|Home | About | Journals | Submit | Contact Us | Français|
Antibody-mediated defense against pathogens typically requires complex interactions between antibodies and other constituents of the humoral and cellular immune systems. However, recent evidence indicates that some antibodies alone can inhibit pathogen function in the absence of complement, phagocytes, or NK cells. In this issue of the JCI, McClelland et al. have begun to elucidate the molecular bases by which antibodies alone can impact pathogen growth and metabolism. They show that mAbs specific for the polysaccharide capsule of the human pathogenic fungus Cryptococcus neoformans elicit diverse effects on fungal gene expression, lipid biosynthesis, susceptibility to amphotericin B, cellular metabolism, and protein phosphorylation. These data suggest that pathogens have the capacity to generate broad metabolic responses as a result of surface binding by pathogen-specific antibodies, effects that may hold therapeutic promise.
The word immunity derives from the Latin “immunitas” meaning exemption and has come to mean protection from disease. Immunity has been observed over the centuries during plagues in Athens and Byzantium, epidemics of bubonic plague and smallpox, as well as with snake bites and vaccinations (1). The identification of specific pathogens in the late 19th century was associated with the concept that such organisms were ultimately inhibited by depleting their environment of required nutrients, by their own metabolic by-products, or by the inhospitableness of infected tissues. Enter host defense. Initial conflicts arose between advocates of a predominantly soluble or humoral basis for immunity and those favoring a cellular basis. These disparate viewpoints were ultimately reconciled in large part when antibodies, the key mediators of humoral immunity, were shown to rely on other soluble factors, particularly complement, and cells known as phagocytes to provide protection against and mediate resolution of infection. For its part, the microbe itself often expresses a range of protective defenses. These microbial virulence factors may bind, mask, or degrade complement components; cleave adherent antibodies (e.g., IgA1 protease); or subvert the activity of antibodies by binding to their effector Fc constant regions (e.g., via staphylococcal protein A or streptococcal protein G) that otherwise direct pathogens to an Fc receptor–bearing phagocyte. The protective effects of antibodies are classically mediated through their specificity for the pathogen (facilitated via their variable regions) and the ability of their Fc constant region to act as a bridge or scaffold. Other host defense mechanisms (e.g., complement, phagocytes, and NK cells) use this foundation to induce the fatal injuries on the pathogen, on which antibody defense is dependent (Figure (Figure1A). 1A).
However, in their study in this issue of the JCI, McClelland et al. advance an intriguing conceptual paradigm that binding of specific antibodies alone can elicit a range of metabolic perturbations in the fungal pathogen Cryptococcus neoformans (2). C. neoformans, widespread in the environment, is well-controlled and rarely symptomatic in healthy individuals, in large part because of antibody-dependent mechanisms. However, the organism causes recalcitrant disease and high mortality in patients with advanced cell-mediated immunodeficiency, such as those individuals with HIV/AIDS who have very low CD4+ T cell counts and patients who have undergone solid organ transplantation (3, 4). McClelland and colleagues show that three antibodies that bind to distinct topological sites on the polysaccharide capsule of C. neoformans elicit varying effects on its gene expression (2). The effects are direct and due to the antibodies in the absence of other soluble or cellular host elements, providing evidence that pathogens can recognize and respond to antibody binding by modulating distinct microbial genetic pathways (Figure (Figure1B).1B). These findings raise the intriguing possibility that the physiology of a pathogen and its susceptibility to clearance may be manipulated by rational antibody design.
Previous studies have revealed that, independent of the presence of complement or phagocytes, antibody-pathogen interactions can disrupt microbial integrity, although the genetic mechanism(s) remained undetermined (5–14). Antibodies raised in mice against several pathogenic species of bacteria (e.g., Borrelia spp.) (5–9) and fungi (e.g., Candida, Fonsecaea, and Cryptococcus species; refs. 10–14) exhibit complement-independent microbicidal (i.e., fatal to microbes) or microbistatic (i.e., growth inhibiting) activities. IgM antibodies to surface-exposed antigens facilitated effective clearance of the Borrelia species in mice, in conjunction with direct injury to the outer bacterial membrane, but internal events were not examined. A human recombinant mAb specific for HSP90 provided broad-spectrum growth inhibition of Candida species and improved the clinical and microbiological outcome of invasive candidiasis in both a murine model (11) and human patients (12) when coadministered with amphotericin B. However, the specific biochemical and/or physiological events underlying these additive effects have not been characterized. Did the complementary effects of the HSP90-specific antibody derive from structural effects on target cells by disrupting cell wall–bound transporters and essential membrane structures? Alternatively, could antibody binding have induced signaling or disruption of signal transduction pathways, thereby interfering with critical gene expression? The work of McClelland et al. begin to address, for the first time to our knowledge, the impact of surface-bound antibodies on gene expression and suggests mechanisms that may underlie the interaction of antibodies and antifungal agents.
McClelland and colleagues undertook an in vitro systems biology analysis of the biochemical and biological effects mediated by mAb binding to the polysaccharide capsule of C. neoformans (2). An important first experiment assessed whether binding of the capsule-specific mAbs to C. neoformans cells induced alterations in gene expression, as assessed by microarray analysis of mRNA. Three C. neoformans capsule–specific mAbs were tested, one IgG mAb (18B7) and two IgM mAbs (12A1 and 13F1), along with nonspecific isotype-matched control mAbs. Relative to the control antibodies, each specific mAb elicited a different pattern of mRNA induction or repression. IgG mAb 18B7 upregulated or downregulated 43 genes, including several encoding proteins involved in fatty-acid biosynthesis. In contrast, mAb 12A1 altered expression of 62 genes, including downregulation of 8 genes encoding proteins involved in ribosome biogenesis and translation. Quantitative reverse-transcription PCR confirmed the microarray results in most but not all cases.
The antibody-altered gene expression patterns detected by McClelland and colleagues suggest interference with biochemical pathways involved in lipid metabolism and/or cell wall synthesis (2), hypotheses supported by several follow-up experiments. First, fungal growth kinetics revealed that the IgG mAb 18B7 enhanced the potency of the antifungal agent amphotericin B, which acts through association with cell membrane ergosterol. Compared with amphotericin B alone, addition of a half-saturating concentration of mAb 18B7 delayed C. neoformans growth. How a control mAb also delayed growth, albeit to a lesser extent, as well as whether the effects on growth were present in the absence of amphotericin B may require further elucidation. Second, binding of mAb 18B7 induced higher levels of lipids, including ergosterol, relative to the nonspecific mAb control. Based on these data, the authors propose that the synergy between amphotericin B and mAb 18B7, in depressing C. neoformans growth rates, resulted from mAb-induced accumulation of amphotericin B binding sites. This model is reminiscent of other cases in which antibodies binding to a polysaccharide capsule alter the surface of the pathogen and host-pathogen interactions under nonopsonizing conditions to the benefit (15) or detriment of the host (16). In the current study (2), the authors took the next step by identifying pathogen-specific gene expression. Finally, mAb-induced reductions in the steady-state levels of protein phosphorylation suggest that intracellular signalling pathways are disrupted by mAb 18B7, specifically those proposed to involve regulation of clathrin-mediated endocytosis.
Overall, the work of McClelland and colleagues (2) provides provocative evidence of what seems to be a novel process, by which antibodies alone may modulate pathogens. A promising narrative framework has been established, providing others with the opportunity to add to the story. Does mAb binding directly impinge on intracellular signaling and gene expression, for example, through specific interactions with membrane-bound signal transduction proteins? Are the observed molecular phenotypes secondary responses to mAb binding, as might be expected if mAb binding were to disrupt structural components at the fungal cell surface, although the antibodies bound at different depths? Do these mAbs act as “catalytic antibodies,” such as those described by Wentworth and colleagues (17), that generated antibody-associated oxidant-mediated injury to Escherichia coli in a process proposed to be effected by H2O2, an ozone-like compound, and •OH hydroxyl radicals (18). In the current study, the source of substrate for generating such products is not clear.
Can the processes revealed in this in vitro study by McClelland and colleagues (2) translate into clinically relevant in vivo effects? Although not an element of normal host microbiota, environmental exposure to C. neoformans appears to be common. Results of serologic data, skin test responsiveness, and lymphocyte proliferation assays suggest high rates of asymptomatic but immunogenic infections over time in healthy individuals. Mechanisms of defense may involve direct interactions between the organism and CD4+ and CD8+ T cells, but much of the protection in healthy individuals likely derives from successful clearance of transient C. neoformans infection by specific antibody in conjunction with complement, phagocytes, and NK cells. However, each of these supportive immune components may be insufficient in patients with advanced cell-mediated immunodeficiency and during the most morbid and frequently fatal manifestation of cryptococcal disease, meningitis.
Indeed, Park, et al. estimate that up to a million cases of meningitis due to C. neoformans occur annually, primarily in patients with AIDS, among whom more that 600,000 die within the first three months (3). In this context, deaths from C. neoformans appear to exceed those from tuberculosis. The profile of the cerebrospinal fluid in most cases reveals limited inflammation, characterized by a low number of white cells (either lymphocytes or phagocytes) and low levels of protein, and complement levels that are typically quite low. Treatment with amphotericin B is only moderately effective (≤50% survival in Africa), and therapy with oral antifungal agents is most often continued for life. Similarly, among solid organ transplant patients, mortality with cryptococcal meningitis ranges from 20% to 50% (4). In these settings, passive administration of capsule-specific mAbs that impair the organism, without requiring effective contributions from complement, phagocytes, or NK cells in the cerebrospinal fluid (i.e., antibodies with the properties of those identified by McClelland and colleagues; ref. 2), early in the course of therapy, may provide a useful and much needed adjuvant to current therapy. That activity of the capsule itself limits the production of specific antibody by B cells further enhances the potential therapeutic contributions of passive immunization.
The current study suggests that C. neoformans has the capacity to generate a broad metabolic, if not specifically cytotoxic, response to binding by pathogen-specific antibodies. The responses were not stereotypic, suggesting that the effects were not due to agglutination or nutritional deficits. Moreover, the authors did not observe substantial changes in genes encoding proteins directly associated with pathogenicity, such as those regulating the capsule or melanin production, although those controlling phospholipase production were affected. Of potential clinical relevance, growth of the fungus in the presence of the antifungal agent amphotericin B, which associates with ergosterol in the lipid membrane, was further impaired in vitro in the presence of one specific antibody. These data are presented by the authors in the context of dysregulation of membrane lipid synthesis. Thus, McClelland and colleagues (2) have taken us on a journey beyond surface phenotype to the central workings of the cell. Building on the genomic information of C. neoformans (19), they have intriguingly revealed that surface binding can modulate fungal metabolism in diverse ways. They indicate that, as observed by others with a range of bacteria, antibodies alone can have direct effects on microbes. Antibodies acting alone provide one more recognized mechanism of host defense, actions that may be most effective clinically in selected environments, such as the cerebrospinal fluid, where other host defense factors are limiting. Nevertheless, effective immunologic control of infection is best achieved when all relevant components are cooperative, even if individual constituents, such as specific antibodies, can stand alone.
This work supported by the Mucosal and Vaccine Research Program Colorado, University of Colorado School of Medicine, the Veterans Affairs Research Service, and NIH grant A1-077069.
Conflict of interest: Edward N. Janoff’s laboratory has received research funding from Pfizer Pharmaceuticals (previously Wyeth Vaccines) and VaxInnate Corporation for vaccine studies unrelated to Cryptococcus.
Citation for this article: J Clin Invest. 2010;120(4):1039–1042. doi:10.1172/JCI42695.
See the related article beginning on page 1355.