Gram-negative Cocci are almost invariably isolated from nasopharyngeal cultures of human beings. The majority of these are commensal species, which are classified principally by sugar-fermentation reactions performed on the organisms following their isolation from the primary throat culture. Thus, Neisseria lactamica derives its name from its ability to ferment lactose. On rare occasions bacteria belonging to a few of these species are able to invade the human host and cause disease. Table 48.1 lists the pathogenic gram-negative cocci.
I. INFECTION BY NEISSERIA MENINGITIDIS AND NEISSERIA GONORRHOEAE
A. Local infection Among the Neisseria and related organisms listed, Neisseria meningitidis and Neisseria gonorrhoeae are the primary pathogens, that is, organisms that are able to cause disease in an otherwise healthy host. Hence, this discussion focuses on these species. DNA hybridization has indicated that Neisseria meningitidis and Neisseria gonorrhoeae are extremely closely related organisms. It is therefore not surprising that the pathogenetic strategies of the organisms are for the most part very similar. Both are able to cause infection of the mucous membranes. Neisseria meningitidis, transmitted by droplets from the respiratory tract of an infected individual, usually infects the mucous membranes of the nasopharynx, but has been isolated on occasion from genitourinary sites. The nasopharyngeal infection with Neisseria meningitidis is most often asymptomatic and is self-limited, with the infection lasting for a few weeks to months. This colonization is referred to as the carrier state and is quite common. During the winter months, the frequency of carriers is usually 10% or greater. In populations that live in close contact, such as in boarding schools and military recruit camps, the carrier rate not infrequently will exceed 50%. Gonococci are transmitted most often by sexual contact and infect the genitourinary tract, but are also quite frequently isolated from rectal and pharyngeal cultures. The infections of the genitourinary tract are most often symptomatic, whereas pharyngeal infections generally cause no symptoms. The conjunctivae are susceptible to gonococcal infection, particularly in neonates who acquire it by passing through an infected birth canal. Historically this infection, ophthalmia neonatorum, was extremely common and was the major cause of acquired blindness until the general acceptance of the preventive Credè procedure consisting of the instillation of drops of a 1% silver nitrate solution in the eyes of all newborns.
This has been replaced by the use of less irritating antibiotic ointments. B. Extension of local infection Genitourinary gonococcal infection in women usually involves the endocervix and with lesser frequency the urethra, the rectum, and the pharynx. As many as 50% of infections may be asymptomatic or with insufficient symptoms to motivate the person to seek medical attention. In at least 10% of infections, there is early extension to the uterine cavity (endometritis), and subsequently ascending infection of the fallopian tubes (pelvic inflammatory disease, PID). PID is the major complication of gonorrhea. Most often it requires hospitalization for differential diagnosis and treatment, and causes tubal scarring resulting in infertility and tubal pregnancies. Infections in men, although sometimes asymptomatic, most often cause sufficient symptoms to drive the affected individual to seek medical attention. In the days prior to availability of antibiotic therapy, epidydimitis occurred in about 10% of cases, but this complication as well as prostatitis are rarely seen today. C. Bacteremic infection Both N. meningitidis and N. gonorrhoeae first colonize the epithelial surface and are able to traverse epithelial cells by mechanisms to be described here. Once the organisms are in the subepithelial space, both the meningococcus and the gonococcus may invade the bloodstream. With N. meningitidis invasion of the bloodstream and the subsequent invasion of the meninges of the brain with resulting meningitis are very dangerous bacterial infections that can very rapidly lead to death, even with optimal antibiotic and supportive therapy. In the United States there are about 3000 reported cases each year of meningococcal meningitis, and now that meningitis due to Haemophilus influenzae type b is a vanishing disease as a result of the widespread acceptance of vaccination, the meningococcus is the leading cause of meningitis.
In contrast with other organisms that cause meningitis, N. meningitidis has the ability to cause epidemic outbreaks with incidences approaching 200 or more per 100,000 per year. Epidemics of meningitis have occurred in all parts of the world, but have been a particular problem in sub-Saharan Africa; for instance in the winter of 1995–1996, there were 250,000 cases in west Africa. In the case of the gonococcus, it is estimated that 1–2% of patients with gonorrhea do have invasion of the bloodstream, which can in rare instances be as fulminant as meningococcal infection, but usually has a much more benign course. The disease most often presents with fever, arthritis, and petechial, hemorraghic, pustular, or necrotic skin lesions. If untreated, this can progress to septic arthritis that may cause severe damage to the affected joint. D. Treatment Meningococcal infection is a medical emergency and the earlier effective antibiotic treatment is initiated the better the prognosis. The mortality rate for meningococcal meningitis is generally about 10%, but it is much higher in meningococcemia and shock. The drug of choice for treatment remains intravenously administered penicillin G in very high doses. However, the pneumococcus and Haemophilus influenzae can at times present a very similar clinical picture, and because these organisms frequently are resistant to penicillin, third-generation cephalosporins are recommended as initial therapy until the meningococcal etiology has been established. Patients need to be hospitalized because the course of the disease is unpredictable and supportive therapy is frequently needed. The treatment of gonococcal infection has changed over the past decades due to the development of partial or complete resistance of this organism to many antibiotics. The list of recommendations proposed by the Centers for Disease Control (CDC) in 1993 take into account the need to treat this disease at the time of the clinic visit with a single dose because many patients are not compliant with regimens that require repeated administration of the medication. In addition, coinfection with Chlamydia trachomatis is a very common occurrence, and the treatment should also eliminate this organism.
II. MOLECULAR MECHANISMS OF INFECTION
One of the problems in the study of neisserial disease is that these organisms are restricted to human beings, and animal models have provided very little information on the pathological events occurring during the various stages of the infection. Because of the excellent response of uncomplicated gonorrhea in men to treatment with modern antibiotics, challenge studies of volunteers have been accepted as ethically justifiable. Such scientific experiments, because of their cost, can be performed only infrequently and therefore correlations have to be made with in vitro models that mimic the in vivo or natural conditions as closely as possible. The human disease has been most closely simulated by an organ culture system employing fallopian tubes that are obtained from women undergoing medically indicated hysterectomies. The epithelium lining the fallopian tubes consists principally of two kinds of cells, mucus-secreting cells, and cells that bear cilia and beat in unison to move the mucus layer. When gonococci are added to the explants in vitro, the first discernible interaction is that the gonococci attach by means of long hair-like projections known as pili to the surface of the mucus-secreting cells, but not to the ciliated cells. This is a distant attachment between the bacteria and the cell surface and occurs about 6 h following inoculation. Then over the next 12–18 h this distant attachment converts to a very close attachment in which the membranes of the host and the parasite come into extensive and intimate contact. The close attachment is believed to be mediated by a set of outer-membrane proteins, named opacity proteins and discussed later. These interactions initiate a signaling cascade that causes the epithelial cells to engulf the gonococci, transport the bacteria through the body of the cell in vacuoles, and egest them in an orderly fashion on the basal part of the cells onto the basement membrane.
Later in the infection (24–72 h) toxic phenomena occur that result in expulsion of the ciliated cells from the epithelium. If meningococci are placed on human fallopian tubes, the same events occur, but over a shorter period of time. However, these events are not seen with commensal Neisseria species or with fallopian tubes that are not of human origin. The appearance of biopsies of cervical tissue taken for cancer-diagnostic purposes that were inadvertently obtained from infected patients show a picture that is quite similar. The events transpiring in the course of the model infection have been the focus of research in order to understand meningococcal and gonococcal disease in molecular terms. Many bacterial species are able to invade epithelial cells, and in the case of Yersinia, Salmonella, Shigella, and Listeria quite a lot is known about this process because of the genetic tractability of these species. In the case of the Neisseria, this exploration is not as far advanced. Obviously the establishment of the mucosal infection depends on It is noteworthy that compared to the other pathogens mentioned, for which invasion occurs quite promptly, there is a long lag in the invasion by the Neisseria as if some slow inductive events need to occur in the host cell or the organism or both. III. ANTIGENS The surface antigens of the Neisseria have been extensively studied to gain an understanding of the molecular steps underlying the pathogenesis of these diseases as well as to identify candidate molecules for inclusion in vaccines. A. Pili Most peripheral on the surface are pili, which are hair-like appendages with a diameter of about 8 nm that emanate from the outer membrane of gonococci and are several bacterial diameters long. Pili consist of the helical aggregation of a single kind of protein subunit of about 18,000MW, known as pilin. The study of pili on gonococci is enormously simplified by the fact that their presence imparts a distinctive appearance to the gonococcal colony as it grows on agar. Piliated colonies are smaller and have sharp edges when viewed with a colony microscope. On laboratory media, isolates with a different colonial appearance, namely larger colonies that are flatter and have an indistinct edge, appear; if the organisms are subcultured nonselectively, these become the predominant colonial form. This colonial form of gonococci is no longer piliated, but in some instances these strains can revert to the piliated state. This ability to turn on and off pilus expression occurs at very high frequency, on the order of 1 in a 1000 cells per generation.
Gonococci that are freshly isolated from patients invariably are piliated. It is known from challenge studies of volunteers that only piliated strains are capable of causing infection. Stable nonreverting pilus-negative strains cannot cause infections. Thus, pili seemed at first to be an ideal vaccine candidate, but it soon became evident that pili are antigenically very variable. It was noted that no two strains of gonococci appeared to have the same pili, and later it was found that the pili expressed by a single strain of gonococcus maintained in the laboratory would over time repeatedly change their antigenicity. Following the cloning of the pilin structural gene, this problem could be approached on a molecular level and the mechanism of antigenic variation is summarized in a very simplified form in Fig. 48.1. Generally, the gonococcus possesses a single genetic locus expressing the pilin, which is called pilE. This locus contains a complete pilin structural gene with its promoter. In addition to the expression locus, gonococci also have eight or more other loci referred to as silent loci. One of these, pilS1, is shown enlarged to indicate that the silent loci contain a large number of distinct pilin genes. This particular locus contains five silent pilin genes. These pilin genes are characterized by the fact that all of them are incomplete. They lack the promoter sequences and portions corresponding to the beginning of the protein coding frame. These incomplete genes are efficiently shuttled by homologous recombination into the pilE expression site, causing the production of a large number of serologically variant pili by a single strain of gonococcus over a period of time. The recombination events can occur in the conserved N-terminal portion of the coding frame and in the conserved region immediately following the termination codon. But it can also occur in several islands of conserved sequences scattered in the variable portion of the pilin genes between the minicassettes. Such antigenic changes occur at a rate of about 1 in a 1000 cells per cell division.
If the new antigenic variant pilin can be assembled into intact pili, then an antigenic variation step has occurred. If the new pilin cannot be assembled into an intact pilus, the organism is pilus negative, but is able to revert to pilus positive as soon as a gene copy that can be assembled is recombined into the expression locus. Thus, the recombinational mechanism accounts for both on–off variation and for antigenic variation. This is obviously a remarkably complex genetic mechanism for varying this protein and the only conceivable evolutionary pressure to force the development of this system is, of course, the human immune system. The volunteer studies have demonstrated that the antigenic variation does occur in vivo and that in fact almost all of the reisolates from the infected volunteers had a different pilus type than that expressed by the infecting strain. Recent crystallographic studies have shown that pili consist of a helical aggregate of five pilin subunits per turn and that, remarkably, the exposed surface of the pilus cylinder consists of the variable domains, whereas the constant regions of the pilin molecule are buried in the cylinder. Pili are associated with a protein of about 110,000 MW, named PilC, which is involved in the assembly of pili and also appears to be present at the tip of the pili and imparts the ability to adhere to epithelial cells. Evidence has been provided that the host antigen recognized by PilC is CD46, a widely distributed cell-surface protein that acts a complement regulatory factor. This antigen also serves as the receptor for measles virus. Meningococci also bear pili that are very similar to gonococcal pili with a similar ability for antigenic variation.
B. Capsules Meningococci are classified into serogroups on the basis of the chemical nature of the capsular polysaccharide they express (see Table 48.2). Epidemiologically, groups A, B, and C are the most important because they are the cause of over 90% of cases of meningitis and meningococcemia. Meningococcal disease occurs worldwide as an endemic disease principally in infants 3 months or older and in young children with a rate of two to five cases per 100,000 population. The incidence is seasonal, with winter and spring having most cases. However, meningococcal disease can also occur in epidemic form where the rate of disease can rise as high as 200–500 cases per 100,000. During epidemics, the peak incidence shifts to an older age group, children 5–7 years of age. During the first half of the twentieth century, epidemics caused by group A meningococci occurred in the United States about every 12 years. Since World War II, there has not been a major epidemic in the United States, but epidemics have occurred in many other parts of the world, notably Brazil, China, Finland, Russia, and Mongolia. However, the area of the world most severely affected is Africa in the meningitis belt that extends through all of the sub-Saharan countries from the Sahel to the rain-forest region. In this region, major epidemics affecting tens of thousands of inhabitants have occurred every 3–4 years, and the problem has worsened recently, so that in the winter season of 1995–1996 there were 250,000 cases in west Africa. The presence of the capsular polysaccharides protects the organism from the natural defense of the host, such as phagocytosis by white cells and killing mediated by complement via the alternative pathway. However, the presence of antibodies to the capsular polysaccharides is protective and these antigens are the basis of the meningococcal vaccines to be discussed later. The locus for the biosynthesis of capsular polysaccharide has been characterized and is encompassed by about 25 kb of the genome. The right and left sides of the locus are conserved among serogroups and are concerned with common biosynthetic steps, such as the addition of lipid carriers and the export of the product from the cytoplasm to the exterior. The middle portion of the locus differs between serogroups and contains the enzymes responsible for the biosynthesis of the activated sugar intermediates for the particular serogroup, as well as the specific polymerase assembling the polysaccharide.
C. Outer-membrane proteins The outer membrane of the pathogenic Neisseria, as is that of other gram-negative bacteria, consists of a lipid bilayer with the outer leaflet consisting principally of lipopolysaccharide (LPS). The outer membrane contains a number of integral membrane proteins, of which the porins are quantitatively predominant. The nomenclature of the neisserial outer-membrane proteins has evolved with the increasing knowledge of these proteins and is summarized in Table 48.3. 1. Porins The porins of the pathogenic Neisseria, as are those of Escherichia coli, are postulated to consist principally of _-pleated sheets arranged perpendicularly to the membrane, with loops exposed to the cytoplasm and eight loops exposed on the surface of the organism. Each functional porin consists of a trimer of the porin subunit. The meningococcus contains two genetic loci that code for the production of outer membrane porins, originally referred to as the class 1 and class 2/3 proteins and now called porA and porB. The gonococcus, although lacking a homolog of the first porin, possesses the porB locus that gives rise to porins that are very similar in amino acid sequence to the class 2/3 proteins. The gonococcal porins vary antigenically to a limited extent primarily in the surface exposed loops and fall into two main classes, referred to as PIA and PIB. PIA strains predominate among gonococci isolated from the bloodstream and apparently have an increased capacity to invade the bloodstream and cause disseminated gonococcal disease. PIA strains also tend to be resistant to the bactericidal action of normal human serum. Strains that cause ascending infection of fallopian tubes are invariably of the PIB type. There are a number of indications that gonococcal porins not only serve as channels through which water and solutes of less than 1000 MW can diffuse through the outer membrane, but also play an active role in pathogenesis.
Biophysical studies in artificial lipid membranes indicate that the gonococcal porins are unusual among gram-negative porins in that they are somewhat anion selective and very voltage sensitive. Voltage sensitivity means that when the protein is in a membrane, the channel will be modulated by the potential across that membrane, such that at low membrane potentials the porin molecules will be open and as the voltage is raised, the probability that the porin molecule is closed increases. In addition, it has been shown that the porins are able to bind GTP and certain other phosphate compounds and that this binding favors the closing of the porin channel. The neisserial porins readily transfer from the outer membrane of living gonococci to foreign membranes, including human cells. It obviously becomes very interesting to ask what the functional consequences of a newly inserted voltage-dependent ion channel in the host-cell membrane may be. This has been studied in detail with human polymorphonuclear leukocytes (PMN) using purified gonococcal porin. Within seconds after the addition of porin to the leukocytes, the membrane potential of these cells becomes hyperpolarized due to chloride ion movement. Shortly thereafter, the membrane potential returns to baseline, presumably because the porin channels adjust to this hyperpolarization by closing, and the active ion pumps of these cells reestablish their baseline potential. Even though the initial effects on the cell’s membrane potential by porin addition are short-lived, the consequences of a foreign voltage-regulated channel in the membrane of these cells are much longer lasting. This is seen when these cells are subsequently exposed to a stimulus such as fMLP (formyl methionyl-leucyl-proline). Normally, fMLP causes an immediate depolarization of the membrane. However, with porin channels present in the membrane, this depolarization is replaced with a prolonged hyperpolarization. Porin also markedly inhibits the aggregation of PMN. Degranulation in response to fMLP, LTB4, or the complement component C5a is also blocked, but is normal when induced with PMA. However, there is no inhibition of superoxide generation in response to these signals.
2. Rmp All strains of gonococci produce an outer-membrane protein originally designated PIII. This protein migrates on a SDS-PAGE with an apparent 31,000MW unreduced, and with an apparent 32,000MW when exposed to reducing agents such as _-mercaptoethanol. Hence, the protein has been named reduction modifiable protein (Rmp). In contrast to other outermembrane proteins, Rmp is a highly preserved antigen showing little if any variation among strains. The sequence of Rmp has substantial homology with OmpA, a protein that is universally present in all enterobacterial species. Rmp is also present in meningococci, where it was originally named class 4 protein and it is almost identical to the Rmp of gonococcus. It has been found that complement-fixing IgG antibodies to Rmp are present in the sera of at least 15% of normal human beings with no history of prior gonococcal infection. These antibodies arise in response to the meningococcal carrier state and also by contact with the enterobacterial flora. Surprisingly, these antibodies do not mediate serum killing or opsonization of gonococci, but instead block the ability of normally bactericidal antibodies directed to other surface antigens to exert their function. These results have been substantiated with monoclonal antibodies. It was found that an anti- Rmp antibody was a powerful blocking antibody, inhibiting the activity of other bactericidal monoclonal antibodies directed to a number of different surface proteins or LPS (lipopolysaccharides). In an epidemiologic study of a population at very high risk for acquiring sexually transmitted diseases, it has been demonstrated that the presence of anti-Rmp antibodies significantly increases the risk of gonococcal infection, demonstrating the inhibitory role of blocking antibodies in the local mucosal infection. The blocking activity of anti-Rmp antibodies is not seen with meningococci, perhaps because this organism expresses quantitatively less Rmp. The molecular mechanism by which anti-Rmp antibodies act as blocking antibodies is not understood. 3. Opa proteins Pathogenic Neisseria express another surface-exposed outer-membrane protein that in the meningococcus was called class 5 protein and in the gonococcus protein II. The presence of this class of protein leads to distinctive changes in the morphology of the colonies on agar. Gonococci that do not express protein II give rise to colonies that are transparent and resemble a beaded water droplet, whereas gonococci expressing this antigen give rise to colonies that are opaque and have a ground-glass appearance. Hence, they have been named opacity or Opa proteins. The expression of Opa can turn off and on at high frequency, and a clone of gonococcus can express at least five or six recognizably different opacity proteins over a period of time. Gonococci freshly isolated from the blood of patients with disseminated gonococcal infection do not express Opas. The same is true of isolates from pelvic inflammatory disease. Strains from males with genitourinary disease usually express Opa protein. Most remarkably, in young women not on birth control pills, the gonococci that can be isolated from the cervix vary so that at the time of ovulation the isolates express Opas, although at the time of menses they do not. So, as with pili, there is phase and antigenic variation both in vitro and in vivo.
However, the mechanism is entirely different than that seen with pili. Gonococci possess 11 copies of complete opa genes in the gonococcal genome, and all of them have a variable number of pentameric repeats of the sequence CTCTT between the ATG initiation codon and the remainder of the protein. This repeat codes for leucine, serine, and phenylalanine, amino acids that are normally contained in the hydrophobic portion of signal sequences. The number of repeats is subject to rapid change due to slipped-strand mispairing during replication. The consequence is that with some number of repeats the beginning of the gene will be in frame with the remainder of the gene, and with a different number of repeats it will be out of frame. Thus, the expression of this class of proteins is controlled at the level of protein translation. The same pertains to meningococci, although they possess only three opa genes. This mechanism of genetic variation has also more recently been described in a number of other mucosal pathogens, notably Haemophilus influenzae and Helicobacter pylori. The different opa genes of a few strains have been sequenced and have been distinguished either by naming them opaA–opaJ or by adding a numerical subscript (see Table 48.3). The loci code for mature proteins about 250 amino acids long. The genes are highly homologous, except for two regions that are very variable and a smaller region that has lesser variation. The differences among the proteins in the content of basic amino acids is noteworthy and the pI of the proteins range from about 7.0–10.0. Since the discovery of this class of proteins, it has been noted that they increased the adhesiveness of gonococci to epithelial cells in tissue culture or to human PMN, and it was inferred that they mediated the close attachment phase in the fallopian tube model. The ligand specificity of the Opa proteins has been defined on a molecular level. It has been shown that a particular Opa protein (OpaA protein of strain MS11) recognizes heparan sulfate on the surface of epithelial cells and that heparin is able to inhibit the binding. The heparan sulfate occurs mainly on the syndecan class of molecules, of which four have been described and two of these (1 and 4) are expressed on epithelial cells. The syndecans are believed to act as receptors or coreceptors for interaction between cells and the extracellular matrix. The other Opa proteins react with several proteins that are members of the carcinoembryonic antigen CEA family.
CEA was originally described as a colon cancer associated antigen and tests for blood levels of CEA antigen are used to clinically monitor the progression of colon cancer. There are now about 20 related proteins known; their genes are clustered on human chromosome 19, they belong to the immunoglobulin (Ig) superfamily, and they have a N-terminal domain homologous to immunoglobulin variable (IgV) domain and a variable number of domains with homology to Ig constant regions. Some are transmembrane proteins with cytoplasmic tails, whereas CEA is GPI-linked. The proteins are heavily glycosylated. Several of the genes are subject to alternative splicing and various family members are expressed on a wide variety of cells, including epithelial cells. The Opa proteins of both the gonococcus and the meningococcus react with the IgV domain of the molecules irrespective of its state of glycosylation. D. Lipopolysaccharide As do other gram-negative bacteria, the pathogenic Neisseria carry lipopolysaccharide (LPS) in the external leaflet of their outer membranes. In contrast to the high-molecular-weight LPS molecules with repeating O-chains seen in many enteric bacteria, the LPS of Neisseria is of modest size and therefore is often referred to as lipooligosaccharide or LOS. Although the molecular size of the LPS is similar to that seen in rough LPS mutants of Salmonella ssp., this substance has considerable antigenic diversity. In the case of the meningococcus, a serological-typing scheme has been developed that separates strains into 12 immunotypes and the detailed structure of the majority of these has been determined. The LPS of the pathogenic Neisseria is heterogeneous and LPS preparations frequently contain several closely spaced bands by SDS-PAGE. Using monoclonal antibodies, it is evident that gonococci are able to change the serological characteristics of the LPS they express and that this antigenic variation occurs at a frequency of 10_2–10_3, indicating that some genetic mechanism must exist to achieve these high frequency variations. The structure of the largest fully characterized gonococcal LPS molecule is shown in Fig. 48.2. To the lipid A are linked two units of keto-deoxyoctulosonic acid (KDO) and two heptoses (HEP).
This inner core region as shown in Fig. 48.2 can carry three oligosaccharide extensions that have been named the _-, _-, and _-chains. The _-chain consisting of N-acetyl glucosamine (GlcNAc) appears to be always present. The _-chain, when present, consists of a lactosyl group; when it is absent, the position is substituted with ethanolamine phosphate. The _-chain in its full form consists of the pentasaccharide shown in Fig. 48.2. An alternative _-chain structure consisting of a trisaccharide is also shown. However, as indicated in Table 48.4 the sugar composition of the _-chain can vary and in every instance it is identical to human cell-surface oligosaccharides, most often part of the glycosphingolipids that in some instances are the determinants of blood-group antigens. Gonococci possess a very unusual sialyl transferase activity, which in vitro is able to use exogenously supplied cytosine monophosphate-NANA (CMPNANA) and add N-acetyl neuraminic acid to the LPS if the organism is expressing the lacto-N-neotetraose _-chain (see Table 48.4). In the human infection in vivo, the concentration of CMP-NANA found in various host environments is sufficient to support this reaction. The sialylation of the LPS causes gonococci to become resistant to the antibody complementdependent bactericidal effect of serum. The resistance is to the bactericidal effect mediated by not only antibodies to LPS, but also to other surface antigens as well. Group B and C meningococci have the capacity to synthesize CMP-NANA as the precursor of their capsule biosynthesis and frequently sialylate their LPS without requiring exogenous CMP-NANA. In the late 1990s, most of the glycosyl transferases responsible for the biosynthesis of gonococcal and meningococcal LPS have been identified, and they are shown in Fig. 48.2. This has provided an understanding of the genetic mechanism that underlies the high frequency variation in the LPS structures expressed by these organisms. Note that four of these genes (lgtA, lgtC, lgtD, and lgtG) are underlined to indicate that they contain in their coding frames homopolymeric tracts of nucleotides. In the case of lgtA, lgtC, and lgtD, these are stretches of 8–17 deoxyguanosines (poly-G) that can vary in size due to errors resulting from slipped-strand mispairing during replication. In lgtG, there is homopolymeric poly-C tract.
When the number of bases in the tracts is such that the coding frames are not disrupted, the respective glycosyl transferases are produced, but, if the number changes, premature termination occurs and no functional enzyme is produced. Thus, the presence of the _-chain depends on whether functional LgtG glucosyl transferase is produced. Similarly in the instance of the _-chain synthesis, if lgtA is on, then the lacto-Nneotetraose chain will be formed and whether the terminal GalNAc is added depends on whether lgtD is on or off. If lgtA is off, then the globoside structure is synthesized if lgtC is on, and only the lactosyl structure is synthesized if lgtC is off. The gonococcus and the meningococcus have evolved a very elegant system to shift readily between a large number of different LPS structures, all of them mimics of human glycolipids. This ability to shift the expression among a number of different LPS structures is not peculiar to the pathogenic Neisseria, but also occurs in Haemophilus influenzae in which at least four genes are subject to phase variation. In this organism, the mechanism is also by slipped-strand mispairing, but occurs in repeated tetrameric sequences that can be either CAAT or GCAA. Thus, it is likely that LPS antigenic variation is important because it is an attribute of a number of mucosal pathogens. How does this molecular mimicry, listed in Table 48.4, benefit the organism? It has been proposed that the human host may find it difficult to produce antibodies to any of these structures and that the ability to change to a different one may compound this problem. Although immune evasion is attractive as an idea, it is clear that the LOS does serve as a target for bactericidal antibodies, and, at least in vitro, perhaps the majority of bactericidal antibodies are directed to this antigen, rather than to other surface structures. In vivo this is, of course, very different because the sialylation of the LOS very effectively inhibits the bactericidal reaction and interferes with phagocytosis as well. However, only the lacto-N-neotetraose structure is effectively sialylated to produce the serum-resistant phenotype. Why does the organism then have a genetic mechanism to alter away from this structure? Perhaps the answer lies in the observation that sialylation of the LOS interferes with invasion of epithelial cells in vitro. There is also evidence that sialylated gonococci are significantly less infectious when used to challenge volunteers. It is clear that the gonococcus can circumvent LOS sialylation, either by the addition of the terminal N-acetyl galactosamine or by the truncation of the chain. It is also possible that the mimicry may benefit the organism by allowing it to be recognized by human carbohydrate-binding molecules such as the C-lectins, the S-lectins, and the sialoadhesins.
IV. NATURAL IMMUNITY A. Bactericidal antibody In the case of the meningococcus, there is clear evidence that the major predisposing factor for bloodstream invasion is the lack of biologically active antibodies to surface components and resultant failure to mediate an antibody–complement bacteriolytic reaction. This was first demonstrated in 1969 by Goldschneider and his colleagues by using two lines of evidence. The first is based on a study done in an adult population. Nearly 15,000 sera were collected from military recruits within the first week of training and stored in anticipation that a number of these would develop meningococcal meningitis during the 8-week basic training. In fact, 60 cases occurred in this cohort and in 54 of these the Neisseria meningitidis causing the infection could be isolated. Each of these sera, as well as 10 sera obtained from unaffected recruits serving in the same training platoons, were tested for bactericidal activity against the strain of Neisseria meningitidis isolated from the patient. Only 5.6% of the patients’ sera were able to kill the disease causing Neisseria meningitidis, whereas 82% of sera obtained from unaffected recruits demonstrated bactericidal activity. The second line of evidence is the demonstration that there is an inverse relationship between the incidence of meningococcal disease and the prevalence of bactericidal antibody, and age. The disease is very rare during the first 3 months of life when maternally derived antibodies are still present. Incidence rises to a peak during between 6 and 12 months of age, when the nadir of bactericidal activity is seen. Thereafter, the incidence progressively diminishes as the prevalence of antibodies rises with age. This is the same relationship that was reported for Haemophilus influenzae meningitis by Fothergill and Wright in 1933. Finally, it is evident that the antibody–complement-dependent bactericidal reaction is clearly important in protection against neisserial systemic infections because deficiencies of late complement components (C6 or C8) impart a specific susceptibility to blood-borne neisserial infections, but not to other bacterial infections. Is there natural immunity to gonococcal infection? It is established that individuals with no known immunological defective can acquire gonorrhea multiple times. In some instances, it has been documented that a single untreated consort may represent the source of the repeated infections. Thus, it has been suggested that there is no such thing as natural immunity to this disease. However, there is another side to the coin and that is the clear evidence that gonococcal infection before the days of antibiotic therapy was as a rule a self-limited disease lasting for a few weeks. This spontaneous elimination of the infection applied not only to the genitourinary disease, but also to disseminated gonococcal infection, to gonococcal arthritis (albeit with bad sequelae), and even in some instances to gonococcal endocarditis. Hence, there is ample evidence that after a period of time gonococci are killed effectively in vivo. In the face of this ability to self-cure, how can we explain the apparent lack of natural immunity? The most likely explanation is that gonococci are inherently so antigenically variable that the immune system requires considerable time to catch up with the repertoire of the gonococcus and eliminate the infection.
V. PREVENTION Since the beginning of the twentieth century, attempts have been made to prepare vaccines for the prevention of meningococcal disease. Vaccines based on whole-cell preparations proved to be ineffective. In the late 1960s and 1970s, methods were developed to purify the capsular polysaccharides of the meningococcus in a form that maintained their high molecular weight. It was shown that injection of school-age children and adults with 25–50_g of group A, C, Y, or W-135 polysaccharide resulted in a strong and longlasting antibody response and that in vitro these antibodies were opsonic and bactericidal. Large-scale field trials both in the United States and overseas demonstrated that both group A and group C polysaccharide vaccines were highly effective in preventing the disease and that the protection lasted for at least 2 years. These vaccines were introduced in the US military over 20 years ago and have essentially eliminated the problem of meningococcal disease among recruits. Vaccination is employed in the military of many other countries and is required for Muslim pilgrims participating in the Hadj. As a general rule, the immune response to purified polysaccharides is age-related, but the response varies with the antigen. Thus, responses to the group C antigen are very low at ages younger than 18 months. Children between 2 and 4 years do respond to the group C antigen, but the response is short-lived, lasting only a few months. After age 6, the responses are similar to adults. As the experience with the Haemophilus influenzae vaccine has demonstrated, the immune responses in this age group can be markedly improved by covalently linking the polysaccharide to a protein carrier to enhance T-cell help in the immune response. Conjugate group C vaccines are being tested. The response to group Aantigen among young children is unusually favorable. Infants who are vaccinated twice, at 3 months and again at 6 months of age will show a brisk booster immune response to the second injection that is sufficient to provide protection. This booster response has not been seen with any other polysaccharide antigen. It has been demonstrated that a protective level of group A antibodies can be maintained by immunization twice in the first year of life, then again at age 2 and upon entry to school. Unfortunately this property of the group antigen has not been taken advantage of in prevention of epidemic disease in Africa. The group B capsular polysaccharide is a homopolymer of _(2–8)-linked N-acetyl neuraminic acid (see Table 48.2). This structure is present in mammalian tissues, notably on the neural cell-adhesion molecule (N-CAM), and the degree of sialylation is particularly elevated during embryonic life. Although the majority of adults have some level of antibodies to this antigen, the injection of the purified antigen generally does not raise additional antibodies. There has also been concern that engendering a strong immune response to this antigen may have deleterious effects on infants during fetal life. Therefore, group B meningococcal vaccines based on partially purified outer membranes with their LPS content reduced by detergent extraction have been prepared and have proved to be able to prevent disease under epidemic conditions. However, as noted before, there is considerable antigenic heterogeneity in meningococcal outer-membrane proteins, and a broadly effective vaccine group B is not available. No vaccine exists for the prevention of gonorrhea, and the problem is formidable because of the extraordinary antigenic variability of this organism. Nevertheless, the experience in several European countries has demonstrated that prevention of this disease can be very effective if public education is combined with rapid treatment of infected individuals and their contacts.
VI. SUMMARY The discrete steps that occur in the mucosal infection by the pathogenic Neisseria are increasingly being explained on molecular and cell biological level. It is evident that the gonococcus has developed very elaborate mechanisms to evade the immune response of human beings. With pili it has chosen the path of antigenic variation. This is an evasion mechanism that is highly developed in eukaryotic parasites such as trypanosomes, and is also seen in prokaryotes such as Borrelia. In the case of Rmp, the gonococcus has chosen the path of antigenic constancy as a target for blocking antibodies. With Opa proteins, the variation may be more a way to succeed in various environments in the host, rather than being an immune evasion. The biological significance of LOS variation is not yet clear, but it must be very useful because Neisseria and Haemophilus influenzae have developed, in principle, the same variation mechanism, although the specific details are quite different. In the era before ready treatment with antibiotics, self-cure of gonorrhea over a period of weeks was commonly seen, and this slow acquisition of natural immunity was probably a reflection of the time needed for the immune response to finally catch up with the variability of the particular strain infecting the human host.
No comments:
Post a Comment