|Home | About | Journals | Submit | Contact Us | Français|
Human monocytotropic ehrlichiosis (HME) is an emerging tick-transmitted zoonosis in the United States caused by Ehrlichia chaffeensis. Ehrlichia canis, E, chaffeensis and E. ewingii have recently been detected in dogs and Rhipicephalus sanguineus ticks from Cameroon; thus the potential exists for human infections. The objective of this study was to determine if Ehrlichia species were associated with acute fevers of unknown etiology in patients from the coastal region of Cameroon. E. chaffeensis was detected in peripheral blood from 12 (10%) of 118 patients using real-time polymerase chain reaction (PCR) amplification of the genus-specific disulfide bond (dsb) formation protein gene. Furthermore, DNA sequencing of PCR amplicons revealed that the dsb gene sequence was identical to E. chaffeensis (Arkansas strain). Patients with detectable E. chaffeensis DNA had clinical manifestations that included fever, headache, myalgia, arthralgia, pulmonary involvement, and diffuse rash.
Human monocytotropic ehrlichiosis (HME) has emerged as an important tick-borne infectious disease in the United States after the discovery of Ehrlichia chaffeensis, an obligately intracellular gram-negative bacterium transmitted primarily by Amblyomma americanum ticks.1–3 E. chaffeensis was isolated in 1991,4 and since that time HME or evidence of E. chaffeensis has been reported in more than 30 states in the US,5 Africa,6,7 Israel,4,8,9 Latin America10,11 and Asia.12–14 A geographically limited serosurvey for human ehrlichioses in Africa suggests that human ehrlichiosis exists, but is an infrequent infection.6,7 Noteworthy is a serologically and clinically well documented case of HME acquired in Mali and diagnosed in the United States,7 which provides the strongest evidence that E. chaffeensis is circulating among yet to be determined reservoirs and vectors in Africa.
A. americanum is found only in the United States;15 however, E. chaffeensis DNA has been detected in other tick species such as Dermacentor variabilis, Ixodes pacificus, A. testudinarium, Haemaphysalis longicornis and H. yeni,16–20 suggesting that the agent is not exclusive to A. americanum. Most recently, E. canis and E. ewingii, were detected in Cameroonian dogs21 and in Rhipicephalus sanguineus ticks obtained from those dogs.22 E. chaffeensis has not been reported in R. sanguineus ticks in the United States, but E. canis and E. ewingii DNA has been detected in R. sanguineus ticks from Oklahoma.23 Although R. sanguineus ticks rarely bite humans in the United States, two stages (larvae and nymph) of these ticks commonly bite humans in Africa and, therefore, may be an important vector in the region with the potential to transmit these zoonotic agents to humans.
In this study, we used a highly sensitive, genus-specific PCR assay to diagnose ehrlichiosis in patients who presented with symptoms of acute febrile illness at local clinics in the South West Province of Cameroon and whose laboratory test results for malaria and typhoid fever, the two known endemic fevers, were negative.
Peripheral blood (3 mL) was collected in sterile tubes containing anticoagulant (EDTA) from patients who presented with febrile illness at the Cameroon Development Corporation Central Clinic in Tiko and the Mount Mary Health Center in Buea, Cameroon between January and June 2003. Patient samples were routinely tested for detection of malaria parasites and for antibodies diagnostic of typhoid fever. Patient samples, which tested negative for both malaria and typhoid fever, were transported on ice to the Rickettsial Laboratory at the University of Buea for diagnosis of ehrlichial infection. Whole blood was collected from 118 patients (77 females and 41 males), and a recent medical history and observed clinical signs were recorded for each patient. Patients also voluntarily provided information on contact with tick-infested domesticated animals. The patients resided in different localities along the coast of Cameroon: Buea (4°9′N, 9°13′E), 29 patients; Limbe (4°2′N, 9°19′E), 38 patients; Muyuka (4°10′N, 9°25′E), 19 patients; and Tiko (4°2′N, 9°19′E), 32 patients. This research was conducted with approval according to the guidelines governing research at the clinical institutions from where patient samples were collected and at the University of Buea.
DNA was extracted from 50 µL of whole blood using the DNeasy Tissue Extraction Kit (Qiagen, Chatsworth, CA) following the manufacturer’s protocol. Purified DNA was quantified using a digital spectrophotometer at 260 nm wavelength (Perkin Elmer MBA 2000, Norwalk, CT) and stored at 4° C until used as template for PCR amplifications.
DNA extracted from blood was quantitated by spectrophotometry (A260) and 250 ng of each sample was added to individual reactions that included the Ehrlichia genus-specific primer pair Dsb-330 (forward) and Dsb-728 (reverse) that amplified a 409 bp of the dsb gene as previously described.24 The amplification reaction, in a final volume of 25 µl, contained 12.5 µl of iQ SYBRGreen Supermix (Bio-Rad, Hercules, CA) and 0.5 µl of each primer at 20 µM (final concentration of 400 nM). PCR cycling conditions consisted of 95° C for 2 min and 50 cycles of 15 s at 95° C, 30 s at 58° C and 30 s at 72° C. In each set of reactions, E. canis genomic DNA was included with each run as a positive control in addition to negative control reactions without DNA template. PCR was performed in 96-well plates using an iCycler iQ multicolor real-time PCR detection system equipped with appropriate filter sets and analyzed with iQ software v 3.1 (BioRad Laboratories). Samples were considered positive and threshold cycle (CT) was determined after it crossed the minimum threshold level above background fluorescence after baseline subtraction.
All PCR amplicons obtained from six patient blood samples in the present study were also confirmed by agarose (1.5%) gel electrophonesis and visualized with ethidium bromide using an ultraviolet transilluminator. PCR amplicons (409-bp) were purified using EXOSAP-IT (USB Corporation, Cleveland, Ohio) according to the manufacturer’s instructions and sequenced directly with the same primers (Dsb-330 and Dsb-728) used for PCR on an ABI automated sequencer (UTMB Protein Chemistry Laboratory). The BLAST program (National Center for Biotechnology Information, Bethesda, MD) was used to compare dsb sequences in order to determine the species.
Patients that had detectable E. chaffeensis DNA were screened for antibodies against E. chaffeensis (Arkansas) antigen slides as previously described.25 For screening, sera were diluted 1:64 in phosphate-buffered saline (PBS) containing 1% bovine serum albumin (BSA) and 0.1% Tween 20.
The 118 patients (77 females and 41 males) whose blood samples did not contain a detectable level of malaria parasites or antibodies to Salmonella species were studied. Their ages ranged from 1 to 64 years (mean 26.2 years). All the patients (100%) included in this study had a fever with onset ranging from one to nine days (average of four days) duration prior to consultation and sample collection. A minority of patients (n=16, 14%) reported having taken antibiotics. Other clinical signs included headache 96 (81%), arthritis 52 (44%), myalgia 48 (41%), pulmonary involvement 36 (31%) and rash 19 (16%); seven patients required hospitalization. Forty–two (36%) patients reported owning dogs or having domesticated animals (goats and sheep) on their property, and 36 (31%) confirmed that their animals were tick infested.
PCR amplification of patient peripheral blood samples detected ehrlichial DNA in 12 (10%) of the 118 patients (6 females and 6 males). One patient was an infant (1 year-old), but the mean age of the patients was 23 years-old. Five patients were from Buea, three from Tiko and two each from Limbe and Muyuka. Clinical symptoms, in addition to fever, included headache (67%), myalgia (42%), arthralgia (58%), pulmonary involvement (17%) and rash (17%) (Table 1). None of the patients with ehrlichial DNA had any antibiotic therapy, and only one required hospitalization. Nine of the patients reared domesticated animals, and eight had contact with tick-infested animals.
Agarose gel electrophoresis of all the above samples that were detected by real-time PCR revealed a single band corresponding to the expected amplicon size (409 bp) (Fig. 1). Sequenced amplicons were 100% identical to each other and to the corresponding sequence of the United States isolates of E. chaffeensis (Arkansas) (GenBank accession numbers CP000236 and AF403711). Sequences from E. chaffeensis dsb PCR amplicons obtained from Cameroonian patients were deposited into GenBank and assigned accession numbers (EF375885-EF375887).
Patients that had detectable E. chaffeensis DNA did not have detectable antibodies to E. chaffeensis as determined by IFA.
This study is the first to provide molecular and clinical evidence of E. chaffeensis infection in a well characterized patient group with undifferentiated febrile illness in Africa. Although HME has been recognized as an important emerging disease causing extensive morbidity among humans in the United States, the geographic distribution of the disease remains undefined due to its recent emergence, challenges in diagnosing the disease, and lack of comprehensive epidemiological studies outside the United States. HME is difficult to diagnose because the clinical findings, which include fever, malaise, headache, myalgia, rigors, cough and rash are nonspecific, serology is often negative at the time of presentation, and isolation of the etiologic agent is difficult and requires weeks to obtain a result.26 These HME patients reported symptoms and presented with clinical signs consistent with HME infection. Although molecular evidence of E. chaffeensis infection was demonstrated in these patients, we did not attempt to isolate the agent from blood or tissue of the suspected patients, nor were additional samples collected after convalescence to confirm infection serologically. However, at the time of presentation (average 4 days after onset of symptoms) antibodies were not present, therefore detection of the agent by PCR provided the best opportunity to identify E. chaffeensis-infected individuals. Molecular detection of E. chaffeensis appears to be most sensitive in HME patients in the acute phase at the time of presentation.27 Nevertheless, future investigations including serologic evaluation and attempts to isolate the agent would be useful towards full characterization the E. chaffeensis strain(s) infecting patients in this country.
The E. chaffeensis dsb gene amplified from the blood of these febrile patients was identical to the gene sequence from US isolates. The homology observed between the E. chaffeensis strains from Cameroon and the United States suggests a common origin or introduction of the pathogen perhaps from imported dogs originating from the United States; however, it may also be endemic to this region, but unrecognized. Previously, we reported that the sequence of dsb genes of E. canis from dog blood and/or R. sanguineus ticks from Cameroon were identical to the isolates from the United States.19,20,25 Although the Cameroonian E. canis and North American E. canis dsb genes were identical, Cameroonian ehrlichiae could be distinguished molecularly based on the number of repeat units of the gp36 gene.28 This observation indicates that the dsb gene is conserved in geographically dispersed isolates and characterization of additional genes that exhibit more diversity, such as VLPT, would potentially identify E. chaffeensis genotypes in Cameroon that differ from genotypes present in the United States.
HME manifests as an undifferentiated flu-like febrile illness, which is not easily clinically differentiated from many other febrile tropical diseases. Currently, diagnosis of HME is still largely based on the combined evaluation of clinical signs, laboratory and epidemiological data. Since most physicians are unfamiliar with HME, this disease is often misdiagnosed and underdiagnosed even in the United States.29,30 The previous lack of documented HME cases in Cameroon may be due to the fact that it is generally considered a pathogen confined to North America and has been rarely diagnosed outside the United States. Furthermore, tick-borne diseases are not routinely considered in a differential diagnosis by local clinicians whose primary diagnostic focus is on endemic malaria and typhoid fever. This situation is complicated by the fact that the diagnosis typically requires specialized skills and equipment which are unavailable in local clinics. However, tick exposure in Cameroon is high and is associated with recreation, occupational, and peridomestic activities. A recent survey of febrile patients whose serum samples did not contain antibodies against Salmonella typhi and did not contain blood malarial parasites revealed a high prevalence of antibodies to Rickettsia, and subsequence molecular studies identified R. africae in the blood of seven of the 118 patients reported in this study.31,32 Notably, E. chaffeensis was not detected in the blood of any of the patients who were infected with R. africae.32 The difficulty in clinical discrimination between these tick-borne bacterial diseases and that caused by the hyperendemic Plasmodium falciparum may also contribute to the lack of recognition of these diseases.
The clinical course of this and other tick-borne diseases requires further investigation especially in sub-Saharan African countries where the prevalence of human immunodeficiency virus (HIV) infection is high. Severe manifestations of HME have been reported to occur more frequently among patients infected with HIV.33 Thus, ehrlichiosis may be a life-threatening illness in HIV-infected persons. Although this study gives conclusive evidence of HME infection in Cameroon, other issues that should be addressed are a full description of the clinical spectrum of ehrlichiosis in African patients, determination of risk factors for severe illness, and isolation of the pathogen.
This research was supported by a National Institutes of Health training grant from the Fogarty International Center (D43-TW00903; DHW) and the Sealy Center for Vaccine Development at the University of Texas Medical Branch (JWM).
Lucy M. Ndip, Department of Biochemistry and Microbiology, University of Buea, Buea, Cameroon.
Marcelo Labruna, Departamento de Medicina Veterinaria Preventiva e Saude Animal, Faculdade de Medicina Veterinaria e Zootecnia, Universidade de São Paulo, São Paulo, Brazil.
Roland N. Ndip, Department of Biochemistry and Microbiology, University of Fort Hare, Alice, 5700, South Africa and Department of Biochemistry and Microbiology, University of Buea, Buea, Cameroon.
David H. Walker, Department of Pathology and Center for Biodefense and Emerging Infectious Diseases, University of Texas Medical Branch, Galveston, Texas 77555-0609.
Jere W. McBride, Departments of Pathology and Microbiology and Immunology, Sealy Center for Vaccine Development, University of Texas Medical Branch, Galveston, Texas 77555-0609.