HIV-1 RNA of 6109. HIV (+) immune success samples were similarly from black non-Hispanic males, with a median age of 59.5, CD4+

T cell counts of 565, and non-detectable HIV-1 RNA. HIV (+) immune failure samples were from predominantly black non-Hispanic (75%) males with a median age of 56, median CD4 + T cell count of 304.5, and non-detectable HIV-1 RNA.

HIV (+) HIV Viremic IF IS (-) n=5 n=3 n=4 n=2 Gender Male (%2) 60 100 100 100 Female (%) 40 0 0 0 Age Median 79 56 56 59.5 44- Range 25-63 51-61 56-63 85 White Non- Demographics 100 0 25 0 Hispanic (%) Black Non- 0 100 75 100 Hispanic (%) Median: HIV Status CD4+ T cell Count ---- Median: 552 Median: 565 304.5 Range: 211- Range: 537- (cells/uL) ---- Range: 392-884 306 593 HIV-1 RNA ---- Median: 6109 ND ND Range: 4411- (copies/uL) ---- ND ND 65953

Table 3.1 Table of Characteristics for both HIV (-) and HIV (+) samples.

ND=Non-detectable

Immunofluorescence

Specific immunofluorescence protocols were designed for TLR3, TLR4, and TLR9 on colonic sections. Uninfected control biopsies were taken from colonic resections from patients with either polyps or adenocarcinoma. HIV (+) biopsies were removed during colonoscopy. Both the uninfected and HIV (+) biopsies were placed in 4%

Paraformaldehyde overnight and then transferred to 80% Ethanol (EtOH). Samples were then placed in paraffin and later cut on a microtome and placed onto slides. Slides were deparaffinized with Xylene for 30 minutes and then rehydrated with 1:1 Xylene and 100%

EtOH, 100% EtOH, 70% EtOH, 50% EtOH, and ddH20. Antigen retrieval was performed in a 95C water bath; It was found that the TLR4 and TLR9 antigen retrieval is optimal with pH6 Citrate buffer, while TLR3 and TLR7 are best revealed using a pH7.5 Tris buffer.

Prior to staining with the primary antibody, slides were blocked with 2% Bovine Serum

Albumin in Tris Buffer Saline with 0.0125% Triton X. TLR4 and TLR9 are detected with monoclonal mouse antibodies (Abcam), so their respective isotype controls (Biolegend) were used in addition to secondary antibody alone controls. TLR3 is detected with a polyclonal rabbit antibody (Abcam), so an isotype control was not used. Slides were stained overnight with the primary TLR antibody, and for an hour with either a Rabbit anti-

Mouse FITC antibody (Novex) or a Goat anti-Rabbit FITC antibody (Novex). Slides were mounted using ProLong Diamond Antifade Mountant with DAPI (ThermoFisher).

Slides were imaged using a Deltavision Microscope at 40X magnification. Exposure times for FITC were chosen such that the greatest range of intensity values could be utilized for quantification. Samples were imaged in batches, and always with a normal control, to limit the possibility of variability between samples. Additionally, our quantification method does not take into account the intensity of the signals relative to each other, rather their intensity relative to a secondary alone control, making our data internally consistent.

Metamorph

With the help of Dr. Scott Howell of the Eye Institue, and using MetaMorph Microscopy automation and Image Analysis Software [Molecular Devices], images were quantified.

The quantification procedure is illustrated in Figure 3.2. Quantification was performed by using the DAPI stain to locate the surface and crypt epithelium, and then encircling either the luminal surface or the lower crypt using the me. The DAPI stain enables you to see the epithelial cell nucleus and the outline of the surface and crypt epithelium. Using the

Secondary antibody staining alone control as a guide, a threshold intensity for positive expression was chosen such that the secondary alone control background staining was negated (Supplemental Figure 3). For example, the threshold for TLR3 was set at 1000 unit, such that any pixel on the image with an intensity higher than 1000 units would be masked in orange (Supplemental Figure 3). Once the threshold was established for an image, the proportion of positive pixels within the area of a circled crypt or surface epithelial region was determined. This is known as Threshold Area %. Quantitation of lamina propria cell TLR expression, described in detail in Supplemental Figure 1, relied on our ability to encircle every cell in the lamina propria, expand that circle to include the

cytoplasm and plasma membrane, and determine their Threshold Area %. Any cell which had a Threshold Area % greater than 1% was considered positive. We then used %

Positive Cells as our readout.

Statistics We performed Mann Whitney U tests between treatment groups and controls, as well as GEE. Mann Whitney U tests were performed in GraphPad prism and compared each region (crypt, surface epithelium, or cell) as an individual data point, therefore, it did not take into account a correlation within or between patients, nor did it account for one patient having more data points than the other.

Dr. Jeffrey Albert generously provided his help with his statistical expertise and in performing the GEE analysis. GEE was performed as a more rigorous analysis of the data, which took into account the correlations within and among groups, and treated the data as clusters. The objective of the data analysis was to compare mean responses among the 4 groups (A: Uninfected, B: HIV (+) Viremic, C: Immune Success, D:

Immune Failure). There were 6 continuous response variables (TLR3-Crypt, TLR3-

Surface Epithelium, TLR4-Crypt, TLR4-Surface Epithelium, TLR9-Crypt, TLR9-Surface

Epithelium), since Threshold Area % could range from 0-100 percent, and 4 dichotomous response variables (TLR3 Lamina Propria, TLR4 Lamina Propria, TLR9 Lamina

Propria), since cells were determined to either be positive or negative (binary response).

To compare means (or proportions) while taking into account multiple (presumably correlated) observations for each person, we used a generalized estimating equations

(GEE) approach. For continuous response variables, we used a log link and a saturated

variance function (allowing different variances for each group). For binary response variables, we used a logit link and the variance function for the Bernoulli distribution. For both types of responses, an independence working correlation structure was used. The

GEE approach uses robust (empirical or ‘sandwich’) estimates for the regression parameter variances. To deal with the relatively small number of individuals (‘clusters’) in the data, we incorporated the Morel, Bokossa, and Neerchal (MBN) correction to the variance estimates[219].

For each group comparison (for each response variable) a t-test based on the GEE

(using the empirical estimates of the variances) was conducted. Regression parameter estimates were exponentiated to provide estimated mean ratios for the continuous responses and estimated odds ratios for the binary responses. 95% t-type confidence limits were obtained for the mean and odds ratios. P-values were not corrected for multiple testing. All analyses were carried out in SAS (Version 9.4).

Chapter 4: Discussion

100

Summary of Findings

In HIV disease, there is damage to the host immune system, particularly in the lamina propria of the intestinal mucosa, which is not restored with successful ART-mediated viral suppression. In addition to an altered lamina propria, the intestinal barrier becomes

“leaky” and allows for microbial translocation and localized inflammation. HIV (+) patients also have increased systemic inflammation, which we partially attribute to the microbial translocation resultant from the intestinal barrier defect.

In these studies, we found an increase in monocyte activation from peripheral blood of HIV (+) patients, and an altered TLR phenotype in the colons of HIV (+) patients, both of which only partially normalize with therapy, possibly providing a mechanism and symptom of the elevated inflammatory status seen in the colon and periphery.

In Chapter 2, we explored the phenotype of monocytes in HIV (+) patients both prior to and after ART and compared them longitudinally and to uninfected controls. We found that monocyte activation, as shown by HLA-DR and CD86 expression, is higher in

HIV (+) patients when compared to controls and only partially reduced with ART.

Additionally, trafficking molecules CCR2 and CX3CR1, are lower in HIV (+) patients and partially normalize with ART. These results indicate that while ART reduces, but does not fully correct these defects found in the monocyte phenotype, residual immune activation remains.

101

In Chapter 3, we studied one potential mechanism for the residual immune activation seen in ART-treated HIV disease, demonstrating an altered abundance and localization of

Toll-Like receptors. We found that TLRs, innate sensors of barrier defect (TLR4) and viral infection (TLR3, TLR9) in the colon, are altered in HIV disease both prior to and after ART. In Viremia, TLR3, TLR4, and TLR9 are all upregulated and the localization of TLR4 is altered toward the crypt (Figure 4.1a, 4.1b). Successful ART normalizes the abundance of TLR4 and TLR9 on the epithelium, and normalizes TLR3 and TLR9 abundance in the lamina propria. However, TLR3 remains elevated in both the surface epithelium and crypt, and TLR4 expression is decreased below healthy levels (Figure

4.1c). Immune failure patients showed increased levels of TLR3 and TLR9 (both viral) in all three compartments when compared to uninfected individuals, while TLR4 expression was shifted towards the surface epithelium and diminished in the lamina propria (Figure 4.1c).

Together these results show that TLR expression is altered during viremia in both the lamina propria and the epithelium, and successful ART either partially normalizes the abundance (TLR4, TLR9) or furthers the disruption (TLR3). Patients with Immune failure have further disrupted TLR expression, likely due to persistent low levels of viral replication, mucosal immune imbalance, and microbial translocation. These TLR alterations provide one potential mechanism for the increased localized inflammation in the colon and the inflammatory monocyte phenotype we found in HIV disease, both prior to and after ART.

102

103

Figure 4.1. Phenotypic alterations in the colon of patients with HIV-1. (A)

Uninfected healthy controls have an intact epithelial cell barrier with normal levels of lamina propria cells, and low levels of inflammation. (B) The HIV-1 infected colon has decreased levels of CD4+ T cells, increased epithelial cell and macrophage activation, decreased TJ proteins, increased microbial translocation, and altered TLR abundance on the colonic epithelium. (C)

Immune Success patients have increased barrier function, CD4+ T cells, and normalized TLR4 and TLR9 expression. Immune Failure patients have normalized TLR4 expression, and increased

TLR3 and TLR9 expression on the colonic epithelium.

104

Discussion

Taken together, our results indicate that while ART has been successful in its ability to diminish viremia, patients continue to have immune dysfunction. Importantly, if treatment is interrupted or there is suboptimal adherence, patient viremia will rebound and the patient could progress to AIDS. This viral rebound indicates viral latency and/or pools of sequestered virus which ART cannot penetrate, so current treatment is not curative. In this treatment era of HIV, it then becomes important to not only treat the viremia, but also provide quality of life to the patient, as well as realize that finding a cure for HIV may not be sufficient.

Prior to our studies, the effect of Viremia and ART on monocyte phenotype was incompletely characterized. In Chapter 2, we found increased monocyte activation in the periphery of HIV (+) patients which did not fully normalize with ART. Specifically, viremic patients have higher HLA-DR and CD86 expression and density on circulating monocytes, lower expression and density of chemokine receptors CCR2 and CX3CR1, and increased patrolling monocytes when compared to uninfected controls. We propose that the increased monocyte activation is likely playing a role in the increased cardiovascular risk seen in HIV (+) patients. This peripheral monocyte activation is accompanied by T cell activation, IL6, sCD14, and LPS, linking the activation in the periphery to microbial translocation (LPS, sCD14). The traditional monocytes, which produce the most IL6 in response to LPS, remain activated (increased HLADR%) despite

ART, which may partially explain the increased levels of IL6 found in the periphery of

HIV (+) patients. Importantly, IL6, one of the many inflammatory increased in HIV

105

disease, is produced after TLR ligation and is correlated with morbidity and mortality[220].

In Chapter 3, we found that Viremic patients, from a different cohort, have increased

TLR3, TLR4, and TLR9 expression in the colonic epithelium and lamina propria, as well as altered localization. These TLRs recognize viral (TLR3-RNA and TLR9-RNA:DNA hybrids) and bacterial (TLR4-LPS) products as well as endogenous ligands, and would, therefore, be activated by both HIV viremia and microbial translocation. These data indicate an overall activated phenotype in HIV disease, likely induced by microbial translocation and viremia.

Today patients have access to ART, but their life expectancy is still lower than the rest of the population and there is persistent peripheral immune activation. As discussed previously, there are patients who successfully restore their CD4+ T cells after 2 years of

ART (immune success) and those who do not (immune failure). Immune Failures have increased morbidity and mortality due to AIDS related and non-AIDS related diseases, so it is important to understand contributing factors to their non-responsiveness in order to provide better quality of care. Of note, both immune successes and immune failures have increased levels of IL6, sCD14, and IFN when compared to their uninfected counterparts, however, immune failures have the highest of these inflammatory indices.

In Chapter 2, we found that after 48 weeks of ART, monocyte activation (HLA-DR and

CD86) diminished and chemokine receptor expression (CCR2 and CX3CR1) increased, though only partially. In addition, CD40 expression on monocytes was further diminished with ART. It is important to note that we could not distinguish these patients

106

into immune successes and failures, since we only tracked them for 48 weeks of therapy, and our definition requires 2 full years of ART, but over half of the patients for which we had week 48 data had already recovered CD4+ cell counts above 500/ul and only a quarter still had <350 CD4+ T cells/ul.

In Chapter 3, we found TLR alterations in both Immune Successes and Immune Failures.

Both IS and IF patients had increased TLR3 in the colonic epithelium. IF patients had increased levels of TLR9 and altered TLR4 localization, while IS patients normalized the expression of TLR4 and TLR9. Together these data illustrate the inability of ART to fully repair the immune dysfunction in HIV (+) patients, and in the case of TLR3, actually cause further disruption.

To reiterate, in viremia there is altered TLRs expression, and likely altered TLR signaling, coupled with decreased barrier function and a decrease in regulatory T cells in the lamina propria which a negative feedback loop that leads to peripheral immune activation, including monocyte activation. Anti-retroviral therapy, whether it is successful in the restoration of CD4+ T cells in the periphery or not, does not fully restore, and in some cases disrupts, immune function.

Our data highlight the need for determining the mechanism behind the changes that occur both in viremia and after ART intervention, in order to provide therapeutics which can help restore mucosal immune function to patients with HIV. Our studies determined the previously unknown expression and localization patterns of TLR in the colons of HIV (+) patients, which can aid in the identification of therapeutics which target specific TLRs

107

which are altered in these patients. TLR agonists could be used to increase TLR activation and aid in the “shock and kill” strategy for HIV eradication. This “shock and kill” strategy involves activating infected cells to bring HIV out of latency (shock) and then CD8+ T cells and NK cells kill the infected cells. If at the same time the patient is receiving ART, neighboring cells will not be infected and this would cure the patient.

Specific TLR ag8onists for TLR2 and TLR4, along with probiotics, may increase T reg and Th17 recovery by increasing the differentiation of these cells. TLR antagonists could also be used both prior to and after the cure to decrease inflammation and reset the balance of the gut.

Future Directions

Confirm TLR expression alterations in Colonic Epithelium of HIV (+) and HIV (-) individuals using Flow cytometry.

Rationale

108

While immunofluorescence can identify one or two TLRs at a time, flow cytometry can analyze significantly more. Developing a protocol for isolating colonic epithelial cells and analyzing TLRs by flow cytometry would greatly increase our understanding of their abundance as well as give us a greater understanding about their co-expression.

Preliminary Findings

In order to gain a further understanding of TLR expression in the colons of HIV (+)

2individuals, both prior to and after ART, we propose to use flow cytometry. While we would not be able to use this technique to assess the localization of the TLR, it will give more insight into their co-expression, as well as introduce a novel isolation technique for the phenotypic analysis of colonic epithelial cells. This isolation technique can then be used in further related and unrelated studies that utilize both flow cytometry and western blotting.

Prior to analyzing HIV (+) colons it was essential to determine the proper technique for human colonic epithelial cell isolation. In Pan et al, they used a rather complex method for isolation of rhesus macaque colonic epithelial cells[215]. We set out to test their method, as well as determine if there were any steps which could be eliminated or improved.

Human intestinal biopsies were obtained from surgical resections using forceps. The full digestion protocol included a DTT wash, two EDTA washes, a triple digest, and then

Percoll isolation. For the DTT in HBSS wash, 10 biopsies were washed for 30 minutes at

37C. After this initial wash, samples were harvested by centrifugation, the supernatant

109

was removed, and the pellet washed in 1mM EDTA in HBSS for 30 minutes at 37C.

This step was then repeated for a second EDTA wash. The tissue was centrifuged, the supernatant was removed, and the samples were minced in digestion media (RPMI with

2% FBS, 20 U/mL DNAse I, and 8ug/mL Liberase DL). The minced tissue was then incubated for 30 minutes with shaking at 37C. After the digestion, single cells are separated with a 70 uM cell strainer, and the remaining tissue digested again. This digestion is performed three times, and then the single cell suspension is fractionated on a

30-60% Percoll gradient and centrifuged at 1000g for 20 minutes. The top layer contains the isolated epithelial cells. Steps were deleted from this protocol to determine optimal conditions for isolating epithelial cells.

We used a live/dead yellow dye (Invitrogen) to determine the effect of each step on cell death, and used a PE conjugated Epithelial cell antigen antibody (Cell Signaling) to detect the proportion of epithelial cells. Gating is shown in Figure 4.2.

We found that using the standard triple digestion protocol for isolating a single cell suspension of colon cells, but which is mainly lamina propria cells, produced a high number of live cells, but less than 40% were epithelial cells. In contrast, the protocol used by Pan et al [215]produced very few cells, many of which were living but only 3% of which were epithelial cells. The best isolation protocol included a 0.15% DTT wash with the triple digest, yielding a higher proportion of dead cells (74.1%), but the highest proportion of living epithelial cells (75.7%) (Figure 4.3a). Neither the single nor the double EDTA wash contributed to a higher proportion of living epithelial cells.

Given that the best protocol for epithelial cell isolation appeared to be highly lethal to the cells, we optimized the concentration of DTT in the wash (Figure 4.2b).

110

Here, again, 0.15% DTT was highly lethal (73.8% cell death) with a reduced success in isolating epithelial cells. The ten-fold dilution, 0.015% DTT wash, was much more successful with 65.3% cell survival and 82.8% epithelial cell purity. Similarly, the 100- fold dilution, 0.0015% DTT wash, was also successful with 69.6% cell survival and

65.7% epithelial cell purity. The best protocol for purity and cell survival, therefore, appeared to be a 0.015% DTT wash with a triple digest (Figure 4.3b).

Using the optimized procedure, five HIV (-) controls were tested for the expression of

TLR2 (PeCy7-eBioscience), TLR4 (AF700-eBioscience), and TLR5 (PE-Abcam). The isolated cells were also stained for CD45 (PerCP-eBioscience), a lymphocyte marker not on epithelial cells, to further distinguish which cells were expressing the TLRs and

EpCAM (AF647-Cell Signaling). We found that TLR2 and TLR5 were moderately expressed on epithelial cells (EpCAM+ CD45-), and highly expressed on both EpCAM-

CD45- cells (which could represent the myeloid lineage) and the EpCAM-CD45+ lymphocytes (Figure 4.4). TLR4 expression was low on all cell types, though we only tried one antibody, so it’s possible that a different antibody may show different TLR4 expression (Figure 4.4).

Goals We must continue to perfect the protocol for studying TLR expression on epithelial cells by flow cytometry and then compare HIV (+) and HIV (-) samples for analysis. Using flow cytometry to examine TLR abundance will allow us to determine the co-expression of these TLRs in the epithelium.

111

- FSC

FSC-A LIVE/DEAD

- SSC

FSC-A EpCAM

Figure 4.2. Gating strategy for Flow Cytometric analysis of Epithelial Cell antigen on

Epithelial cells after Isolation protocol. Cells were gated on FSC-H vs FSC-A (P1) to remove doublets and then gated on live cells (P2) using a live/dead yellow dye. Cells were then gated on FSC-A vs SSC-A (P3) to remove debris from analysis. Finally, cells were gated on EpCAM (P4). This is an example of EpCAM expression after a triple digest.

112

0 Normal 3x Digest 0.15% DTT wash, 0.15% DTT wash, 0.15% DTT wash, 0.15% DTT wash, 3x Digest EDTA wash, 3x 2x EDTA wash, 2x EDTA wash, Digest 3x Digest Percoll, 3x Digest

Live Cell % Live Epithelial Cells %

90 80 70 60 50 40 30 20 10 0 Normal 3x Digest 0.15% DTT wash, 3x 0.015% DTT wash, 0.0015% DTT was, Digest 3x Digest 3x Digest

Live Cell % Live Epithelial Cells %

Figure 4.3. Proportion of Live Cells and Live Epithelial Cells with various wash steps prior to triple digestion. A) Epithelial cells were isolated with either the full protocol from Pan et al [215](0.15% DTT wash, two EDTA washes, Percoll isolation and a triple digest-far right), or the indicated modifications, one being just the triple digest (far left).

B) Having found that 0.15% DTT with triple digestion was the best at retrieving isolated epithelial cells, with an unacceptable level of cell death, the proportion of DTT was diluted to 0.015% or 0.0015%.

113

TLR panel Isotype Panel TLR2 PeCy7 mIgG1 PeCy7 TLR4 AF700 mIgG2a AF700 TLR5 PE mIgG2a PE CD45 PerCP mIgG1 APC EpCAM APC

Figure 4. 4. Preliminary results of TLR expression on partially purified epithelial cells

(EpCAM+ CD45-) and immune cells (EpCAM- CD45- and EpCAM- CD45+).

114

TLR ligands and cytokines alter the abundance and localization of TLRs on the colonic epithelium and lamina propria

Rationale The effect of TLR ligands on TLR abundance and localization in the colon has not yet been studied. Evidence in the brain suggests that TLR3 ligation by poly I:C can downregulate TLR4 expression[221]. Furthermore, TLR9 ligation by CPG in HT29 cells

(a colonic cell line) can increase the expression of TLR9. There are currently no studies with primary colon tissue, representing an important area of study.

In addition to understanding the effect of TLR activation, we want to determine the effect of TNF, IFNa/b, IL-6, IL10, IL-22, and IL-17. It’s already been shown in the brain that

IFN can alter TLR3 signaling through Poly I:C[217], so it is important to see what happens in the colon. Previous studies were in different tissues or in cell lines, so it is essential to be able to culture primary tissue to elucidate these effects.

Preliminary Findings

An organotypic tissue culture system was developed as described in Figure 4.5. Our first attempts included complete RPMI media supplemented with 20% Fetal Bovine Serum

(FBS), 1% Hepes, and several antibiotics/antifungals (Pen/Strep, Piperacillin,

Tazobactam, and Fungizone), as was used in Fletcher et al[214], but after 12 hours of

115

incubation, the epithelial cells sloughed off (Figure 4.6). Upon searching the literature which discussed organotypic colonic tissue culture[222], we tested a few different medias that were suggested, and we found that both CMRL and L15 media were much better at maintaining epithelial integrity (Figure 4.7). We could culture biopsies in either CMRL and L15 media for as long as 48 hrs without substantial epithelial cell loss.

Goals Our goal is to increase the number of days we can culture the punch biopsies and maintain an intact epithelium, enabling us to stimulate these organotypic cultures with

TLR ligands or cytokines (IL-17, IL-22, IL-6, IL-10, TNF, IFNalpha). Organoid culture medium typically contains Epidermal Growth Factor and kinase inhibitors (Y-27632-

Sigma; A 83-01-Tocris; SB202190-Sigma), so we propose to test the effect of these on epithelial cell survival in the organotypic culture. After stimulation with a variety of agonists, the organotypic cultures will be fixed in 4% PFA, embedded in paraffin, and stained using IF protocols we perfected in Chapter 3. We also propose to develop colonic organoids, a three- dimensional colon model derived from primary intestinal biopsies.

We aim to further understand the influence of TLR ligand binding in the colon on the expression and localization of the primary target TLR, as well on other non-target TLRs similar to the TLR3/TLR4 interaction in the brain. We also aim to further understand the influence of pro- and anti-inflammatory cytokine signaling on TLR expression and localization, both on their own and in consort with TLR ligands.

116

Figure 4.5. Schematic of Organotypic tissue culture. A piece of gelfoam is placed on a 12 mm 3.0uM polycarbonate, TC treated transwell (costar). Punch biopsies are made using 6mm Biopsy punch (Miltex) and placed on the gelfoam (pfiser) which is slightly larger than the tissue. The gelfoam and tissue are then surrounded with matrigel

(Invivogen). Complete media is then placed in the bottom well.

117

Figure 4.6. Histological H&E stains of punch biopsies magnified 20X after 12 (A), 24

(B), and 48 (C) hrs of incubation in complete RPMI media. Punch biopsies were taken from separate surgical colonic resections for Panels A, B, and C. Evidence of the depletion of the epithelial cells can be seen as early as 12 hrs, as seen in comparison to how the epithelium appeared prior to the incubation.

118

Figure 4.7. Histological H&E stains of punch biopsies magnified 20X after 24, 48, 72 hrs, and 7 days of incubation in complete media. Punch biopsies are all taken from the same patient’s surgical resections. A punch biopsy that was not cultured was immediately fixed in 4% PFA and transferred to 80% EtOH the next day (A). Punch biopsies were cultured in either CMRL media (B), L15 media (C), or RPMI (D).

Biopsies were fixed in 4% PFA after 24, 48, 72 hrs or 7days and transferred to 80%

EtOH the next day.

119

Evaluate the toxicity of an HIV virion and ART compounds on TLR abundance and localization in Colonic Epithelium and Lamina Propria

Rationale As we showed in Chapter 3, it appears that both viremia and ART are modulating mucosal TLR expression and location. Components of HIV can ligate TLR3 and TLR9; so one might expect exposing organotypic mucosal cultures to HIV may result in alterations in the TLR abundance and/or localization, as observed in vivo. Additionally, anti-retroviral compounds, particularly protease inhibitors have been previously shown to alter TLR signaling in HIV (+) patients as well as increase activation and inflammation.

We, therefore, propose that the treatment of organotypic cultures with protease inhibitors would alter the abundance and/or localization of the TLRs.

Goals We will use the organotypic culture system to determine if TLR expression and localization is altered directly by viremia or by ART. This would then provide a potential mechanism for the alterations observed in Chapter 3, such as increased TLR3, TLR4, and TLR9 expression in viremia and increased TLR3 expression after ART.

120

Concluding Remarks Currently, one of the main goals at the National Institutes of Health (NIH) is to find a cure for

HIV, but our data suggests that finding a cure for the virus may not be sufficient. We found alterations both peripherally, as shown in Chapter 2, and locally in the intestinal mucosa, as shown in Chapter 3, which occurred during viremia and which were not fully restored despite the absence of detectable viral replication and successful recovery of CD4+ T cell counts in the blood. The longitudinal effects of ART on the phenotype of monocytes in HIV (+) patients had not been fully characterized prior to our studies, and the phenotype of TLRs in the colonic epithelium was entirely unknown, leading to new questions about the mechanisms behind these alterations. Is it the viremia, ART, or inflammation which is causing these alterations? It is most likely a combination of all three, and, therefore, removing virus from the patient would solve the problem of viremia and toxic effects of ART, but the damage done to the immune system and the sustained immune activation would still need to be addressed, and it is likely that there will still be Immune Failures in the HIV Cure Era. Pursuing the mechanisms behind the alterations we found will aid in the efforts to address residual immune dysfunction in HIV (+) patients and lead to therapeutics which will help these patients become both devoid of virus and have a healthy and functional immune system.

121

References

1. Haverkos, H.W. and J.W. Curran, The current outbreak of kaposi's sarcoma and opportunistic infections. CA: A Cancer Journal for Clinicians, 1982. 32(6): p. 330-339. 2. AVERT. History of HIV and AIDS Overview. 2016 [cited 2016; Available from: http://www.avert.org/professionals/history-hiv-aids/overview#footnote6_c06u148. 3. Gallo , R.C. and L. Montagnier The Discovery of HIV as the Cause of AIDS. New England Journal of Medicine, 2003. 349(24): p. 2283-2285. 4. Nyamweya, S., A. Hegedus, A. Jaye, S. Rowland-Jones, K.L. Flanagan, and D.C. Macallan, Comparing HIV-1 and HIV-2 infection: Lessons for viral immunopathogenesis. Reviews in Medical Virology, 2013. 23(4): p. 221-240. 5. Hemelaar, J., The origin and diversity of the HIV-1 pandemic. Trends in Molecular Medicine. 18(3): p. 182-192. 6. Frankel, A.D. and J.A. Young, HIV-1: fifteen proteins and an RNA. Annu Rev Biochem, 1998. 67: p. 1-25. 7. Wilen, C.B., J.C. Tilton, and R.W. Doms, HIV: Cell Binding and Entry. Cold Spring Harbor Perspectives in Medicine, 2012. 2(8): p. a006866. 8. Okoye, A.A. and L.J. Picker, CD4(+) T cell depletion in HIV infection: mechanisms of immunological failure. Immunological reviews, 2013. 254(1): p. 54-64. 9. Jayappa, K.D., Z. Ao, and X. Yao, The HIV-1 passage from cytoplasm to nucleus: the process involving a complex exchange between the components of HIV-1 and cellular machinery to access nucleus and successful integration. Int J Biochem Mol Biol, 2012. 3(1): p. 70-85. 10. Kimata, J.T., A.P. Rice, and J. Wang, Challenges and strategies for the eradication of the HIV reservoir. Curr Opin Immunol, 2016. 42: p. 65-70. 11. Freed, E.O., HIV-1 assembly, release and maturation. Nat Rev Micro, 2015. 13(8): p. 484- 496. 12. Bryant, M. and L. Ratner, Myristoylation-dependent replication and assembly of human immunodeficiency virus 1. Proc Natl Acad Sci U S A, 1990. 87(2): p. 523-7. 13. Gottlinger, H.G., J.G. Sodroski, and W.A. Haseltine, Role of capsid precursor processing and myristoylation in morphogenesis and infectivity of human immunodeficiency virus type 1. Proc Natl Acad Sci U S A, 1989. 86(15): p. 5781-5. 14. Gallay, P., S. Swingler, J. Song, F. Bushman, and D. Trono, HIV nuclear import is governed by the phosphotyrosine-mediated binding of matrix to the core domain of integrase. Cell, 1995. 83(4): p. 569-76. 15. Campbell, E.M. and T.J. Hope, HIV-1 capsid: the multifaceted key player in HIV-1 infection. Nat Rev Micro, 2015. 13(8): p. 471-483. 16. Harrison, G.P. and A.M. Lever, The human immunodeficiency virus type 1 packaging signal and major splice donor region have a conserved stable secondary structure. J Virol, 1992. 66(7): p. 4144-53. 17. Lapadat-Tapolsky, M., H. De Rocquigny, D. Van Gent, B. Roques, R. Plasterk, and J.L. Darlix, Interactions between HIV-1 nucleocapsid protein and viral DNA may have important functions in the viral life cycle. Nucleic Acids Res, 1993. 21(4): p. 831-9.

122

18. Paxton, W., R.I. Connor, and N.R. Landau, Incorporation of Vpr into human immunodeficiency virus type 1 virions: requirement for the p6 region of gag and mutational analysis. J Virol, 1993. 67(12): p. 7229-37. 19. Gulnik, S., J.W. Erickson, and D. Xie, HIV protease: enzyme function and drug resistance. Vitam Horm, 2000. 58: p. 213-56. 20. Sarafianos, S.G., B. Marchand, K. Das, D. Himmel, M.A. Parniak, S.H. Hughes, and E. Arnold, Structure and function of HIV-1 reverse transcriptase: molecular mechanisms of polymerization and inhibition. Journal of molecular biology, 2009. 385(3): p. 693-713. 21. Beilhartz, G.L. and M. Götte, HIV-1 Ribonuclease H: Structure, Catalytic Mechanism and Inhibitors. Viruses, 2010. 2(4): p. 900-926. 22. Esposito, D. and R. Craigie, HIV integrase structure and function. Adv Virus Res, 1999. 52: p. 319-33. 23. McCune, J.M., L.B. Rabin, M.B. Feinberg, M. Lieberman, J.C. Kosek, G.R. Reyes, and I.L. Weissman, Endoproteolytic cleavage of gp160 is required for the activation of human immunodeficiency virus. Cell, 1988. 53(1): p. 55-67. 24. Sáez-Cirión, A., J.L.R. Arrondo, M.J. Gómara, M. Lorizate, I. Iloro, G. Melikyan, and J.L. Nieva, Structural and Functional Roles of HIV-1 gp41 Pretransmembrane Sequence Segmentation. Biophysical Journal, 2003. 85(6): p. 3769-3780. 25. Du Toit, A., Structural biology: The many faces of the HIV-1 spike. Nat Rev Micro, 2014. 12(12): p. 792-792. 26. Feinberg, M.B., D. Baltimore, and A.D. Frankel, The role of Tat in the human immunodeficiency virus life cycle indicates a primary effect on transcriptional elongation. Proc Natl Acad Sci U S A, 1991. 88(9): p. 4045-9. 27. Rasty, S., P. Thatikunta, J. Gordon, K. Khalili, S. Amini, and J.C. Glorioso, Human immunodeficiency virus tat gene transfer to the murine central nervous system using a replication-defective herpes simplex virus vector stimulates transforming growth factor beta 1 gene expression. Proc Natl Acad Sci U S A, 1996. 93(12): p. 6073-8. 28. Brother, M.B., H.K. Chang, J. Lisziewicz, D. Su, L.C. Murty, and B. Ensoli, Block of Tat- mediated transactivation of tumor necrosis factor beta gene expression by polymeric- TAR decoys. Virology, 1996. 222(1): p. 252-6. 29. Sharma, V., M. Xu, L.M. Ritter, and N.M. Wilkie, HIV-1 tat induces the expression of a new hematopoietic cell-specific transcription factor and downregulates MIP-1 alpha gene expression in activated T-cells. Biochem Biophys Res Commun, 1996. 223(3): p. 526-33. 30. Sastry, K.J., M.C. Marin, P.N. Nehete, K. McConnell, A.K. el-Naggar, and T.J. McDonnell, Expression of human immunodeficiency virus type I tat results in down-regulation of bcl- 2 and induction of apoptosis in hematopoietic cells. Oncogene, 1996. 13(3): p. 487-93. 31. Pollard, V.W. and M.H. Malim, The HIV-1 Rev protein. Annu Rev Microbiol, 1998. 52: p. 491-532. 32. Strebel, K., HIV-1 Vpu — an ion channel in search of a job. Biochimica et Biophysica Acta (BBA) - Biomembranes, 2014. 1838(4): p. 1074-1081. 33. Romani, B. and S. Engelbrecht, Human immunodeficiency virus type 1 Vpr: functions and molecular interactions. J Gen Virol, 2009. 90(Pt 8): p. 1795-805. 34. Rose, K.M., M. Marin, S.L. Kozak, and D. Kabat, The viral infectivity factor (Vif) of HIV-1 unveiled. Trends Mol Med, 2004. 10(6): p. 291-7. 35. Basmaciogullari, S. and M. Pizzato, The activity of Nef on HIV-1 infectivity. Front Microbiol, 2014. 5: p. 232.

123

36. FDA-Approved HIV Medicines | HIV/AIDS Fact Sheets | Education Materials | AIDSinfo. 2015; Available from: http://www.ncbi.nlm.nih.gov/pubmed/. 37. Chowers, M., B.S. Gottesman, L. Leibovici, J.M. Schapiro, and M. Paul, Nucleoside reverse transcriptase inhibitors in combination therapy for HIV patients: systematic review and meta-analysis. Eur J Clin Microbiol Infect Dis, 2010. 29(7): p. 779-86. 38. Usach, I., V. Melis, and J.E. Peris, Non-nucleoside reverse transcriptase inhibitors: a review on pharmacokinetics, pharmacodynamics, safety and tolerability. J Int AIDS Soc, 2013. 16: p. 1-14. 39. Poveda, E., V. Briz, and V. Soriano, Enfuvirtide, the first fusion inhibitor to treat HIV infection. Aids Reviews, 2005. 7(3): p. 139-147. 40. Podany, A.T., K.K. Scarsi, and C.V. Fletcher, Comparative Clinical Pharmacokinetics and Pharmacodynamics of HIV-1 Integrase Strand Transfer Inhibitors. Clin Pharmacokinet, 2016. 41. Briz, V., E. Poveda, and V. Soriano, HIV entry inhibitors: mechanisms of action and resistance pathways. Journal of Antimicrobial Chemotherapy, 2006. 57(4): p. 619-627. 42. Temesgen, Z., Cobicistat, a pharmacoenhancer for HIV treatments. Drugs Today (Barc), 2013. 49(4): p. 233-7. 43. Abram, M.E., A.L. Ferris, K. Das, O. Quinones, W. Shao, S. Tuske, W.G. Alvord, E. Arnold, and S.H. Hughes, Mutations in HIV-1 reverse transcriptase affect the errors made in a single cycle of viral replication. J Virol, 2014. 88(13): p. 7589-601. 44. Chen, T.K. and G.M. Aldrovandi, Review of HIV antiretroviral drug resistance. Pediatr Infect Dis J, 2008. 27(8): p. 749-52. 45. WHO. HIV/AIDS. 2015; Available from: http://www.who.int/mediacentre/factsheets/fs360/en/. 46. Wong, J.K. and S.A. Yukl, Tissue reservoirs of HIV. Curr Opin HIV AIDS, 2016. 11(4): p. 362-70. 47. Grabmeier-Pfistershammer, K., P. Steinberger, A. Rieger, J. Leitner, and N. Kohrgruber, Identification of PD-1 as a unique marker for failing immune reconstitution in HIV-1- infected patients on treatment. J Acquir Immune Defic Syndr, 2011. 56(2): p. 118-24. 48. Kelley, C.F., C.M. Kitchen, P.W. Hunt, B. Rodriguez, F.M. Hecht, M. Kitahata, H.M. Crane, J. Willig, M. Mugavero, M. Saag, J.N. Martin, and S.G. Deeks, Incomplete peripheral CD4+ cell count restoration in HIV-infected patients receiving long-term antiretroviral treatment. Clin Infect Dis, 2009. 48(6): p. 787-94. 49. Tirumalasetti, N. and P. Prema Latha, Lymph nodes cytology in HIV seropositive cases with haematological alterations. The Indian Journal of Medical Research, 2014. 139(2): p. 301-307. 50. Funderburg, N.T., A. Andrade, E.S. Chan, S.L. Rosenkranz, D. Lu, B. Clagett, H.A. Pilch- Cooper, B. Rodriguez, J. Feinberg, E. Daar, J. Mellors, D. Kuritzkes, J.M. Jacobson, and M.M. Lederman, Dynamics of Immune Reconstitution and Activation Markers in HIV+ Treatment-Naïve Patients Treated with Raltegravir, Tenofovir Disoproxil Fumarate and Emtricitabine. PLoS ONE, 2013. 8(12): p. e83514. 51. Moore, R.D., J.G. Bartlett, and J.E. Gallant, Association between use of HMG CoA reductase inhibitors and mortality in HIV-infected patients. PLoS One, 2011. 6(7): p. e21843. 52. Cros, J., N. Cagnard, K. Woollard, N. Patey, S.Y. Zhang, B. Senechal, A. Puel, S.K. Biswas, D. Moshous, C. Picard, J.P. Jais, D. D'Cruz, J.L. Casanova, C. Trouillet, and F. Geissmann, Human CD14dim monocytes patrol and sense nucleic acids and viruses via TLR7 and TLR8 receptors. Immunity, 2010. 33(3): p. 375-86.

124

53. Westhorpe, C.L., A. Maisa, T. Spelman, J.F. Hoy, E.M. Dewar, S. Karapanagiotidis, A.C. Hearps, W.J. Cheng, J. Trevillyan, S.R. Lewin, D. Sviridov, J.H. Elliott, A. Jaworowski, A.M. Dart, and S.M. Crowe, Associations between surface markers on blood monocytes and carotid atherosclerosis in HIV-positive individuals. Immunol Cell Biol, 2014. 92(2): p. 133- 8. 54. Funderburg, N.T., D.A. Zidar, C. Shive, A. Lioi, J. Mudd, L.W. Musselwhite, D.I. Simon, M.A. Costa, B. Rodriguez, S.F. Sieg, and M.M. Lederman, Shared monocyte subset phenotypes in HIV-1 infection and in uninfected subjects with acute coronary syndrome. Blood, 2012. 120(23): p. 4599-4608. 55. Linton, L., M. Karlsson, J. Grundström, E. Hjalmarsson, A. Lindberg, E. Lindh, H. Glise, R. Befrits, I. Janczewska, P. Karlén, O. Winqvist, and M. Eberhardson, HLA-DR(hi) and CCR9 Define a Pro-Inflammatory Monocyte Subset in IBD. Clinical and Translational Gastroenterology, 2012. 3(12): p. e29. 56. Andersen, N.N. and T. Jess, Risk of cardiovascular disease in inflammatory bowel disease. World Journal of Gastrointestinal Pathophysiology, 2014. 5(3): p. 359-365. 57. Farquhar, M.G. and G.E. Palade, Junctional complexes in various epithelia. J Cell Biol, 1963. 17: p. 375-412. 58. Ulluwishewa, D., R.C. Anderson, W.C. McNabb, P.J. Moughan, J.M. Wells, and N.C. Roy, Regulation of tight junction permeability by intestinal bacteria and dietary components. J Nutr, 2011. 141(5): p. 769-76. 59. Neunlist, M., L. Van Landeghem, M.M. Mahe, P. Derkinderen, S.B. des Varannes, and M. Rolli-Derkinderen, The digestive neuronal-glial-epithelial unit: a new actor in gut health and disease. Nat Rev Gastroenterol Hepatol, 2013. 10(2): p. 90-100. 60. Sosinsky, G.E. and B.J. Nicholson, Structural organization of gap junction channels. Biochim Biophys Acta, 2005. 1711(2): p. 99-125. 61. Abreu, M.T., Toll-like receptor signalling in the intestinal epithelium: how bacterial recognition shapes intestinal function. Nat Rev Immunol, 2010. 10(2): p. 131-44. 62. Brown, E.M., M. Sadarangani, and B.B. Finlay, The role of the immune system in governing host-microbe interactions in the intestine. Nat Immunol, 2013. 14(7): p. 660-7. 63. Kucharzik, T., N. Lugering, K. Rautenberg, A. Lugering, M.A. Schmidt, R. Stoll, and W. Domschke, Role of M cells in intestinal barrier function. Ann N Y Acad Sci, 2000. 915: p. 171-83. 64. Chistiakov, D.A., Y.V. Bobryshev, E. Kozarov, I.A. Sobenin, and A.N. Orekhov, Intestinal mucosal tolerance and impact of gut microbiota to mucosal tolerance. Frontiers in Microbiology, 2014. 5: p. 781. 65. Koboziev, I., F. Karlsson, and M.B. Grisham, Gut-associated lymphoid tissue, T cell trafficking, and chronic intestinal inflammation. Annals of the New York Academy of Sciences, 2010. 1207(Suppl 1): p. E86-E93. 66. Strauch, U.G., F. Obermeier, N. Grunwald, S. Gurster, N. Dunger, M. Schultz, D.P. Griese, M. Mahler, J. Scholmerich, and H.C. Rath, Influence of intestinal bacteria on induction of regulatory T cells: lessons from a transfer model of colitis. Gut, 2005. 54(11): p. 1546-52. 67. Kinugasa, T., T. Sakaguchi, X. Gu, and H.C. Reinecker, Claudins regulate the intestinal barrier in response to immune mediators. Gastroenterology, 2000. 118(6): p. 1001-11. 68. Blaschitz, C. and M. Raffatellu, Th17 Cytokines and the Gut Mucosal Barrier. Journal of Clinical Immunology, 2010. 30(2): p. 196-203. 69. Lycke, N.Y., IgA B Cell Responses to Gut Mucosal Antigens: Do We Know it all? Frontiers in Immunology, 2013. 4: p. 368.

125

70. Fasano, A. and T. Shea-Donohue, Mechanisms of disease: the role of intestinal barrier function in the pathogenesis of gastrointestinal autoimmune diseases. Nat Clin Pract Gastroenterol Hepatol, 2005. 2(9): p. 416-22. 71. Hunt, P.W., E. Sinclair, B. Rodriguez, C. Shive, B. Clagett, N. Funderburg, J. Robinson, Y. Huang, L. Epling, J.N. Martin, S.G. Deeks, C.L. Meinert, M.L. Van Natta, D.A. Jabs, and M.M. Lederman, Gut epithelial barrier dysfunction and innate immune activation predict mortality in treated HIV infection. J Infect Dis, 2014. 210(8): p. 1228-38. 72. Estes, J., J.V. Baker, J.M. Brenchley, A. Khoruts, J.L. Barthold, A. Bantle, C.S. Reilly, G.J. Beilman, M.E. George, D.C. Douek, A.T. Haase, and T.W. Schacker, Collagen deposition limits immune reconstitution in the gut. J Infect Dis, 2008. 198(4): p. 456-64. 73. Lackner, A.A., M. Mohan, and R.S. Veazey, The gastrointestinal tract and AIDS pathogenesis. Gastroenterology, 2009. 136(6): p. 1965-78. 74. Stubblefield Park, S., H. Sung, N. Funderburg, J. Meddings, and A. Levine, Increased small intestinal and colonic permeability, and loss of villus tip surface area, correlates with microbial translocation and immune activation in HIV (71.5). The Journal of Immunology, 2012. 188(1 Supplement): p. 71.5. 75. Keating, J., I. Bjarnason, S. Somasundaram, A. Macpherson, N. Francis, A.B. Price, D. Sharpstone, J. Smithson, I.S. Menzies, and B.G. Gazzard, Intestinal absorptive capacity, intestinal permeability and jejunal histology in HIV and their relation to diarrhoea. Gut, 1995. 37(5): p. 623-629. 76. Lima, A.A., T.M. Silva, A.M. Gifoni, L.J. Barrett, I.T. McAuliffe, Y. Bao, J.W. Fox, D.P. Fedorko, and R.L. Guerrant, Mucosal injury and disruption of intestinal barrier function in HIV-infected individuals with and without diarrhea and cryptosporidiosis in northeast Brazil. Am J Gastroenterol, 1997. 92(10): p. 1861-6. 77. Chung, C.Y., S.L. Alden, N.T. Funderburg, P. Fu, and A.D. Levine, Progressive Proximal-to- Distal Reduction in Expression of the Tight Junction Complex in Colonic Epithelium of Virally-Suppressed HIV+ Individuals. PLoS Pathog, 2014. 10(6): p. e1004198. 78. Chun, T.W., D.C. Nickle, J.S. Justement, J.H. Meyers, G. Roby, C.W. Hallahan, S. Kottilil, S. Moir, J.M. Mican, J.I. Mullins, D.J. Ward, J.A. Kovacs, P.J. Mannon, and A.S. Fauci, Persistence of HIV in gut-associated lymphoid tissue despite long-term antiretroviral therapy. J Infect Dis, 2008. 197(5): p. 714-20. 79. Lee, J., J.M. Gonzales-Navajas, and E. Raz, The "polarizing-tolerizing" mechanism of intestinal epithelium: its relevance to colonic homeostasis. Semin Immunopathol, 2008. 30(1): p. 3-9. 80. Sato, M., H. Suemori, N. Hata, M. Asagiri, K. Ogasawara, K. Nakao, T. Nakaya, M. Katsuki, S. Noguchi, N. Tanaka, and T. Taniguchi, Distinct and essential roles of transcription factors IRF-3 and IRF-7 in response to viruses for IFN-alpha/beta gene induction. Immunity, 2000. 13(4): p. 539-48. 81. Takaoka, A., H. Yanai, S. Kondo, G. Duncan, H. Negishi, T. Mizutani, S. Kano, K. Honda, Y. Ohba, T.W. Mak, and T. Taniguchi, Integral role of IRF-5 in the gene induction programme activated by Toll-like receptors. Nature, 2005. 434(7030): p. 243-9. 82. Honda, K., H. Yanai, H. Negishi, M. Asagiri, M. Sato, T. Mizutani, N. Shimada, Y. Ohba, A. Takaoka, N. Yoshida, and T. Taniguchi, IRF-7 is the master regulator of type-I interferon- dependent immune responses. Nature, 2005. 434(7034): p. 772-7. 83. Honda, K. and T. Taniguchi, IRFs: master regulators of signalling by Toll-like receptors and cytosolic pattern-recognition receptors. Nat Rev Immunol, 2006. 6(9): p. 644-658.

126

84. Hu, X., P.K. Paik, J. Chen, A. Yarilina, L. Kockeritz, T.T. Lu, J.R. Woodgett, and L.B. Ivashkiv, IFN-gamma suppresses IL-10 production and synergizes with TLR2 by regulating GSK3 and CREB/AP-1 proteins. Immunity, 2006. 24(5): p. 563-74. 85. Kunsch, C. and C.A. Rosen, NF-kappa B subunit-specific regulation of the interleukin-8 promoter. Mol Cell Biol, 1993. 13(10): p. 6137-46. 86. Kang, H.B., Y.E. Kim, H.J. Kwon, D.E. Sok, and Y. Lee, Enhancement of NF-kappaB expression and activity upon differentiation of human embryonic stem cell line SNUhES3. Stem Cells Dev, 2007. 16(4): p. 615-23. 87. Libermann, T.A. and D. Baltimore, Activation of interleukin-6 gene expression through the NF-kappa B transcription factor. Mol Cell Biol, 1990. 10(5): p. 2327-34. 88. Shimizu, Y., G.A. van Seventer, K.J. Horgan, and S. Shaw, Costimulation of proliferative responses of resting CD4+ T cells by the interaction of VLA-4 and VLA-5 with fibronectin or VLA-6 with laminin. J Immunol, 1990. 145(1): p. 59-67. 89. Son, Y.H., Y.T. Jeong, K.A. Lee, K.H. Choi, S.M. Kim, B.Y. Rhim, and K. Kim, Roles of MAPK and NF-kappaB in interleukin-6 induction by lipopolysaccharide in vascular smooth muscle cells. J Cardiovasc Pharmacol, 2008. 51(1): p. 71-7. 90. Murphy, T.L., M.G. Cleveland, P. Kulesza, J. Magram, and K.M. Murphy, Regulation of interleukin 12 p40 expression through an NF-kappa B half-site. Mol Cell Biol, 1995. 15(10): p. 5258-67. 91. Homma, Y., S. Cao, X. Shi, and X. Ma, The Th2 transcription factor c-Maf inhibits IL-12p35 gene expression in activated macrophages by targeting NF-kappaB nuclear translocation. J Interferon Cytokine Res, 2007. 27(9): p. 799-808. 92. Shakhov, A.N., M.A. Collart, P. Vassalli, S.A. Nedospasov, and C.V. Jongeneel, Kappa B- type enhancers are involved in lipopolysaccharide-mediated transcriptional activation of the tumor necrosis factor alpha gene in primary macrophages. J Exp Med, 1990. 171(1): p. 35-47. 93. Collart, M.A., P. Baeuerle, and P. Vassalli, Regulation of tumor necrosis factor alpha transcription in macrophages: involvement of four kappa B-like motifs and of constitutive and inducible forms of NF-kappa B. Mol Cell Biol, 1990. 10(4): p. 1498-506. 94. Shive, C.L., A. Biancotto, N.T. Funderburg, H.A. Pilch-Cooper, H. Valdez, L. Margolis, S.F. Sieg, G.A. McComsey, B. Rodriguez, and M.M. Lederman, HIV-1 is not a major driver of increased plasma IL-6 levels in chronic HIV-1 disease. J Acquir Immune Defic Syndr, 2012. 61(2): p. 145-52. 95. Stone, S.F., P. Price, N.M. Keane, R.J. Murray, and M.A. French, Levels of IL-6 and soluble IL-6 receptor are increased in HIV patients with a history of immune restoration disease after HAART. HIV Med, 2002. 3(1): p. 21-7. 96. Matsumoto, T., T. Miike, R.P. Nelson, W.L. Trudeau, R.F. Lockey, and J. Yodoi, Elevated serum levels of IL-8 in patients with HIV infection. Clin Exp Immunol, 1993. 93(2): p. 149- 51. 97. Hardy, G.A.D., S. Sieg, B. Rodriguez, D. Anthony, R. Asaad, W. Jiang, J. Mudd, T. Schacker, N.T. Funderburg, H.A. Pilch-Cooper, R. Debernardo, R.L. Rabin, M.M. Lederman, and C.V. Harding, Interferon-α Is the Primary Plasma Type-I IFN in HIV-1 Infection and Correlates with Immune Activation and Disease Markers. PLoS ONE, 2013. 8(2): p. e56527. 98. Takeuchi, O., S. Sato, T. Horiuchi, K. Hoshino, K. Takeda, Z. Dong, R.L. Modlin, and S. Akira, Cutting edge: role of Toll-like receptor 1 in mediating immune response to microbial lipoproteins. J Immunol, 2002. 169(1): p. 10-4.

127

99. Takeuchi, O., K. Hoshino, T. Kawai, H. Sanjo, H. Takada, T. Ogawa, K. Takeda, and S. Akira, Differential roles of TLR2 and TLR4 in recognition of gram-negative and gram- positive bacterial cell wall components. Immunity, 1999. 11(4): p. 443-51. 100. Alexopoulou, L., A.C. Holt, R. Medzhitov, and R.A. Flavell, Recognition of double- stranded RNA and activation of NF-kappaB by Toll-like receptor 3. Nature, 2001. 413(6857): p. 732-8. 101. Poltorak, A., X. He, I. Smirnova, M.Y. Liu, C. Van Huffel, X. Du, D. Birdwell, E. Alejos, M. Silva, C. Galanos, M. Freudenberg, P. Ricciardi-Castagnoli, B. Layton, and B. Beutler, Defective LPS signaling in C3H/HeJ and C57BL/10ScCr mice: mutations in Tlr4 gene. Science, 1998. 282(5396): p. 2085-8. 102. Hayashi, F., K.D. Smith, A. Ozinsky, T.R. Hawn, E.C. Yi, D.R. Goodlett, J.K. Eng, S. Akira, D.M. Underhill, and A. Aderem, The innate immune response to bacterial flagellin is mediated by Toll-like receptor 5. Nature, 2001. 410(6832): p. 1099-103. 103. Takeuchi, O., T. Kawai, P.F. Muhlradt, M. Morr, J.D. Radolf, A. Zychlinsky, K. Takeda, and S. Akira, Discrimination of bacterial lipoproteins by Toll-like receptor 6. Int Immunol, 2001. 13(7): p. 933-40. 104. Heil, F., H. Hemmi, H. Hochrein, F. Ampenberger, C. Kirschning, S. Akira, G. Lipford, H. Wagner, and S. Bauer, Species-specific recognition of single-stranded RNA via toll-like receptor 7 and 8. Science, 2004. 303(5663): p. 1526-9. 105. Hemmi, H., O. Takeuchi, T. Kawai, T. Kaisho, S. Sato, H. Sanjo, M. Matsumoto, K. Hoshino, H. Wagner, K. Takeda, and S. Akira, A Toll-like receptor recognizes bacterial DNA. Nature, 2000. 408(6813): p. 740-5. 106. Jensen, S.B. and S.R. Paludan, Sensing the hybrid--a novel PAMP for TLR9. Embo j, 2014. 33(6): p. 529-30. 107. Schaefer, L., A. Babelova, E. Kiss, H.J. Hausser, M. Baliova, M. Krzyzankova, G. Marsche, M.F. Young, D. Mihalik, M. Gotte, E. Malle, R.M. Schaefer, and H.J. Grone, The matrix component biglycan is proinflammatory and signals through Toll-like receptors 4 and 2 in macrophages. J Clin Invest, 2005. 115(8): p. 2223-33. 108. Vabulas, R.M., S. Braedel, N. Hilf, H. Singh-Jasuja, S. Herter, P. Ahmad-Nejad, C.J. Kirschning, C. Da Costa, H.G. Rammensee, H. Wagner, and H. Schild, The endoplasmic reticulum-resident heat shock protein Gp96 activates dendritic cells via the Toll-like receptor 2/4 pathway. J Biol Chem, 2002. 277(23): p. 20847-53. 109. Curtin, J.F., N. Liu, M. Candolfi, W. Xiong, H. Assi, K. Yagiz, M.R. Edwards, K.S. Michelsen, K.M. Kroeger, C. Liu, A.K. Muhammad, M.C. Clark, M. Arditi, B. Comin-Anduix, A. Ribas, P.R. Lowenstein, and M.G. Castro, HMGB1 mediates endogenous TLR2 activation and brain tumor regression. PLoS Med, 2009. 6(1): p. e10. 110. Vabulas, R.M., P. Ahmad-Nejad, S. Ghose, C.J. Kirschning, R.D. Issels, and H. Wagner, HSP70 as endogenous stimulus of the Toll/interleukin-1 receptor signal pathway. J Biol Chem, 2002. 277(17): p. 15107-12. 111. Asea, A., M. Rehli, E. Kabingu, J.A. Boch, O. Bare, P.E. Auron, M.A. Stevenson, and S.K. Calderwood, Novel signal transduction pathway utilized by extracellular HSP70: role of toll-like receptor (TLR) 2 and TLR4. J Biol Chem, 2002. 277(17): p. 15028-34. 112. Zhang, P., C.J. Cox, K.M. Alvarez, and M.W. Cunningham, Cutting edge: cardiac myosin activates innate immune responses through TLRs. J Immunol, 2009. 183(1): p. 27-31. 113. Jiang, D., J. Liang, J. Fan, S. Yu, S. Chen, Y. Luo, G.D. Prestwich, M.M. Mascarenhas, H.G. Garg, D.A. Quinn, R.J. Homer, D.R. Goldstein, R. Bucala, P.J. Lee, R. Medzhitov, and P.W. Noble, Regulation of lung injury and repair by Toll-like receptors and hyaluronan. Nat Med, 2005. 11(11): p. 1173-9.

128

114. Scheibner, K.A., M.A. Lutz, S. Boodoo, M.J. Fenton, J.D. Powell, and M.R. Horton, Hyaluronan fragments act as an endogenous danger signal by engaging TLR2. J Immunol, 2006. 177(2): p. 1272-81. 115. Tesar, B.M., D. Jiang, J. Liang, S.M. Palmer, P.W. Noble, and D.R. Goldstein, The Role of Hyaluronan Degradation Products as Innate Alloimmune Agonists. American Journal of Transplantation, 2006. 6(11): p. 2622-2635. 116. Liu-Bryan, R., P. Scott, A. Sydlaske, D.M. Rose, and R. Terkeltaub, Innate immunity conferred by toll-like receptors 2 and 4 and myeloid differentiation factor 88 expression is pivotal to monosodium urate monohydrate crystal–induced inflammation. Arthritis & Rheumatism, 2005. 52(9): p. 2936-2946. 117. Liu-Bryan, R., K. Pritzker, G.S. Firestein, and R. Terkeltaub, TLR2 signaling in chondrocytes drives calcium pyrophosphate dihydrate and monosodium urate crystal- induced nitric oxide generation. J Immunol, 2005. 174(8): p. 5016-23. 118. Vabulas, R.M., P. Ahmad-Nejad, C. da Costa, T. Miethke, C.J. Kirschning, H. Hacker, and H. Wagner, Endocytosed HSP60s use toll-like receptor 2 (TLR2) and TLR4 to activate the toll/interleukin-1 receptor signaling pathway in innate immune cells. J Biol Chem, 2001. 276(33): p. 31332-9. 119. Kariko, K., H. Ni, J. Capodici, M. Lamphier, and D. Weissman, mRNA is an endogenous ligand for Toll-like receptor 3. J Biol Chem, 2004. 279(13): p. 12542-50. 120. Cavassani, K.A., M. Ishii, H. Wen, M.A. Schaller, P.M. Lincoln, N.W. Lukacs, C.M. Hogaboam, and S.L. Kunkel, TLR3 is an endogenous sensor of tissue necrosis during acute inflammatory events. J Exp Med, 2008. 205(11): p. 2609-21. 121. Chiron, D., I. Bekeredjian-Ding, C. Pellat-Deceunynck, R. Bataille, and G. Jego, Toll-like receptors: lessons to learn from normal and malignant human B cells. Blood, 2008. 112(6): p. 2205-13. 122. Tang, A.H., G.J. Brunn, M. Cascalho, and J.L. Platt, Pivotal advance: endogenous pathway to SIRS, sepsis, and related conditions. J Leukoc Biol, 2007. 82(2): p. 282-5. 123. Apetoh, L., F. Ghiringhelli, A. Tesniere, M. Obeid, C. Ortiz, A. Criollo, G. Mignot, M.C. Maiuri, E. Ullrich, P. Saulnier, H. Yang, S. Amigorena, B. Ryffel, F.J. Barrat, P. Saftig, F. Levi, R. Lidereau, C. Nogues, J.P. Mira, A. Chompret, V. Joulin, F. Clavel-Chapelon, J. Bourhis, F. Andre, S. Delaloge, T. Tursz, G. Kroemer, and L. Zitvogel, Toll-like receptor 4- dependent contribution of the immune system to anticancer chemotherapy and radiotherapy. Nat Med, 2007. 13(9): p. 1050-9. 124. Ohashi, K., V. Burkart, S. Flohe, and H. Kolb, Cutting edge: heat shock protein 60 is a putative endogenous ligand of the toll-like receptor-4 complex. J Immunol, 2000. 164(2): p. 558-61. 125. Lehnardt, S., E. Schott, T. Trimbuch, D. Laubisch, C. Krueger, G. Wulczyn, R. Nitsch, and J.R. Weber, A vicious cycle involving release of heat shock protein 60 from injured cells and activation of toll-like receptor 4 mediates neurodegeneration in the CNS. J Neurosci, 2008. 28(10): p. 2320-31. 126. Dybdahl, B., A. Wahba, E. Lien, T.H. Flo, A. Waage, N. Qureshi, O.F. Sellevold, T. Espevik, and A. Sundan, Inflammatory response after open heart surgery: release of heat-shock protein 70 and signaling through toll-like receptor-4. Circulation, 2002. 105(6): p. 685- 90. 127. Chase, M.A., D.S. Wheeler, K.M. Lierl, V.S. Hughes, H.R. Wong, and K. Page, Hsp72 induces inflammation and regulates cytokine production in airway epithelium through a TLR4- and NF-kappaB-dependent mechanism. J Immunol, 2007. 179(9): p. 6318-24.

129

128. Wheeler, D.S., M.A. Chase, A.P. Senft, S.E. Poynter, H.R. Wong, and K. Page, Extracellular Hsp72, an endogenous DAMP, is released by virally infected airway epithelial cells and activates neutrophils via Toll-like receptor (TLR)-4. Respir Res, 2009. 10: p. 31. 129. Okamura, Y., M. Watari, E.S. Jerud, D.W. Young, S.T. Ishizaka, J. Rose, J.C. Chow, and J.F. Strauss, 3rd, The extra domain A of fibronectin activates Toll-like receptor 4. J Biol Chem, 2001. 276(13): p. 10229-33. 130. Johnson, G.B., G.J. Brunn, Y. Kodaira, and J.L. Platt, Receptor-mediated monitoring of tissue well-being via detection of soluble heparan sulfate by Toll-like receptor 4. J Immunol, 2002. 168(10): p. 5233-9. 131. Lau, C.M., C. Broughton, A.S. Tabor, S. Akira, R.A. Flavell, M.J. Mamula, S.R. Christensen, M.J. Shlomchik, G.A. Viglianti, I.R. Rifkin, and A. Marshak-Rothstein, RNA-associated autoantigens activate B cells by combined B cell antigen receptor/Toll-like receptor 7 engagement. J Exp Med, 2005. 202(9): p. 1171-7. 132. Kelly, K.M., H.Y. Zhuang, D.C. Nacionales, P.O. Scumpia, R. Lyons, J. Akaogi, P. Lee, B. Williams, M. Yamamoto, S. Akira, M. Satoh, and W.H. Reeves, "Endogenous adjuvant" activity of the RNA components of lupus autoantigens Sm/RNP and Ro 60. Arthritis and Rheumatism, 2006. 54(5): p. 1557-1567. 133. Sioud, M., Innate sensing of self and non-self RNAs by Toll-like receptors. Trends Mol Med, 2006. 12(4): p. 167-76. 134. Leadbetter, E.A., I.R. Rifkin, A.M. Hohlbaum, B.C. Beaudette, M.J. Shlomchik, and A. Marshak-Rothstein, Chromatin-IgG complexes activate B cells by dual engagement of IgM and Toll-like receptors. Nature, 2002. 416(6881): p. 603-7. 135. Viglianti, G.A., C.M. Lau, T.M. Hanley, B.A. Miko, M.J. Shlomchik, and A. Marshak- Rothstein, Activation of autoreactive B cells by CpG dsDNA. Immunity, 2003. 19(6): p. 837-47. 136. Tian, J., A.M. Avalos, S.Y. Mao, B. Chen, K. Senthil, H. Wu, P. Parroche, S. Drabic, D. Golenbock, C. Sirois, J. Hua, L.L. An, L. Audoly, G. La Rosa, A. Bierhaus, P. Naworth, A. Marshak-Rothstein, M.K. Crow, K.A. Fitzgerald, E. Latz, P.A. Kiener, and A.J. Coyle, Toll- like receptor 9-dependent activation by DNA-containing immune complexes is mediated by HMGB1 and RAGE. Nat Immunol, 2007. 8(5): p. 487-96. 137. Marius, T., S. Anders, and N. Piotr, High Mobility Group Box Protein-1 in HIV-1 Infection: Connecting Microbial Translocation, Cell Death and Immune Activation. Current HIV Research, 2011. 9(1): p. 6-10. 138. Cassetta, L., O. Fortunato, L. Adduce, C. Rizzi, J. Hering, P. Rovere-Querini, M.E. Bianchi, M. Alfano, and G. Poli, Extracellular high mobility group box-1 inhibits R5 and X4 HIV-1 strains replication in mononuclear phagocytes without induction of chemokines and cytokines. AIDS, 2009. 23(5): p. 567-577. 139. Nowak, P., B. Barqasho, and A. Sönnerborg, Elevated plasma levels of high mobility group box protein 1 in patients with HIV-1 infection. AIDS, 2007. 21(7): p. 869-871. 140. Cario, E. and D.K. Podolsky, Differential alteration in intestinal epithelial cell expression of toll-like receptor 3 (TLR3) and TLR4 in inflammatory bowel disease. Infect Immun, 2000. 68(12): p. 7010-7. 141. Wonderlich, E.R. and S.M. Barratt-Boyes, SIV infection of rhesus macaques differentially impacts mononuclear phagocyte responses to virus-derived TLR agonists. J Med Primatol, 2013. 42(5): p. 247-53. 142. Cho, I. and M.J. Blaser, The Human Microbiome: at the interface of health and disease. Nature reviews. Genetics, 2012. 13(4): p. 260-270.

130

143. Round, J.L. and S.K. Mazmanian, Inducible Foxp3+ regulatory T-cell development by a commensal bacterium of the intestinal microbiota. Proc Natl Acad Sci U S A, 2010. 107(27): p. 12204-9. 144. Frank, D.N., A.L. St. Amand, R.A. Feldman, E.C. Boedeker, N. Harpaz, and N.R. Pace, Molecular-phylogenetic characterization of microbial community imbalances in human inflammatory bowel diseases. Proceedings of the National Academy of Sciences of the United States of America, 2007. 104(34): p. 13780-13785. 145. Toh, Z.Q., A. Anzela, M.L.K. Tang, and P.V. Licciardi, Probiotic Therapy as a Novel Approach for Allergic Disease. Frontiers in Pharmacology, 2012. 3: p. 171. 146. Bisgaard, H., N. Li, K. Bonnelykke, B.L. Chawes, T. Skov, G. Paludan-Muller, J. Stokholm, B. Smith, and K.A. Krogfelt, Reduced diversity of the intestinal microbiota during infancy is associated with increased risk of allergic disease at school age. J Allergy Clin Immunol, 2011. 128(3): p. 646-52.e1-5. 147. Johansson, M.A., Y.M. Sjogren, J.O. Persson, C. Nilsson, and E. Sverremark-Ekstrom, Early colonization with a group of Lactobacilli decreases the risk for allergy at five years of age despite allergic heredity. PLoS One, 2011. 6(8): p. e23031. 148. van Nimwegen, F.A., J. Penders, E.E. Stobberingh, D.S. Postma, G.H. Koppelman, M. Kerkhof, N.E. Reijmerink, E. Dompeling, P.A. van den Brandt, I. Ferreira, M. Mommers, and C. Thijs, Mode and place of delivery, gastrointestinal microbiota, and their influence on asthma and atopy. J Allergy Clin Immunol, 2011. 128(5): p. 948-55.e1-3. 149. Vujkovic-Cvijin, I., R.M. Dunham, S. Iwai, M.C. Maher, R.G. Albright, M.J. Broadhurst, R.D. Hernandez, M.M. Lederman, Y. Huang, M. Somsouk, S.G. Deeks, P.W. Hunt, S.V. Lynch, and J.M. McCune, Dysbiosis of the Gut Microbiota Is Associated with HIV Disease Progression and Tryptophan Catabolism. Science Translational Medicine, 2013. 5(193): p. 193ra91-193ra91. 150. Li, S.X., A.J.S. Armstrong, C.P. Neff, M. Shaffer, C.A. Lozupone, and B.E. Palmer, Complexities of Gut Microbiome Dysbiosis in the Context of HIV Infection and Antiretroviral Therapy. Clinical Pharmacology & Therapeutics, 2016. 99(6): p. 600-611. 151. McCausland, M.R., S.M. Juchnowski, D.A. Zidar, D.R. Kuritzkes, A. Andrade, S.F. Sieg, M.M. Lederman, and N.T. Funderburg, Altered Monocyte Phenotype in HIV-1 Infection Tends to Normalize with Integrase-Inhibitor-Based Antiretroviral Therapy. PLoS ONE, 2015. 10(10): p. e0139474. 152. Wilson, E.M., A. Singh, K.H. Hullsiek, D. Gibson, W.K. Henry, K. Lichtenstein, N.F. Onen, E. Kojic, P. Patel, J.T. Brooks, I. Sereti, and J.V. Baker, Monocyte-activation phenotypes are associated with biomarkers of inflammation and coagulation in chronic HIV infection. J Infect Dis, 2014. 210(9): p. 1396-406. 153. Zawada, A.M., K.S. Rogacev, B. Rotter, P. Winter, R.R. Marell, D. Fliser, and G.H. Heine, SuperSAGE evidence for CD14++CD16+ monocytes as a third monocyte subset. Blood, 2011. 118(12): p. e50-61. 154. Ziegler-Heitbrock, L., P. Ancuta, S. Crowe, M. Dalod, V. Grau, D.N. Hart, P.J. Leenen, Y.J. Liu, G. MacPherson, G.J. Randolph, J. Scherberich, J. Schmitz, K. Shortman, S. Sozzani, H. Strobl, M. Zembala, J.M. Austyn, and M.B. Lutz, Nomenclature of monocytes and dendritic cells in blood. Blood, 2010. 116(16): p. e74-80. 155. Weinberg, A., L.Y. Song, C. Wilkening, A. Sevin, B. Blais, R. Louzao, D. Stein, P. Defechereux, D. Durand, E. Riedel, N. Raftery, R. Jesser, B. Brown, M.F. Keller, R. Dickover, E. McFarland, and T. Fenton, Optimization and limitations of use of cryopreserved peripheral blood mononuclear cells for functional and phenotypic T-cell characterization. Clin Vaccine Immunol, 2009. 16(8): p. 1176-86.

131

156. Meijerink, M., D. Ulluwishewa, R.C. Anderson, and J.M. Wells, Cryopreservation of monocytes or differentiated immature DCs leads to an altered cytokine response to TLR agonists and microbial stimulation. Journal of Immunological Methods, 2011. 373(1–2): p. 136-142. 157. Silveira, G.F., P.F. Wowk, A.M.B. Machado, C.N.D. dos Santos, and J. Bordignon, Immature Dendritic Cells Generated from Cryopreserved Human Monocytes Show Impaired Ability to Respond to LPS and to Induce Allogeneic Lymphocyte Proliferation. PLoS ONE, 2013. 8(7): p. e71291. 158. Jiang, W., M.M. Lederman, J.R. Salkowitz, B. Rodriguez, C.V. Harding, and S.F. Sieg, Impaired Monocyte Maturation in Response to CpG Oligodeoxynucleotide Is Related to Viral RNA Levels in Human Immunodeficiency Virus Disease and Is at Least Partially Mediated by Deficiencies in Alpha/Beta Interferon Responsiveness and Production. Journal of Virology, 2005. 79(7): p. 4109-4119. 159. Cheadle, W.G., The human leukocyte antigens and their relationship to infection. Am J Surg, 1993. 165(2A Suppl): p. 75s-81s. 160. McDevitt, H.O., Regulation of the immune response by the major histocompatibility system. N Engl J Med, 1980. 303(26): p. 1514-7. 161. McLeish, K.R., S.R. Wellhausen, and W.L. Dean, Biochemical basis of HLA-DR and CR3 modulation on human peripheral blood monocytes by lipopolysaccharide. Cell Immunol, 1987. 108(1): p. 242-8. 162. Engel, P., J. Gribben, G. Freeman, L. Zhou, Y. Nozawa, M. Abe, L. Nadler, H. Wakasa, and T. Tedder, The B7-2 (B70) costimulatory molecule expressed by monocytes and activated B lymphocytes is the CD86 differentiation antigen. Vol. 84. 1994. 1402-1407. 163. Freeman, G.J., V.A. Boussiotis, A. Anumanthan, G.M. Bernstein, X.Y. Ke, P.D. Rennert, G.S. Gray, J.G. Gribben, and L.M. Nadler, B7-1 and B7-2 do not deliver identical costimulatory signals, since B7-2 but not B7-1 preferentially costimulates the initial production of IL-4. Immunity, 1995. 2(5): p. 523-32. 164. Pearson, L.L., B.E. Castle, and M.R. Kehry, CD40-mediated signaling in monocytic cells: up-regulation of tumor necrosis factor receptor-associated factor mRNAs and activation of mitogen-activated protein kinase signaling pathways. Int Immunol, 2001. 13(3): p. 273-83. 165. Yang, J., L. Zhang, C. Yu, X.-F. Yang, and H. Wang, Monocyte and macrophage differentiation: circulation inflammatory monocyte as biomarker for inflammatory diseases. Biomarker Research, 2014. 2: p. 1-1. 166. Mina-Osorio, P., B. Winnicka, C. O'Conor, C.L. Grant, L.K. Vogel, D. Rodriguez-Pinto, K.V. Holmes, E. Ortega, and L.H. Shapiro, CD13 is a novel mediator of monocytic/endothelial cell adhesion. J Leukoc Biol, 2008. 84(2): p. 448-59. 167. Fitch, K.V., S. Srinivasa, S. Abbara, T.H. Burdo, K.C. Williams, P. Eneh, J. Lo, and S.K. Grinspoon, Noncalcified coronary atherosclerotic plaque and immune activation in HIV- infected women. J Infect Dis, 2013. 208(11): p. 1737-46. 168. Williams, D.W., K. Anastos, S. Morgello, and J.W. Berman, JAM-A and ALCAM are therapeutic targets to inhibit diapedesis across the BBB of CD14+CD16+ monocytes in HIV-infected individuals. J Leukoc Biol, 2015. 97(2): p. 401-12. 169. Petrov, V., N. Funderburg, A. Weinberg, and S. Sieg, Human β defensin-3 induces chemokines from monocytes and macrophages: diminished activity in cells from HIV- infected persons. Immunology, 2013. 140(4): p. 413-420. 170. Gascon, R.L., A.B. Narvaez, R. Zhang, J.O. Kahn, F.M. Hecht, B.G. Herndier, and M.S. McGrath, Increased HLA-DR expression on peripheral blood monocytes in subsets of

132

subjects with primary HIV infection is associated with elevated CD4 T-cell apoptosis and CD4 T-cell depletion. J Acquir Immune Defic Syndr, 2002. 30(2): p. 146-53. 171. Espinosa, E., C.E. Ormsby, G. Reyes-Teran, R. Asaad, S.F. Sieg, and M.M. Lederman, Dissociation of CD154 and cytokine expression patterns in CD38+ CD4+ memory T cells in chronic HIV-1 infection. J Acquir Immune Defic Syndr, 2010. 55(4): p. 439-45. 172. Olvera-Garcia, G., E. Espinosa, S.F. Sieg, and M.M. Lederman, Cytomegalovirus-specific responses of CD38+ memory T cells are skewed towards IFN-gamma and dissociated from CD154 in HIV-1 infection. Aids, 2014. 28(3): p. 311-6. 173. Zhang, R., C.J. Fichtenbaum, D.A. Hildeman, J.D. Lifson, and C. Chougnet, CD40 ligand dysregulation in HIV infection: HIV glycoprotein 120 inhibits signaling cascades upstream of CD40 ligand transcription. J Immunol, 2004. 172(4): p. 2678-86. 174. Hearps, A.C., A. Maisa, W.J. Cheng, T.A. Angelovich, G.F. Lichtfuss, C.S. Palmer, A.L. Landay, A. Jaworowski, and S.M. Crowe, HIV infection induces age-related changes to monocytes and innate immune activation in young men that persist despite combination antiretroviral therapy. Aids, 2012. 26(7): p. 843-53. 175. Martin, G.E., M. Gouillou, A.C. Hearps, T.A. Angelovich, A.C. Cheng, F. Lynch, W.J. Cheng, G. Paukovics, C.S. Palmer, R.M. Novak, A. Jaworowski, A.L. Landay, and S.M. Crowe, Age- associated changes in monocyte and innate immune activation markers occur more rapidly in HIV infected women. PLoS One, 2013. 8(1): p. e55279. 176. Baker, J.V., K.H. Hullsiek, A. Singh, E. Wilson, W.K. Henry, K. Lichtenstein, N. Onen, E. Kojic, P. Patel, J.T. Brooks, H.N. Hodis, M. Budoff, I. Sereti, and C.D.C.S.U.N.S.I. for the, Immunologic Predictors of Coronary Artery Calcium Progression in a Contemporary HIV Cohort. AIDS (London, England), 2014. 28(6): p. 831-840. 177. Burdo, T.H., M.R. Lentz, P. Autissier, A. Krishnan, E. Halpern, S. Letendre, E.S. Rosenberg, R.J. Ellis, and K.C. Williams, Soluble CD163 made by monocyte/macrophages is a novel marker of HIV activity in early and chronic infection prior to and after anti-retroviral therapy. J Infect Dis, 2011. 204(1): p. 154-63. 178. Krishnan, S., E.M. Wilson, V. Sheikh, A. Rupert, D. Mendoza, J. Yang, R. Lempicki, S.A. Migueles, and I. Sereti, Evidence for innate immune system activation in HIV type 1- infected elite controllers. J Infect Dis, 2014. 209(6): p. 931-9. 179. Hattab, S., A. Guihot, M. Guiguet, S. Fourati, G. Carcelain, F. Caby, A.G. Marcelin, B. Autran, D. Costagliola, and C. Katlama, Comparative impact of antiretroviral drugs on markers of inflammation and immune activation during the first two years of effective therapy for HIV-1 infection: an observational study. BMC Infect Dis, 2014. 14: p. 122. 180. Hileman, C.O., B. Kinley, V. Scharen-Guivel, K. Melbourne, J. Szwarcberg, J. Robinson, M.M. Lederman, and G.A. McComsey, Differential Reduction in Monocyte Activation and Vascular Inflammation With Integrase Inhibitor-Based Initial Antiretroviral Therapy Among HIV-Infected Individuals. J Infect Dis, 2015. 212(3): p. 345-54. 181. Grinspoon, S.K., C. Grunfeld, D.P. Kotler, J.S. Currier, J.D. Lundgren, M.P. Dube, S.E. Lipshultz, P.Y. Hsue, K. Squires, M. Schambelan, P.W. Wilson, K.E. Yarasheski, C.M. Hadigan, J.H. Stein, and R.H. Eckel, State of the science conference: Initiative to decrease cardiovascular risk and increase quality of care for patients living with HIV/AIDS: executive summary. Circulation, 2008. 118(2): p. 198-210. 182. Grover, S.A., L. Coupal, N. Gilmore, and J. Mukherjee, Impact of dyslipidemia associated with Highly Active Antiretroviral Therapy (HAART) on cardiovascular risk and life expectancy. Am J Cardiol, 2005. 95(5): p. 586-91.

133

183. Currier, J.S., J.D. Lundgren, A. Carr, D. Klein, C.A. Sabin, P.E. Sax, J.T. Schouten, and M. Smieja, Epidemiological evidence for cardiovascular disease in HIV-infected patients and relationship to highly active antiretroviral therapy. Circulation, 2008. 118(2): p. e29-35. 184. Tseng, Z.H., E.A. Secemsky, D. Dowdy, E. Vittinghoff, B. Moyers, J.K. Wong, D.V. Havlir, and P.Y. Hsue, Sudden cardiac death in patients with human immunodeficiency virus infection. J Am Coll Cardiol, 2012. 59(21): p. 1891-6. 185. Andrade, A., S.L. Rosenkranz, A.R. Cillo, D. Lu, E.S. Daar, J.M. Jacobson, M. Lederman, E.P. Acosta, T. Campbell, J. Feinberg, C. Flexner, J.W. Mellors, D.R. Kuritzkes, and A.C.T.G.A.T. for the, Three Distinct Phases of HIV-1 RNA Decay in Treatment-Naive Patients Receiving Raltegravir-Based Antiretroviral Therapy: ACTG A5248. The Journal of Infectious Diseases, 2013. 208(6): p. 884-891. 186. Hanley, J.A., A. Negassa, M.D.d. Edwardes, and J.E. Forrester, Statistical Analysis of Correlated Data Using Generalized Estimating Equations: An Orientation. American Journal of Epidemiology, 2003. 157(4): p. 364-375. 187. R Development Core Team, R: A language and environment for statistical computing. 2010, R Foundation for Stastical Computing. Retrieved from http://www.R-project.org: Vienna, Austria. 188. RStudio, RStudio: Integrated development environment for R (Version 0.98.1062). 2012: Boston, MA. Retrieved September, 2014. 189. Carey, V.J., gee: Generalized Estimation Equation solver. 2012: CRAN. 190. Hunt, P.W., E. Sinclair, B. Rodriguez, C. Shive, B. Clagett, N. Funderburg, J. Robinson, Y. Huang, L. Epling, J.N. Martin, S.G. Deeks, C.L. Meinert, M.L. Van Natta, D.A. Jabs, and M.M. Lederman, Gut Epithelial Barrier Dysfunction and Innate Immune Activation Predict Mortality in Treated HIV Infection. The Journal of Infectious Diseases, 2014. 210(8): p. 1228-1238. 191. Gordon, S.N., B. Cervasi, P. Odorizzi, R. Silverman, F. Aberra, G. Ginsberg, J.D. Estes, M. Paiardini, I. Frank, and G. Silvestri, Disruption of Intestinal CD4(+) T Cell Homeostasis Is a Key Marker of Systemic CD4(+) T Cell Activation in HIV-Infected Individuals. Journal of immunology (Baltimore, Md. : 1950), 2010. 185(9): p. 5169-5179. 192. Guadalupe, M., E. Reay, S. Sankaran, T. Prindiville, J. Flamm, A. McNeil, and S. Dandekar, Severe CD4+ T-cell depletion in gut lymphoid tissue during primary human immunodeficiency virus type 1 infection and substantial delay in restoration following highly active antiretroviral therapy. J Virol, 2003. 77(21): p. 11708-17. 193. Verhoeven, D., S. Sankaran, M. Silvey, and S. Dandekar, Antiviral therapy during primary simian immunodeficiency virus infection fails to prevent acute loss of CD4+ T cells in gut mucosa but enhances their rapid restoration through central memory T cells. J Virol, 2008. 82(8): p. 4016-27. 194. George, M.D., E. Reay, S. Sankaran, and S. Dandekar, Early antiretroviral therapy for simian immunodeficiency virus infection leads to mucosal CD4+ T-cell restoration and enhanced gene expression regulating mucosal repair and regeneration. J Virol, 2005. 79(5): p. 2709-19. 195. Mehandru, S., M.A. Poles, K. Tenner-Racz, P. Jean-Pierre, V. Manuelli, P. Lopez, A. Shet, A. Low, H. Mohri, D. Boden, P. Racz, and M. Markowitz, Lack of mucosal immune reconstitution during prolonged treatment of acute and early HIV-1 infection. PLoS Med, 2006. 3(12): p. e484. 196. Guadalupe, M., S. Sankaran, M.D. George, E. Reay, D. Verhoeven, B.L. Shacklett, J. Flamm, J. Wegelin, T. Prindiville, and S. Dandekar, Viral suppression and immune restoration in the gastrointestinal mucosa of human immunodeficiency virus type 1-

134

infected patients initiating therapy during primary or chronic infection. J Virol, 2006. 80(16): p. 8236-47. 197. Tincati, C., M. Biasin, A. Bandera, M. Violin, G. Marchetti, L. Piacentini, G.L. Vago, C. Balotta, M. Moroni, F. Franzetti, M. Clerici, and A. Gori, Early initiation of highly active antiretroviral therapy fails to reverse immunovirological abnormalities in gut-associated lymphoid tissue induced by acute HIV infection. Antivir Ther, 2009. 14(3): p. 321-30. 198. Klatt, N.R. and J.M. Brenchley, Th17 cell dynamics in HIV infection. Curr Opin HIV AIDS, 2010. 5(2): p. 135-40. 199. Brenchley, J.M., D.A. Price, T.W. Schacker, T.E. Asher, G. Silvestri, S. Rao, Z. Kazzaz, E. Bornstein, O. Lambotte, D. Altmann, B.R. Blazar, B. Rodriguez, L. Teixeira-Johnson, A. Landay, J.N. Martin, F.M. Hecht, L.J. Picker, M.M. Lederman, S.G. Deeks, and D.C. Douek, Microbial translocation is a cause of systemic immune activation in chronic HIV infection. Nat Med, 2006. 12(12): p. 1365-71. 200. Baroncelli, S., C.M. Galluzzo, M.F. Pirillo, M.G. Mancini, L.E. Weimer, M. Andreotti, R. Amici, S. Vella, M. Giuliano, and L. Palmisano, Microbial translocation is associated with residual viral replication in HAART-treated HIV+ subjects with <50copies/ml HIV-1 RNA. J Clin Virol, 2009. 46(4): p. 367-70. 201. Ancuta, P., A. Kamat, K.J. Kunstman, E.Y. Kim, P. Autissier, A. Wurcel, T. Zaman, D. Stone, M. Mefford, S. Morgello, E.J. Singer, S.M. Wolinsky, and D. Gabuzda, Microbial translocation is associated with increased monocyte activation and dementia in AIDS patients. PLoS One, 2008. 3(6): p. e2516. 202. Canipe, A., T. Chidumayo, M. Blevins, M. Bestawros, J. Bala, P. Kelly, S. Filteau, B.E. Shepherd, D.C. Heimburger, and J.R. Koethe, A 12 week longitudinal study of microbial translocation and systemic inflammation in undernourished HIV-infected Zambians initiating antiretroviral therapy. BMC Infectious Diseases, 2014. 14: p. 521. 203. Marchetti, G., G.M. Bellistri, E. Borghi, C. Tincati, S. Ferramosca, M. La Francesca, G. Morace, A. Gori, and A.D. Monforte, Microbial translocation is associated with sustained failure in CD4+ T-cell reconstitution in HIV-infected patients on long-term highly active antiretroviral therapy. Aids, 2008. 22(15): p. 2035-8. 204. Valdez, Y., E.M. Brown, and B.B. Finlay, Influence of the microbiota on vaccine effectiveness. Trends in Immunology. 35(11): p. 526-537. 205. Kawai, T. and S. Akira, The role of pattern-recognition receptors in innate immunity: update on Toll-like receptors. Nat Immunol, 2010. 11(5): p. 373-384. 206. Glavan, T.W., C.A. Gaulke, C. Santos Rocha, S. Sankaran-Walters, L.A. Hirao, M. Raffatellu, G. Jiang, A.J. Baumler, L.R. Goulart, and S. Dandekar, Gut immune dysfunction through impaired innate pattern recognition receptor expression and gut microbiota dysbiosis in chronic SIV infection. Mucosal Immunol, 2016. 9(3): p. 677-688. 207. Lederman, M.M., L. Calabrese, N.T. Funderburg, B. Clagett, K. Medvik, H. Bonilla, B. Gripshover, R.A. Salata, A. Taege, M. Lisgaris, G.A. McComsey, E. Kirchner, J. Baum, C. Shive, R. Asaad, R.C. Kalayjian, S.F. Sieg, and B. Rodriguez, Immunologic Failure Despite Suppressive Antiretroviral Therapy Is Related to Activation and Turnover of Memory CD4 Cells. Journal of Infectious Diseases, 2011. 204(8): p. 1217-1226. 208. Lederman, M.M., N.T. Funderburg, R.P. Sekaly, N.R. Klatt, and P.W. Hunt, Residual Immune Dysregulation Syndrome in Treated HIV infection. Advances in immunology, 2013. 119: p. 51-83. 209. Baker, J.V., G. Peng, J. Rapkin, D.I. Abrams, M.J. Silverberg, R.D. MacArthur, W.P. Cavert, W.K. Henry, J.D. Neaton, and f.t.T.B.C.P.f.C.R.o. AIDS, CD4+ count and risk of non-AIDS diseases following initial treatment for HIV infection. AIDS, 2008. 22(7): p. 841-848.

135

210. Metsch, L.R., M. Pereyra, D.W. Purcell, C.A. Latkin, R. Malow, C.A. Gómez, M.H. Latka, and f.t.I.S. Team, Correlates of Lending Needles/Syringes Among HIV-Seropositive Injection Drug Users. JAIDS Journal of Acquired Immune Deficiency Syndromes, 2007. 46: p. S72-S79. 211. Gutierrez, F., S. Padilla, M. Masia, J.A. Iribarren, S. Moreno, P. Viciana, J. Hernandez- Quero, R. Aleman, F. Vidal, M. Salavert, J.R. Blanco, M. Leal, F. Dronda, S. Perez Hoyos, and J. del Amo, Patients' characteristics and clinical implications of suboptimal CD4 T-cell gains after 1 year of successful antiretroviral therapy. Curr HIV Res, 2008. 6(2): p. 100-7. 212. Février, M., K. Dorgham, and A. Rebollo, CD4(+) T Cell Depletion in Human Immunodeficiency Virus (HIV) Infection: Role of Apoptosis. Viruses, 2011. 3(5): p. 586- 612. 213. Shive, C.L., B. Clagett, M.R. McCausland, J.C. Mudd, N.T. Funderburg, M.L. Freeman, S.A. Younes, B.M. Ferrari, B. Rodriguez, G.A. McComsey, L.H. Calabrese, S.F. Sieg, and M.M. Lederman, Inflammation Perturbs the IL-7 Axis, Promoting Senescence and Exhaustion that Broadly Characterize Immune Failure in Treated HIV Infection. J Acquir Immune Defic Syndr, 2016. 71(5): p. 483-92. 214. Fletcher, P.S., J. Elliott, J.C. Grivel, L. Margolis, P. Anton, I. McGowan, and R.J. Shattock, Ex vivo culture of human colorectal tissue for the evaluation of candidate microbicides. Aids, 2006. 20(9): p. 1237-45. 215. Pan, D., A. Das, D. Liu, R.S. Veazey, and B. Pahar, Isolation and Characterization of Intestinal Epithelial Cells from Normal and SIV-Infected Rhesus Macaques. PLoS ONE, 2012. 7(1): p. e30247. 216. Elmaagacli, A.H., N. Steckel, M. Ditschkowski, Y. Hegerfeldt, H. Ottinger, R. Trenschel, M. Koldehoff, and D.W. Beelen, Toll-like receptor 9, NOD2 and IL23R gene polymorphisms influenced outcome in AML patients transplanted from HLA-identical sibling donors. Bone Marrow Transplant, 2011. 46(5): p. 702-8. 217. Agarwal, S.K., M. Wu, C.K. Livingston, D.H. Parks, M.D. Mayes, F.C. Arnett, and F.K. Tan, Toll-like receptor 3 upregulation by type I interferon in healthy and scleroderma dermal fibroblasts. Arthritis Res Ther, 2011. 13(1): p. R3. 218. Poles, M.A., W.J. Boscardin, J. Elliott, P. Taing, M.M. Fuerst, I. McGowan, S. Brown, and P.A. Anton, Lack of decay of HIV-1 in gut-associated lymphoid tissue reservoirs in maximally suppressed individuals. J Acquir Immune Defic Syndr, 2006. 43(1): p. 65-8. 219. Morel, J.G., M.C. Bokossa, and N.K. Neerchal, Small Sample Correction for the Variance of GEE Estimators. Biometrical Journal, 2003. 45(4): p. 395-409. 220. Kuller, L.H., R. Tracy, W. Belloso, S. De Wit, F. Drummond, H.C. Lane, B. Ledergerber, J. Lundgren, J. Neuhaus, D. Nixon, N.I. Paton, and J.D. Neaton, Inflammatory and coagulation biomarkers and mortality in patients with HIV infection. PLoS Med, 2008. 5(10): p. e203. 221. Wang, P.F., H. Fang, J. Chen, S. Lin, Y. Liu, X.Y. Xiong, Y.C. Wang, R.P. Xiong, F.L. Lv, J. Wang, and Q.W. Yang, Polyinosinic-polycytidylic acid has therapeutic effects against cerebral ischemia/reperfusion injury through the downregulation of TLR4 signaling via TLR3. J Immunol, 2014. 192(10): p. 4783-94. 222. Randall, K.J., J. Turton, and J.R. Foster, Explant culture of gastrointestinal tissue: a review of methods and applications. Cell Biol Toxicol, 2011. 27(4): p. 267-84.

136