Emerging Infectious Diseases [Volume 5 No.1 / January - February 1999] Suggested Citation Letters Navigational Instinct: A Reason Not to Live Trap Deer Mice in Residences To the Editor: Although the rodent that most often invades homes in North America is the house mouse, Mus musculus, the deer mouse, Peromyscus maniculatus, principal vertebrate host of Sin Nombre virus (SNV) (1), also invades homes (2), particularly in rural areas. Barring deer mice from human habitations would prevent domiciliary acquisition of SNV. Current recommendations (3) are to prevent wild rodents from entering homes or to snap trap (kill) them should they enter. To conduct longitudinal studies of hantaviruses in southeastern Colorado on a former cattle ranch now returning to its natural condition as short-grass prairie, we often stay in an old bunkhouse, used by many research groups at irregular intervals. The house, furnished with beds and full kitchen facilities, is well maintained but has openings through which mice can pass to and from the outside. For safety and cleanliness, we removed mice we found inside the house, but between April 1996 and April 1998, we live trapped and released them rather than snap trapping them. Before release the rodents were identified to species; were measured and assessed regarding general appearance and health, sexual preparedness, and presence of wounds; were bled for antibody tests; and were ear-tagged. Nineteen deer mice and one pinyon mouse (a P. truei, which did not return) were examined and tagged. At first, we simply released these animals approximately 50 m from the house, but when we realized that they were returning, we released them at increasing distances (50 m to 1,500 m) from the house; the distances were measured by pace counts by at least two investigators. Three deer mice had been captured multiple times in our test grid (as far as 250 m from the house) before they were first captured in the house. Once captured in the house, however, they were not captured in traps of the grid (i.e., outside the house). The mean distance traversed by the five deer mice that returned to the house was at least 394 m; one mouse returned after being released 500 m and 1,000 m, then 750 m, and 1,200 m from the house at consecutive daily trapping sessions of 3 days. Sometime within the subsequent 6 weeks, this mouse returned to the house from the 1,000-m release point and then from 750 m and 1,200 m away on consecutive days within our 3-day trapping period. Each of the mice returning to the house did so within 24 hours of release, two as few as 6 hours after release from 500 m and 750 m away. Nine mice were captured once; six of eight mice captured twice were captured at least once more; one was captured 10 times, one 7 times, one 6 times, one 4 times, and two 3 times. Equal numbers of male and female, adult and juvenile mice were captured in the house, but only adult mice (5 of 5) returned to the house. Returning deer mice maintained or gained weight between captures and grew in length at approximately the same rate as deer mice captured in the test grid. Some rodents have been documented to move similar distances (e.g., 1,200 m), but they took more than 2 weeks to complete the trek (4). Homing ability, site fidelity, and navigational proficiency of rodents are well documented (5,6). Teferi and Millar (7) studied the homing ability of deer mice in Alberta, Canada; 50% of deer mice in that study returned to their home sites (a short-grass prairie habitat). The mice traveled 650 m to 1,980 m (mean 1,500 m) and had to cross a river and pass optimal habitat patches to reach their home sites. Deer mice with previous homing experience were more successful in returning home (100%) than inexperienced mice (60%) and faster in doing so (8). Teferi and Millar (7) suggest that these deer mice were able to navigate in a direct route to their home sites. We released mice in locations where they had no direct route to the house; they had to follow a winding road, climb over rocky outcroppings nearly 17 m high, or otherwise surmount obstacles and dangers, such as predators (7). None of the mice we captured had immunoglobulin G (IgG) antibody to SNV. However, infected deer mice released and then returning to a house or uninfected deer mice released, infected, and then returning to a house would increase the likelihood of human contact with an SNV-infected mouse. The risk would be the same for other hantaviruses infecting other peridomestic rodents. Against current recommendations that rodents in homes be snap trapped, some homeowners live trap and release them outside their homes. Our data strongly support snap trapping mice in homes and provide evidence that released wild mice return and may place the residents at risk. Acknowledgments We thank T. Davis, S.B. Calisher, and E. Kuhn for their assistance in completing these studies. This work was partially funded by contract U50-CCU-813420-01 from the U.S. Centers for Disease Control and Prevention. Charles H. Calisher,* William P. Sweeney,† J. Jeffrey Root,* and Barry J. Beaty* *Colorado State University, Fort Collins, Colorado, USA; and †University of Texas Medical Branch, Galveston, Texas, USA References 1. Childs JE, Ksiazek TG, Spiropoulou CF, Krebs JW, Morzunov S, Maupin GO, et al. Serologic and genetic identification of Peromyscus maniculatus as the primary rodent reservoir for a new hantavirus in the southwestern United States. J Infect Dis 1994;169:1271-80. 2. Glass GE, Johnson JS, Hodenbach GA, DiSalvo LJ, Peters CJ, Childs JE, et al. Experimental evaluation of rodent exclusion methods to reduce hantavirus transmission to humans in rural housing. Am J Trop Med Hyg 1997;56:359-64. 3. Centers for Disease Control and Prevention. Hantavirus infection—southwestern United States: interim recommendations for risk reduction. MMWR Morb Mortal Wkly Rep 1993;42(RR-11):1-13. 4. Ostfeld RS, Manson RH. Long-distance homing in meadow voles, (Microtus pennsylvanicus). Journal of Mammalogy 1996;77:870-3. 5. August PV, Ayvazian SG, Anderson JGT. Magnetic orientation in a small mammal, Peromyscus leucopus. Journal of Mammalogy 1989;70:1-9. 6. Fluharty SL, Taylor DH, Barrett GW. Sun compass orientation in the meadow vole, Microtus pennsylvanicus. Journal of Mammalogy 1976;57:1-9. 7. Teferi T, Millar JS. Long distance homing by the deer mouse, Peromyscus maniculatus. Canadian Field-Naturalist 1993;107:109-11. 8. Robinson WL, Falls JB. A study of homing of meadow mice. American Midland Naturalist 1965;73:188-224. Bartonella quintana in Body Lice Collected from Homeless Persons in Russia To the Editor: Lice are obligate blood-feeding insects; three lice species (Pediculus humanus var capitatis, P. humanus var corporis, and Phtirus pubis) have been connected with humans throughout history. The body louse (P. humanus corporis) is the vector for three infectious diseases: epidemic typhus caused by R. prowazekii, trench fever caused by B. quintana, and relapsing fever caused by Borrelia recurrentis (1-3). Infestation with the body louse is associated with cold weather, poverty, and poor hygiene. In Russia, louse-transmitted diseases have caused more deaths than any other infectious disease in recent centuries (4). During the last decade, pediculosis (infestation with P. humanus) has increased markedly throughout the world (5,6), especially in developing countries and in areas (e.g., Eastern Europe, Russia) that have undergone vast social and economic changes. The incidence of pediculosis in Russia is approximately 220 to 300 cases per 100,000 inhabitants(7). Social and economic upheavals in the former Soviet Union have increased the number of homeless people, among whom pediculosis is highly prevalent (6). A disease of the past, epidemic typhus, has reemerged as a public health concern after a 1996 outbreak in Burundi, the largest outbreak of the disease since World War II (5,8). During World War II, a huge typhus epidemic caused illness in more than 20,000,000 people in Russia. R. prowazekii infection can persist in a latent form in convalescent typhus patients, remanifesting itself in a recrudescent form (Brill-Zinsser disease) in patients under stress (1). Sporadic cases of Brill-Zinsser disease are reported every year in all regions of the former Soviet Union (9) and because most of the population has no immunity to R. prowazekii, the risk for a typhus outbreak is increased. In a recent outbreak in the Lipetsk region, 360 km from Moscow, 24 louse-infested, febrile patients in an unheated psychiatric institution had serologically diagnosed typhus (10). The great epidemics of trench fever in Europe took place during World War I (2). However, recently a large outbreak of trench fever associated with epidemic typhus has been reported in Burundi (5). Sporadic cases of B. quintana infection have occurred during the last decade in Europe and the United States, mainly in HIV-infected patients, the homeless, and persons with chronic alcoholism; the infection has manifested itself as trench fever, bacteremia, bacillary angiomatosis, or endocarditis (11-16). Relapsing fever has not been reported in Russia for more than 50 years, despite a high prevalence after the 1917 revolution and during World War II (17). We studied the presence of typhus, trench fever, and relapsing fever agents in body lice collected from homeless persons in Moscow. The lice were collected at the Moscow Municipal Disinfection Center, where the homeless wash and delouse themselves, as well as disinfect or change their clothes. Only participants who gave informed consent were included. Lice were collected from the participant's clothing (from the inner surface and seams of t-shirts, shirts, and sweaters); 3 to 25 lice were found on each volunteer. Lice were collected from May to October 1996 (214 samples) and from June to September 1997 (54 samples). From June to September 1997, 300 homeless male attendees were examined; 57 (19%) had body lice or louse eggs (three had only eggs) on their clothing. Lice were identified as P. humanus corporis, according to standard taxonomic keys (6,18). Lice from each person were split into pools of three to eight insects, and DNA was extracted from each pool and tested for R. prowazekii, B. quintana, and B. recurrentis by polymerase chain reaction (PCR) analysis. Primers used for PCR analysis and conditions for DNA amplification have been described (5,19-22). Uninfected, laboratory-reared lice served as negative controls, and DNA of R. rickettsii, B. elizabethae, and B. burgdorferi were used as positive controls. Results of each amplification were resolved in 1% agarose gels (type LE; Sigma-Aldrich Chimie, St. Quentin Fallavier, France) and were visualized under UV light after ethidium bromide staining. The sizes of amplicons were determined by comparison with the DNA molecular weight marker VI (Boehringer, Mannheim, Germany). To confirm the identity of amplicons, their nucleotide base sequence was determined by using an AmpliCycle sequencing kit (Perkin-Elmer Corp., Foster City, CA) according to the manufacturer's instructions. PCRs incorporating rickettsia- and borrelia-specific primers did not yield products from any DNA extracts derived from the louse samples. Positive controls in both reactions yielded bands of the expected size. Thus, louse samples were not infected with R. prowazekii or B. recurrentis. Initial screening with PCR incorporating nonspecific primer pairs for Bartonella species yielded products of the estimated amplicon size of approximately 1,200 bp for 33 (12.3%) of the 268 louse samples. These results were confirmed by PCR incorporating primers (CS.443p–CS.979n) specific for the gltA gene. Products of this reaction were characterized by base-sequence determination. All 33 Bartonella-positive samples yielded a partial gltA sequence identical to that of B. quintana (22). Persons infested with infected lice were younger than 30 years to older than 60 years of age. A recent report indicates that 11% of the homeless in Russia are infested with lice (23); in our limited study, we observed a prevalence as high as 19%. With widespread louse infestation and overcrowding, a single case of Brill-Zinsser disease can cause an outbreak of epidemic typhus. A patient more than 50 years of age with Brill-Zinsser disease was the suspected primary source of typhus infection during the 1997 Lipetsk outbreak. Presence of lice in the hospital permitted disease dissemination (10). Although our data showed that none of the 268 louse pools were infected with R. prowazekii, the serious threat of an outbreak requires continued surveillance. No samples were found to contain B. recurrentis DNA, yet dissemination of body lice could also cause relapsing fever to reemerge. Interest in bartonellosis has recently increased, particularly in association with HIV infection, because Bartonella species can cause bacteremia in the immunocompromised (15). Recent investigations have demonstrated that B. quintana cause bacillary angiomatosis, lymphadenopathy (16), endocarditis (24), and infections of the central nervous system (25,26) in healthy persons. Recent reports of B. quintana infection outbreaks in the United States (14,27), Africa (5), and Europe (11,13,28) suggest either greater awareness or a reemergence of this infection. Persons who are homeless or alcoholic are particularly at risk (11-13,27,29). In all recently reported cases, the role of a possible arthropod vector has remained unclear (30,31), although lice exposure, together with homelessness, is a risk factor for B. quintana-induced bacillary angiomatosis (15). The fact that 12.3% of studied lice samples were B. quintana-positive confirms the role of this arthropod vector in the contemporary life cycle of the agent. A similar prevalence of B. quintana in body lice was reported in Burundi (5) and has been observed in France (D. Raoult, unpub. data). On the basis of data from our study, Moscow should be considered an area at high risk for an outbreak of bartonellosis. Elena B. Rydkina,*† Véronique Roux,* Eugenia M. Gagua,‡ Alexandre B. Predtechenski,§ Irina V. Tarasevich,† and Didier Raoult*† *Université de la Méditerranée, Marseille, France; †Russian Academy of Medical Sciences, Moscow, Russia; ‡Moscow Municipal Disinfection Center, Moscow, Russia; and §Research Center of Virology,Russia References 1. Raoult D, Roux V. Rickettsioses as paradigms of new or emerging infectious diseases. Clin Microbiol Rev 1997;10:694-719. 2. Maurin M, Raoult D. Bartonella (Rochalimaea) quintana infections. Clin Microbiol Rev 1996;9:273-92. 3. Johnson WD. Borrelia species (relapsing fever). In: Mandell GL, Douglas RG, Bennet JE, editors. Principles and practice of infectious diseases. 3rd ed. Edinburgh: Churchill Livingstone; 1990. p. 1816-8. 4. Patterson KD. Typhus and its control in Russia, 1870-1940. Med History 1993;37:361-81. 5. Raoult D, Ndihokubwayo JB, Tissot-Dupont H, Roux V, Faugere B, Abegbinni R, et al. Outbreak of epidemic typhus associated with trench fever in Burundi. Lancet 1998;352:353-8. 6. Tarasevich IV, Zemskaya AA, Dremova VP, Frolova AI, Hudobin VV, Lange AB. Human lice (diagnosis, medical significanse, methods of elimination) [in Russian]. Moscow: Medzdrav USSR; 1990. p. 5-7. 7. Tarasevich IV, Fetisova NF. Classical typhus [in Russian]. ZniSO (Health of Population and Environment) 1995;2:9-13. 8. Raoult D, Roux V, Ndihokubwayo JB, Bise G, Baudon D, Martet G, et al. Jail fever (epidemic typhus) outbreak in Burundi. Emerg Inf Dis 1997;3:357-60. 9. Eremeeva ME, Balayeva NM, Raoult D. Serological response of patients suffering from primary and recrudescent typhus: comparison of complement fixation reaction, Weil-Felix test, microimmunofluorescence, and immunoblotting. Clin Diagn Lab Immunol 1994;1:318-24. 10. Tarasevich IV, Rydkina E, Raoult D. An outbreak of epidemic typhus in Russia. Lancet 1998;352:1151. 11. Brouqui P, Houpikian P, Tissot Dupont H, Toubiana P, Obadia Y, Lafay V, et al. Survey of the seroprevalence of Bartonella quintana in homeless people. Clin Infect Dis 1996;23:756-9. 12. Comer JA, Flynn C, Regnery RL, Vlahov D, Childs JE. Antibodies to Bartonella species in inner-city intravenous drug users in Baltimore, MD. Arch Intern Med 1996;156:2491-5. 13. Drancourt M, Mainardi JL, Brouqui P, Vandenesch F, Carta A, Lehnert F, et al. Bartonella (Rochalimaea) quintana endocarditis in three homeless men. N Engl J Med 1995;332:419-23. 14. Jackson LA, Spach DH, Kippen DA, Sugg NK, Regnery RL, Sayers MH, et al. Seroprevalence to Bartonella quintana among patients at a community clinic in downtown Seattle. J Infect Dis 1996;173:1023-6. 15. Koehler JE, Sanchez MA, Garrido CS, Whitfeld MJ, Chen FM, Berger TG, et al. Molecular epidemiology of bartonella infections in patients with bacillary angiomatosis-peliosis. N Engl J Med 1997;337:1876-83. 16. Raoult D, Drancourt M, Carta A, Gastaut JA. Bartonella (Rochalimaea) quintana isolation in patient with chronic adenopathy, lymphopenia, and a cat. Lancet 1994;343:977. 17. Gromashevski LB, Vaindrach GM. Relapsing typhus [in Russian]. Moscow: Medgiz; 1946. p. 78-96. 18. Kim KC, Ludwig HW. The family classification of the Anoplura. Systematic Entomology 1978;3:249-84. 19. Regnery RL, Spruill CL, Plikaytis BD. Genotypic identification of rickettsiae and estimation of intraspecies sequence divergence for portions of two rickettsial genes. J Bacteriol 1991;173:1576-89. 20. Roux V, Rydkina E, Eremeeva M, Raoult D. Citrate synthase gene comparison, a new tool for phylogenetic analysis, and its application for the Rickettsiae. Int J Syst Bacteriol 1997;47:252-61. 21. Roux V, Raoult D. The 16S-23S rRNA intergenic spacer region of Bartonella (Rochalimaea) species is longer than usually described in other bacteria. Gene 1995;156:107-11. 22. Birtles RJ, Raoult D. Comparison of partial citrate synthase gene (gltA) sequences for phylogenetic analysis of Bartonella species. Int J Syst Bacteriol 1996;46:891-7. 23. Mitko E. To count homeless in autumn [in Russian]. Vechernyaya Moskva (Evening Moscow) Newspaper 1997; 137. 24. Raoult D, Fournier PE, Drancourt M, Marrie TJ, Etienne J, Cosserat J, et al. Diagnosis of 22 new cases of Bartonella endocarditis. Ann Intern Med 1996;125:646-52. 25. Parrott JH, Dure L, Sullender W, Buraphacheep W, Frye TA, Galliani CA, et al. Central nervous system infection associated with Bartonella quintana: a report of two cases. Pediatrics 1997;100:403-8. 26. Case Records of the Massachusetts General Hospital. N Engl J Med 1998;338:112-9. 27. Spach DH, Kanter AS, Dougherty MJ, Larson AM, Coyle MB, Brenner DJ, et al. Bartonella (Rochalimaea) quintana bacteremia in inner-city patients with chronic alcoholism. N Engl J Med 1995;332:424-8. 28. Stein A, Raoult D. Return of trench fever. Lancet 1995;345:450-1. 29. Jackson LA, Spach DH. Emergence of Bartonella quintana infection among homeless persons. Emerg Infect Dis 1996;2:141-4. 30. Relman DA. Has trench fever returned? N Engl J Med 1995;332:463-4. 31. Walker DH, Barbour AG, Oliver JH, Lane RS, Dumler JS, Dennis DT, et al. Emerging bacterial zoonotic and vector-borne diseases. Ecological and epidemiological factors. JAMA 1996;275:463-9. Tick-Transmitted Infections in Transvaal: Consider Rickettsia africae To the Editor: We report a case of African tick-bite fever (ATBF) in a 54-year-old French hunter returning to France on 21 April 1997, after a 15-day visit to Transvaal, South Africa. While traveling in the veld, the hunter removed (but did not keep) two ticks from his left leg. Two days later, he observed eschars at the bite sites. Within 5 days, he had high fever (39.5°C) and headache and decided to fly back to France, where he was admitted to the Infectious Diseases Department in the Hotel Dieu Hospital in Clermont-Ferrand. The patient's clinical symptoms were persistent fever, severe headache, and two inflammatory eschars on the left leg. Laboratory results were normal. On 22 April, an acute-phase serum sample and eschar biopsy were sent to our laboratory. The patient was treated with 200 mg per day doxycycline for 10 days. His symptoms resolved. A second serum sample was collected on 13 May. Microimmunofluorescence was performed as previously described (1). Although the acute-phase serologic results were negative, the convalescent-phase serum exhibited anti–R. africae and anti–R. conorii titers of 16 for immunoglobulin (Ig) G and 8 for IgM. Sera were adsorbed with R. conorii and R. africae antigens (2), and serologic testing and Western blot analysis (1) were performed on the resultant supernatants. Cross-adsorption of the convalescent-phase serum caused the homologous and heterologous antibodies to disappear when adsorption was performed with R. africae antigens; only homologous antibodies disappeared when adsorption was performed with R. conorii. Western immunoblot, performed with the same adsorbed serum, indicated R. africae infection by demonstrating a specific reactivity pattern with R. africae–specific antigens in the 110-kDa to 145-kDa region (2). An inoculation eschar biopsy specimen was injected into human embryonic lung fibroblasts, according to the centrifugation shell-vial technique (3). After 6 days' incubation at 32°C, a Gimenez staining of methanol-fixed human embryonic lung fibroblasts showed rickettsialike bacilli. The strain was identified by direct immunofluorescence performed on the cells with an anti–R. africae monoclonal antibody (4). Moreover, DNA was extracted from the ground eschar biopsy specimen and from 200 µL of shell-vial supernatant, by using a QIAmp Tissue kit (QIAGEN GmbH, Hilden, Germany), according to the manufacturer's instructions. These extracts were used as templates with primers complementary to a portion of the coding sequence of the rOmpA encoding gene in a polymerase chain reaction (PCR) assay (5), and the base sequences of the resulting PCR products were determined (5). The sequence obtained by both methods was the same as the R. africae sequence in Genbank (100% similarity). Since first described in Africa in 1910, tick-transmitted rickettsioses have been imputed to a single rickettsial species, Rickettsia conorii, although two distinct clinical illnesses have been observed (6): an urban form in patients in contact with dogs and their ticks (Rhipicephalus spp.) characterized by fever, headache, myalgia, cutaneous rash, and a lesion at the site of the tick bite (7), and a rural form in patients in contact with cattle or game and their ticks (Amblyomma spp.) characterized by mild signs and frequent lack of rash (8). Although R. africae was initially isolated from Amblyomma cattle ticks in 1973, the first evidence of its pathogenic role in humans was seen in 1992 in a patient who, after a tick bite, had fever, an inoculation eschar, regional lymphadenopathy, but no cutaneous rash (9). Since then, an additional 20 cases of R. africae–related infections have been reported in travelers returning from Zimbabwe and South Africa (2,10). R. conorii has long been considered the only African spotted fever group rickettsia, responsible for both Mediterranean spotted fever and ATBF. Since the first case was described (9), most of the 20 reported cases of ATBF occurred as outbreaks (2,10) in Europeans returning from Zimbabwe and South Africa. The occurrence of concomitant ATBF cases is unusual since Mediterranean spotted fever is generally sporadic and is likely related to the biologic characteristics of the recognized vector of R. africae, Amblyomma spp. ticks. While both are nonnidicolous ticks, Amblyomma spp. and Rhipicephalus spp. exhibit very different host-seeking behavior (11). Amblyomma spp. are ticks of cattle and wild ungulates, are not host-specific, and can readily feed on humans; they are "hunter ticks" and exhibit an "attack strategy" (in response to stimuli they specifically converge on nearby hosts). Rhipicephalus spp. are dog ticks and vectors of R. conorii; very host-specific, they exhibit an "ambush strategy" (they are passive and remain quiescent in their habitat until a vertebrate host passes). Up to 72% of A. hebraeum are infected with Rickettsia-like organisms, in particular R. africae (12); Amblyomma spp. are widely distributed in rural areas in sub-Saharan Africa (13) and prevalence of A. hebraeum ticks, incidence of ATBF cases, and prevalence of R. africae antibodies have been strongly linked (14). Rural Africans are also commonly infected with R. africae, usually at a young age (14). In Zimbabwe, Kelly et al. (15) demonstrated that 55% of the tested human sera had antibodies against R. africae. ATBF usually has specific clinical features: shorter incubation period than for Mediterranean spotted fever, multiple inoculation eschars (related to the host-seeking behavior and host-specificity of Amblyomma spp. ticks, which are "attack ticks" [15]), regional lymphadenopathies, frequent lack of cutaneous rash or a pale vesicular eruption, and absence of complications (2). Although only 22 proven cases have been described so far (including the present case), ATBF has been recognized as a commonly encountered disease in southern Africa since 1900 (8,16). Epidemiologic and clinical features indicate that several cases previously diagnosed on the basis of serology results only as R. conorii-caused may have been caused by R. africae. Given the serologic cross-reactivity among spotted fever group rickettsiae, microimmunofluorescence, the easiest serologic method, may not be sufficient for the etiologic diagnosis of a rickettsial spotted fever. A definitive diagnosis of ATBF requires either additional serologic procedures, such as cross-adsorption or Western blot, or the use of PCR or culture. As for PCR, rOmpA-amplification possesses sufficient sequence heterogeneity among the spotted fever group rickettsiae to be used as an identification tool (5). The centrifugation-shell vial-cell culture (3), used routinely in our laboratory, reliably isolates strictly intracellular bacteria, including rickettsia, from blood and tissue specimens, especially eschar biopsies (the specimen of choice for isolation procedures or genomic detection). We noted cross-reactions between R. africae and R. conorii. Cross-adsorption between anti—R. africae and antiR. conorii antibodies and Western blots confirmed that the antibodies we detected were directed specifically at R. africae. Furthermore, both PCR and cell culture confirmed the diagnosis of R. africae infection. ATBF appears to be an important emerging disease in visitors to rural areas of southern Africa. R. africae should be considered a potential pathogen in patients returning from such areas who have fever, headache, multiple inoculation eschars, or regional lymphadenopathy after a tick bite. Pierre-Edouard Fournier,* Jean Beytout,† and Didier Raoult* *Université de la Méditerranée, Marseille, France; and †Centre Hospitalier Régional Hotel Dieu, Clermont-Ferrand, France References 1. Teysseire N, Raoult D. Comparison of Western immunoblotting and microimmunofluoresence for diagnosis of Mediterranean spotted fever. J Clin Microbiol 1992;30:455-60. 2. Brouqui P, Harle JR, Delmont J, Frances C, Weiller PJ, Raoult D. African tick bite fever: an imported spotless rickettsiosis. Arch Int Med 1997;157:119-24. 3. Marrero M, Raoult D. Centrifugation-shell vial technique for rapid detection of Mediterranean spotted fever rickettsia in blood culture. Am J Trop Med Hyg 1989;40:197-9. 4. Xu W, Beati L, Raoult D. Characterization of and application of monoclonal antibodies against Rickettsia africae, a newly recognized species of spotted fever group rickettsia. J Clin Microbiol 1997;35:64-70. 5. Roux V, Fournier PE, Raoult D. Differentiation of spotted fever group rickettsiae by sequencing and analysis of restriction fragment length polymorphism of PCR amplified DNA of the gene encoding the protein rOmpA. J Clin Microbiol 1996;34:2058-65. 6. Conor A, Bruch A. Une fièvre éruptive observée en Tunisie. Bull Soc Pathol Exot Filial 1910;8:492-6. 7. Sant'Anna JF. On a disease in man following tick-bites and occurring in Lourenço Marques. Parasitology 1912;4:87-8. 8. Troup JM, Pijper A. Tick-bite fever in Southern Africa. Lancet 1938;ii:1183-6. 9. Kelly P, Matthewman LA, Beati L, Raoult D, Mason P, Dreary M, et al. African tick-bite fever——a new spotted fever group rickettsiosis under an old name. Lancet 1992;340:982-3. 10. Fournier PE, Roux V, Caumes E, Donzel M, Raoult D. An outbreak of Rickettsia africae infections among participants in an adventure race from South Africa. Clin Infect Dis. In press 1998. 11. Sonenshine DE. Ecology of non-nidicolous ticks. In: Sonenshine DE, editor. Biology of ticks. Oxford (NY): Oxford University Press; 1993. p. 3-65. 12. Beati L, Kelly PJ, Matthewman LA, Mason P, Raoult D. Prevalence of Rickettsia-like organisms and spotted fever group Rickettsiae in ticks (Acari: Ixodidae) from Zimbabwe. J Med Entomol 1995;32:787-92. 13. Kelly PJ, Beati L, Mason PR, Matthewman LA, Roux V, Raoult D. Rickettsia africae sp nov, the etiological agent of African tick bite fever. Int J Syst Bacteriol 1996;46:611-4. 14. Tissot-Dupont H, Brouqui P, Faugere B, Raoult D. Prevalence of antibodies to Coxiella burnetii, Rickettsia conorii, and Rickettsia typhi in seven African countries. Clin Infect Dis 1995;21:1126-33. 15. Kelly PJ, Mason PR, Matthewman LA, Raoult D. Seroepidemiology of spotted fever group rickettsial infections in human in Zimbabwe. Am J Trop Med Hyg 1991;94:304-9. 16. Gear JHS, Bevan C. An outbreak of tick-bite fever. S Afr Med J 1936;10:485-8. Extended-Spectrum Beta-Lactamase-Producing Salmonella Enteritidis in Trinidad and Tobago To the Editor: Salmonella Enteritidis, a predominantly localized pathogen of the human gastrointestinal tract, can become invasive in very young, very old, malnourished, and immunocompromised patients. In recent years, S. Enteritidis has emerged as a major intestinal pathogen in Trinidad and Tobago (population 1.2 million); in 1997, S. Enteritidis caused 79 (66%) of 119 culture-confirmed salmonella infections, in contrast to 18 (18%) of 99, 48 (47%) of 102, and 107 (61%) of 178 in 1994, 1995, and 1996, respectively. Increased incidence of S. Enteritidis infections has been reported worldwide (1,2). Of 216 human S. Enteritidis isolates tested for antimicrobial susceptibility between 1994 and 1996 in Trinidad, none were resistant to cephalosporins, aminoglycosides, ampicillin, trimethoprim-sulphamethoxazole, chloramphenicol, and norfloxacin/ciprofloxacin by the Kirby-Bauer disk diffusion method, which uses the National Committee for Clinical Laboratory Standards (NCCLS) breakpoints (3). Here we report an unusual isolate of S. Enteritidis resistant to all penicillins and cephalosporins——including third-generation cephalosporins, gentamicin, tobramicin, and trimethoprim-sulphamethoxazole——by the Kirby-Bauer disk diffusion method. Amoxicillin-clavulanate and piperacillin-tazobactam disks gave zone sizes of 15 mm and 19 mm, respectively, which are classified as intermediate in the NCCLS guidelines. This isolate was recovered from the blood culture of a febrile, nonneutropenic patient with multiple myeloma on two occasions 24 hours apart in March 1998. The isolate was sensitive only to ofloxacin and imipenem. Admitted to the hospital with compressed fracture of the spine for physiotherapy in December 1997, the patient had several febrile episodes and received several courses of multiple empirically prescribed antibiotics (cefotaxime, gentamicin, and piperacillin). The patient had not traveled abroad during the previous 6 months. Because cephalosporin resistance in salmonellae has not been reported before in the Caribbean, we investigated the mechanism behind this third-generation cephalosporin resistance further. Using amoxicillin-clavulanate in combination with ceftazidime, ceftriaxone, and aztreonam, we performed the double disk synergy test to determine whether this strain was an extended-spectrum beta-lactamase producer as described elsewhere (3); augmentation of the zone at the junction of amoxicillin-clavulanate and aztreonam/ceftriaxone/ceftazidime zones confirmed that indeed it was. In the past few years, third-generation cephalosporin resistance in S. Enteritidis has been described in Europe (4), the United States (5), Turkey (6), India (7,8), and Argentina (9). Few reports exist of extended-spectrum beta-lactamasemediated third-generation cephalosporin resistance in Salmonella spp. To our knowledge, this is the first report of this type of resistance among S. Enteritidis in the Caribbean. This patient was treated with ciprofloxacin for 1 week; subsequent blood cultures were negative. This unusual isolate highlights the need to establish an antimicrobial resistance surveillance network for Salmonella isolates, including S. Enteritidis, to monitor the trends and new types of resistance mechanisms in the Caribbean. An epidemiologic study of S. Enteritidis infections is being planned to describe the extent of the problem and to define risk factors and vehicles of human infections in three Caribbean countries, including Trinidad and Tobago. B.P. Cherian,* Nicole Singh,* W. Charles,* and P. Prabhakar*† *Port of Spain General Hospital, Port of Spain, Trinidad; and †Caribbean Epidemiology Center (CAREC), Port of Spain, Trinidad References 1. Centers for Disease Control and Prevention. Salmonella surveillance annual tabulation summary, 1993-1995. Atlanta: U.S. Department of Health and Human Services; 1997. 2. Communicable Disease Surveillance Center. Salmonella in humans: PHLS salmonella data set, England and Wales, 1981-1996. London: The Center; 1997. 3. National Committee for Clinical Laboratory Standards. Performance standards for the anti-microbial disk susceptibility tests for bacteria that grow aerobically. Approved standard M7 - A4. Villanova (PA): The Committee; 1997. 4. Fantin B, Pangon B, Potel G, Caron F, Vallee E, Vallois JM, et al. Activity of sulbactam in combination with ceftriaxone in vitro and in experimental endocarditis caused by Escherichia coli producing SHV-2-like beta- lactamases. Antimicrob Agents Chemother 1990:34;581-6. 5. Morosini MI, Blasquez J, Negri MC, Canton R, Loza E, Baquero F. Characterization of a nosocomial outbreak involving an epidemic plasmid encoding for TEM-27 in Salmonella enteritidis. J Infect Dis 1996;174:1015-20. 6. Herikstad H, Hayes PS, Hogan J, Floyd P, Snyder L, Augulo FJ. Ceftriaxone resistant Salmonella in the United States. Pediatr Infect Dis J 1997;16:904-5. 7. Vahaboglu H, Hall LM, Mulazimoglu L, Dodanli S, Yildirim I, Livermore DM. Resistance extended spectrum cephalosporins caused by PER-1 beta lactamase, in Salmonella typhimurium in Istanbul, Turkey. J Med Microbiol 1995;43:294-9. 8. Kumar A, Nath G, Bhatia BD, Bhargava V, Loiwal V. An outbreak of multidrug resistant Salmonella typhimurium in a nursery. Indian Pediatr 1995;(3)980:881-5. 9. Wattal C, Kaul V, Chigh TD, Kler N, Bhandari SK. An outbreak of multi drug resistant Salmonella typhimurium in Delhi (India). Indian J Med Res 1994;100:266-7. 10. Rossi A, Lopardo H, Woloj M, Picanet A, Marino M, Galds M, et al. Non-typhoid Salmonella spp. resistant to cefotaxime. J Antimicrob Chemother 1995;36:697-702. New emm (M Protein Gene) Sequences of Group A Streptococci Isolated from Malaysian Patients To the Editor: We analyzed the M-type–specific emm gene sequences of 24 random Streptococcus pyogenes isolates from sterile- and nonsterile-site clinical specimens of Malaysian patients. In contrast to isolates in the United States, which rarely have new emm sequences, 6 of these 24 Malaysian isolates had new emm gene sequences, which suggests a large reservoir of group A streptococci expressing new M-type specificities in Malaysia. The M protein is a surface-exposed principal virulence factor of group A streptococci (GAS) and a potential vaccine candidate. The hypervariable M-type–specific N-terminal portion of the M molecule extends from the cell wall and evokes protective antibodies. Approximately 75 M antigenic types of GAS are recognized, and several provisional types have been proposed (1). Formulation of a universally effective vaccine is complicated by the M-typespecific nature of protective anti-GAS antibodies, temporal and geographic variations in GAS M-type prevalence (2), and lack of information on GAS M types from areas where rheumatic fever and rheumatic heart disease, sequelae of GAS pharyngitis, are endemic (3). The lack of specific M-typing antisera is also a limiting factor in determining type distribution. Recently, Beall and colleagues (4,5) demonstrated that sequence analysis of the hypervariable portion of the emm gene encoding M-type specificity (emm typing) was an alternative when M-typing antisera were not available. Attempts to type selected Malaysian strains of GAS by M protein status have yielded poor results. Fewer than 16% of strains were typable with standard M-typing antisera (6). The existence of new M types in Southeast Asia was suggested as an explanation. To investigate this possibility, we subjected 27 selected strains (6 from blood, 15 from pharyngitis, 3 from pus, and 3 pharyngeal carrier cultures) collected between January 1994 and December 1996 to emm typing. Initial isolation, serogrouping, T typing, and determination of opacity factor production were performed in Kuala Lumpur, by standard techniques, commercial media, reagents, and antisera (7). Strains were transported to the Centers for Disease Control and Prevention in Atlanta, Georgia, USA, for emm typing, where serogrouping, T typing, and opacity factor determinations were repeated, and emm typing was performed (4,5). DNA sequences were subjected to homology searches against all known emm sequences by Genetics Computer Group Software, Version 9. (Most sequences in this database were found in strains isolated from patients living in Europe and North America.) Of the 27 cultures analyzed, 24 were GAS, 2 were group G streptococci, and 1 was nongroupable Streptococcus. Ten of the 24 GAS strains were standard emm types emm3, emm12, emm22, emm60, and emm76 (encoding the classic M types M3, M12, M22, M60, and M76, respectively); 4 were the provisional emm types pt180, pt2841, and pt5757; and 3 were previously identified emm sequence types st64/14 and st2034 (GenBank accession numbers X72932 and U74320, respectively). The st2034 sequence, originally identified in children from Papua New Guinea, has also been found in Brazil, California, and Hawaii (B. Beall, R. Facklam, unpub. data). One GAS had a sequence previously found in group G streptococci (emmLG6, accession number U25741). Finally, 6 were of five new emm sequence types discovered in this study (st4529, st4547, st4532, st4545, and st3018, with accession numbers AF060368, AF052426, AF077666, AF077668, and AF077669, respectively). The newly found group A st4545 sequence was more similar to various group G streptococcal emm sequences than to known group A emm sequences. One group G isolate had a previously found group G 5' emm sequence (stLG6, accession number U25741). The nongroupable Streptococcus had an emm sequence previously associated with group L Streptococcus (Beall and Facklam, unpub. data). These results demonstrate the usefulness of emm typing in areas where specific M-typing antisera are not available. Identifying 6 (25%) of 24 GAS with new emm types provides further evidence of new M serotypes of GAS in Malaysia. The deduced amino acid sequences of the mature hypervariable N termini of ST4529, ST4532, ST4547, and ST3018 ranged from 43% to 82% identity to M proteins of known sequence (data not shown). The M nontypability of these isolates suggests unique serologic specificity. ST4547, ST4532, and ST3018 had 70% to 82% identity over the first 55-variable-region amino acids, with their closest matching known M proteins (M70, M27, and M22, respectively), but whether antibodies against any of these proteins would cross-protect against strains expressing these M proteins is unknown. Even though the M70 protein is 70% identical over its first 50 variable N terminal amino acids to the M33 protein, antibodies against the M70 and M33 proteins do not cross-protect, which suggests that no cross-protection would occur. The new deduced M protein with the lowest similarity to any known M protein was ST4529, whose closest match had a 43% identity over the N-terminal 55 residues of the M5 protein. st4529 likely encodes a new serospecifically unique M protein. These findings potentially affect vaccine development. Although new emm sequences were identified in a survey in the United States (5), the percentage of new strains with new emm sequences was much lower (6%) than was found with these Malaysian isolates. emm typing of a larger number of strains from rheumatic fever–and rheumatic heart disease–endemic areas is required to deduce amino acid sequences for the development of a suitable M protein-based vaccine. Acknowledgments We thank Sukeri Kasni and Theresa Hoenes for technical assistance. This work was supported by University Kebangsaan Malaysia sabbatical leave grant and Fulbright Fellowship Award (1997) to Farida Jamal. Farida Jamal,* Sabiha Pit,* Richard Facklam,† and Bernard Beall† *University Kebangsaan Malaysia, Jalan Raja Muda Abdul Aziz, Kuala Lumpur, Malaysia; and †Centers for Disease Control and Prevention, Atlanta, Georgia, USA References 1. Fraser CAM, Colman G. Some provisional M-types among Streptococcus pyogenes (Lancefield group A). In: Recent advances in streptococci and streptococcal diseases. Kimura Y, Kotami S, Shiokawa Y, editors. Bracknell (UK): Reedbooks Ltd; 1985. p. 35-6. 2. Musser JM, Kapur V, Kanjilal S, et al. Geographical and temporal distribution and molecular characterization of highly pathogenic clones of Streptococcus pyogenes expressing allelic variants of pyogenic exotoxin A (scarlet fever toxin). J Infect Dis 1993;167:337-46. 3. Kaplan EL. Global assessment of rheumatic fever and rheumatic heart disease at the close of the century. Circulation 1993;88:1964-72. 4. Beall B, Facklam R, Thompson T. Screening emm-specific PCR products for routine and accurate typing of group A streptococci. J Clin Microbiol 1996;34:953-8. 5. Beall B, Facklam R, Hoenes T, Schwartz B. Survey of emm gene sequences and T-antigen types from systemic Streptococcus pyogenes infection isolates collected in San Francisco, CA, Atlanta, GA and Connecticut in 1994 and 1995. J Clin Microbiol 1997;35:1231-5. 6. Jamal F, Pit S, Johnson DR, Kaplan EL. Characterization of group A streptococci isolated in Kuala Lumpur. J Trop Med Hyg 1995;98:343-6. 7. Johnson DR, Sramsk J, Kaplan EL, Bicova R, Havlicek J, Havlickova H, et al. Laboratory diagnosis of group A streptococcal infection. Geneva: World Health Organization; 1996. Mutant Chemokine Receptor (CCR-5) and Its Relevance to HIV Infection in Arabs To the Editor: Approximately 10% of HIV-infected patients may remain AIDS-free for a long time; moreover, some persons do not become infected with HIV despite multiple high-risk sexual exposures (1,2). Factors responsible for this relative resistance to infection and disease include cytotoxic T cells, neutralizing antibodies, high concentrations of certain chemokines (e.g., RANTES, MIP-1), human leukocyte antigen haplotype, mannose-binding protein, and tumor necrosis factor alpha, C4, and TAP polymorphism (2-4). One of the chemokine receptors, CCR-5, which along with CD4 acts as co-receptor for HIV entry into macrophages, provides upon mutation a genetic restriction to HIV infection in homozygous persons and control of disease progression in heterozygous persons (5,6). Thus a 32bp deletion in the open reading frame of the region encoding the second extracellular loop of this receptor causes synthesis of a highly truncated protein that fails to express on the cell surface, leading to loss of HIV-1 co-receptor activity. Studies in healthy Caucasian Europeans and North Americans show that approximately 1% of the population are homozygous for this deletion ([delta]32), whereas 15% to 20% are heterozygous (5-9); surprisingly, a higher percentage (up to 20%) of persons at high risk for HIV but HIV- negative are homozygous for this deletion. However, no such mutation is seen in Japanese, Native Americans, Chinese, Africans, and Tamil Indians, which suggests that in these groups either resistance to HIV infection is not present or factors other than mutated CCR-5 are in operation. African-Americans and Hispanics show a low rate of mutation, possibly because of intermarriage with Caucasians (4). The low frequency of CCR-5 mutation in Arabs with close contacts with Turks in the Eastern Province of Saudi Arabia may also be due to intermarriage. However, certain persons with mutated CCR-5 can become HIV-infected (10); in such cases other chemokine receptors (e.g., CXCR-4, CCR-2, and CCR-3) are believed to facilitate infection. HIV infection in Saudi Arabia (population 18 million) is uncommon; the World Health Organization has so far (1985 to 1997) documented 334 cases of AIDS in this region (11). We, therefore, studied for the first time the mutation of CCR-5 in Arabs residing in Saudi Arabia. DNA was isolated from the peripheral blood mononuclear cells of 105 male blood donors not infected with HIV and nine HIV-infected patients (seven male and two female). The latter were divided into three groups, according to published criteria (2): four persons whose infection did not progress over the long term and who showed only modest decline of CD4 count after several years of infection, one person whose infection progressed normally, and four persons whose infection progressed rapidly and CD4 count fell below 100/µl within 2 years. Primers flanking 32 nucleotide deletion of CCR-5 were used to generate wild type (W) and deleted ([delta]32) fragments of 189 bp and 157 bp, respectively (5). Amplification was done in a Perkin-Elmer thermal cycler 9,600, by a 20 µl reaction mixture that contained 0.25mM of dNTPs, 20 pM of each primer ('5-CAAAAAGAAGGTCTTCATTACACC-3, 5-CCTGTGCCTCTTCTTCTCATTTCG-3'), and 0.5 units of Taq polymerase in 1x reaction buffer. All reagents were obtained from Pharmacia (Sweden). The amplified product was separated on 2% agarose at 120 V for 45 min and examined under UV light. Of the uninfected blood donors, 104 (99%) were homozygous for the wild type, and 1 (0.96) was heterozygous for the mutation. None of the HIV-infected patients had the mutation. Thus, the mutation is present, albeit infrequently, in Arabs. A review of 68 HIV-infected patients in our files showed that, as in Caucasians, infection progressed rapidly in 8%, did not progress over the long term in 6%, and progressed normally in 86% (2). Therefore, other hitherto unknown protective factors must be operative in Arabs. Iman H. Al-Sheikh, Amjad Rahi, and Mohammed Al-Khalifa King Faisal University and Regional Laboratory, Ministry of Health, Dammam, Saudi Arabia References 1. Malkovsky M. HLA and natural history of HIV infection. Lancet 1996;348:142-3. 2. Haynes BF, Pantaleo G, Fauci AS. Toward an understanding of the correlates of protective immunity to HIV infection. Science 1996;271:324-8. 3. Cocchi F, DeVico AL, Garzine-Demo A, Arya SK, Gallo RC, Lusso P. Identification of RANTES, MIP-1a and MIP-1b as the major HIV suppressive factors produced by CD8+ T-cells. Science 1995;270:1811-5. 4. McNicholl JM, Smith DK, Qari SH, Hodge T. Host genes and HIV: the role of the chemokine receptor gene CCR-5 and its alleles (32 CCR-5). Emerg Infect Dis 1997:3:261-71. 5. Huang VY, Paxton WA, Wollinsky SM, Neumann AU, Zhang L, He T, et al. The role of mutant CCR-5 allele in HIV-1 transmission and disease progression. Nature Med 1996;2:1240-3. 6. Dean M, Carrington M, Winkler C, Huttley GA, Smith MW, Allikmets R, et al. Genetic resistance of HIV infection and progression to AIDS by deletion of the CKR5 structural gene. Science 1996;273:1856-62. 7. Liu R, Paxton WA, Choe S, Ceradini D, Martin SR, Horuk R, et al. Homozygous defect in HIV co-receptor accounts for resistance of some multiply-exposed individuals to HIV-1 infection. Cell 1996;86:367-77. 8. Samson M, Libert F, Duranz BJ, Rucker J, Liesnard C, Farber CM, et al. Resistance to HIV-1 infection in caucasian individuals bearing mutant alleles of the CCR-5 chemokine receptor gene. Nature 1996;382:722-5. 9. Zimmerman PA, Bucklewhite A, Alkhatib G, Spalding T, Kubofcik J, Combadiere C. Inherited resistance to HIV-1 conferred by an inactivating mutation in CC chemokine receptor 5- studies in populations with quantified risk. Mol Med 1997;3:23-36. 10. O'Brien TR, Winkler C, Dean M, Nelson, JAE, Carrington M, Michael NL, et al. HIV-1 infection in man homozygous for CCR-5 (delta)32. Lancet 1997;349:1219-20. 11. World Health Organization. AIDS update: reported AIDS cases in East Mediterranean Region. EMR AIDS News 1997;1:8. Emerging Infectious Diseases National Center for Infectious Diseases Centers for Disease Control and Prevention Atlanta, GA URL: ftp://ftp.cdc.gov/pub/EID/vol5no1/ascii/letters.txt Please note that figures and equations are not available in ASCII format; their placement within the text is noted by [fig] and [eq], respectively. Greek symbols are spelled out. The following codes are used: (ft) for footnote; (sup) for superscript; (sub) for subscript; >/= for greater than or equal to.