DMID/ORD Workshop on Human Ehrlichiosis

National Institute of Allergy and Infectious Diseases (NIAID)

Workshop on Human Ehrlichiosis
September 25, 1996 • Rockville, MD

Agenda

Wednesday, September 25
8:30 a.m.	Introduction and Welcome John R. La Montagne, Ph.D. Phillip J. Baker, Ph.D.
	Clinical, Microbiological, and Diagnostic Aspects of Human Ehrlichiosis
8:40 a.m.	*Human infections with Ehrlichia chaffeensis: clinical, pathological,and immunologic aspects* David H. Walker, M.D.
9:15 a.m.	Diagnosing human granulocytic ehrlichiosis (HGE): Clinical and laboratory findings that aid in differentiation from other infectious entities J. Stephen Dumler, M.D.
9:50 a.m.	Recent studies on the biology of granulocytic ehrlichiosis Jesse L. Goodman, M.D.
10:40 a.m.	*Characterization of Ehrlichia chaffeensis* phagosome** Yasuko Rikihisa, Ph.D.
11:15 a.m.	A new look at human granulocytic ehrlichiosis: The New York State experience Susan J. Wong, Ph.D.
	Epidemiology, Ecology, and Vector Biology
1:00 p.m.	*Descriptive epidemiology of ehrlichiosis caused by Ehrlichia chaffeensis, 1985–1995* James G. Olson, Ph.D.
1:35 p.m.	*Investigations on the natural history of Ehrlichia chaffeensis* and related organisms** William R. Davidson, Ph.D.
2:10 p.m.	Studies on human granulocytic ehrlichiosis in Connecticut Louis A. Magnarelli, Ph.D.
2:45 p.m.	*Polymerase chain reaction studies on Ehrlichia chaffeensis, an etiologic agent of human ehrlichiosis, in dogs from southeast Virginia* Jacqueline E. Dawson, M.S.
	*Models of Ehrlichia Infection: Animals and In Vitro* Cultivation**
3:30 p.m.	*Coinfection with B. burgdorferi* and granulocytic ehrlichia** David H. Persing, M.D., Ph.D.
4:05 p.m.	The tick phase of the human granulocytic ehrlichiosis agent in vitro Ulrike G. Munderloh, D.V.M., Ph.D.
	Workshop Summary and Roundtable Discussion
4:40 p.m.	Focused discussion J. Stephen Dumler, M.D. Abdu F. Azad, Ph.D.
5:30 p.m.	Workshop adjournment

Abstracts

Recent Studies on the Biology of Human Granulocytic Ehrlichiosis

Jesse L. Goodman, M.D.; Curt Nelson; Marina Klein, M.D.; Division of Infectious Diseases, University of Minnesota School of Medicine, Minneapolis, MN 55455

HGE is an acute febrile illness characterized by myalgias, arthralgias, headache, leukopenia, thrombocytopenia, and elevations in serum transaminases. To provide background for this workshop, we will first review the clinical and laboratory manifestations, diagnosis, and treatment of infection with HGE. We will then describe in detail our laboratory's initial cultivation of the agent in human promyelocytic leukemia cells and the culture and PCR methodologies utilized. Recent studies to be presented will include those that have so far identified specific rifamycin and quinolone antibiotics with high activity against the agent. These drugs represent important potential alternative therapies for HGE for those individuals (e.g., children less than 9 years old) unable to take tetracyclines. In addition, we have demonstrated the lack of activity against HGE for those drugs, other than tetracyclines, used to treat Lyme disease, a common co-infecting pathogen. Finally, as an update from ongoing studies, we will describe subsequent isolations of HGE and similar agents from humans and animals, their variability, and the results of studies that aim to elucidate their basic biologic characteristics and interactions with the host.

Characterization of Ehrlichia chaffeensis Phagosome

Yasuko Rikihisa and Roy E. Barnewell, Department of Veterinary Biosciences, College of Veterinary Medicine, The Ohio State University, Columbus, Ohio 43210-1093

Ehrlichia chaffeensis parasitizes host monocytes or macrophages and resides in a phagosome that does not fuse with lysosomes. Double immunofluorescence labeling was used to study the nature of E. chaffeensis inclusion compartment in human monocyte cell line, THP-1. All ehrlichial inclusions were strongly positive for transferrin receptor but negative for clathrin heavy chain. Early endocytic markers such as HLA-DR, HLA-ABC, and 2-microglobulin were found in only some ehrlichial inclusions. Ehrlichial inclusions were negative for 73-kDa subunit of vacuolar type proton ATPase or lysosomal markers such as LAMP-1 or CD63. Ehrlichiae, therefore, take advantage of the transferrin-transferrin receptor recycling pathway to avoid lysosomal fusion.

Diagnosing HGE—Clinical and Laboratory Findings that Aid in Differentiation from Other Infectious Entities

J. Stephen Dumler, M.D., Division of Medical Microbiology, Department of Pathology, The Johns Hopkins Medical Institutions, Baltimore, MD

HGE is acute undifferentiated febrile illness caused by an E. phagocytophila- and E. equi-like ehrlichia that is transmitted by Ixodes ricinus-complex ticks. While many patients present with a typical clinical syndrome of fever, headache, and myalgias with leukopenia, thrombocytopenia, and elevated liver function tests, diverse clinical manifestations may confuse an otherwise straightforward diagnosis. A high seroprevalence rate in many locations indicates that subclinical infections probably occur. On the far end of the severity spectrum, fatalities associated with opportunistic infections are well documented. A high degree of confusion exists among primary care physicians about what ehrlichioses are, how they are diagnosed, and what constitutes appropriate treatment and follow-up. This situation is confounded by the potential co-occurrence of other tick-borne infections such as Lyme borreliosis or babesiosis. The classic and not-so-classic clinical and laboratory features of HGE will be presented with a discussion of the current methods for diagnosis, pitfalls of diagnosis, and emerging data about some biological aspects of the disease and its causative agent(s).

Human Infections with Ehrlichia chaffeensis: Clinical, Pathological, and Immunologic Aspects

David H. Walker, M.D., Department of Pathology, The University of Texas Medical Branch at Galveston, Galveston, TX 77555-0609

The spectrum of syndromes and distribution of severity of human monocytic ehrlichiosis (HME) is incompletely defined. Two prospective active surveillance studies suggest the contradictory views that HME is usually either asymptomatic or requires hospitalization. The largest series reporting 237 passively collected cases includes 62% hospitalized patients, a median duration of illness of 23 days, and a systemic disease often with gastrointestinal, hepatic, neurologic, and hematopoietic involvement. Bone marrow hyperplasia, granulomas, and erythrophagocytosis; multifocal hepatocellular necrosis; perivascular lymphohistiocytic infiltrates; and meningitis are known pathologic lesions. The pathology of HME has yet to be investigated adequately. Pathologic study of fatal cases suggests that HME can occur as an opportunistic infection or conversely can induce immunosuppression.

A serious flaw in our current state of knowledge is the laboratory basis for the diagnosis of HME, which is overwhelmingly serologic. A small fraction of cases have been documented by Ehrlichia chaffeensis-specific PCR. Only three isolates of E. chaffeensis have been established, all from cases of human illness. Each isolate differs from the others genetically and antigenically. Ehrlichia canis was isolated from a healthy seropositive person. It is quite likely that all of the Ehrlichia species capable of causing human infection have not yet been discovered. The genetic and antigenic diversity of E. chaffeensis itself is incompletely known as are their potential strain-determined pathogenicity. The ehrlichical virulence factors and the host factors that determine host resistance and the severity of illness have yet to be identified. Likewise, the mechanisms of immunity, or indeed the existence of protective immunity to E. chaffeensis, have yet to be established. There are serious deficiencies in each of the animal models of ehrlichiosis, particularly for the investigation of immune mechanisms against E. chaffeensis. The problems with the models include distant genetic relationship of E. chaffeensis with the E. risticii-E. sennetsu genogroup, lack of characterization of some models for the target cells and athologic lesions, unrealistic route of inoculation (certainly not via tick-bite transmission), and lack of quantitation of the ehrlichical inoculum and the time course of the organ infectivity titers.

HME poses substantial challenges in diagnosis, pathogenesis, and immunity. Research—optimally collaborative, interactive, and multidisciplinary—holds the answers.

A New Look at Human Granulocytic Ehrlichiosis: The New York State Experience

Susan J. Wong, Ph.D., Diagnostic Immunology Laboratory, Wadsworth Center, NYSDOH, P.O. Box 22002, Albany, NY 12201-2002

The Wadsworth Center began providing diagnostic serology for ehrlichiosis on July 18, 1995. Since tick vectors for HGE and HME are both found in NYS, we looked for antibodies to Ehrlichia equi, E. chaffeensis, and in all human clinical specimens submitted with a clinical suspicion of HGE. Cross-reaction between E. equi and E. chaffeensis was identified using sera from HGE cases in NYS. Examination of sera from HME cases from several other states demonstrated no reactivity on HGE substrate slides. The cross-reaction seen in NYS HGE cases is uni-directional to E. chaffeensis. E. chaffeensis-reactive sera from other states had very low reactivity on Lyme screening tests and only anti-63kD and anti-75kD reactivity on Borrelia immunoblots. These data suggest probable concurrent HME and Lyme disease in some NYS residents. HGE reactive sera are strongly reactive on Lyme screening tests with multiple Borrelia-specific antibodies as well as nonspecific (e.g., heat-shock) antibodies. These results point out the need for better diagnostic reagents and better diagnostic tests. We examined cytokine profiles (IL-4 and IFN- ) in sera positive to E. equi, B. burgdorferi, or both to see if questions about possible modulation of the human immune response could be answered, but the preliminary results are inconclusive. We are currently examining acute sera from HGE cases for anti-platelet antibodies. A dual study of detection by serology and PCR at Wadsworth Center and Dr. J.S. Dumler's laboratory at Johns Hopkins University College of Medicine is under way. We are awaiting convalescent specimens, but over 200 specimens have been analyzed at both centers using serologic and DNA amplification mechanisms. Preliminary analysis of the data again emphasizes the need for more than one laboratory assay to confirm a clinical case and for the development of more specific diagnostic tests. In no way can the laboratory methods for ehrlichiosis yet be considered standardized. There exists a shortage of high-quality, consistent reagents and technical staff with appropriate expertise to interpret the results. The NYSDOH and NYSCVM prepared HGE-infected horse neutrophil slides from two ponies inoculated with PCR-positive HGE whole blood from Westchester County. This project ensures a uniform supply of reagent slides for clinical testing of NYS residents. Both the E. chaffeensis and HGE bacteria from Westchester County are currently in tissue culture at the Wadsworth Center. The panel approach to ehrlichiosis testing at NYS (serology and PCR analysis for both HGE and HME) has just identified the first case of HME acquired in New York State.

Descriptive Epidemiology of Ehrlichiosis caused by Ehrlichia chaffeensis, 1985–1995

James G. Olson, Ph.D., Division of Viral and Rickettsial Diseases, Centers for Disease Control and Prevention, Atlanta, GA 30333

A total of 415 cases of ehrlichiosis caused by Ehrlichia chaffeensis were confirmed by the Centers for Disease Control and Prevention from 1985 through 1995. The geographic distribution of cases is concentrated in the southeast quarter of the United States. States with the highest number of confirmed cases include Missouri, Tennessee, Oklahoma, Texas, Arkansas, Virginia, and Georgia. With the exception of 1985, the number of cases confirmed each year ranges between 20 and 79. The cases in 1985 were retrospectively confirmed after the case description was first published in 1986. The peak number of cases (79) occurred in 1993, followed closely by 1992 (67). An active investigation of an epidemic of cases (n = 51) in a Tennessee community during these 2 years added to the number of cases identified through regular surveillance. The frequency of ehrlichiosis cases among males exceeds that among females by almost threefold. The frequency of cases by 10-year age groups generally increases with age. Prior to 1994, ehrlichiosis patients under 10 years of age were underrepresented, but recent data have shown that cases occur in this age group more frequently than those in the next two 10-year age groups. The frequency increases with each age group through 70 years and then decreases in the next two age groups. Ehrlichiosis is not a nationally notifiable disease, but several states have made it reportable. The Council of State and Territorial Epidemiologists has developed a draft case description to aid in surveillance activities. The new description includes the use of polymerase chain reaction among the laboratory criteria for confirmation of a case.

Investigations on the Natural History of Ehrlichia chaffeensis and Related Organisms

William R. Davidson, J.M. Lockhart, D.E. Stallknecht, S.E. Little, Southeastern Cooperative Wildlife Disease Study, College of Veterinary Medicine, The University of Georgia, Athens, GA 30602-7387; J.E. Dawson and C.K. Warner, Division of Viral and Rickettsial Diseases, National Center for Infectious Diseases, Centers for Disease Control and Prevention, Atlanta, GA 30333

Since 1992, we have conducted a series of research projects designed to help elucidate the natural history and field epidemiology of Ehrlichia chaffeensis. These studies have disclosed: (1) white-tailed deer over a broad region of the eastern United States have E. chaffeensis-reactive antibodies; (2) the occurrence of E. chaffeensis-reactive antibodies among deer populations has a significant site-specific geographic association with infestations of lone star ticks (Amblyomma americanum); (3) E. chaffeensis-reactive antibodies in one deer population were temporally associated with the appearance of lone star tick infestations; (4) white-tailed deer are susceptible to experimental infection with E. chaffeensis, seroconvert, and develop detectable rickettsemia; (5) blood, spleen, and lymph node of deer from populations with lone star tick infestations routinely are PCR positive for E. chaffeensis; (6) 2–12% of adult lone star ticks collected from sites supporting deer populations with serologic or PCR evidence of infection are PCR positive; (7) lone star tick infested white-tailed deer populations are naturally infected and rickettsemic based on six isolates of E. chaffeensis from three deer populations in Georgia; (8) other species of mammals at endemic sites have limited or no serologic evidence of exposure; and (9) the host associations of ticks parasitizing the wild mammal community at a known endemic site suggest little or no transmission by other species of ticks. In addition, these studies have: (1) disclosed a novel Ehrlichia-like organism most closely related to the E. phagocytophila genogroup; (2) resulted in the development of a PCR protocol to specifically detect this novel agent; (3) shown that this novel agent is common among deer and is associated with lone star tick infestations; (4) demonstrated this novel agent in lone star ticks via PCR; (5) shown that primers routinely used to detect the HGE agent will amplify product when the novel deer agent is used as template; and (6) detected the HGE agent in wild deer from Georgia based on sequence-confirmed PCR products. Related research is ongoing and a long-term experimental E. chaffeensis infection trial in deer will begin in the near future.

Studies on Human Granulocytic Ehrlichiosis in Connecticut

Louis A. Magnarelli, Ph.D., Connecticut Agricultural Experiment Station, P.O. Box 1106, New Haven, CT 06504

Field and laboratory studies are being conducted to determine prevalence of Ehrlichia equi infections in ticks and mammals. Based on results of polymerase chain reaction and DNA detection methods, E. equi was present in 50% of the 118 Ixodes scapularis females analyzed. Serologic testing for antibodies to E. equi revealed exposure to this bacterium or to related agents in human beings, equids, dogs, and Peromyscus leucopus. There is evidence of other tick-borne infections (i. e., Rocky Mountain spotted fever, babesiosis, and Lyme borreliosis) in human beings, dogs, and white-footed mice. In areas where Ixodes scapularis and Dermacentor variabilis are abundant, multiple tick-borne pathogens probably infect a multitude of hosts.

Polymerase Chain Reaction Evidence of Ehrlichia chaffeensis, an Etiologic Agent of Human Ehrlichiosis, in Dogs from Southeast Virginia

Jacqueline E. Dawson, Kalen Cookson, Suzanne Jenkins, Jay F. Levine, and James G. Olson, Viral and Rickettsial Zoonoses Branch, Division of Viral and Rickettsial Diseases, National Center for Infectious Diseases, Centers for Disease Control and Prevention, Public Health Service, U.S. Department of Health and Human Resources, Atlanta, Georgia 30333 (Dawson, Olson); Department of Microbiology, Pathology, and Parasitology, North Carolina Department State University (Cookson, Levine); Virginia Department of Health, Richmond, Virginia 23219 (Jenkins)

In 1986, a new form of human ehrlichiosis was recognized in the United States. Clinical signs of this disease were similar to Rocky Mountain spotted fever and included fever, headache, malaise, myalgia, arthralgia, nausea, and/or vomiting. Several studies confirmed association of the disease with history of exposure to either Amblyomma americanum (lone star tick) or Dermacentor variabilis (American dog tick). Analysis of 16S ribosomal RNA gene sequences produced phylograms indicating that E. chaffeensis was most closely related to E. canis, the cause of canine ehrlichiosis, and E. ewingii, the cause of canine granulocytic ehrlichiosis. Indeed, initial investigators believed that E. canis was infecting human beings. Dogs, in addition to being naturally infected with E. canis and E. ewingii, have been documented to be experimentally susceptible to E. chaffeensis infection. These experimentally exposed dogs did not have signs of disease, leading to the suggestion that they may act as natural reservoirs. To ascertain if dogs are naturally infected with Ehrlichia chaffeensis, blood was drawn from 74 dogs; 73 were tested serologically for antibodies reactive to E. chaffeensis and E. canis, and 38 were tested for the presence of E. chaffeensis, E. canis, and E. ewingii by polymerase chain reaction (PCR). 74 dogs from 5 animal shelters and 1 kennel in 3 cities and 3 counties in southeastern Virginia were sampled. Twenty-eight (38.4%) dogs had a positive titer (minimum > 1:64) for antibodies reactive to E. chaffeensis, and 28 (38.4%) had a positive titer reactive to E. canis. PCR analysis indicated that 8 dogs (42.1%) were positive for E. chaffeensis and 6 dogs (31.6%) were positive for E. ewingii. All dogs were negative by the PCR test for E. canis. We conclude that canine E. chaffeensis infection may be more prevalent than E. canis or E. ewingii infection in this region of the United States.

Coinfection with Borrelia burgdorferi and a Granulocytic Ehrlichia

Erik K. Hofmeister, Manuel M. Moro, and David H. Persing, Division of Experimental Pathology and Department of Immunology, Mayo Clinic Foundation, Rochester, MN 55905

Our laboratory has been interested in coinfection with Babesia microti and the HGE agent as possible contributors to biologic variation in Lyme disease. All three organisms are transmitted by the same tick, and active coinfections with B. burgdorferi and B. microti or E. phagocytophila have now been demonstrated in humans. Seroprevalence studies show that evidence of exposure to both organisms can be found in sera from Lyme disease patients from the Upper Midwest, the northeastern United States, and Europe; serologic data suggest that exposure to the HGE agent predated the discovery of the pathogen by nearly a decade. Both B. microti and E. phagocytophila have been shown to produce substantial immunsuppressive effects relative to superimposed infection in animal models. To study the immunologic effects of coinfection with the HGE agent and Borrelia burgdorferi, we have been studying mouse models of acute and chronic infection and coinfection. Initial studies have shown that: (1) C3H, Balb/C, and Bl0 mice infected with the HGE agent alone develop acute and chronic infection, with organisms detectable in the spleen 6 months after inoculation; (2) mice infected with the HGE agent alone develop serologic responses to a subset of B. burgdorferi antigens by Western blotting; (3) specific antibody responses to a recombinant HSP60 protein derived from the HGE agent can be detected a year or more after mouse inoculation; (4) when infections with the HGE agent and B. burgdorferi peak simultaneously, humoral immune responses to B. burgdorferi are markedly suppressed and spirochetal burden increases in tissues two- to threefold by quantitative PCR. The latter findings suggest that within the transmission cycle of Lyme disease, B. burgdorferi and the HGE agent may occupy a unique ecological niche characterized by immunologic commensalism.

The Tick Phase of the Human Granulocytic Ehrlichiosis Agent in Vitro

U.G. Munderloh, S.F. Hayes, C.M. Nelson, T.J. Kurtti, and J.L. Goodman, University of Minnesota, Department of Entomology, St. Paul, MN; Rocky Mountain Laboratories, NIH, Hamilton, MT; University of Minnesota Department of Medicine, Minneapolis, MN

The human granulocytic ehrlichiosis (HGE) agent is a newly recognized member of a genus in which the type species, Ehrlichia canis, was first described more than 60 years ago. Nevertheless, little is known about the biology and behavior of these pathogens in their vectors. By analogy with other tick-borne agents, it is reasonable to expect that the dramatic differences in environmental conditions experienced by the HGE agent when switching hosts may induce differential gene expression and morphologic changes. Growth and development of ehrlichiae in the tick culminate in the elaboration of forms responsible for initiating infection of the mammalian host, and as such are of biomedical importance. Indeed, mammalian antibodies recognize antigens produced by the HGE agent in tick cell culture as demonstrated by immunofluorescence and Western blot. This has potentially important implications for the future development of diagnostics and vaccines. Ehrlichiae are difficult to demonstrate in ticks, let alone to study in detail. Cell cultures derived from the natural vector offer an in vitro system for analysis of elusive associations between pathogen and vector. We have demonstrated that the morphologic forms of E. equi in cultured tick cells deviate markedly from those observed in mammalian cells in vitro and in vivo. In this workshop, we will present evidence indicating that the HGE agent, when transferred from human promyelocytic leukemia cells into vector tick cell culture, undergoes similar transformation. In order to correlate the events that take place in vitro with those that occur in vivo, we first established the pathogen in vitro in human as well as in vector tick cell culture. Either culture system produces organisms infectious for mammals. Moreover, laboratory-reared, infected ticks successfully transmitted the agent to small rodent hosts, causing infection of bone marrow cells as well as peripheral blood leukocytes, and was recovered from them in culture. We will present evidence from studies of the forms that develop in tick cell culture at the light and electron microscopic level and correlate them with those observed in ticks.

Summary

The Division of Microbiology and Infectious Diseases of NIAID, with support from the Office of Rare Diseases, convened a workshop on human ehrlichiosis on September 25, 1996, at the Solar Building (Room 1AO4) in Rockville, Maryland. About 20 individuals were invited to the workshop; approximately 25 were in attendance. This report includes the agenda for the workshop as well as copies of abstracts of papers presented by invited speakers, all of whom are internationally known experts in the field. The meeting was considered to be a great success by all who attended. It provided not only a comprehensive summary of the current state of knowledge but also served to identify the following major gaps in our knowledge of human ehrlichiosis:

the need to characterize more fully the clinical symptoms, pathogenesis, and epidemiology of human ehrlichiosis in order to improve diagnosis
the lack of sufficient knowledge concerning the entire spectrum of strains involved in the expression of human ehrlichiosis
the need to evaluate the influence of co-infection with Ehrlichia on the expression and severity of other tick-borne diseases, especially Lyme borreliosis
the need to develop more sensitive and reliable diagnostic procedures for the rapid detection of ehrlichial infections as well as to train laboratory personnel to perform such tests and evaluate the results obtained
the need for improved and more efficient methods for the isolation of Ehrlichia from infected host tissues and vectors
the need to isolate and characterize relevant ehrlichial surface antigens for use in diagnostic tests and in the development of vaccines
the need to define the chemical and molecular basis for the antigenic and genetic diversity encountered between clinical isolates as well as to evaluate their contribution to pathogenesis
the need to identify those factors that determine virulence as well as host factors that influence resistance, protective immunity, and the severity of disease
the need to characterize, in great detail, the response of the host's immune system to Ehrlichia as well as the mechanisms involved in the development and expression of host immunity
the need to devise more effective therapeutic approaches as well as alternative therapies for the treatment of ehrlichiosis, especially in the presence of other co-infecting agents
the need to acquire more precise information on the epidemiology and prevalence of ehrlichiosis as well as the risk factors involved so that more rational and effective approaches for disease control can be implemented
the need to acquire more information on the host range and vectors responsible for transmitting this infection as well as on the mode of transmission of various ehrlichial species

It was recognized that the ability of Goodman and co-workers to culture the agent of human granulocytic ehrlichiosis in vitro represents a major breakthrough (New Eng. J. Med.:334; 209–215, 1996). This now enables one (a) to examine, in great detail, the mechanisms by which Ehrlichia invade and damage hosts cells; (b) to characterize specific cell surface antigens for use in diagnostic procedures; and (c) to isolate specific genetic material (DNA) for use either in the preparation of recombinant diagnostic reagents or vaccines or as primers in diagnostic polymerase chain reaction (PCR) procedures. All agreed that a workshop on the standardization of diagnostic tests based on these approaches should be convened in the near future in order to acquire more precise information on the incidence of ehrlichiosis as well as its emergence.

The organizers and participants of this workshop also expressed their gratitude to the ORD for providing the funds to convene this timely and most informative workshop.

Report Submitted by:
Phillip J. Baker, Ph.D.
Program Officer, Vector-Borne Disease Program
Division of Microbiology and Infectious Diseases (DMID)
National Institute of Allergy and Infectious Diseases
Through: John LaMontagne, Ph.D.
Director, Division of Microbiology and Infectious Diseases
National Institute of Allergy and Infectious Diseases

Back to Scientific Conferences Home

Last Reviewed: January 9, 2006