Complement activation-‐ good or evil in HIV-‐1 infection?

(1)

Linköping University Medical Dissertations No. 1281

Complement activation-‐ good or evil in

HIV-‐1 infection?

Interaction of Free and Complement Opsonized HIV-‐1 with

Monocyte Derived Dendritic Cells and Immune Cells in the

Cervical Mucosa

Veronica Tjomsland

Department of Clinical and Experimental Medicine Linköping University, Sweden

(2)

Department of Clinical and Experimental Medicine Linköping University

SE-581 85 Linköping

Cover: The Human immunodeficiency virus

The pictures in this thesis are illustrated by Rada Ellegård. Published articles have been reprinted with permission from respective copyright holder.

Printed by LiU-Tryck, Linköping, Sweden, 2011 ISBN: 978-91-7393-010-9

(3)

”They don't actually see the real world, where 95% of the people with HIV are not treated and are dying. And even though we have some blue sky now in our country, the sky could become cloudy again very soon”

Luc Montagnier

“The world needs people who dare to think differently, you don’t change anything by walking in other peoples footsteps”

Veronica Tjomsland

Dedicated to my husband and children for their unending love and

support

(4)

Linköping 2011

Supervisor

Marie Larsson, Associate Professor Division of Molecular Virology Department of Clinical and Experimental Medicine Linköping University, Sweden

Faculty opponent

Barbara L. Shacklett, Associate professor Department of Medical Microbiology and Immunology

University of California, Davis, USA

Co-‐supervisors

Committee Board

Jorma Hinkula, Professor

Kristina Broliden, Professor

Division of Molecular Virology Unit of Infectious Diseases Department of Clinical and Department of Medicine

Experimental Medicine Karolinska Institute, Sweden

Linköping University, Sweden

Karl-‐Eric Magnusson, Professor Maria Jenmalm, Associate professor

Division of Molecular Virology AIR/Clinical Immunology

Department of Clinical and Department of Clinical and

Experimental Medicine experimental Medicine

Linköping University, Sweden Linköping University, Sweden

Sven Hammarström, Professor Division of Cell Biology

Department of Clinical and Experimental Medicine Linköping University, Sweden

(5)

PREFACE

This thesis describes the results of my research carried out during my PhD study at the University of Linköping. The thesis gives you first a general introduction to the world of HIV-‐1, the complement system, dendritic cells (DCs), and antigen presentation. This is followed by a presentation of the papers. Not much is known about the MHC class I and II antigen presentation pathways used by immature and mature DCs to present antigens from whole HIV-‐1 particles and the first project focused on this topic. In the second project we studied the initial interactions of free and opsonized HIV-‐1 with DCs with the focus on receptor families involved in the viral binding. Since our results had shown that opsonized HIV-‐1 interacted with DCs in a unique way we continued in the third project to study the receptors and pathways used by DCs to process and present antigens derived from both free and complement opsonized HIV-‐1. In addition, this project also studied the effects these viral sources had on the antigen presentation machinery. In the final project we used the knowledge acquired from our in vitro experiments with free and complement opsonized HIV-‐1 and applied it on an ex vivo study. The HIV-‐1 interactions and infection of immune cells located in cervical mucosa were studied using an explant model and we examined if infection could be prevented by targeting different receptors expressed by immune cells and mucosa. Finally, I want to thank my supervisor Marie Larsson for making this thesis possible.

Veronica Tjomsland November 2011

(6)

ABSTRACT

Worldwide, the heterosexual route is the most common mode of sexual transmission of HIV-‐1 and women are particularly susceptible to this infection. After penetration of the mucosal epithelium HIV-‐1 interacts with potential target cells, i.e. dendritic cells (DCs) and CD4+ T cells. The complement system, a key component of the innate immune system, is immediately activated by HIV-‐1 in vivo. However, HIV-‐1 can resist complement mediated lysis and become coated with complement fragments and this opsonization influences the viral interaction with immune cells. The DCs are the most potent antigen presenting cell. This cell effectively links the innate recognition of viruses to the generation of an adaptive immune response. However, HIV-‐1 exploits the function of the DCs to facilitate viral spread and infection. HIV-‐1 interacts with a range of receptors expressed by the DCs including C-‐type lectins, integrins and complement receptors (CRs). The uptake of virions by DCs leads to their activation and migration to the lymph nodes. At this site DCs present HIV-‐1 derived antigen on MHC class I and II molecules and trigger an HIV-‐1 specific T cell response. The interplay between the virus and the DCs is complex and the initial receptor binding may affect antigen uptake, infection, and antigen presentation.

The fundamental questions of this thesis are the following: How is free and opsonized HIV-‐1 internalized, processed, and presented on MHC class I and II molecules by DCs and how do free and opsonized HIV-‐1 particles interact with immune cells in the cervical mucosa?

Our results indicate that opsonization of HIV-‐1 plays a critical role in the interaction with immune cells. Complement opsonization of HIV-‐1 (C-‐HIV) significantly enhanced the internalization by the DCs compared to free HIV (F-‐HIV). Both C-‐HIV and F-‐HIV interacted with the CD4 receptor, C-‐type lectins and integrins. In addition, opsonization of HIV-‐1 favored an MHC class I presentation by DCs compared to F-‐HIV. However, the endocytic receptors macrophage mannose receptor, β7 integrin, and CR3 guided the antigens to different compartments with distinct properties and efficiencies for degradation and MHC class I and II presentation of viral antigens. MHC class I presentation of F-‐HIV and C-‐HIV was dependent of viral fusion in a CD4/coreceptor dependent manner. Moreover, MHC class II presentation of antigens derived from HIV-‐1 required endocytosis and proteolysis in acidified compartments. HIV-‐1 infection of cervical mucosa immune cells and tissue was assessed in a cervical tissue explant model.

(7)

C-‐HIV significantly enhanced infection of DCs compared to F-‐HIV, whereas C-‐HIV decreased the infection of CD4+ T cells. Blocking the viral use of integrins in the cervical tissue explants significantly decreased the HIV-‐1 infection of both emigrating DCs and CD4+ T cells and the establishment of founder populations in these tissues. This thesis work has brought forward new facts that can be used to facilitate the development of an effective vaccine or microbicide.

(8)

LIST OF PAPERS INCLUDED IN THE THESIS

I

Pathways utilized by dendritic cells for binding, uptake, processing and presentation of antigens derived from HIV-‐1.

Sabado RL, Babcock E, Kavanagh DG, Tjomsland V, Walker BD, Lifson JD, Bhardwaj N, Larsson M.

Eur J Immunol. 2007 Jul; 37(7):1752-‐63.

II Complement Opsonization of HIV-‐1 Enhances the Uptake by Dendritic Cells and Involves the Endocytic Lectin and Integrin Receptor Families.

Tjomsland V, Ellegård R, Che K, Hinkula J, Lifson JD, Larsson M. PLoS One. 2011; 6(8):e23542. Epub 2011 Aug 11.

III Complement opsonization of HIV-‐1 results in a different intracellular processing efficiency and pattern leading to an enhanced MHC I class presentation by dendritic cells.

Tjomsland V, Ellegård R, Burgener A, Hinkula J, Lifson JD, Larsson M. Manuscript

IV Blocking of integrins significantly inhibits HIV-‐1 infection of human cervical mucosa immune cells and development of founder populations.

Tjomsland V, Ellegård R, Kjölhede P, Hinkula J, Lifson JD, Larsson M.

Manuscript

(9)

ABBREVIATIONS

Ab Antibody

ABC Avidin biotin complex

AIDS Acquired immunodeficiency syndrome

APC Antigen presenting cell

APOBEC3G Apoplipoprotein B mRNA-‐editing, enzyme-‐catalytic, polypeptide-‐like 3G

ART Antiretroviral therapy

AT-‐2 Aldrithiol-‐2

AZT Azidothymidine

CCR5 CC chemokine receptor 5 CXCR4 CXC chemokine receptor 4 C-‐HIV Complement opsonized HIV-‐1

C-‐IgG-‐HIV Complement opsonized HIV-‐1 in combination with immune complex

DAPI 4’,6’-‐diamidino-‐2-‐phenylindole

DCs Dendritic cells

DC-‐SIGN Dendritic cell-‐specific ICAM-‐3-‐grabbing non-‐integrin dsDNA Double stranded DNA

EDTA Ethylene-‐diamine-‐tetra-‐acetic acid

ER Endoplasmic reticulum

F-‐HIV Free-‐HIV

fH factor H

FITC Fluorescein isothiocyanate

gp41 HIV-‐1 glycoprotein 41 gp120 HIV-‐1 glycoprotein 120

HAART Highly active anti-‐retroviral therapy HIV-‐1 Human immunodeficiency virus-‐1 ICAM Intercellular adhesion molecule IgG-‐HIV IgG opsonized HIV-‐1

IDCs Immature dendritic cells

IFN Interferon

IFRs Interferon regulatory factors

IL Interleukin

(10)

LFA-‐1 lymphocyte function-‐associated antigen 1

LCs Langerhans cells

LTR Long terminal repeats

MAC Membrane attack complex

MHC Major histocompatibility complex

MDC Mature dendritic cells

MDDC Monocyte derived dendritic cells

MMR Macrophage mannose receptor

Nef Negative factor

PAMPS Pathogen associated molecular patterns PBMC Peripheral blood mononuclear cells

PBS Phosphate-‐buffered saline

PDCs Plasmacytoid dendritic cells

PE Phycoerythrin

PFA Para formaldehyde

PHS Pool human serum

PIC Pre-‐integration complex

PR HIV-‐1 protease

RNA Ribonucleic acid

RT Reverse transcriptase

SIV Simian immunodeficiency virus

SAMHD-‐1 SAM domain and HD domain containing protein 1 ssRNA Single stranded RNA

TAR Transactivation response element

TLR Toll like receptor

TRIM Tripartite motif-‐ containing protein Vif Viral infectivity factor

(11)

9.1 Paper I………..40 9.1.1 Background………..40 9.1.2 Principal findings………...41 9.1.3 Discussion/ Conclusion………41 9.2 Paper II……….42 9.2.1 Background………..42 9.2.2 Principal findings………...42 9.2.3 Discussion/ Conclusion………43 9.3 Paper III……….…….……….43 9.3.1 Background………..43 9.3.2 Principal findings………...………..44 9.3.3 Discussion/ Conclusion………45 9.4 Paper IV………...45 9.4.1 Background………..45 9.4.2 Principal findings……….………46 9.4.3 Discussion/ Conclusion………....47

(13)

10. CONCLUSIONS AND FUTURE DIRECTIONS ……….48 10.1 Complement activation-‐ good or evil in HIV-‐1 infection?...48 10.2 Future Challenges………...49 11. POPULÄRVETENSKAPLIG SAMMANFATTNING………...50 12. ACKNOWLEDGEMENTS………...53 13. REFERENCES………...57

14. REPRINTS OF ORIGINAL PAPERS AND MANUSCRIPT 14.1 Paper I 14.2 Paper II 14.3 Paper III 14.4 Paper IV

(14)

Introduction

1

1. INTRODUCTION

In 1981 a new syndrome appeared in the United States. The patients had an acquired immune deficiency with a marked depletion of the CD4+ T cell count. Two years later HIV-‐1 was identified by Luc Montagnier and Françoise Barré-‐Sinoussi as the causative agent of acquired immune deficiency syndrome (AIDS) (1). Currently more than 30 million people are infected with HIV-‐1 and an estimated 2.6 million are newly infected every year in the world and millions have died from AIDS (2). This makes this infection one of the worst epidemics of this century. Moreover, the HIV/AIDS epidemic is accompanied by many tragic and difficult social challenges like discrimination, stigma, denial and a growing number of children who have lost parents to AIDS (3). In 2005, thirteen million children younger than 15 years of age had already lost one or both of their parents to AIDS (4).

The natural history of HIV-‐1 infection involves a long period of clinical latency with a gradual loss of CD4+ T cells before the infection progresses to AIDS. AIDS are defined by a CD4+ T cell count below 400cells/µl blood and without treatment this will lead to opportunistic infections, the appearance of rare malignancies and ultimately death. The most prevalent route of sexual transmission is by heterosexual intercourse. Women are particularly at high risk to acquire HIV-‐1 infection due to social and biological factors and therefore bear the greatest burden (5). However, much is still unknown about the biological factors in the female genital tract contributing to resistance against HIV-‐1 infection.

HIV-‐1 is a retrovirus that belongs to the genus Lentiviridae. Lentivirus is characterized by a long incubation period, however it is now clear from studies in Macaques that local events important to establish an systemic infection take place quickly in the early stages of simian immunodeficiency virus (SIV) infection (6). Following entry of HIV-‐1 through the mucosa epithelium founder populations are established in the submucosa and the dendritic cells (DCs) will transfer the virus to CD4+ T cells in the mucosal stroma and lymph nodes (7). In the lymph nodes the DCs will efficiently present HIV-‐1 antigens to T cells via MHC class I and II restricted pathways and mount a specific immune response against HIV-‐1. MHC class I and II presentation and activation of CD4+ and CD8+ T cells are important events that will determine the outcome of the infection. Most individuals control the viremia poorly in the absence of antiretroviral therapy. Today the only

(15)

Introduction

2

effective approach against HIV-‐1 infection is antiretroviral therapy but many limitations exist such as toxicity, costs, distribution in developing countries, and resistance. Unfortunately strategies to prevent HIV-‐1 transmission have had limited success over the past three decades (6). Vaccines or microbicides have not proven efficient and have even in some cases enhanced HIV-‐1 infection (8, 9). An effective HIV-‐1 vaccine will probably require activation of CD4+ and CD8+ T cell responses directed against crucial HIV-‐1 epitopes (10).

There exists an urgent need today for an HIV-‐1 vaccine or microbicides to prevent HIV-‐1 transmission and constrain the ongoing pandemic.

(16)

HIV-‐1

3

2. HIV-‐1

HIV-‐1 belongs to the genus Lentivirus and is further divided into the family Retroviridae. HIV-‐1 has a spherical morphology with a diameter of 100-‐120 nm and is surrounded by a lipid bilayer, an envelope. This envelope is acquired from the host cell during the process of viral budding and contains approximately 72 spikes of the viral receptor gp120 bound together with the transmembrane spanning glycoprotein gp41(11). The envelope may also express many other receptors like ICAM-‐1 and HLA class I and II molecules, acquired from the infected cell during the budding process (12). The nucleocapsid, which has a conical shape, contains a viral protease (PR), reverse transcriptase (RT), integrase (IN), and two copies of a single stranded RNA (ssRNA) molecule (13) (Fig. 1).

Figure 1. Structure of the HIV-‐1 particle.

The HIV-1 is composed of two copies of positive ssRNA encoding the 9 viral genes. The viral genome is enclosed by a conical nucleocapsid composed of 2000 copies of the viral protein gag p24 (14). In the nucleocapsid are the pol encoded enzymes, integrase (IN), reverse transcriptase (RT), and protease (PR), all needed by the virus for infection. Surrounding the nucleocapsid is a matrix composed of the p17 gag protein and the matrix is in turn surrounded by a viral envelope. The HIV protein Env protrudes from the viral envelope and is composed of gp120 and gp41 proteins. gp41 is an anchor protein, attaching gp120 to the viral envelope and HIV-1 uses this glycoprotein complex to attach and fuse with target cells (15).

(17)

HIV-‐1

4

2.1 HIV-‐1 life cycle

The infection begins with the binding of HIV-‐1 to the target cells by the viral receptor gp120 to a 58kDa glycoprotein, the CD4 receptor. The CD4 receptor is expressed on T cells, monocytes, macrophages, DCs, eosinophils, and microglia cells (16). Upon binding to CD4, gp120 undergoes a conformational change and is able to bind the coreceptor CC-‐ chemokine receptor 5 (CCR5) or CXC-‐chemokine receptor 4 (CXCR4). The binding of gp120 to both CD4 and coreceptor leads to further conformational changes that allow gp41 to penetrate the cell membrane (17, 18). Following membrane fusion the virus capsid is uncoated in the cytoplasm of the host cell and the viral RNA is released. The capsid undergoes a progressive destabilization during its transport towards the nucleus to ensure productive infection as uncoating should not occur too early or too late in the process (19) (Fig. 2). The viral RNA is transcribed into a double stranded DNA (dsDNA) by RT, but this transcription is negatively affected by the presence of the host cell protein APOBEC3G. However, the HIV-‐1 protein Vif counteracts the cell’s antiviral effect by down regulation of APOBEC3G and prevents incorporation of this protein into progeny virions (20). The pre-‐integration complex navigates through the pores of the nucleolus. In the nucleus the viral DNA can be found in three different forms, linear, a circular form of 2-‐ long terminal repeats (LTR), or a circle of 1-‐LTR (21). None of the circular forms lead to the production of infectious virus but the viral genes Tat and Nef can be transcribed from them (22). The linear dsDNA of the pre-‐integration complex is integrated in the host cell genome and this is mediated by IN (23). The integration might lead to a latent infection, i.e. nonproductive (24), but if cellular proteins bind to the viral LTR, transcription of Nef, Tat, and Rev can occur and these HIV-‐1 proteins are normally expressed very shortly after infection. When sufficient amount of Tat protein has been produced, Tat proteins start to control further transcription of HIV-‐1 genes by binding to the TAR site (Transactivation response element). In the early phase of replication only multiply spliced mRNA are produced, but when sufficient amounts of Rev proteins are produced, non-‐spliced or single spliced mRNA can be generated as well (25) (Fig. 2). The core of the maturing HIV particle is formed by the gene products pol and gag. The gene products coded by the env gene form the glycoprotein 120/41 spikes in the viral envelope (Fig. 3). The proteins Gag and Pol are also derived from a big precursor polyprotein. The formation of a new viral particle occurs in several steps; two copies of ssRNA associate together with the RT enzymes, while core proteins assemble around them forming the viral capsid. The immature particles migrate toward the cell surface and assemble, the

(18)

HIV-‐1

5

large precursor polyproteins are then cleaved resulting in the viral budding from the cell plasma membrane and thereby the acquiring of a lipid envelope. The budding of HIV-‐1 virions is believed to occur through areas in the host cell membrane rich in cholesterol (26). During the budding it is essential that the expression of CD4 receptors are downregulated in the host cell membrane to avoid the interaction with gp120 (27). Nef (negative factor) is important for replication and the pathogenesis of HIV. Many functions have been described for Nef, including the down regulation of CD4, coreceptors, MHC class I and II molecules by inducing endocytosis of these molecules, consequently affecting antigen presentation and recognition by the HIV-‐1 specific immune response (28-‐30). Later in the replication cycle the env gene product trap CD4 in the endoplasmic reticulum (ER) (31).

(19)

HIV-‐1

6

Figure 2. Life cycle of HIV-‐1

The life cycle of HIV-‐1 begins when the virus binds to CD4 and coreceptor on a target cell. When HIV-‐1 have bound to the infection receptors the envelope complex undergoes a structural change resulting in fusion with the cell membrane and the virus inject its contents into the cytosol (17, 18). The viral genetic material is transcribed from ssRNA into dsDNA by the use of the HIV-‐1 enzyme RT. The viral dsDNA is then integrated into the host genome by the help of IN. From the integrated DNA the cell produces RNA and viral proteins (32). The HIV-‐1 protease cleaves the newly synthesized proteins, enabling them to join the RNA and assemble by the cell membrane. Finally, new viral particles bud off from the cell membrane and can infect new target cells (32).

(20)

HIV-‐1

7

Figure 3. Organization of the HIV-‐1 genome.

HIV-‐1 has nine genes coding for 15 viral proteins. The structural genes gag, pol and env are the same in all retroviruses and these genes contain information necessary to make new viral particles. The other six genes, tat, rev, nef, vif, vpr, and vpu, are regulatory genes for proteins that control the ability of HIV-‐1 to infect and replicate in a host cell. Long terminal repeats (LTR) are regions controlling the production of new virions and is triggered by HIV-‐ 1 proteins or host cell proteins (16).

2.2 Relevant aspects of HIV-‐1 innate and adaptive immunity

It is well established that HIV-‐1 infection results in strong activation of the immune system (6). The innate immunity conducts the first line of defense followed by the adaptive immunity. The innate and adaptive responses are closely interlinked and a strong initial innate response is likely to lead to potent adaptive immunity. Several components of the innate defense are activated by HIV-‐1, e.g. the complement cascade, type I IFNs, and inflammatory cytokines (33). HIV-‐1 is transmitted through the mucosa and targets specific immune cells, i.e. CD4+CCR5+ T cells and DCs (34, 35). The adaptive immune response is incapable to mount a defense sufficient to clear the infection and the onset is too late to stop the massive destruction of the CD4+CCR5+ T cells that occurs within two weeks after onset of infection (34, 35).

The first line of defense does not require previous antigen encounter and may if strong enough limit replication of the microbe giving the adaptive immunity enough time to mount a potent and efficient immune response (36). The innate immune response can be divided in to three groups; cellular, intracellular, and extracellular (37). The cellular

(21)

HIV-‐1

8

components of the innate immunity are for instance Langerhans cells (LCs), DCs, monocytes, γδ T cells, and natural killer cells (NK cells) (38). To begin with these cells have innate effector functions but later they may play a part in the induction of adaptive immunity (36). For instance, DCs produce factors important for the initial innate defense but they also prime the naïve T cells in the lymph nodes and activate the adaptive immune response (39). In the initial immune response two families of transcription factors play a major role in the innate anti-‐viral defense, the NFkB family and the interferon regulatory factors (IRFs). The IRFs play a central role in the induction and regulation of proteins, type I IFNs, and chemokines mediating antiviral responses. The production of type I IFNs has an important role in the innate antiviral response, they attract immune cells to the site of infection, increase the function of macrophages, T cells, NK cells, and B cells and induce maturation of plasmacytoid DCs (PDCs) (40-‐42). IRF-‐3 plays a central role in the induction of antiviral response. The viral activation of this factor leads to production of IFNβ, which stimulates the transcription of IRF-‐7 that further augments the synthesis of IFNβ. The antiviral effect of IFN is mediated by the induction of a large amount of cellular genes, i.e. IFN-‐stimulatory genes (ISG), ISG15 was one of the first ISG identified and has been shown to have antiviral effects (43).

Toll like receptors (TLRs) is a family of receptors important in the innate immune response. TLRs detect microbes and induce antimicrobial host defense responses by recognizing conserved regions on pathogens, denoted as pathogen-‐associated molecular patterns (PAMPS) (44). TLRs are involved in the destruction of pathogens, coordinating the immune response, and regulating the functionality of DCs (42). The presence of ssRNA activates TLR7/8 while dsRNA activates TLR3 (45). HIV-‐1 is recognized mainly through TLR7 on PDCs and TLR8 on blood myeloid DCs (MDCs) and monocyte derived DCs (MDDCs). PDCs are an important component of the innate immune defense and a main producer of type I IFNs (46). Another part of the innate immune defense is the restriction factors including, tripartite motif-‐containing protein (TRIM), 5α, 1, 19 and 22, tetherin, SAM domain and HD domain-‐containing protein 1 (SAMHD-‐1), and apoplipoprotein B mRNA-‐editing, enzyme-‐catalytic, polypeptide-‐like 3G (APOBEC3G) (47-‐50). APOBEC3G is found in T cells, monocytes, macrophages, and DCs. The incorporation of APOBEC3G into the HIV-‐1 genomes leads to extensive mutations in the viral DNA, rendering them nonfunctional and inhibiting viral replication (51). However, HIV-‐1 counteracts this defense mechanism by the production of the viral protein Vif. Vif decrease the synthesis of APOBEC3G and enhances the 26S proteasome mediated

(22)

HIV-‐1

9

degradation making APOBEC3G unavailable for budding virions (47). Innate factors that exert their effects in an extracellular manner are produced as a part of the innate defense and include large amounts of type I interferons (IFNs), i.e. IFN-‐α and IFN-‐β. Type I IFNs are produced by mainly by PDCs but also by MDCs, and macrophages during the early phase of a viral infection and they promote TH1 cell development by activating the transcription factor STAT4. In addition IFNs also prevent activated T cells from undergoing apoptosis (52, 53). The CC chemokines CCL5 (RANTES), CCL3 (MIP-‐1α) and CCL4 (MIP-‐1β) are secreted by activated DCs, macrophages, NK cells, and γδ T cells and these factors can block the CCR5 coreceptors and prevent HIV-‐1 infection (54). However, some cellular proteins downregulate the antiviral response, among them are the cellular DNAse TREX1, which degrades unintegrated proviral DNA and thereby helping the virus to be undetected by TLR9 or cytoplasmic DNA sensors (55). Defensins are extracellular innate peptides that can contribute to protection against HIV-‐1 infection in the mucosa. Another essential component of the innate immune response is the complement system (56) and this part of the innate immunity is described and discussed in depth below.

Figure 4. Approaches by HIV-‐I to circumvent the cell mediated antiviral responses.

Complement factors, type I IFNs and the intrinsic cellular proteins TRIM, tetherin, APOBEC3G, and SAMHD-‐ 1 contribute to the inhibition of viral replication inside the host cells. On the other hand, some of host cell proteins, e.g. TREX1, contribute to the down regulation of the antiviral response. In addition, the virus has genes encoding for proteins that can impair the antiviral defense.

APOBEC3G SAMHD1 TRIM 5α, 1, 19, 22 Tetherin ISG15 Type I interferons Vif Vpx Vpu, Nef IRF-3 Vpr, Vif TREX1 Complement

(23)

The complement system

10

3. THE COMPLEMENT SYSTEM

3.1 Overview

The complement system is composed of more than 30 cell surface and serum components (57) and around 90% of them are produced by hepatocytes but complement proteins can also be produced by monocytes, macrophages, endothelial, and epithelial cells (58, 59). The human complement system is the first line of the defense against pathogens by inducing complement mediated lysis and tagging targets for phagocytosis. However, lately it has been shown that complement also plays an important role in induction and maintenance of the adaptive immune responses, i.e. antigen presentation, and T cell activation (60). In addition, the complement system is involved in the enhancement of the antibody induced responses via complement receptors (CRs) and Fc receptors (FcRs) (60).

The complement system can be activated in three different ways dependent on the trigger. All pathways; the classical pathway, the lectin pathway, and the alternative pathway converge at the activation and triggering of complement component 3 (C3). The classical pathway is sometimes also referred to as the antibody dependent classical pathway and is activated by the binding of complement component 1q (C1q), a subcomponent of the C1 complex, to IgG/IgM clusters bound to cell walls of pathogens or apoptotic cells, or by the pentraxin family members. Alternatively, direct interaction by C1q with some types of pathogens can also trigger this pathway. The C1 complex attracts C2 and C4 and generates the C2C4 convertase, which is able to cleave the C3 protein and results in C3a and C3b (61, 62).

The lectin binding pathway or the mannose binding pathway (MBP) is initiated by the recognition of characteristic carbohydrate patterns expressed on the surface of microorganisms. Binding occurs via the mannose-‐binding lectin (MBL) protein family and ficolins and activates MBP associated serine proteases (MASPs) (62). The different MASPs are similar to C1r and C1q, therefore the following cascade resembles the classical pathway and will converge at the activation and cleavage of C3 (63, 64).

The alternative pathway of the complement cascade represents a process that needs no exogenous trigger. By spontaneous C3 hydrolysis, new binding sites are exposed and factor B binds to hydrolyzed C3 and is cleaved by factor D and results in formation of C3

(24)

11

convertase, which is cleaved into C3a and C3b. C3b interacts with factor B and this factor in turn is cleaved by factor D, creating a full C3 convertase (C3bBb) that is stabilized by the binding of properdin (65, 66). Subsequently, more and more C3b is drawn to this multiprotein complex attached to the surface of the microbe leading to an effective opsonization (60).

After opsonization of the pathogen, the terminal complement pathway is triggered resulting in formation of a terminal membrane attack complex (MAC). The MAC is a pore like structure created in the membrane of the pathogen leading to its lysis and destruction (60). The complement system is strictly controlled to protect the host from complement mediated damage. This is mediated by soluble and cell bound complement regulators.

Among the regulators is C1 esterase inhibitor (C1-‐INH). This inhibitor have an effect on several proteases in the classical and lectin binding pathway. The abundantly expressed factor H (fH) acts on the C3 convertase or serve as a cofactor for degradation of C3b, but can also prevent self attack. The C3 convertase is also regulated by factor I (fI), factor H like protein, and C4 binding protein. In addition, most cells in the body express receptors that function as convertase regulators, e.g. complement receptor 1 (CR1) and CD55, but they also express receptors working as cofactors for fI, e.g. CR1 and CD49. The plasma membrane bound protein protectin (CD59), a complement regulatory protein, inhibits the formation of the MAC complex (67, 68). The inactivation and degradation of C3b leads to the production of inactivated C3 fragments iC3b, iC3dg, and iC3d and these complement fragments do not have any further function in the lytic cascade but are ligands to complement receptors.

Complement receptor 1 (CR1: CD35) is a cell membrane receptor expressed on leucocytes, erythrocytes, and podocytes. CR1 binds C3b and C4b and plays an important role in the regulation of the complement cascade but CR1 also binds immune complexes coated with C3b and remove them from circulation by transporting them to the liver or spleen (69). Complement receptor 2 (CR2: CD21) is predominantly expressed on B cells, T cells, and follicular dendritic cells (FDCs) and interacts mainly with C3dg and C3d. Complement receptor 3 (CR3: MAC-‐1) and complement receptor 4 (CR4: pl 150,95) are both members of the β2-‐integrin family. CR3 consists of two chains, an 165 kDa αM-‐chain (CD11b) and an 95 kDa β-‐chain (CD18) and is expressed primarily on myeloid cells but also on NK cells , microglia, osteoblasts, and some epithelial cells (70). CR4 has the same β2-‐chain but instead this chain is linked to a 150 kDa αX-‐chain (CD11c) and the CR4 is

(25)

12

basically found on the same cell types as CR3. CR3 has been shown to be involved in many coordinating and adhesion functions in the immune system, e.g. adhesion and migration of leucocytes during homing, and the binding and phagocytosis of opsonized particles (70-‐72). CR3 can bind to several ligands with high affinity including iC3b, ICAM-‐ 1, fibrinogen, and clotting factor X and with low affinity to C3b and C3bg (70, 73). The binding site for iC3b, C3b, and C3bg are located on the α-‐chain (CD11b) and the binding is Ca2+_{dependent
(73).
Several
studies
have
reported
that
cells
expressing
CR3
and
CR4}

have an enhanced HIV-‐1 replication. The CR3 and CR4 expressed by DCs are involved in trans infection of HIV-‐1 (74). In addition, an increasing amount of evidence indicates that CR3 and CR4 also play a role in antigen presentation and CD8+ T cell activation (75).

3.2 Complement opsonization of HIV-‐1

Several viruses including HIV-‐1, Vaccinia virus, Herpes simplex virus (HSV), and Epstein-‐ Barr virus have been shown to directly activate the complement system (76). HIV-‐1 is able to activate all three pathways of the complement system already in the initial phase of infection (76). The lectin pathway is activated by the binding of MBL to high mannose carbohydrates on HIV-‐1 gp120 (77) and the classical pathway is activated by the binding of viral gp41 to the A-‐chain of C1q (78). The activation occurs in the absence of antibodies. However, after seroconversion the presence of HIV-‐1 specific antibodies further enhances the activation of the classical complement pathway (79, 80). Of note, due to mechanisms developed by HIV-‐1, virions resist complement mediated lysis and the activation of the complement cascade result in deposition of inactivated C3 fragments on the viral surface, i.e. opsonization (81, 82) (Fig. 5 and 6). HIV-‐1 acquires complement lysis resistance factors during the budding from the host cell plasma membrane and these receptors are incorporated in the viral envelope. These factors that inhibit the complement cascade are the membrane cofactor protein (MCP: CD46), decay accelerating factor (DAF: CD55), and CD59 (83). In addition, HIV-‐1 can bind soluble fH, which further protects virions from destruction (64, 84). There are many other pathogens besides HIV-‐ 1 that have developed different methods to escape the complement system (81, 82, 85). However, HIV-‐1 is not only spared from lysis it also uses the deposition of complement fragments on the surface to its own advantage (86).

The interaction of HIV-‐1 with cells is mediated by the viral receptor gp120 binding to multiple receptors including CD4 and coreceptors (87). However when HIV-‐1 is covered with C3 fragments the carbohydrates expressed on gp120 may be partly or completely

(26)

13

covered by complement fragments and thereby poorly accessible for receptor binding. Experiments in macaques and in vitro T cell experiments have shown that opsonization of virions by C3 fragments masks epitopes on the viral envelope leading to reduced infection of T cells, which are CR3 negative (88-‐90). Moreover, virions also use the complement fragments to increase their infectivity by interacting with cells expressing CRs. The complement fragment iC3b is the major ligand for CR3, but this receptor also binds to other ligands like ICAM-‐1, which is an adhesion molecule acquired by the virions from the host cell plasma membrane during the process of budding. In addition the gp41 part of the HIV-‐1 envelope receptor can also interact with CR3 (91). Finally, complement opsonized HIV-‐1 have been found throughout the body, e.g. in blood, breast milk, mucosa, seminal fluid, and lymph nodes (64), and should be taken in consideration when studying HIV-‐1.

Figure 5. Free and opsonized HIV-‐1.

HIV-‐1 immediately activates the complement cascade but is protected from complement mediated lysis leading to deposition of C3 fragments on the surface of HIV-‐1 (C-‐HIV) (92). After seroconversion, HIV-‐1 can be covered with HIV-‐1 specific antibodies (IgG-‐HIV) and HIV-‐1 specific antibodies in combination with complement fragments (C-‐IgG-‐HIV) (93). Seroconversion enhances the activation of the classical pathway and increases the amount of C3 cleavage products deposited on the surface of HIV-‐1 (64, 94).

(27)

14

Figure 6. Complement activation on the viral surface of HIV-‐1.

HIV-‐1 can activate all three pathways of the complement system, classical, mannose-‐ binding-‐lectin (MBL) and alternative pathway. The initiation of the classical pathway can occur in the absence of HIV-‐ specific antibodies but they enhance the activation of the classical pathway after seroconversion (95, 96). The classical pathway is initiated by the binding of C1q to gp41 (97). However, activation by the mannose-‐binding-‐lectin (MBL) pathway is triggered by the binding of MBL to carbohydrate side chains expressed on gp120 (98).

The alternative pathway is independent of antibodies and starts by the hydrolyzation of C3 to C3(H2O). All three pathways result in the formation of C3 convertase, which cleaves C3 into C3b and C3a. However, HIV-‐ 1 escape compliment mediated lysis by MAC (C5b6789), owing to factors acquired during the budding from the host cell. These factors are incorporated in the viral envelope and include CD55, CD59, and CD46 (99). CD55 dissociates the C3 convertase and CD59 blocks the formation of the MAC complex by the polymerization of C9. CD46 interacts with factor I (fI), which cleaves C3b to inactive C3b (iC3b) and subsequently to C3c and C3d. Factor H (fH), incorporated in the viral envelope, interacts with gp120 and gp41 and this protects the virions from complement mediated lysis (86, 100-‐102). However, fH also plays role in the inactivation of C3b by working as an additional cofactor for fI (73, 103).

(28)

15

3.3 Outcomes after complement activation by HIV-‐1

A fraction of the HIV-‐1 particles trigger the terminal activation pathway and are lysed by the MAC, but a substantial amount of the virions remains opsonized and mediates their effects on the immune system by interacting with CRs and FcRs (60). The complement opsonized virions affect the immune system in many ways (60). For instance, interaction of complement opsonized HIV-‐1 with CR1 on erythrocytes might facilitate the spread of opsonized HIV-‐1 to the liver and spleen where HIV-‐1 can be transferred to target cells (104). CR2 is involved in trapping HIV-‐1 in the centers in the lymphoid organs by binding complement and immune complex opsonized HIV-‐1 to FDCs. In fact, CR2 is the main HIV-‐ 1 binding receptor on FDCs in vivo, no involvement of CR1 or CR4 (105). HIV-‐1 opsonized with complement and/or immune complex binds to the surface of the FDCs and can stay trapped there for months without infecting the FDCs (106). During this time the trapped virions are highly infectious for CD4+ T cells even in the presence of neutralizing antibodies (107).

Virions opsonized by complement fragments and immune complexes mark them for uptake by phagocytosis and destruction. Phagocytes like DCs and macrophages internalize the opsonized virus mainly via FcRs or CRs. The presence of iC3b on the viral surface leads to the interaction with CR3 and CR4 and several studies have shown a highly increased HIV-‐1 infection in cells expressing these CRs (73). For instance, DCs infected with HIV opsonized with complement and anti HIV-‐IgG had a 10-‐fold increased infection compared to cells infected with free virions (108). Of note, viral replication increased in latently infected monocytes following stimulation of CR3 (109). A twofold increase in HIV-‐1 infection was seen in an epithelial cells line when infected with seminal fluid opsonized virions compared to free virions and this enhanced infection was due to CR3 engagement (110). We have previously shown in our group that complement opsonized virions are more efficiently internalized via receptor mediated endocytosis than free viral particles (111).

Complement activation-­‐ good or evil in HIV-­‐1 infection?