6
Pathogenesis

The pathogenesis of a disease describes the mechanisms by which it develops, progresses, and either persists or is resolved. Understanding pathogenesis of an infectious disease at the cellular and molecular levels is critical for discovering, developing, and implementing methods to prevent infection, and to improve patient outcomes after treatment.

By determining which microbial molecules establish infection by binding to and entering human cells or tissues, for example, scientists can develop vaccines against tick-borne diseases (TBDs)—as they already have for influenza. The pathogenesis of tick-borne diseases can also reveal why some individuals are more prone to severe disease, or fail to resolve infection.

Scientists rely on several methods to study the pathogenesis of TBDs. These include in vitro studies, based on cultured cells; animal studies, based on tracking animals with a disease; and patient studies, based on clinical trials and specimens from biopsies and autopsies. While no one approach can represent the full spectrum and complexity of human disease, the ability to “reduce” or “control” the number of variables by using in vitro and animal models allows more rapid and less equivocal determination of key variables in disease progression—knowledge required to improve prevention, diagnosis, and treatment of TBD in patients.

Animal models have been especially useful in shedding light on the key features of tick-borne infectious diseases. These models include naturally occurring infectious disease, such as neuroborreliosis in horses and Rocky Mountain spotted fever in dogs, and infections introduced into animals such as mice. Mice are particularly helpful in revealing the pathogenesis of infectious disease because scientists can study mice that differ only in a



The National Academies | 500 Fifth St. N.W. | Washington, D.C. 20001
Copyright © National Academy of Sciences. All rights reserved.
Terms of Use and Privacy Statement



Below are the first 10 and last 10 pages of uncorrected machine-read text (when available) of this chapter, followed by the top 30 algorithmically extracted key phrases from the chapter as a whole.
Intended to provide our own search engines and external engines with highly rich, chapter-representative searchable text on the opening pages of each chapter. Because it is UNCORRECTED material, please consider the following text as a useful but insufficient proxy for the authoritative book pages.

Do not use for reproduction, copying, pasting, or reading; exclusively for search engines.

OCR for page 97
6 Pathogenesis The pathogenesis of a disease describes the mechanisms by which it develops, progresses, and either persists or is resolved. Understanding pathogenesis of an infectious disease at the cellular and molecular levels is critical for discovering, developing, and implementing methods to prevent infection, and to improve patient outcomes after treatment. By determining which microbial molecules establish infection by bind- ing to and entering human cells or tissues, for example, scientists can de- velop vaccines against tick-borne diseases (TBDs)—as they already have for influenza. The pathogenesis of tick-borne diseases can also reveal why some individuals are more prone to severe disease, or fail to resolve infection. Scientists rely on several methods to study the pathogenesis of TBDs. These include in vitro studies, based on cultured cells; animal studies, based on tracking animals with a disease; and patient studies, based on clinical trials and specimens from biopsies and autopsies. While no one approach can represent the full spectrum and complexity of human disease, the abil- ity to “reduce” or “control” the number of variables by using in vitro and animal models allows more rapid and less equivocal determination of key variables in disease progression—knowledge required to improve preven- tion, diagnosis, and treatment of TBD in patients. Animal models have been especially useful in shedding light on the key features of tick-borne infectious diseases. These models include naturally occurring infectious disease, such as neuroborreliosis in horses and Rocky Mountain spotted fever in dogs, and infections introduced into animals such as mice. Mice are particularly helpful in revealing the pathogenesis of infectious disease because scientists can study mice that differ only in a 97

OCR for page 97
98 CRITICAL RESEARCH NEEDS IN TICK-BORNE DISEASES single gene, and because they can use imaging to track the progression of infection and cellular trafficking in real time. In this chapter, six scientists presented the state of the science regard- ing the pathogenesis of tick-borne infections—specifically those caused by pathogens in the Anaplasma, Borrelia, Ehrlichia, and Rickettsia genera. PATHOGENESIS OF BORRELIA BURGDORFERI INFECTION AND DISEASE Janis J. Weis, Ph.D., Department of Pathology, University of Utah School of Medicine In humans, the bite of the infected tick is required for introduction of the pathogen through healthy skin. This extracellular pathogen starts in the der- mal tissue where it begins to adapt to life in the mammalian host by changing the expression of its surface glycoproteins. At the same time, the bacterium stimulates responses of inflammatory cells and their secreted mediators that cause acute-phase lesions such as the classical erythema migrans (EM) lesion. The bacterium also activates proteases and other induced host cell molecules to allow for dissemination through the blood and into other tissues, including secondary skin lesions, joints, the heart, and nervous tissue (Coleman et al., 1997; Gebbia et al., 2004; Rosa et al., 2005). Differences in the severity and spectrum of disease among patients in- fected with Borrelia burgdorferi is one of the hallmarks of Lyme disease (Steere and Glickstein, 2004). The reasons for this variation include both genetic differences among strains of the bacterium and differences in the host responses. On the bacterial side, genetically distinct strains, identified by ribosomal spacer types and outer surface protein C (OspC) heterogeneity, have been associated with invasive versus localized cutaneous disease (Wang et al., 2002; Wormser et al., 2008a). Furthermore, B. burgdorferi is charac- terized by a large and complex plasmid content, some of which are essential for infection and others which can vary among strains (Rosa et al., 2005). Similarly, the host response has significant differences in the host response. Among human patients, approximately 60 percent of infected patients who do not receive early treatment develop clinical arthritis (Steere et al., 1987). This difference between those patients who do and those who do not develop arthritis reflects, at least in part, genetic differences in host response. These effects can be studied using inbred strains of mice with defined genetic differ- ences and with clearly reproducible difference in the severity of carditis and arthritis following B. burgdorferi infection (Barthold et al., 1990). The use of gene knockout mice has begun to unravel the genetic contri- bution to the spectrum of the disease. Severely combined immunodeficient C3H/HeJ mice (SCID)—which lack B and T lymphocytes—develop severe

OCR for page 97
99 PATHOGENESIS arthritis and carditis independent of the number of B. burgdorferi used to ex- perimentally infect the mice. In contrast, C57BL/6 mice with the same SCID mutation develop only mild arthritis and carditis, again independent of the infectious dose (Barthold et al., 1992). These results suggest that although B cells and T cells are important in clearing B. burgdorferi, the adaptive im- mune response (mediated by B and T lymphocytes) itself does not drive severe disease. Furthermore, C57BL/6 mice with mild disease and severely affected C3H mice have equal numbers of bacteria in their ankle tissues, evidence that arthritis severity does not correlate with the bacterial load (Ma et al., 1998; Morrison et al., 1999). Understanding how B. burgdorferi traffic to and colonize various tis- sues is important in shedding light on the reasons for differences in the organ-specific manifestations and severity of disease. B. burgdorferi ex- presses outer surface proteins that selectively interact with endothelial cells, platelets, chondrocytes, and extracellular matrix via specific interactions with integrins, glycosaminoglycans, fibronectin, and collagen (Coleman et al., 1997; Gebbia et al., 2004; Coburn et al., 2005). These interactions are important in homing to and colonization of tissues, including the skin, joint, and heart. Bacterial ligands, such as DBPA/B, p66, BBk32, and OspC, promoting heart and joint invasion have been identified by genetic and im- munological techniques (Coburn et al., 2005). These receptor-ligand inter- actions also contribute to inflammatory responses in resident cells (Behera et al., 2005, 2006a). The host response to B. burgdorferi plays a key role in disease patho- genesis. B. burgdorferi does not produce toxins or proteases that are di- rectly responsible for tissue damage upon colonization. In contrast, the bacterium produces multiple molecules that activate host responses and can lead to localized and generalized inflammatory pathogenic responses. Most of these host responses normally function to contain or clear infections and are components of the innate defense and/or inflammatory response (Liu et al., 2004; Benhnia et al., 2005; Behera et al., 2006a; Oosting et al., 2010). Although their purpose is to clear infection, if continually activated, they lead to lesion development and disease. Numerous signaling pathways have been identified that are responsible for Lyme disease arthritis (see Table 6-1). The Pam3Cys-lipid-modified pro- teins of B. burgdorferi, which are abundantly expressed by the bacteria, are the best characterized. These lipoproteins interact with the host, specifically through toll-like receptor (TLR)-2 and TLR-1 heterodimers, and activate signaling through the adaptor molecule MyD88. This signaling pathway results in activation of numerous proinflammatory cytokines, chemokines, and matrix metalloproteinases (Hirschfeld et al., 1999; Alexopoulou et al., 2002). In addition, the bacterial flagellin and peptidoglycan also activate host TLRs, again connecting to the MyD88 pathway (Bolz et al., 2004;

OCR for page 97
100 CRITICAL RESEARCH NEEDS IN TICK-BORNE DISEASES TABLE 6-1 Numerous Signaling Pathways Have Been Implicated in Pathogenesis Signaling B. burgdorferi Ligand Host Receptor Pathway Type of Response Pam3Cys- TLR2/TLR1 MyD88 Pro-inflammatory cytokines, outer surface & NF-kB MMPs chemokines, anti- lipoproteins dependent, MAP inflammatory (1, 2, 18) (Osps) Kinases Flagellin TLR5 MyD88 Cytokines, etc. (33) Peptidoglycan TLR2, NOD2 MyD88 Cytokines, etc. (29) Type I IFN (α/β)(26, 30) RNA TLR7 & 2nd MyD88 unidentified PRR dependent and IRF 3 dependent Type I IFN (α/β)(26) Secreted Unknown IRF3 molecules α3β1 integrin BBB07 Endosome Cytokines, MMP(5) IL-2, IFNγ(19, 20) Diacyglycolipid- CD1d iNKT-TCR BbGL-IIc Liu et al., 2004; Behera et al., 2006a; Shin et al., 2008). The result of this MyD88 stimulation is production of pro-inflammatory products such as cytokines, chemokines, and matrix metalloproteinases. Using knockout mice that lack individual components of the pathways, the contribution of specific pathways to control of infection and resolu- tion of disease can be studied. The TLR/MyD88 pathway is important for host defense and for controlling the bacteria numbers in tissues. However, knockout mice that lack either TLR2 or MyD88 still develop arthritis (Wooten et al., 2002; Bolz et al., 2004; Liu et al., 2004; Behera et al., 2006a). These results suggest that although this pathway is important for host defense and control of the bacterial numbers, it is not essential for arthritis development. Consequently, global gene expression profile analysis was used to find pathways specific to arthritis development (rather than host defense) in C3H and C57BL/6 mice. These experiments focused on early time points in pathogenesis, 1 week after infection prior to the ar- rival of inflammatory cells in joint tissue. As noted previously, C3H mice had severe ankle swelling beginning at week 1, while C57BL/6 mice had minimal swelling. In C3H mice, there is an early and transient induction of genes associated with an interferon signature profile at one week that decreases by week 2. Furthermore, immunodeficient (lacking interleukin-10 or IL-10) C57BL/6 mice showed a delayed increase in the same interferon panel that remained elevated through the infection (Crandall et al., 2006). One hypothesis was that type I interferons (IFNs), which are normally associated with host response to viral infections, are important for the devel- opment of arthritis following B. burgdorferi infection. To test that hypothesis,

OCR for page 97
101 PATHOGENESIS the C3H mice were treated with an antibody that blocked the receptor for type I IFNs. Arthritis was reduced by 50 percent in these C3H mice given a single injection of interferon-blocking antibody before infection (Miller et al., 2008). A second study with mutant CH3 mice deficient in interferon receptors confirmed the involvement of type I IFNs. These studies provide functional evidence for the involvement of type I IFNs in the development of arthritis. This result was unexpected as most bacteria known to induce host type I inter- feron are intracellular, whereas B. burgdorferi are extracellular. Notably, this type I interferon pathway can be induced by at least three distinct ligands from B. burgdorferi, some of which function independently from the MyD88 adaptor molecule pathway (Petzke et al., 2009; Miller et al., 2010). Importantly, the host responses and inflammatory pathways have organ- specific differences. In contrast to arthritis, which is characterized by infil- tration of neutrophils, carditis is characterized by influx of macrophages and T lymphocytes at the base of the heart where B. burgdorferi infiltrates connective tissue (Barthold et al., 1990; Ruderman et al., 1995; Bockenstedt et al., 2001). Also unlike arthritis, the numbers of infectious bacteria in the heart are correlated with the severity of inflammation (Morrison et al., 1999). Furthermore, T lymphocytes help resolve Lyme disease carditis by produc- tion of interferon gamma and other cytokines, which in turn activate the macrophages to clear B. burgdorferi from the heart, and therefore suppress the carditis (McKisic et al., 2000; Olson et al., 2009). Disruption of this interferon gamma pathway in C57BL/6 mice results in more severe carditis. Knowledge Gaps and Research Opportunities Weis noted three key questions for future study: • What is responsible for the variability in individuals’ response to infection by Borrelia burgdorferi? • Why do some symptoms persist in some patients? • What is responsible for the pathogenesis of neuroborreliosis in patients? DURATION OF SPIROCHETE INFECTION FOLLOWING ANTIBIOTIC TREATMENT IN ANIMALS Linda K. Bockenstedt, M.D., Yale University School of Medicine Bacterial infections have a number of outcome determinants, including pathogen factors, host genes, host co-morbid conditions, host immunity, and the effects of antibiotics. A current debate is how effective antibiot- ics are in vivo against B. burgdorferi. Antibiotic treatment failures occur

OCR for page 97
102 CRITICAL RESEARCH NEEDS IN TICK-BORNE DISEASES occasionally in all animal models of Lyme borreliosis and in humans. Evi- dence for treatment failure comes from persistence of B. burgdorferi DNA after treatment for Lyme disease arthritis and, in animal models, the ability to culture spirochetes from tissues. Several studies have found that B. burg- dorferi DNA can be detected in tissues for extended periods of time after antibiotic treatment of laboratory animals even though cultures of tissues may be negative (Straubinger et al., 1997; Bockenstedt et al., 2002; Hodzic et al., 2008). This raises the question of how to interpret the significance of B. burgdorferi DNA in tissues. In our published study (Bockenstedt et al., 2002), we used xenodiagno- sis with ticks to determine whether bacterial DNA detected in mouse tissues after antibiotic treatment indicated the presence of spirochetes that could replicate and cause infection. In this approach, uninfected laboratory-reared ticks were allowed to feed on mice that had previously been infected with B. burgdorferi and treated with antibiotics (doxycycline or ceftriaxone). Immunofluorescent staining was then used to determine whether the ticks had acquired B. burgdorferi spirochetes. The results were equivocal in that spirochete forms were detected microscopically in the tick midguts, but, on further study, these spirochetes appeared to be attenuated because genes on specific plasmids required for B. burgdorferi infectivity could not be detected by polymerase chain reaction (PCR) (Bockenstedt et al., 2002). Similarly, spirochetes could not be cultured from tissues of mice that had been treated with antibiotics. No other method was used to assess vi- ability and infectivity of the spirochetes visualized in ticks. However, larval ticks that had fed on antibiotic-treated mice could not transmit B. burg- dorferi infection to uninfected mice. In a subsequent study (Hodzic et al., 2008), ticks used for xenodiagnosis of ceftriaxone-treated mice were able to transmit B. burgdorferi DNA to uninfected immunodeficient SCID mice, which lack T and B cells, and are highly susceptible to B. burgdorferi in- fection. In addition, B. burgdorferi DNA could be detected in some SCID mice that had received tissue transplants of skin from antibiotic-treated mice. Although viable spirochetes could not be cultured from the ticks, the antibiotic-treated donor mice, or the recipient mice, rare spirochete forms were visualized microscopically in specific connective tissues of some donor mice (Hodzic et al., 2008). These findings raised two questions. First, does the persistent DNA indicate continued infection, or is it simply residual debris? Second, are the rare residual spirochete forms viable and infectious? To begin to address these questions, we used two-photon (multiphoton) confocal microscopy to directly visualize the location and motility of spiro- chetes in living, anesthetized mice in real time. C57BL/6 MyD88-deficient mice were infected with B. burgdorferi strain 297, a strain previously iso- lated from a Lyme disease patient but subsequently genetically modified to express a green fluorescent protein. The C57BL/6 mice are relatively disease resistant, but deficiency in MyD88 results in higher bacterial load so that

OCR for page 97
103 PATHOGENESIS infected mice have 100- to 1,000-fold more spirochetes in the skin and other organs, thus enhancing the ability to image the spirochetes. After 21 days of infection to allow dissemination of spirochetes throughout the tis- sues, mice were treated with either ceftriaxone or doxycycline. Multiphoton microscopy was then used to image spirochetes in ear skin and tendons, two easily accessible sites where spirochetes often reside. After imaging, mouse tissues were tested by culture to detect viable organisms, PCR to detect residual spirochete DNA, and direct fluorescent antibody (DFA) staining to detect residual spirochete antigen. In the experiments using ceftriaxone, mice were treated twice daily with ceftriaxone for 5 days or sham treated. We began analyzing mice during the antibiotic treatment period and up to 9 days after completion. Spiro- chetes could not be cultured from tissues of mice treated with antibiotics at any time point. Multiphoton microscopy revealed a large number of spirochetes moving in the ear skin of sham-treated mice, but after just two doses of ceftriaxone (1 day of treatment), only a few spirochetes remained in antibiotic-treated mice. Fewer spirochetes were seen in the tendons of sham-treated mice, and only two stationary spirochetes were found in the tendon of one mouse treated with two doses of ceftriaxone (Bockenstedt et al., 2011). The spirochetes in the tendons of sham-treated mice were less motile than those in the skin. With the exception of mice analyzed after one day of antibiotic treatment, no spirochetes could be visualized in ear skin or tendons of antibiotic-treated mice at any time. At the end of the experiment, however, DFA revealed green fluorescent material adjacent to the ear cartilage in all of the antibiotic-treated mice. In the doxycycline experiments, mice were given a one-month course of antibiotics supplied in drinking water to maintain serum levels above the minimal concentration necessary to inhibit B. burgdorferi growth in vitro. This method of antibiotic administration sustains therapeutic serum drug levels analogous to levels achieved in humans treated with oral doxy- cycline. Mice were analyzed between 2 and 10 weeks after the last dose of antibiotics. Similar to the ceftriaxone-treated mice, spirochetes could not be cultured from mice treated with doxycycline. Ticks used for xenodiagnosis also tested negative by culture after feeding on antibiotic-treated mice. In sham-treated (control) mice, multiphoton microscopy revealed motile spi- rochetes in the skin, as well as large, amorphous collections of fluorescent debris near ear cartilage, a finding not seen in uninfected mice. Similar non- motile fluorescent material adjacent to the ear cartilage was also visualized in the treated mice (Bockenstedt et al., 2011), and these mice tested positive for B. burgdorferi DNA by PCR in both skin and joints. To determine whether the amorphous fluorescent material contained viable and infectious spirochetes, ear tissue was transplanted into MyD88- deficient mice, which may provide a “permissive” environment for at- tenuated organisms. When analyzed up to 5 months after the transplant,

OCR for page 97
104 CRITICAL RESEARCH NEEDS IN TICK-BORNE DISEASES sera from mice transplanted with tissues from antibiotic-treated mice only showed reactivity to single bands on immunoblots. Sera from mice trans- planted with tissue from sham-treated mice, in contrast, showed a banding pattern typically found in mice and humans infected with B. burgdorferi. Neither viable spirochetes nor spirochete DNA were observed in mice that had received tissue transplants from treated mice. In contrast, mice that received ear transplants from sham-treated mice did have spirochete DNA and were culture positive (Bockenstedt et al., 2011). In a separate experi- ment, infected immunocompetent C57BL/6 mice given a month-long course of oral doxycycline were evaluated similarly for persistent infection by skin transplantation into MyD88-deficient mice. Only one of the five mice that received ear transplants from an antibiotic-treated donor mouse developed a serologic response to B. burgdorferi as indicated by a single immunoblot band. Mice that received transplants from sham-treated donor mice, in con- trast, developed evidence of infection, as revealed by tissue culture, PCR, and serologic conversion (Bockenstedt, unpublished observations). Bockenstedt noted that a number of conclusions can be drawn from this work: • Antibiotics are effective in eliminating B. burgdorferi infection in im- munocompetent C57BL/6 mice and even immunodeficient C57BL/6 MyD88-deficient mice. Because C57BL/6 mice are relatively disease resistant, similar studies are in progress in immunocompetent and MyD88-deficient C3H mice, a mouse strain background that is more susceptible to B. burgdorferi-induced infection and disease. • Spirochete debris may persist for some time after B. burgdorferi- infected MyD88-deficient mice are treated with antibiotics. More extensive analyses need to be performed to determine whether such debris occurs in different tissues throughout the host and whether this debris could serve as a nidus for stimulating inflammation. • Tissue transplants containing the debris may elicit an antibody re- sponse in the new host, but if the donor mice were treated with antibiotics, the tissue transplants induce responses to only one or two B. burgdorferi proteins. Only mice that receive transplants from sham-treated mice become infected, as shown by tissue culture, PCR, and seroconversion. Transplants from mice treated with antibiotics do not introduce infection into recipient mice. DISCUSSION The discussion session focused on the pathogenesis of Lyme disease and how these studies inform us about human disease. A participant questioned if the spirochetes were able to change their morphology and “hibernate”

OCR for page 97
105 PATHOGENESIS under periods of stress. Bockenstedt noted that there are changes in the morphology of spirochetes when they are exposed to different stressful conditions in culture. For example, if the spirochetes are nutrient deprived, they can actually stop making peptidoglycan and fold up on themselves. However, the same phenomenon has not been observed in vivo. Schutze questioned whether the observed residual pieces of organisms could stimulate an inflammatory cascade. Bockenstedt noted that this ma- terial does contain DNA, which might stimulate an inflammatory cascade through TLRs. The fluorescent material in the tissue specimens has not been isolated for testing, and there may be other components in the sample that would trigger inflammation. However, the current research has not been able to answer these questions. Another participant inquired about the genetic differences between the C57BL/6 and the C3Hmice and what these differences meant for TBDs. Dr. Weis noted that some genes transcend the genetic differences among the mice strains, for example, MyD88, TLR-2, and interleukin 10. Mutations in those genes on either the C57BL/6 or the C3H background would result in a compromise either in host defense, resulting in elevated levels of bacteria in tissues, or in the case of IL-10, an increase in inflammation (Brown et al., 1999, 2008; Wooten et al., 2002). The MyD88-deficient knockout mice are particularly interesting because the bacterial number in tissues is significantly elevated compared to bacteria loads in wild-type mice (Bolz et al., 2004). She further noted that there is no indication of a change in the bacteria, but rather that they persist longer because the host cannot clear the bacteria. Research is currently being done to understand the difference in ar- thritis severity between C57BL/6 and C3H mice through the use of genetic intercross populations. Weis noted that she has identified at least 12 differ- ent loci that are different between these two mice strains (Ma et al., 2009). One area of research will be to look for genes that regulate type I interferon because there is a difference in the induction of type I interferon in infected joints between the two mouse lines. ANTIGENIC VARIATION AS A MECHANISM FOR PERSISTENT BORRELIA INFECTION Steven J. Norris, Ph.D., Department of Pathology & Laboratory Medicine, University of Texas Medical School at Houston Pathogens can vary in their ability to be invasive and toxic to an or- ganism (see Figure 6-1). For example, Clostridium botulinum is very toxi- genic and produces a powerful neurotoxin, but is not invasive. In contrast, C. perfringens, which causes gas gangrene, is both highly invasive and

OCR for page 97
106 CRITICAL RESEARCH NEEDS IN TICK-BORNE DISEASES FIGURE 6-1 The relationship between invasiveness and toxigenesis. SOURCE: Norris et al., 2010. highly toxigenic. A group of pathogens, including Treponema pallidum and Mycobacterium tuberculosis that produce no known toxins, are highly Figure 6.1 invasive organisms that can persist for the lifetime of the host. R01965 Lyme disease Borrelia is a highly motile organism. In animal models, bitmapped B. burgdorferi disseminates early during infection into numerous tissues, including skin, joints, heart, bladder, and spleen, and persists in the tissues for up to 2 years. The persistence of infection in humans is not well under- stood, but likely can last months to years; however, the Borrelia produces no known toxins or enzymes that cause tissue damage. Thus, Lyme disease Borrelia falls into the group of highly invasive, non-toxigenic pathogens. To cause persistent infection, B. burgdorferi must have multiple ways of evading a host’s immune response (see Norris et al., 2010). One common mechanism is protective niches through sequestration of the pathogen in dense tissue, such as tendons. A second cellular process is through down- regulation of antigen expression. During infection of the mammalian host, B. burgdorferi down-regulates the expression of the surface protein antigen OspA. This protein is important during the tick part of the B. burgdorferi life cycle, but the organism usually does not express OspA at high levels during mammalian infection. A third mechanism is the inhibition of a

OCR for page 97
107 PATHOGENESIS host’s innate immune response. For example, B. burgdorferi inhibits the complement cascade by complement regulator-acquiring surface proteins. A fourth mechanism is antigenic variation, which causes a change in a surface structure that usually occurs at a higher rate than expected from mutation. Antigenic variation and, specifically, the vls gene system will be the focus of this presentation. A 28-kilobase linear plasmid of B. burgdorferi B31 called lp28-1 con- tains a variable membrane protein-like sequence locus, which resembles a similar system in the relapsing fever spirochete, a prototypical antigeni- cally variant pathogen. The plasmid contains both an expression site called vlsE and a set of silent cassettes upstream from vlsE. Alignment between the expression site and the silent cassettes reveals regions of sequence identity or relative invariant sequence and other regions of variation. Ap- proximately 92 percent of the genetic sequences of the silent cassettes are identical to those of the central part of the expression site. In contrast, the areas of variation are important in determining the structure of the antigen expressed by the vls system. The initial hypothesis was that each of the silent cassettes could ex- change into the expression site, and therefore result in approximately 15 variants of the antigen. Subsequently, however, segmental recombination via a gene conversion mechanism was discovered in which the silent cas- settes donate genetic sequences of different lengths and locations into the expression site. This recombination event appears to occur only within the mammalian host and has not been detected in standard liquid culture or in ticks. The recombination process continues as long as mammals are infected and can theoretically produce as many as 1032 different sequences of amino acids. Most variations in sequences consist of only one or two amino acids. In fact, so much variation occurs that it is rare to find the same vlsE sequence twice in a given tissue 28 days after infection (Zhang et al., 1997; Zhang and Norris, 1998; Coutte et al., 2009). The VlsE pro- tein is anchored to the outer membrane of the organism, with the variable regions being accessible on the surface of the protein. Thus, the sequence differences in those regions provide a mechanism so that the organism can effectively evade the immune response through continuously changing the amino acid sequence of the exposed region. The invariant regions also elicit a host immune response during infec- tion, one of which is now used to diagnose Lyme disease (see Chapter 8). The IR6 region of the protein, also called C6, induces a particularly high antibody response in humans and other animals. This region is not the only reactive invariant region, but the one that is best characterized. Overall, it is not understood how this protein can permit evasion of the immune system while also inducing a high antibody response. In a landmark study, the portion of the B. burgdorferi plasmid lp28-1

OCR for page 97
114 CRITICAL RESEARCH NEEDS IN TICK-BORNE DISEASES These studies have provided clues to genes involved in hypertension, diabe- tes, and other conditions regulated by a complexity of traits. PATHOGENESIS OF EHRLICHIA AND ANAPLASMA INFECTION AND DISEASE Nahed Ismail, Ph.D., M.Sc., Department of Pathology, Meharry Medical College Ehrlichia and Anaplasma are small obligate, intracellular gram-neg- ative bacteria with a characteristic dimorphic appearance and cell wall ultrastructure. They reside in cytoplasmic endosomes generally within he- mopoietic cells that have evolved in close association with ticks and res- ervoir hosts. There are several species of Ehrlichia and Anaplasma, with E. chaffeensis being the causative agent of human monocytic ehrlichiosis (HME) and A. phagocytophilum being the causative agent for human gran- ulocytic anaplasmosis (HGA). In the mammalian hosts, including infected humans, the primary target cells of E. chaffeensis and A. phagocytophilum are, respectively, monocytes and neutrophils. E. chaffeensis is a small bacterium with a 1.2–1.5 mb genome. Unlike classical gram-negative bacteria, this pathogen lacks both peptidoglycan and lipopolysaccharide (LPS), but it uses cholesterol acquired from the host to maintain membrane integrity. A similar mechanism is used by Anaplasma. The E. chaffeensis P28 outer membrane protein family stimu- lates specific antibody responses in humans. P28 is also immunoprotec- tive: Antibodies against it protect mice from fatal infection. However, the presence of a large family of P28 proteins may also enable the bacteria to evade the host’s immune system and adapt to different hosts such as ticks and mammals. Several secreted Ehrlichia proteins have tandem repeats as- sociated with interaction between the pathogen and the host. Furthermore, several proteins with eukaryote-like ankyrin domains, which influence tran- scription and translation of genes in the host, have been described. The mechanism for delivering secreted proteins into the host cell cytosol is not completely understood but in part uses the type IV secretion system (TFSS). Ehrlichia and Anaplasma have developed mechanisms for evading a host’s immune response. For example, Ehrlichia down-regulates cytokines essential for stimulating a protective Th1 phenotype of acquired immune response and subsequent elimination of the bacteria. These cytokines in- clude IL-12, IL-15, IL-18, and MHC class II. Ehrlichia and Anaplasma also down-regulate the TLR2 and TLR4 receptors that the innate immune system uses to recognize and respond to Ehrlichia. Furthermore, Ehrlichia and Anaplasma also down-regulate several bactericidal mechanisms of monocytes and neutrophils, including degradation of p22phox, inhibition

OCR for page 97
115 PATHOGENESIS of superoxide generation, and inhibition of phagolysosomal fusion. To survive and replicate inside cells, the bacteria induce apoptotic inhibitors or decrease expression of apoptotic inducers. As discussed in the previous chapter, HME can manifest as either a mild, self-limited disease or a severe fulminate disease with a toxic shock– like syndrome. Patients with severe HME usually have multiorgan dysfunc- tion that progresses to multiorgan failure. Although their symptoms may be nonspecific, patients with severe HME also present with marked leukope- nia, lymphopenia, marked thrombocytopenia, and elevated liver enzymes. There is a disconnection between the number of bacteria in the blood of HME patients and the severity of the disease. This suggests that the pathogenesis of the disease and the outcome of infection have a significant immune-mediated component. The first task is therefore to characterize the molecular and cellular immune mechanisms that contribute to Ehrlichia- induced toxic shock. The long-term goal is to develop both a vaccine and an immune-based therapy. A well-established fact is that protective immunity against several in- tracellular bacteria is mediated by Th1 cells that promote cell mediated immune responses (O’Garra and Murphy, 2009). Stimulation of T cells occurs when bacteria is phagocytosed by the host antigen-presenting cells (APCs), processed into small peptides and presented to naïve CD4+T cells and CD8+ T cells in the context of MHC class II and I, respectively. Fol- lowing activation, T cells differentiate into either Type-1 or Type-2 cells depending on costimulatory signals cytokine environment. Intracellular bacteria stimulate IL-12 production by APCs to induce Th1-type cells that produce large amounts of IFN-γ. IFN-γ produced by Th1 cells activate macrophages to kill the bacteria, activate the bactericidal mechanisms of neutrophils, and enhance or stimulate an antibody response, mainly IgG2a antibodies. The latter allows opsonization (i.e., engulfing and digesting) of extracellular bacteria, and the killing of intracellular bacteria. Immunocompetent mice have been used to understand how host de- fenses interact with the bacteria and contribute to resolution or progression of disease. Although E. chaffeensis does not accurately recapitulate human infection and disease, the related Ixodes ovatus Ehrlichia (IOE), which is highly virulent, and Ehrlichia muris, which is mildly virulent, do recapitu- late key features and have been used in a series of elucidating studies. The disease is dose dependent. For example, mice receiving a large dose of IOE intradermally died on day 10 post-infection, while mice receiving a low dose of IOE survived (Stevenson et al., 2006). Specifically, the mice that died developed focal necrosis, apoptosis, and toxic shock–like syndrome. However, as in humans, despite severe tissue injury, these mice did not have evidence of overwhelming infection. The immune mechanisms responsible for fatal disease were examined.

OCR for page 97
116 CRITICAL RESEARCH NEEDS IN TICK-BORNE DISEASES CD4 T cell proliferation and the frequency of CD4 Th1 cells were de- creased, which, as noted, are very important in clearing Ehrlichia and intra- cellular bacteria from the host. The mice that died also had a concomitant increase in proinflammatory and anti-inflammatory cytokines TNF-alpha and IL-10—both implicated in tissue injury (Ismail et al., 2004, 2006). Fur- thermore, there was a marked expansion of CD8 T cells producing TNF- alpha in the mice that died. Mice that lacked CD8+T cells survived a lethal low-dose infection with IOE compared to similarly infected wild-type mice (Ismail et al., 2007). Survival of IOE-infected CD8+T cell deficient mice was associated with enhanced bacterial elimination, increased numbers of CD4+Th1 cells, decreased TNF-alpha production, and decreased tissue injury. These data suggest that CD8 T cells play a pathogenic role during severe and fatal monocytic ehrlichiosis by mediating apoptosis of CD4 T cells, decreasing Th1-type responses, and immunopathology. Although fatal ehrlichiosis is associated with an increase in patho- genic CD8+T cells, which possibly mediate leukopenia and low CD4+T cell count, the mechanism by which Ehrlichia induced this pathogenic response is not yet known. It is well known that early interactions occur between the host’s antigen-presenting cells and innate lymphocytes, such as when natural killer (NK) and natural killer T (NKT) cells influence the subsequent acquired immune response against intracellular pathogens. Unlike conventional CD4+ and CD8+T cells, NKT cells recognize endog- enous host self-ligands as well as foreign microbial ligands (e.g., glycolip- ids, lipoprotein, or even cholesterol) presented by antigen-presenting cells through a receptor called the CD1d molecule, which is a non-polymorphic MHC class I-like molecule. For gram-negative bacteria that have LPS such as Salmonella, NKT cells are stimulated via signals mediated by endogenous self-ligands presented in the context of CD1d and signals generated by toll-like receptors (TLRs are pattern recognition receptors). In contrast, alpha protobacteriae, including Ehrlichia that lack LPS, ap- pear to have a specific bacterial ligand that directly stimulates NKT cells (Mattner et al., 2005). NKT cells are essential for eliminating the bacteria, thus NKT-deficient mice succumb to an overwhelming bacterial infection (Stevenson et al., 2006). Moreover, NKT cells prevent chronic joint in- flammation after infection with Borrelia. A recent study has shown that Lyme disease patients seem to have a low number of NKT cells and low migration of those cells to joints, which was postulated to be an etiologic factor that contributes to arthritis in Lyme disease patients. One recom- mendation from this study was to propose enhancing the stimulation of NKT cells, and their migration to peripheral tissues, so they would sup- press joint inflammation (Tupin et al., 2008). One question that remained was whether NK cells are functionally

OCR for page 97
117 PATHOGENESIS similar to NKT cells, or whether they have different roles during infection. In fatal ehrlichiosis, NK cells expand in the liver by day 7 post-infection and produce most of the cytokines produced during fatal ehrlichiosis, in- cluding TNF-α, IFN-γ, and IL-10 (Stevenson et al., 2010). Furthermore, these NK cells are also highly cytotoxic. The next step was to show a causal association between NK cells and development of immunopathol- ogy and fatal disease. After depleting NK cells from mice, there was a significant decrease of the systemic cytokine production, mainly IL-10 and TNF-alpha, and a decreased number of apoptotic cells and necrotic foci, suggesting that NK cells directly or indirectly mediate tissue injury during fatal ehrlichiosis. Even more surprising, the absence of NK cells enhanced the elimination of bacteria (Stevenson et al., 2010). That suggested, con- versely, that NK cells inhibit effective elimination of bacteria. Together, those findings suggest that interaction of virulent Ehrlichia with antigen- presenting cells following high-dose lethal infections results in strong stimulation of cytotoxic NK cells. These data suggest that NK cells are possibly the main inducers of the harmful/pathogenic immune responses seen in ehrlichiosis, including generation of pathogenic CD8+T cells and development of CD4+Th1 hyporesponsiveness. The mechanism by which NK cells promote pathogenic responses following ehrlichial infection is not completely known, however, it is possible that this occurs via stimu- lation of IL-10 production, as well as via pro-inflammatory cytokines. Human patients with fatal HME had increased Th2 immunosuppressive cytokines, mainly IL-10, as compared to those with mild disease (Ismail et al., unpub. data). Patients with fatal outcomes also had a higher level of NK and monocyte chemokines, IP-10 and MCP-1, and decreased T cell chemokines, including RANTES. In addition, these patients had increased pro-inflammatory cytokines, including IL-1 alpha, IL-6, and TNF-alpha, and increased neutrophil chemokines, including IL-8 and granulocytic colony-stimulating factor. In conclusion, cytokine dysregulation and expansion of pathogenic NK and CD8 T cells are the main immunopathological mechanisms in ehrlichi- osis that mediate tissue injury and organ dysfunction. On the other hand, hyporesponsiveness of CD4+ T cells, decreased number of CD4+T cells, and a late-stage apoptosis (programmed cellular death) of CD4+T cells also contribute to severity of disease, possibly by failure to control continuous microbial stimulation of NK and CD8+T cells. These findings are consistent with Dumler’s findings in a murine model of HGA, in which pathogenic in- nate responses consisting of uncontrolled macrophage activation, NK, and NKT play a role in the immunopathology caused by Anaplasma phagocy- tophilum infection in mice.

OCR for page 97
118 CRITICAL RESEARCH NEEDS IN TICK-BORNE DISEASES Knowledge Gaps and Research Opportunities Ismail noted the following areas are critical ones for future study: Understanding the Bacteria • The regulatory mechanisms that control the developmental cycle of E. chaffeensis; • Proteomic analysis of the biphasic forms of E. chaffeensis, to identify the determinants of invasiveness and virulence; • The mechanistic details of how the T4SS and other secretion mecha- nisms secrete Ehrlichia and Anaplasma effectors, and their subcel- lular sites of action; and • Identification of effector candidates—including ankyrin-motif bear- ing proteins and cognate partners secreted via T4SS or other secre- tion apparatus. This will provide a molecular basis for understanding pathogen subversion of host defense, and disease. Understanding the Host • Immune defense mechanisms and regulation at the peripheral sites of tick-borne Ehrlichia infection, such as the skin, liver, and lung; • The relative contribution of specialized Langerhans cells, hepato- cytes, Kupffer cells, and endothelial cells to immune surveillance, immunity, and pathology; • Local factors influencing dendritic cell, NK, and T cell recruitment and differentiation; • The mechanisms controlling the cross-presentation of endosome/ phagosome-derived Ehrlichia antigens to CD8+T cells; and • The role of regulatory T cells in controlling immune responses to Ehrlichia. Potential Therapeutics • Molecular and cellular profiles of mild and fatal infections in pa- tients with HME; • Collection of human samples, such as blood, cerebral spinal fluid, and tissues; • Development of screening tests, including biomarkers, to identify individuals at early stages of infection, and those at risk for progres- sive disease; • Studies of the efficacy of highly promising interventions in animal models of disease; and

OCR for page 97
119 PATHOGENESIS • Characterization of host defenses and immune responses in models of tick-transmitted Ehrlichia and Anaplasma infections that mimic mild and severe HME and HGA. PATHOGENESIS OF RICKETTSIAL INFECTIONS Gustavo Valbuena, M.D., University of Texas Rickettsia are small, obligate intracellular bacteria in the Class α-Proteobacteria. Ticks serve as both vectors and primary reservoirs of spotted fever group Rickettsia as they can transmit the Rickettsia between stages and transovarially to the next tick generation. In general, small mam- mals act as amplifying hosts and human infections are accidental. The main target tissue in mammalian hosts, including humans, is the endothelium, which lines the interior of the vascular system. Taxonomically, Rickettsia can be subdivided into four groups: typhus, spotted fever, transitional, and ancestral. In North America, the spotted fever group and the typhus group are of most concern. Rickettsia rickettsii cause the most severe rickettsiosis in North America, Rocky Mountain spotted fever. However, the newly discovered R. parkerii also produces an important disease syndrome, although apparently less severe than that produced by R. rickettsia. Ticks can survive for long periods while harboring Rickettsia, although the organisms may decrease the fitness of the tick. When a tick attaches to a host, the Rickettsia are “reactivated”—a poorly understood process that requires 12 to 18 hours and results in the Rickettsia acquiring an infectious phenotype. Because hard ticks take several days to feed on a vertebrate host, they produce substances that inhibit the host’s immune and coagula- tion systems, possibly allowing infection to become established. Rickettsia enter endothelial cells rapidly through a process of recep- tor-induced endocytosis: enzymes produced by the bacteria rapidly lyse the endocytic vacuole and move into the cytoplasm, where they replicate (Weiss, 1973). Several mechanisms are likely involved in damaging the en- dothelium. The first is cell death, necrosis (Silverman, 1984), in which the replicating Rickettsia lyse the cell. Second, there is evidence of increased oxidative stress as cells respond to the intracellular infection (Rydkina et al., 2004). Third, although cells could also die through apoptosis, there is evi- dence that rickettsia can inhibit apoptosis to favor its own survival (Bechelli et al., 2009). In addition, during the infection, Rickettsia induce increased production of nitric oxide and several lipid mediators derived from the cyclooxygenase system, particularly COX-2 (Rydkina et al., 2010). Once infection is established, endothelial cells acquire an activated

OCR for page 97
120 CRITICAL RESEARCH NEEDS IN TICK-BORNE DISEASES phenotype that can trigger coagulation and activate the host’s inflammatory response. The endothelial cells can express cytokines and other immuno- modulatory substances, and adhesion molecules, which recruit leukocytes to the infection sites (Valbuena and Walker, 2009). The inflammatory response is mediated, in part, through NF-kappa B, an important transcription factor that regulates many immune response genes (Sahni et al., 1998). Other mecha- nisms in the host’s inflammatory response include synthesis of inflammatory cytokines, including IL-1 and IL-8. When endothelial cells interact with im- mune cells, especially if the latter are producing interferon gamma and TNF- alpha, as in the case of NK cells or CD8 T cells, the endothelial cells become activated and kill Rickettsia. A goal is the harness the mechanisms that allow the endothelium to kill the pathogen for use in treating disease. There are a number of reasons why Rocky Mountain spotted fever often becomes severe and results in a high case fatality rate. Endothelial cells normally form a barrier in the vasculature and balance the movement of fluid between the intravascular and extravascular spaces. The disruption of this barrier due to rickettsial infection of the endothelium affects these functions and results in leakage of fluid. When this occurs in organs such as the brain or lungs, the disease can rapidly progress. Rocky Mountain spot- ted fever may also be severe because it is a systemic infection involving cells that regulate the coagulation and immune systems. Furthermore, clinicians often confuse Rocky Mountain spotted fever with viral illnesses—for ex- ample, influenza in North America and dengue fever in Latin America. This confusion can have severe consequences because early suspicion of spot- ted fever can result in effective treatment with the antibiotic doxycycline. However, once patients develop the full spectrum of disease, physicians may refer them to higher level hospitals, which may treat the patients with newer broad-spectrum antibiotics—to which R. rickettsia are frequently constitutively resistant. Another barrier to combating Rocky Mountain spotted fever is that current diagnostic tests rely on antibodies, which are produced after the infection has already disseminated. Knowledge Gaps and Research Opportunities Valbuena noted that the number of key areas for future study include: • Determination of the mechanism by which Rickettsia are reactivated in the tick to an infectious state. (The fact that the bacteria must be reactivated allows for public health intervention. For example, be- cause ticks must remain attached to a host for at least 6 to 8 hours to transmit R. rickettsii, people at risk for exposure could prevent infection by checking their bodies daily for ticks.)

OCR for page 97
121 PATHOGENESIS • Further definition of the cells that are initially infected and the un- derlying early pathology. Rickettsia could be transmitted directly into vessels and cause rapid systemic infection or, alternatively, the bacteria could move into lymphatic vessels, and from there into local lymph nodes, triggering an early response of the immune system. • Understanding of the preference of Rickettsia rickettsii to infect en- dothelial cells in vivo, given that they can infect numerous cell types in vitro. • Better understanding of the roles of autophagy and of the activated innate intracellular mechanisms is needed. • Identification of the metabolic pathways used by Rickettsia during growth and replication in the cytosol. • Identification of genes and proteins differentially expressed and re- quired for growth in mammalian hosts versus tick vectors. • Identification of the R. rickettsii antigens that stimulate a protective immune response. This will be essential for development of a vaccine against Rocky Mountain spotted fever. • Development of better animal models including those that better re- capitulate the natural mode of transmission via tick bite and include human tissue and immune systems. • Study of Rickettsia from a systemic approach that considers the response of the vector to the host, the response of the host to the vector, the response of the vector and host to the bacteria, and the response of the bacteria to the vector and host. Three overall priorities for addressing rickettsial diseases were highlighted: • Diagnostics. People now die of these diseases because clinicians have no effective way to diagnose them at the early stages of disease when antibiotic treatment is most effective. • More studies of the disease pathogenesis. These would allow scien- tists to develop treatments for severe disease, such as when complica- tions arise because of delayed treatment. • Vaccine development. This would allow prevention of infection and disease in endemic areas and for individuals at high risk of exposure. DISCUSSION One participant asked how much is known about the intracellular tick- borne pathogens in the tick and the triggers for reactivation. Valbuena noted that in the Rickettsia, research has observed that all tissues in the ticks are infected. Thus when the ticks bite, their salivary glands are already infected.

OCR for page 97
122 CRITICAL RESEARCH NEEDS IN TICK-BORNE DISEASES Those rickettsiae in the salivary glands undergo a mechanism of reactivation by a process that is still poorly understood. He further noted that in nature Rickettsia can be transmitted through stages of the tick life cycle, as well as to the next generation by transovarial passage. In general, ticks do not require an amplifying host, unlike Ehrlichia, to maintain the infection in nature. Ismail noted that studies in animal models of HME and patients infected with E. chaffeensis have identified infection-induced production of chemo- kines, which are attractants for monocytes and other host target cells such as neutrophils. The influx of these cells into the skin would provide a niche for further replication of ehrlichiae within the mammalian host. However, it is not yet clear whether a similar process occurs in infected ticks. Another participant asked if there may be a potential therapeutic ap- proach by targeting the mechanism where the infected host has to get the pathogen back to the tick. Ismail noted that in the knowledge gaps there is a need to study different parameters of the immune responses at multiple time points in humans, reservoir host (e.g., white-tailed deer), and vector host. This would include examining the different stages of infection to see how the bac- teria progress from initial infection until the tick again acquires the bacterium. Gerber asked if Valbuena had done dose responses and then looked at the CD4 or CD8 T cell response in mice. Valbuena noted that these experiments were done in Walker’s laboratory. In the mouse model, when R. conorii, which is similar to R. rickettsii, is injected at a relatively low dose into the mouse, the animals become ill but do survive and establish a solid immunity. If these mice are subsequently challenged with extremely high doses, they will not succumb to the disease. In contrast, a very high dose will result in death of a naïve animal. These results may imply that a dysregulation in the immune response occurs when the dose is too high but also indicate that protective immunity, as a basis for a vaccine, is feasible. Ismail stated that for Ehrlichia infection, they noted that the dose can control the magnitude of immune responses and determine the outcome of infection. For example, a strong correlation between the dose of Ehrlichia and decreased CD4+T cell count and apoptotic death of activated CD4 T cells and NKT cells had been established in murine models of fatal mono- cytotropic ehrlichiosis. In fact, when mice were treated with doxycycline, the number of NKT cells could be restored. CONCLUDING THOUGHTS ON PATHOGENESIS Guy Palmer, D.V.M., Ph.D., College of Veterinary Medicine, Washington State University The overriding lesson from these scientists is the yin and yang of the immune response to tick-borne pathogens in the Anaplasma, Borrelia,

OCR for page 97
123 PATHOGENESIS Ehrlichia, and Rickettsia genera. On the one hand, the immune response controls the number of pathogens and helps people and other animals avoid massive systemic infection. On the other hand, the immune effectors themselves—especially those of the innate immune system—cause inflam- mation and tissue damage. These lessons apply to both early and persistent phases of infection, corresponding to acute and chronic disease. Scientists have clearly identified the innate immune system, and specific immune effectors, as mediating inflammation in Lyme disease. In fact, the mechanisms underlying inflammation and damage can be organ specific. That is, the mechanism producing arthritis differs from that leading to carditis. Studies based on genetically defined lines of mice have clear relevance to pathogenesis in humans, as individual patients may suffer severe symp- toms in some organs and not in others. Comments from patient advocates throughout the discussion—and indeed throughout the workshop— underscored this variation in symptoms. Presenting scientists and discus- sants alike emphasized the need for better markers of the progression in severity and chronicity of tissue damage. This is a notable translational gap between basic research on the science of Lyme disease and help for patients. The persistence of B. burgdorferi infection is complex and involves both antigenic variation and sequestration. That is, the bacteria’s ability to generate novel variants that display new antigens on their surface allows them to escape the host’s adaptive immune system. How these variants may alter the response of cells and tissues and inflammatory immune responses remain unanswered questions. Other knowledge gaps include how the bac- teria’s repertoire of surface proteins varies among strains, and how those variants affect disease. The evidence that infectious spirochetes sequester in sites protected from antibodies also raises important questions. Are these spirochetes truly quiescent in replicating and in stimulating inflammation? How similar are these spirochetes to “persister” cells described in bacterial infections? Do these sequestered bacteria reemerge during actual infection—as suggested by passive serum transfer experiments? Experimental approaches can likely help close these knowledge gaps, but applying the findings to human infec- tion will again prove challenging. Unlike passive techniques such as PCR, scientists can use imaging to detect viable organisms. Imaging has therefore provided new answers to vexing questions regarding whether or not infection persists even after antibiotic therapy. These questions concern the migration of bacteria to and from transmission sites, and the responses of cells to viable as well as nonviable bacteria. Indeed, the detection of “remnant” material at infection sites raises questions about whether an antigenic stimulus persists even after viable, replicating bacteria are killed. The use of imaging technology may

OCR for page 97
124 CRITICAL RESEARCH NEEDS IN TICK-BORNE DISEASES also allow scientists to examine host pathogen interactions concerning the progression of Lyme disease in deeper tissues such as the joints and heart. Infection with rickettsial pathogens, including those in the Anaplasma, Ehrlichia, Orientia, and Rickettsia genera, can progress so rapidly that patients require immediate hospitalization and intensive care—along with antimicrobial therapy—to prevent death. The acuteness and severity of these infections highlight the need for better educating medical profession- als in regions where the organisms are endemic. Investigators also need to better define these endemic regions and determine the risk that infectious bacteria and their animal hosts as they shift their range and distribution, and the likelihood that new pathogens will emerge. Finally, we need more accurate tools for clinical and laboratory diag- nosis of these diseases. The reasons underlying differences in the severity and rapidity of progression in patients is a major scientific gap—both on the pathogen side (diversity of species and strains) and the host side (genetic background and immune status). With only a few exceptions, the pathogenesis of the broad group of rickettsial diseases is understudied—typical of many neglected diseases of significant but underappreciated significance for public health. Workshop presenters and discussants emphasized all these challenges. However, two experimental models reveal the progress that scientists can achieve. Experiments using the Ixodes ovatus Ehrlichia are some of the best so far and underscore the dominant theme of the session: that the immune system is responsible for controlling infection but also producing the severe toxic shock–like syndrome when that control gets out of hand. A better understanding of immune mechanisms and effectors is critical to improving therapy once infection has progressed to severe acute disease. Research on Rickettsia in the spotted fever group has similarly begun to elucidate the pathogenesis of severe disease. Progress in developing animal models illustrates the possibilities. Still, the knowledge gaps regarding the pathogenesis of the rickettsial pathogens are numerous and wide, and the need for experiments that lay the groundwork for translating that knowl- edge to human disease is strong. In fact, such translational studies are essential for the full spectrum of tick-borne pathogens. To avoid a “translational canyon” between ex- perimental studies and human treatment and prevention, scientists should consider studying B. burgdorferi in naturally occurring models, such as neuroborreliosis in horses and Rickettsia rickettsii and Ehrlichia canis in dogs. The use of “humanized” organs such as human skin in mouse models—as noted by Valbuena—can accelerate scientists’ understanding of pathogenesis, and speed the application of that understanding to treat- ing patients and preventing infection.