Microbial Translocation and B Cell Dysfunction in Human Immunodeficiency Virus Disease

The gut mucosal barrier disrupted in HIV disease, resulting in increased systemic exposure to microbial products such as Lipo Polys Accharide (LPS). The association of enhanced microbial translocation and B cell dysfunction in HIV disease is not fully understood. High dose and short term exposure of microbial Toll-Like Receptor (TLR) agonists were used as vaccine adjuvants, however, low dose and long term exposure of TLR agonists could be harmful. The characteristics of B cell dysfunction in HIV disease included B cell, especially memory B cell depletion, enhanced levels of autoimmune antibodies and impaired vaccine or antigen responsiveness. This review discusses and explores the possibility of the effect of microbial translocation on memory B cell depletion and impaired vaccine responses in HIV infection. By determining the mechanisms of B cell depletion and perturbations in HIV disease, it may be possible to design interventions that can improve immune responses to vaccines, reduce selected opportunistic infections and perhaps slow disease progression.

Innate and adaptive immunities: There are two types of immune responses in the human system, the innate and adaptive immune responses. The innate immune system has the function to immediately and nonspecifically respond to invading pathogens, including inflammation, complement system and cellular innate responses. For example, dendritic cells produce interferon-alpha in response to virus infection and interferon-alpha has the function of killing the virus (Yoshikai, 2006). When innate immunity cannot eliminate pathogens, adaptive immunity plays a role to control infection. Adaptive immune response defends the host from infection by a specific manner. This means that certain cells of the adaptive immune system recognize and respond to specific pathogens and have specific memories, next time these memory cells can attack the same invading pathogens with a fast and strong response. Innate immunity defenses are immediate and short lasting. Adaptive immunity defenses mount often 1-2 weeks and last longer (Yoshikai, 2006).

Toll-like receptor and its ligands:
Toll-Like Receptor (TLR) families are responsible for the recognition of Pathogen-Associated Molecular Patterns (PAMPs) expressed by pathogens and distinguishable from host molecules. TLRs recognize structurally conserved molecules derived from microbes. PAMPs activation links to the innate immune system, results in the production of proinflammatory cytokines and the expression of antimicrobial genes. Activation of PAMPs also plays a key role in shaping adaptive immune response. There are more than 10 families of TLR ligands found until now; these TLR ligands include DNA of bacteria, RNA of the virus and Lipo Poly Saccharide (LPS) produced by most Gramnegative bacteria and Gram-positive bacteria. TLR ligands have the function of maintaining normal immune function, however, too much TLR ligands may result in immune activation and cause disease pathogenesis. A certain and low level of TLR ligands in human body is a consequence of living in a non sterile environment, however, the systemic heightened level of TLR ligands are believed to be a consequence of the impaired mucosal integrity, especially in the gut (Abreu, 2010;Fukata and Abreu, 2009).
TLR ligands can be found in bacteria, fungi and also viruses and provide, via TLR binding, early recognition of microbial invasion. TLR1, TLR2 and TLR6 are triggered by peptidoglycan and other microbial products, TLR3 by double-stranded RNA, TLR4 by Lipo Poly Saccharide (LPS), TLR5 by flagellin, TLR7 and TLR8 by imidazoquinolines and TLR9 by unmethylated CpG DNA (Barton and Medzhitov 2003). All TLRs, except TLR3, activate the adaptor Molecule (MyD88) -dependent pathways that culminate in the activation of nuclear factor-kB (NF-kB) transcription factors, as well as the Mitogen-Activated Protein Kinases (MAPKs) extracellular signal-regulated kinase (ERK), p38 and c-Jun N-terminal Kinase (JNK) ( Barton and Medzhitov 2003;O'Neill, 2000;2002a;2002b). These transcription factors function in concert to promote inflammatory responses (e.g., IL-6, IL-10 and TNF-a) (Lamping et al., 1998;Shen et al., 2008;Soboll et al., 2006). HIV itself is also functional as TLR ligands since RNA sequences derived from the HIV genome are capable of signaling through TLR 7 and TLR 8 (Meier et al., 2007).

Human B lymphocytes and TLRs:
Human B lymphocytes produce antibodies against invading pathogens that are key to induce a humoral immune response. Human B cells express high levels of TLR1, TLR6, TLR9 and TLR10, intermediate levels of TLR7 and low levels of TLR2 and TLR4 (Hornung et al., 2002). Most B cell studies focus on TLR9, the main TLR on human B cells, which binds to bacterial DNA (Mansson et al., 2006;Chuang et al ., 2002;Jiang et al., 2007;Martinson et al., 2007;Vollmer et al., 2004). It has been demonstrated recently that switched and IgMpositive memory B cells constitutively express Toll-Like Receptor 9 (TLR9) (Bernasconi et al., 2003) and it has been suggested that the repeated stimulation through unmetylated CpG may function to continuously and specifically restimulate B cells and maintain serologic memory in the absence of traditional protein antigens (Bernasconi et al., 2002). In the model, serum IgG antibodies constitute the specific memory (with B cell receptor signals) built up by previous experience and vaccination and IgM antibodies may represent first-line memory, indispensable against pathogens never encountered before and T cell Independent (TI) antigens. On the other hand, naïve B cells are also reported to proliferate and produce IgM (not IgG) in response to CpG ODN (TLR9 ligand) alone from our previous study and others (Jiang et al., 2007;Bekeredjian-Ding et al., 2005). Furthermore, IgG class switch DNA recombination is induced among human naïve B cells in response to CpG ODN and IL-10 (He et al., 2004). Thus, TLRs may play a role in the autoimmune diseases those are present in HIV infection. However, the role of TLRs in autoimmune diseases is not clear and is controversial (Lamphier et al., 2006;Papadimitraki et al., 2006). Importantly, TLR ligands enhance antigen-specific B cell responses through B Cell Receptor (BCR) stimulation (Poeck et al., 2004). However, It remains to be demonstrated whether TLRs are important in specific recall reactions induced by either infection or vaccination (TLR ligands as adjuvants). (Grossman et al., 2002). Although the HIV virus infects CD4+ T cell, monocytes/macrophages and dendritic cells, the exact mechanisms of loss of CD4+ T cells, key cells to achieve adaptive humoral and cellular immune responses, are still not fully understood. In HIV infection, there is increasing consensus that immune activation is central to disease pathogenesis (Giorgi et al., 1993). Recent studies suggest that increased microbial products from the damaged gut in chronic HIV infection are at least partially responsible for chronic immune activation (Brenchley et al., 2006;Jiang et al., 2009) (Titanji et al., 2005;Milito et al., 2000;2004;Guan et al., 2009). Although the lack of CD4+ T cell help may explain some of these deficiencies, there also appear to be intrinsic defects in B lymphocytes that can be demonstrated in functional assays that do not require T helper cells (Malaspina et al., 2003). The functional defects that have been described in B cells from

Chronic immune activation in HIV infection: HIV infection induces broad immune perturbation, including dysfunction and loss of CD4+ T cells and B cells
HIV-infected persons include impaired proliferation responses (to B cell antigen receptor stimulation, CD40L and CpG ODNs), reduced antibody production following vaccination, B cell hyperactivation and hypergammaglobulinemia and increased susceptibility to spontaneous apoptosis (Conge et al., 1998;Malaspina et al., 2008;Moir et al., 2004). B cell hyperactivation in HIV disease may lead to the increased susceptibility of these cells to apoptosis and may also contribute to impaired immune responsiveness.
The effects of B cell depletion and impaired HIVspecific antibodies on SIV/HIV pathogenesis and disease progression are not clear and the results are contradictory (Mao et al., 2005;Miller et al., 2007;McKay et al., 2003;Mestecky et al., 2004;Gaufin et al., 2009). However, B cells are depleted and functionally impaired in pathogenic SIV-infected rhesus macaques, but not in non-pathogenic African green monkeys (Titanji et al., 2005;Dykhuizen et al., 1998;Holznagel et al., 2002;Steger et al., 1998). Nonpathogenic SIV-infected animal models also do not demonstrate gut damage or increased systemic levels of microbial products (Brenchley et al., 2006;Pandrea et al., 2007). B cell apoptosis is rare in non-pathogenic SIV-infected monkeys in the absence of gut enteropathy, B cell apoptosis is present in pathogenic SIV-infected monkeys with microbial translocation, suggesting that B cell death may be induced by HIV infection and microbial translocation.
B cells may also be activated and functionally impaired by HIV itself. It was demonstrated that B cells from HIV-infected viremic patients carry replicationcompetent virus on their surface through CD21, a complement receptor. It was also found that virus bound to B cells could efficiently infect activated CD4 T cells and cause B cell dysfunction (Moir et al., 2001;2000). There are other mechanisms involved in HIV-associated B cell dysfunction. HIV interacts with CXCR4 (Feng et al., 2006) on the B cell surface and induces B cell apoptosis. HIV-1 nef protein can activate and stimulate B cells to differentiate (Chirmule et al., 1994;Swingler et al., 2003) as reflected by polyclonal activation (hyperimmunoglobulinmia) in HIV infection (Milito et al., 2004;Nagase et al., 2001). In contrast, a recent study by Qiao et al. (2006) shows that HIV nef protein directly inhibits B cell functional class switches. However, the mechanisms of HIV-associated B cell defects are not completely understood.
Microbial translocation may play an important role in HIV-associated B cell perturbations. Loss of memory B cells and reduced production of antigen-specific antibody is seen in the majority of chronic HIV infection even though the humoral system is subject to repeated and long-term stimulation through TLR agonists released from the gut (Brenchley et al., 2006;Titanji et al., 2006;Milito et al., 2004). This is not only due to desensitization since at the same time there is a B cell polyclonal activation as reflected by increased total IgM and IgG levels Nagase et al., 2001). Short-term exposure to TLR ligands (e.g., CpG ODNs) enhances immune responses and has adjuvant effects (Jiang et al., 2007;Malaspina et al., 2008;McCluskie and Krieg, 2006;Krug et al., 2001;Jiang et al., 2005). However, chronic systemic exposure to microbial TLR ligands in HIV disease may have deleterious effects. Nevertheless, humoral immune dysfunction is present differently in HIV disease as reflected by enhanced ex vivo B cell apoptosis with reduced antigen-specific antibody production and polyclonal activation, as compared to other diseases occurring with microbial translocation (e.g., Inflammatory bowel disease) where the autoimmune disease appears to play an important role in immunopathogenesis and gut damage (Kazemi-Shirazi et al., 2002;Roggenbuck et al., 2009). As neither CD4/B lymphopenia nor cell-mediated immune deficiency is recognized concomitants of untreated inflammatory bowel disease (Melmed et al., 2010), it would appear that the virus maintains a central role in cellular and humoral immunodeficiency in HIV infection.

The loss of memory B cells may be related to increased susceptibility of these cells to apoptosis:
Spontaneous B cell apoptosis ex vivo as measured by binding of annexin V is increased in acute and chronic HIV infection (Titanji et al., 2005;Samuelsson et al., 1997). Several cell death signaling pathways has been implicated in HIV infection, such as TNFα/TNFR, TRAIL and Fas/FasL (Lichtner et al., 2004;Gasper-Smith et al., 2008;Katsikis et al., 1997;Stylianou et al., 2002;Petrovas et al., 2005;Mueller et al., 2001Nunnari et al., 2005. Moreover, studies by Susan Moir and others indicate that enhanced CD95/Fas expression on B cells in treatment-naïve HIV+ donors is related to B cell apoptosis by exogenous Fas ligand in vitro (Moir et al., 2004). Fas is expressed at low levels on the surface of naïve B cells and enhanced levels in memory B cells (Miyawaki et al., 1992;Schattner and Friedman 1996). In contrast with Fas expression, the expression of Fas ligand is reported to be much more restricted and often requires cell activation. Monocytes or macrophages are capable of producing Fas ligand after activation by opsonizedzymosan or HIV infection in vitro (Badley et al., 1996;Brown and Savill, 1999). Importantly, in vivo treatment of anti-Fas ligand Ab (RNOK203) reduces cell death in circulating B cells from SIV-infected individuals and increases antibody responses to viral proteins (Salvato et al., 2007). Thus, a Fas/FasL-induced cell signal may be involved in B cell death in HIV infection.
Enhanced memory B cell apoptosis may result in impaired antibody responsiveness to vaccination in HIV infection.
A remaining gap in knowledge is the effect of antiretroviral therapy on microbial translocation and B cell restoration. Data from previous studies have shown that the levels of LPS and the 16s RDNA in plasma are significantly reduced after initiation of antiretroviral therapy, but never decrease to normal even among patients with restored normal CD4 counts (Brenchley et al., 2006). Consistent with this, B cell recovery was slower than CD4 T cell recovery after antiretroviral therapy and was also never restored to normal (Milito, 2004;Terpstra et al., 1989). Although the data relating to HIV-specific IgA are conflicting, it remains clear that the majority of chronically HIV-infected individuals do not mount vigorous HIV-specific IgA antibody responses either locally at mucosal sites or systemically (Mestecky et al., 2004;Broliden et al., 2001Clerici et al., 2002Devito et al., 2000a;2000b).
Although short-term administration of HAART may improve antibody responses (Melvin and Mohan, 2003), long-term administration is still unable to maintain protective levels of antibodies against vaccination antigens like measles, tetanus, influenza and pneumococcus (Titanji et al., 2006;Hart et al., 2007). It suggests that low levels of microbial translocation and HIV RNA in patient plasma after HAART may contribute to the incomplete recovery of antibody responses. The further studies should be designed to be better understood the mechanisms of memory B cell apoptosis in HIV disease. This knowledge would be valuable to improve vaccine responsiveness, decrease opportunistic infections and slow down disease progression.

ACKNOWLEDGEMENT
This study is supported by grant NIAID R01AI091526.