Serologic and Molecular Detection of Ehrlichia chaffeensis and Anaplasma phagocytophila (Human Granulocytic Ehrlichiosis Agent) in Korean Patients

Abstract

Sera from 491 Korean patients with acute febrile diseases were tested for Ehrlichia chaffeensis and Anaplasma phagocytophila antibodies by indirect immunofluorescence assay (IFA), Western blotting, and TaqMan real-time PCR. Overall, 0.4% of sera reacted with E. chaffeensis, and 1.8% reacted with A. phagocytophila in IFAs. This is the first report of detection of antibodies to A. phagocytophila and E. chaffeensis in Korea and suggests the presence of A. phagocytophila and E. chaffeensis or antigenically similar species.

Human granulocytic ehrlichiosis (HGE) and human monocytic ehrlichiosis (HME) are emerging, tick-transmitted infectious diseases described in many areas of the United States and (for HGE) Europe. During a recent reclassification, the names “HGE agent” and “Ehrlichia equi” were merged with Ehrlichia phagocytophila, which was changed to Anaplasma phagocytophila, and the etiological agents of HGE and HME, Anaplasma phagocytophila and Ehrlichia chaffeensis, respectively, were moved into the family Anaplasmataceae (9).

HME was first reported in 1987, and E. chaffeensis was isolated 4 years later (8). HGE was identified in 1990 and reported in 1994 (6). HME and HGE were first described in parts of Europe in 1991 and 1995, respectively, and serologic evidence reveals that A. phagocytophila infections occur broadly across Europe (2, 4, 14, 15). Although definitive cases have not been identified outside of North America and Europe, additional seroepidemiologic and molecular studies suggest their presence elsewhere, including Africa and Asia (5, 20). There have been no reports of ehrlichiosis in both humans and animals in Korea. In this study, we describe the first serologic evidence for E. chaffeensis and A. phagocytophila infections in Korean subjects and confirm these findings with Western blots and TaqMan real-time PCR.

Sera of 491 febrile patients from Jeonbuk and Jeonnam Public Health and Environmental Research Institute (PHERI) and the National Institute of Health in Korea (KNIH) were collected from June through November 2000 and were submitted for routine tests for hemorrhagic fever with renal syndrome (HFRS), Orienta tsutsugamushi, leptospirosis, Lyme disease, and Rickettsia typhi. A total of 172 individuals were seropositive for O. tsutsugamushi (35.0%), and 21 (3.8%) were seropositive for Leptospira spp. The remainder were seronegative for these infectious agents. The intervals after fever when samples were obtained were not known.

Serologic tests for A. phagocytophila and E. chaffeensis were conducted at The Johns Hopkins University School of Medicine (21). A. phagocytophila was propagated in the HL-60 cell line in RPMI 1640 medium supplemented with 1% fetal bovine serum (FBS) and 2 mM l-glutamine, whereas E. chaffeensis was propagated in the DH82 cell line in EMEM (Gibco) medium supplemented with 5% FBS and 2 mM l-glutamine (11, 12, 21). The indirect immunofluorescence assay (IFA) was performed as described previously (21). Briefly, Teflon-coated slides with heavily infected HL-60 and DH82 cells were acetone fixed before use. Human sera were diluted 1:80 in phosphate-buffered saline (PBS) with 0.5% nonfat dry milk (PBSM), added to the antigen slides, and incubated at room temperature (RT) for 1 h. After washing, fluorescein isothiocyanate (FITC)-labeled goat anti-human immunoglobulin G (IgG), IgA, and IgM (Kirkegaard & Perry Laboratories, Gaithersburg, Md.) were added at a dilution of 1:50 and incubated at RT for 1 h. After washing, the antigen slides were counterstained with 0.005% Evans blue and mounted for examination by fluorescence microscopy. All of the sera were tested for O. tsutsugamushi antibodies by IFA at PHERI or KNIH.

For the Western blot assay, A. phagocytophila and E. chaffeensis used as antigens were purified by Renografin density gradient centrifugation (1, 3, 22). Antigens were separated by sodium dodecyl sulfate-polyacrylamide gel electrophoresis, transferred to nitrocellulose, and cut into antigen strips. Membranes were blocked with Tris-buffered saline (TBS; 20 mM Tris, 500 mM NaCl [pH 7.5]) containing 0.5% nonfat dry milk (NFDM) for 1 h. Human sera diluted 1:80 in 1% normal goat serum diluted with 0.1 M PBS, with 0.05% Tween 20 and 0.5% nonfat dry milk (PBSTM), were added to individual antigen strips and allowed to react at RT for 1 h. Following washes, alkaline phosphatase-labeled goat anti-human IgG, IgA, and IgM (KPL) were added, and this mixture was incubated for 1 h. After additional washes, BCIP-NBT (5-bromo, 4-chloro, 3-indolyl phosphate-nitroblue tetrazolium chloride) was used as a substrate for color development.

Real-time PCR was conducted by a procedure modified from that of Pusterla et al. (16). Genomic DNA was extracted from 200 μl of IFA-reactive sera with a QIAmp blood kit (Qiagen, Chatsworth, Calif.) according to the manufacturer's directions. The TaqMan probe and primers for the 16S rRNA gene of Ehrlichia and Anaplasma spp. were ESP-P (5′-6-FAM-ACGCGTTAAGCACTCCGCCTGG-TAMRA-3′), ESP-F (5′-ACGATGAGTGCTGAATGTGG-3′), and ECH-R (5′-TAACTGAGTTGTCGATCCAATGAAA-3′). PCR was performed in a 20-μl volume containing 450 nM each primer and 250 nM fluorescent-labeled probe; TaqMan Universal PCR master mix (Applied Biosystems) was added to each reaction and cycled in a Sequence Detection System 7700 (Applied Biosystems) thermal cycler under the following conditions: uracil N-deglycosylase digestion at 50°C for 2 min, AmpliTaq Gold preactivation at 95°C for 10 min, and 45 cycles at 95°C for 20 s and 55°C for 60 s. Positive and negative DNAs from organisms of related tick-borne diseases were tested to examine the specificity for TaqMan real-time PCR with the probe and primers designed for Ehrlichia and Anaplasma species (data not shown).

Of the 491 Korean sera, 9 reacted with A. phagocytophila and 2 reacted with E. chaffeensis by IFA; no cross-reactions between A. phagocytophila and E. chaffeensis were detected in these samples. Seven of the 11 samples were also positive for O. tsutsugamushi antibodies (Table 1). The reactive sera were from individuals between 43 and 80 years old (Table 1). Western blotting revealed that both E. chaffeensis IFA-positive sera reacted with the 28-kDa antigen of E. chaffeensis, and one of the nine A. phagocytophila IFA-positive sera reacted with the A. phagocytophila 44-kDa protein (Fig. 1).

TABLE 1.

Sex and age characteristics and positive results of serological testing for A. phagocytophila and E. chaffeensis confirmed by Western blotting and real-time PCR

City and/or province	Sexa	Age (yr)b	IFA titer (Western blot result)c		Real-time PCR resultd	IFA result for O. tsutsugamushie
			A. phagocytophila	E. chaffeensis
Seongbuk, Seoul	M	U	320 (−)	N	−	−
Paju, Gyeounggi	M	43	640 (+)	N	−	−
Youngcheon, Gyeongbuk	M	56	N	320 (+)	−	−
Muju, Jeonbuk	M	80	N	2,560 (+)	−	+
Jangsu, Jeonbuk	M	72	640 (−)	N	−	−
Gokseong, Jeonnam	M	47	2,560 (−)	N	−	+
Naju, Jeonnam	M	75	2,560 (−)	N	−	+
Muan, Jeonnam	F	53	1,280 (−)	N	−	+
Suncheon, Jeonnam	F	54	640 (−)	N	−	+
Yeongam, Jeonnam	F	49	1,280 (−)	N	−	+
Kangjin, Jeonnam	F	80	80 (−)	N	+	+

^aM, male; F, female.

^bU, unknown.

^cN, negative in IFA test; −, negative in Western blot; +, positive in Western blot. There were no significant cross-reactions among A. phagocytophila, E. chaffeensis, and O. tsutsugamushi.

^dDNA was purified and amplified with the Ehrlichia or Anaplasma sp. 16S rRNA gene fragment by real-time TaqMan PCR with genus-specific primers.

^eAll of the sera were tested by IFA for O. tsutsugamushi before being tested for A. phagocytophila and E. chaffeensis.

FIG. 1.

Western blot with A. phagocytophila (A) and E. chaffeensis (B) as antigens reacted with IFA-positive serum of Korean patients with acute fever. (A) Lanes: M, size marker; 1, uninfected HL-60 cell antigen reacted with A. phagocytophila IFA-negative serum; 2, uninfected HL-60 cell antigen reacted with A. phagocytophila IFA-positive serum; 3, purified A. phagocytophila antigens reacted with monoclonal antibody 20B4, which detects A. phagocytophila Msp2 (44-kDa protein antigen; 4, purified A. phagocytophila antigens reacted with A. phagocytophila IFA-positive Korean human serum. (B) Lanes: M, size marker; 1, DH82 cell antigens reacted with E. chaffeensis IFA-negative serum; 2, DH82 cell antigens reacted with E. chaffeensis IFA-positive control serum from a U.S. patient; 3, purified E. chaffeensis antigen reacted with E. chaffeensis IFA-positive control serum from a U.S. patient; 4 to 5, purified E. chaffeensis antigen reacted with E. chaffeensis IFA-positive Korean human sera.

An Anaplasma or Ehrlichia species 16S rRNA gene fragment was amplified by TaqMan real-time PCR with genus-specific primers from DNA of the nine positive sera. This serum was from an individual from Kangjin, Jeonnam, Korea. Sequence analysis was not performed.

In this report, we evaluated exposure to A. phagocytophila and E. chaffeensis or antigenically similar bacteria in human patients in Korea. The distribution of infections seems to be mainly in agricultural regions of Korea, where various ticks are commonly present. Since there is a high seropositive rate (35.0%) in Korea, especially in Jeonnam province, many patients are also O. tsutsugamushi seropositive. The next most common disease diagnosed was leptospirosis, with 3.8% of the sera positive. The serologic tests for A. phagocytophila and E. chaffeensis were positive mainly in patients who were middle age or older (43 to 80 years old).

Eleven (2.2%) of the 491 sera reacted with the antigens of A. phagocytophila and E. chaffeensis at an antibody titer of ≥1:2,560. The seropositive rate of A. phagocytophila (1.8%) was higher than that of E. chaffeensis (0.4%), but the difference was not significant.

We found that 6 of 11 A. phagocytophila-positive sera were also seropositive for O. tsutsugamushi, suggesting a cross-reaction. To evaluate this possibility, four samples seropositive only for O. tsutsugamushi were retested as seronegative in IFA with A. phagocytophila and E. chaffeensis antigens. This is further supported by the lack of E. chaffeensis or A. phagocytophila antibodies in the remaining 166 sera that were found reactive with O. tsutsugamushi. Thus, serologic cross-reactions between E. chaffeensis or A. phagocytophila and O. tsutsugamushi are probably not significant, and human ehrlichiosis sometimes exists as a mixed infection with scrub typhus. Serologic cross-reactions between A. phagocytophila and E. chaffeensis are frequent (10), but were not detected in any of these Korean samples.

Although two samples were positive with E. chaffensis antigen, we cannot exclude infection by serologically related species, such as E. canis or E. muris, which are known to be present in Asian animals. Further serological studies with a variety of antigens of Ehrlichia spp. may allow more precise identification of the infecting agent in these and similar cases.

The IFA is the most sensitive serodiagnostic method; however, inconsistent results may occur because of antigenic diversity and technical differences among laboratories. For example, A. phagocytophila (E. equi) MRK strain antigen revealed a high degree of variability, even when performed in the same laboratory, depending upon antigen preparation and method of analysis (21). Because of the potential for nonspecificity with serologic assays, we attempted confirmation by amplification of Ehrlichia and Anaplasma 16S rRNA gene fragments from the positive sera. The single Taqman PCR-positive sample had an A. phagocytophila IFA titer of 80, but was negative by A. phagocytophila Western blot. We suspect that the low titer represents early infection when circulating bacteria or DNA might be found. Nested PCR for the 16S rRNA gene of A. phagocytophila did not confirm this result, a phenomenon previously observed when serum is used as the template for amplification of E. chaffeensis or A. phagocytophila (7).

Immunoblot analyses showed specific antibodies for the agents, including one of nine sera reactive with the diagnostic 44-kDa A. phagocytophila antigen (1, 17) and both E. chaffeensis IFA-positive sera reactive with a diagnostic 28-kDa E. chaffeensis antigen band (3, 22). The reactions were generally weak and absent for most of the A. phagocytophila IFA-positive samples, perhaps due to the existence of antigenically divergent strains in Korean ticks. Although prior studies suggest that the A. phagocytophila 44-kDa antigen is a sensitive marker, previous studies have been conducted with sera obtained from a single specific location or that included only a small number of samples overall (13, 17, 22). Cultivation of the causative agents would provide definitive evidence and would help to devise the best serologic tests.

Although rickettsial diseases (in particular, those caused by O. tsutsugamushi) are well studied in Korea (18), there are no reports regarding the incidence of Ehrlichia spp. Although no definitive, culture-confirmed cases were identified, this report provides the first evidence that Koreans may be infected by A. phagocytophila and E. chaffeensis or antigenically similar organisms (19). Additional studies are needed to isolate and identify Ehrlichia and Anaplasma spp. from humans and animals in Korea to confirm their presence and capacity to cause disease.