David A. Relman MD
Among the vast diversity of microscopic life on this planet, only a small subset is believed to be capable of infecting the human body and causing disease. For example, only 7 of the 38–40 extant bacterial divisions on the earth contain members that are recognized pathogens in humans. The features that distinguish this subset of microbes (bacteria, fungi, parasites, and viruses) from all others have been revealed in increasing detail over the past few decades through the development of methods in microbial genetic manipulation, host cell imaging, structural biology, and, more recently, genomics. The findings indicate adaptation and coevolution of host and pathogen. Common themes are apparent among the strategies and mechanisms used by some microbes in their behavior as pathogens; these themes are the focus of this chapter. From an evolutionary perspective, pathogenicity appears to have “arisen” on multiple occasions, among diverse putative microbial ancestors. The commonality of themes and strategies that we witness today in extant pathogens probably reflects convergent evolutionary processes, operating in part over long periods of time, as host and pathogen adapt and counteradapt. Pathogenic capabilities are also acquired over much shorter time periods through mechanisms of horizontal genetic exchange. A comparison of a pathogen and closely related nonpathogen often identifies only small discrete differences between the two, sometimes in the form of a single contiguous chromosomal segment, a plasmid, or a gene.
The intimate dynamic between human host and microbe is easily disrupted by changes in either participant, especially when strong selective pressures are brought to bear on the microbe, whose capabilities for adaptation are enormous. This is the concept behind the “emergence” of newly recognized infectious diseases, an issue that has received a great deal of attention over the past decade. Emergence occurs either when a microbe gains new relevant genetic traits (discussed in the next section) or when the host or environment provides new opportunities and niches that enhance microbial access to or survival within a host. Some of these opportunities can be appreciated from an understanding of normal host defenses to infection (see Chapter 2), as in the loss of humoral or cellular immunity (eg, from cytotoxic chemotherapy or malnutrition), or normal anatomic barriers (eg, from trauma). Others arise from human behavior at the individual (eg, sexual) or societal (eg, from war or migration) level. A better understanding of this dynamic might lead to more enlightened therapeutic and preventive strategies against infectious disease.
MICROORGANISMS AS PATHOGENS
The last few decades of experimental investigation in microbial pathogenesis have taught us that virulence is a polygenic attribute of certain microorganisms that chose to persist or multiply within privileged anatomic niches of a host. A pathogen is a microbe that causes damage to a host as part of its (the pathogen's) strategy for multiplication within the host or for transmission to or from a host. Some microbes routinely cause disease (damage) within a given host and are called primary pathogens; other microbes cause disease only when normal host defenses are impaired, and these are called opportunistic pathogens. The difference between these two types of pathogens is related to the particular skills of the first group in either finding a privileged anatomic niche within the host as a site for multiplication or competing with the normal endogenous microflora for a crowded niche (eg, mucosal surface or skin). Virulence is a measure of the frequency with which a microbe causes disease in a specific host, the severity of the resulting disease, the efficiency with which the organism is transmitted to or from a host, or a combination of these factors.
The mechanisms and strategies adopted by pathogens that result in disease are often encoded by genes that are clustered together in a pathogen's genome. The examination of genome structure in many bacterial pathogens reveals so-called pathogenicity islands, 10- to 200-kilobase segments of DNA with as many as tens to hundreds of genes that often have a significantly different guanine-plus-cytosine nucleotide composition from the rest of the genome. This difference in island nucleotide composition is believed to reflect its origin in a distantly related microorganism. Pathogenicity islands are usually flanked by direct repeats of DNA or transfer RNA genes, suggesting that they may have been transduced into the bacterium and into specific sites of the bacterial genome by a bacteriophage. Remnant plasmid mobilization genes suggest that in some cases the island may have existed as an extrachromosomal element at some time in the distant past (tens to hundreds of millions of years ago). Most islands contain genes that encode specialized secretion systems, adhesins, or toxins that are necessary for disease causation. It is not uncommon for bacterial pathogens to contain multiple islands; for example, Salmonella serovar Typhimurium has one island, SPI1, whose gene products are required for bacterial entry into host cells, and another island, SPI2, whose products mediate survival and replication inside a phagosomal vacuole within host cells.
Given this evidence for large-scale genetic exchange and the presence of clusters of specialized genes within pathogens, it should not be surprising to find frequent horizontal transmission of virulence-associated genes among contemporaneous microbes, on a shorter time scale. This form of horizontal transmission is mediated by bacteriophages, plasmids, and transposable elements such as transposons. The number of plasmid- and phage-associated virulence genes is substantial; a partial list is provided in Table 3-1. The clinical significance of these transmission events should not be underestimated. Outbreaks of diphtheria may result from corynephage infection of nontoxigenic Corynebacterium diphtheriae strains within hosts and the conversion of these strains to toxin producers after phage delivery of the diphtheria toxin gene (see Chapter 51). The same type of event appears to be responsible for the dissemination of Shiga-like toxin genes among otherwise harmless Escherichia coli serotypes and the subsequent creation of enterohemorrhagic strains. Furthermore, recent data indicate that fluoroquinolone exposure may boost Shiga-like toxin production by these phage-infected strains through phage regulatory pathways and, therefore, may be detrimental to the host.
Table 3-1. Examples of virulence factors encoded on mobile genetic elements
All microbial pathogens follow the same steps in causing disease. After entry into a susceptible host, they find a specialized niche for attachment and subsequent colonization. To multiply and successfully colonize, they must confront and counter host defenses. Toxins and other secreted products often play a role in this effort, as well as in the effort to find an appropriate anatomic niche in this and other hosts. Transmissibility and transmission are a trait and process less well understood, except in those circumstances that involve a mechanical or biological vector. Pathogens, like other microbes, constantly sense their changing environment and regulate their behavior accordingly. The regulation of virulence-associated traits is complex and well integrated into systems used to regulate other microbial characteristics. Despite the universal nature of these steps, each pathogen displays a unique set of mechanisms, products, and responses that defines its pathogenic signature. Two members of the poxvirus family, smallpox virus and molluscum contagiosum virus, cause radically distinct clinical syndromes (skin lesions with and without a fulminant systemic inflammatory response, respectively). Although they share 103 genes, molluscum contagiosum virus and smallpox virus possess 59 and 83 genes, respectively, that the other does not contain; these virus-specific genes are predicted to encode proteins that subvert host innate immune defenses in a contrasting manner.
REGULATION OF VIRULENCE
Precisely timed and measured responses characterize the expression of microbial virulence attributes. Timing and measurement are critical to a successful pathogenic strategy, owing to the highly changing environments faced by microbial pathogens and the energy costs associated with expression of virulence, as well as the complex set of host defenses designed to recognize and neutralize pathogens. For these reasons, pathogens recognize multiple environmental cues, use diverse regulatory mechanisms and systems, and integrate these cues and systems in a well-tuned network. A regulon is a group of physically unlinked genes that respond in a coordinated fashion to a given stimulus. Common features of microbial virulence-associated regulatory schemes are apparent.
Some of the most important environmental signals recognized by pathogens are extracellular iron concentration, temperature, osmolarity, oxygen and CO2partial pressures, specific signaling factors secreted by themselves, and contact with a host cell surface. Extracellular iron concentration, for example, tells pathogens when they have arrived within an animal host, since the latter expends great efforts to sequester and keep this critical element from microorganisms at mucosal surfaces. Pathogens and other microbes use a wide variety of sensing systems to read these signals and transduce the information into an appropriate response. Iron is recognized in some bacteria by intracellular iron-binding proteins that regulate the expression of iron acquisition proteins and receptors (eg, transferrin-binding proteins), as well as a number of bacterial toxins (eg, diphtheria and Shiga toxins) at the level of transcription. Other sensing systems that transcriptionally regulate virulence genes include the AraC class of transcriptional activators (eg, V cholerae ToxT), alternative RNA polymerase sigma factors, histonelike proteins, quorum-sensing systems whereby bacterial population density and growth phase are recognized in a species-specific manner, and the two-component sensor-response regulator family.
The two-component family is one of the most well-studied bacterial virulence-associated regulatory systems. Each of these systems comprises a sensor protein that spans the cytoplasmic membrane and a cytoplasmic response regulator protein with which it interacts, usually by means of high-energy phosphate transfer. Phosphorylation induces activation of the response regulator, thereby allowing it to bind DNA promoter sequences and control gene expression. It is not surprising that the genomes of pathogens whose lifestyles require adaptation to multiple diverse environments contain a large number of such systems (eg, E coli, > 40), whereas those pathogens that face a restricted number of environments contain few (eg, Helicobacter pylori, only 5). With closer examination, variations on this theme become apparent. Vibrio spp., for example, express a protein in the cytoplasmic membrane, ToxR, that acts as both sensor (of osmolarity) and response regulator (of porin expression) (Figure 3-1). This protein is expressed by most marine vibrios. For Vibrio cholerae, the expansion of its preferred environments to include the human intestinal tract has brought with it the means to acquire cholera toxin genes (via a bacteriophage) and to acquire an intestinal attachment organelle that also serves as the toxin phage receptor (also via a phage), and this expansion placed both of these virulence attributes under the control of ToxR.
Regulation of virulence genes also takes place at the level of genome structure. Amplification of gene copy number enables up-regulation of gene product expression. Gene rearrangements are responsible for antigenic variation in such diverse pathogens as Neisseria gonorrhoeae (see Chapter 52), Borrelia recurrentis (see Chapter 65), and Trypanosoma brucei (see Chapter 85). Changes in DNA or RNA conformation can also effect modulation of gene expression.
ADHERENCE & COLONIZATION
Attachment to host substrates is an early and critical step for most pathogens. Those organisms that remain outside a host cell often rely on redundant and complementary attachment systems to ensure continued colonization of the host. In some cases, the expression of an attachment factor (ie, adhesin) alone confers on a microbe the ability to cause disease. In many other cases, adhesins and their cognate host receptors determine host species specificity for the pathogen, as well as tissue or organ tropism, and cell specificity. For many of the reasons described above, adhesin expression is regulated. The dominant adhesin for Bordetella pertussis is the 220-kDa protein filamentous hemagglutinin. Because filamentous hemagglutinin is the most abundant secreted protein for this organism, consuming a considerable proportion of the cell's resources for expression, it is important for B pertussis to make this protein only when it is needed. Thus, it is placed under the control of a two-component system.
Microbial adherence is critical for microbe-microbe interactions and the formation of biofilms. Bacteria such as Staphylococcus spp. secrete a polysaccharide slime substance that enables them to colonize foreign material and exclude antimicrobial agents. Surfaces of foreign and natural acellular material in the human body provide a substrate for microbial community development. On the tooth surface, for example, a diverse but well-ordered community of bacteria form a biofilm that begins on a layer of host proline-rich proteins and relies on many different binding interactions between bacterial receptors and ligands. The streptococci and Actinomyces spp. are among the “early colonizers,” whose presence is necessary before “late colonizers” such as Porphyromonas gingivalis and treponemes can attach.
Figure 3-1. Schematic drawing of Vibrio cholerae, its chromosomally integrated prophages, and the regulation of cholera toxin (CTX) expression. ToxR is a transmembrane protein that activates expression of chromosomal genes encoding the porins OmpU and OmpT under certain osmotic conditions. In collaboration with the transmembrane protein TcpP, it also regulates expression of the transcriptional activator ToxT, which in turn is responsible for cholera toxin expression, and the pilin subunit TcpA expression. ToxT and TcpA are encoded by the integrated VPIΦ prophage, and CTX by the CTXΦ prophage. The CTXΦ phage uses the TcpA pilus as a receptor for entry into V cholerae.
Complex multimeric structures are critical for the attachment of many microbial pathogens. The hairlike bacterial appendage known as a pilus (or fimbria) is a well-characterized example. Pili are composed of structural subunits arranged in a helical fashion, extending out from the bacterial surface. They are flexible and are often capped by adhesin proteins that provide diverse binding specificities. Pili mediate lectinlike interactions with host cells and are also recognized by protein receptors. Regulated expression and assembly of pili are highly orchestrated and involve clusters of cotranscribed genes organized as an operon. Because pilin proteins are often antigenic, their expression and structure are variable. A class of protozoa shares a different sort of attachment organelle, known as the apical complex. This polar structure found on Toxoplasma spp., Cryptosporidium spp., and other “apicomplexans” controls host cell attachment through an active mechanism and secretes factors necessary for entry of the protozoan into host cells in some cases.
Pathogens often co-opt host attachment proteins and host receptors for their own adherence-related purposes. A wide variety of pathogens coat themselves with host extracellular matrix proteins such as fibronectin in order to bind to extracellular matrix receptors such as integrins. In this way, extracellular matrix proteins act as a molecular bridge between microbe and host cell. Rhinoviruses bind to intercellular adhesion molecule-1, a critical host cell attachment receptor, and HIV binds to CD4 and chemokine receptors on lymphocytes and macrophages, which are critical for antigen recognition and chemotactic responses, respectively.
The particular selection of ligand-receptor combinations by any given microbe dictates the subsequent course of events. These events include induction of up-regulated binding affinity, microbial internalization, and host cell intoxication. In a dramatic example, enteropathogenic E coli (and other bacterial pathogens) inject their own receptor into a host cell, using a specialized (“type” III) secretion system, where the receptor subsequently becomes activated by phosphorylation, localized to the host cell membrane and then promotes formation of an unusual host cell cytoskeletal rearrangement, culminating in formation of a “pedestal” on which the bacterium forms an intimate attachment.
Host cell intoxication is one of the oldest-known mechanisms associated with microbial virulence. Toxins block host cellular defenses, elicit or create microbial nutrients, break down anatomic barriers, and facilitate microbial transmission. In some extreme cases, virulence can be entirely ascribed to the action of a secreted toxin; examples include diphtheria and tetanus. In these cases, antitoxin immunity is sufficient for protection against disease. But most infectious diseases and pathogenic strategies are not that simple. Microbial toxins exist in many forms and with many different activities. Some are intrinsic to the microbial cell wall, such as the lipopolysaccharide (or endotoxin) of gram-negative bacteria, while other toxins are secreted (exotoxins), either nonspecifically from the cell or through specialized secretion systems, such as the type III and type IV systems, directly into a host cell. Exotoxins of the first type usually facilitate their own entry into a host cell by means of a binding subunit that may take advantage of host cell internalization pathways. In a manner and for reasons similar to those described for adhesins, toxin expression is often tightly regulated.
Microbial toxins can be classified on the basis of shared mechanisms of action. The nature of the host cell receptor or molecular target, the mechanism of activation, or the intracellular trafficking of a microbial toxin may determine its specificity of effect. One of the most well-known families of microbial exotoxins is the group of bacterial ADP-ribosylating toxins, which includes diphtheria toxin, cholera toxin, pertussis toxin, E coli heat-labile enterotoxin, and Pseudomonas exotoxin A. Each catalyzes the transfer and covalent binding of ADP-ribose from NAD to a target protein. The different targets explain the diverse biological outcomes of these toxins. Sometimes, the explanation for difference in effect lies elsewhere. Tetanus and botulinum toxins are zinc metalloproteases that enter motor neurons at the neuromuscular junction. They cleave the same group of proteins within neurotransmitter-containing vesicles. But the clearly distinct neurological results of this action, flaccid and spastic paralysis, can be explained by differential intracellular trafficking of these two toxins: botulinum toxin remains in the terminal neuron at the neuromuscular junction where excitatory neurotransmitters are normally released, while tetanus toxin is transported in a retrograde fashion across the interneuron synapse where inhibitory neurotransmitters are normally released. Other families of microbial toxins include pore-forming hemolysins, secreted adenylate cyclase toxins, RNA glycosidases, and superantigens.
ENTRY INTO & LIFE WITHIN HOST CELLS
For viruses and more specialized bacterial and eukaryotic pathogens, entry into a host cell is an essential feature of their survival or replication strategy within the host. Intracellular entry can be either a passive or active process on the part of either participant. Although most microbes that routinely enter host cells find the intracellular compartment a favored site for replication, this is not always the case; some may be simply passing through on their way to deeper portions of a tissue or the bloodstream (eg, Neisseria meningitidis and the Whipple's disease bacillus Tropheryma whippelii) or may remain inside a cell in a nonreplicating state. Only those agents that replicate inside a host cell or exist primarily within host cells rather than outside them are properly defined as intracellular pathogens. The inside of a host cell offers a number of significant features for a microbe. First, there may be essential nutrients, macromolecular synthetic machinery, and sources of energy. Organisms that can replicate only when inside a host cell (eg, viruses or Coxiella burnetii) are obligate intracellular pathogens, while those that can replicate both inside and outside (eg, Salmonella serovar Typhimurium) are facultative intracellular pathogens. Second, the intracellular compartment offers protection from a variety of extracellular microbicidal factors. Third, it may provide a ready means for dissemination. Pathogens (eg, Salmonella spp. or HIV) that persist or grow within leukocytes may be carried to distant sites within the host by these cells. For Legionella spp., the internal compartment of a free-living amoeba provides a protected niche in external freshwater environments. Encystment of the amoebae allows persistence of the two symbionts under environmental conditions that are unfavorable for growth.
Intracellular pathogens take advantage of many naturally occurring internalization pathways in professional phagocytic cells, as well as creating their own opportunities and pathways in nonprofessional phagocytes. Leukocyte integrin receptors mediate host cell-cell contact, communication, and development; they also mediate internalization or attachment of complement-coated microbes, as well as those that express their own integrin ligand (eg, B pertussis filamentous hemagglutinin). Entry through this route does not necessarily trigger an oxidative burst; ie, microorganisms can survive this process. Fc receptors on phagocytes offer another route for entry, but they target the subsequent phagocytic compartment for fusion with lysosomes. Most intracellular pathogens seek to avoid this event. For example, the highly successful intracellular pathogen Toxoplasma gondii, if forced to bind to Fc receptors, will die within a host cell.
Entry into nonprofessional phagocytes is usually either induced by the pathogen or forced by the pathogen through an active process. These entry processes require actin polymerization in most cases. Induced entry involves manipulation of the host cell actin cytoskeleton, through a zipperlike mechanism. Host cell receptors that are coupled to the host cell cytoskeleton, such as dystroglycan, E-cadherin, and integrins, provide internalization pathways of this sort for microbes (Mycobacterium leprae and arenaviruses, Listeria monocytogenes, and Yersinia spp., respectively), even though the usual role for these receptors may involve cell adhesion. On the other hand, induced entry is sometimes a more dramatic event, when microbes elicit membrane ruffling to become internalized. Shigella and Salmonella spp. both secrete proteins into nonprofessional host cells that cause this type of event (a trigger mechanism) through interactions between these effector proteins (eg, Salmonella SopE) and host low-molecular-weight GTP-binding proteins of the Rho family. Active, forced entry is described for the apicomplexan parasites, which rely on their own actin-based contractile skeleton and attachment proteins localized near the apical complex to pull the host membrane around them.
Intracellular pathogens choose different sorts of anatomic compartments for themselves, once inside (Table 3-2). To a large degree they create or modify these compartments by selecting entry receptors and then by secreting a variety of factors. All pathogens are encompassed by a vacuolar membrane immediately upon entry. Some remain within this vacuole and allow it to mature through the endocytic pathway and fuse with lysosomes. Coxiella burnettirequires the acidic environment of a phagolysosome for completion of its lifecycle. Others block phagosomal development and modify intravacuolar conditions to suit their needs. Legionella pneumophila, for example, recruits mitochondria and other organelles so that they surround the vacuole. Finally, some pathogens destroy the vacuolar membrane quickly and escape into the host cell cytoplasm. L monocytogenes illustrates this last strategy. Listeriolysin-O lyses the vacuole membrane and is required for bacterial virulence. Once in the cytoplasm, L monocytogenes induces actin polymerization at one bacterial pole and is thereby pushed through the cytoplasm at the end of a comet tail-like structure (see Chapter 51). The bacterium moves from the inside of one cell to the inside of an adjacent cell through protrusions of host cell membrane and subsequent engulfment by the next cell. Shigella and Rickettsia spp. and vaccinia virus also induce the same kind of actin-based movement within host cell cytoplasm.
Table 3-2. The preferred intracellular niches for some microbial pathogens.
MANIPULATION OR AVOIDANCE OF THE HOST IMMUNE RESPONSE
Coevolution and adaptation of host with pathogen are well illustrated by the extensive efforts of pathogens to evade, subvert, or co-opt the wide array of host immune defenses. These efforts include the establishment of latency, antigenic variation (described above), inhibition of antigen processing or presentation, inhibition of phagocytosis or oxidative burst, inhibition or stimulation of cytokine responses, anticomplement strategies, induction or inhibition of apoptosis, and inhibition of lymphocyte homing (Table 3-3). Viruses provide many examples, although they are by no means the exclusive operators of these mechanisms. Apoptosis (programmed cell death) is a common host cell antivirus defense mechanism. Some viruses, such as papillomaviruses and adenoviruses, prevent host cell apoptosis by expressing proteins that bind to the host cell cycle control proteins p53 and Rb. It makes sense that papillomavirus would want to promote the growth of its host cell, the squamous cell of the skin, since this cell is terminally differentiated and poorly able to replicate.
Epstein-Barr virus and other herpesviruses express homologs of Bcl-2, a human antiapoptotic protein, whereas molluscum contagiosum virus and herpesviruses produce proteins with death effector domains that interfere with Fas- and tumor necrosis factor receptor-1-associated apoptotic signaling pathways. In contrast, a number of bacterial pathogens induce apoptosis as a means of eliminating host defense cells (eg, activated macrophages).
Table 3-3. Strategies used by pathogens to subvert normal hostdefenses.1
Disruption or co-optation of cytokine-mediated host responses is also a common pathogen counter-defensive strategy. Kaposi's sarcoma herpesvirus (human herpesvirus 8) encodes homologs of the human chemoattractant cytokines macrophage inflammatory protein-1α and RANTES that are biologically active. They are angiogenic and also compete with other ligands for their cognate chemokine receptors. Kaposi's sarcoma herpesvirus produces a version of interleukin-6 that signals through the same proinflammatory pathways as does the human homolog. Finally, Kaposi's sarcoma herpesvirus expresses a G-protein-coupled receptor that is homologous to the human interleukin-8 receptor. It stimulates constitutive host cell proliferation. Poxviruses encode a soluble form of the human interleukin-1 receptor. Deliberate disruption of the viral gene in animal models of poxvirus infection causes more fulminant disease. It is believed that this soluble receptor down-regulates the effects of host interleukin-1, which is elicited by the virus and which would otherwise kill the host before the virus has had adequate time for replication.
TRANSMISSION & LIFE OUTSIDE THE MAMMALIAN HOST
Transmission of pathogens is essential to their ultimate survival. Some rely on arthropod (biological) or mechanical vectors; this is especially true of pathogens that reside in nonhuman animal reservoirs (zoonotic agents; see Chapter 90). Adaptation to and recognition of vector niches are increasingly well understood for these microbes. The agent of Lyme disease, Borrelia burgdorferi, for example, expresses certain outer surface proteins only when it is in the tick and expresses others only when it is in a human. Other pathogens rely on human disease symptoms such as cough and diarrhea for transmission. Any pathogen that spends significant time in an external inanimate environment must find a strategy for surviving harsh conditions. Sporulation or encystation is one strategy. Another is incorporation within the integument or internal structures of another life form; for example, V cholerae attaches to algae and copepods during the periods that it spends in brackish estuarine aquatic environments. The bridging of the environmental sciences with medical microbiology will be an important step in further elucidating this poorly understood phase in the life of microbial pathogens.
Cotter PA, Miller JF: In vivo and ex vivo regulation of bacterial virulence gene expression. Curr Opin Microbiol 1998;1:17.
Finlay BB, Falkow S: Common themes in microbial pathogenicity revisited. Microbiol Mol Biol Rev 1997;61(2): 136.
Groisman EA, Ochman H: Pathogenicity islands: bacterial evolution in quantum leaps. Cell 1996;87:791.
Guiney DG: Regulation of bacterial virulence gene expression by the host environment. J Clin Invest 1997;99:565.
Hueck CJ: Type III protein secretion systems in bacterial pathogens of animals and plants. Microbiol Mol Biol Rev 1998;62:379.
Hultgren SJ, et al: Pilus and nonpilus bacterial adhesins: assembly and function in cell recognition. Cell 1993;73(5): 887.
Marrack P, Kappler J: Subversion of the immune system by pathogens. Cell 1994;76:323.
Miao EA, Miller SI: Bacteriophages in the evolution of pathogen-host interactions. Proc Natl Acad Sci USA 1999;96:9452.
Wilson M, Seymour R, Henderson B: Bacterial perturbation of cytokine networks. Infect Immun 1998;66:2401.