Human herpesvirus 8 seroconversion in Kenyan women by enzyme-linked immunosorbent assay and immunofluorescence assay

Journal of Clinical Virology 30 (2004) 137–144

Human herpesvirus 8 seroconversion in Kenyan women by enzyme-linked immunosorbent assay and immunofluorescence assay Bhavna H. Chohan a,c,1 , Heather Taylor b,2 , Rosemary Obrigewitch b,2 , Ludo Lavreys c,3 , Barbra A. Richardson d,4 , Kishorchandra N. Mandaliya f,5 , Job J. Bwayo a,6 , Joan K. Kreiss c,e,7 , Rhoda Ashley Morrow b,∗ a

Department of Medical Microbiology, University of Nairobi, P.O. 19676, Nairobi, Kenya b Department of Laboratory Medicine, University of Washington, Seattle, WA, USA c Department of Epidemiology, University of Washington, Seattle, WA, USA d Department of Biostatistics, University of Washington, Seattle, WA, USA e Department of Medicine, University of Washington, Seattle, WA, USA f Coast Provincial General Hospital, P.O. 91276, Mombasa, Kenya Received 30 April 2003; accepted 1 August 2003

Abstract Background: Human herpesvirus 8 (HHV-8) antibody tests vary in reported sensitivity and specificity, depending on the population tested and the assay. Objective: The purpose of this study was to compare the ability to detect seroconversion to HHV-8 in a cohort of HHV-8 seronegative female commercial sex workers in Kenya using three tests: HHV-8 viral lysate-based enzyme-linked immunosorbent assay (ELISA), an immunofluorescence assay for HHV-8 lytic antigens (IFA-lytic) and IFA for latent nuclear antigens (IFA-LANA). Study design: By ELISA, 16 women from a prospective cohort of commercial sex workers were identified as seroconverting to HHV-8. A total of 124 post-enrollment samples from these 16 women as well as the enrollment samples were tested for HHV-8 antibodies by all three assays to monitor seroconversion. Results: Of 16 women with apparent seroconversion by ELISA, 8 had a rise in IFA-lytic titers either concomitant with or prior to the first positive ELISA sample and no initial LANA by IFA. Five of the 16 women were IFA-LANA positive at entry, indicating prior infection with HHV-8. Three women had no evidence of seroconversion by either IFA-lytic or IFA-LANA and two of these three had increased ELISA reactivity concomitant with HIV-1 infection. Conclusions: Conversion from a negative to a positive ELISA result for HHV-8 antibody indicated seroconversion in only half of the study cohort of 16 women when IFA-lytic and IFA-LANA results were considered. The IFA-lytic assay was more sensitive than ELISA for early antibody responses. The IFA-LANA was positive in some women who had neither IFA-lytic nor ELISA antibodies suggesting it may be a marker for latent infections. Presumptive identification of incident HHV-8 infection by ELISA screening followed by IFA-lytic testing to confirm the positive test and IFA-LANA to rule out prior infection provides the most accurate documentation of HHV-8 seroconversion. © 2003 Elsevier B.V. All rights reserved. Keywords: Human herpesvirus 8; Human immunodeficiency virus; Kaposi’s sarcoma; Enzyme-linked immunosorbent assay; Immunofluorescence assay

∗ Corresponding author. Present address: Children’s Hospital & Regional Medical Center, Virology Office, G-815, 8G-3, 4800 Sand Point Way N.E., Seattle, WA 98105, USA. Tel.: +1-206-987-2117; fax: +1-206-987-3885. E-mail addresses: [email protected] (B.H. Chohan), [email protected] (H. Taylor), [email protected] (R. Obrigewitch), [email protected] (L. Lavreys), [email protected] (B.A. Richardson), [email protected] (K.N. Mandaliya), [email protected] (J.J. Bwayo), [email protected] (J.K. Kreiss), [email protected] (R.A. Morrow). 1 Present address: International AIDS Research and Training Program, Box 359909, 325 Ninth Ave., Seattle, WA 98104-2499, USA. Tel.: +1-206-731-2822; fax: +1-206-731-2427. 2 Present address: Children’s Hospital & Regional Medical Center, Clinical Virology Lab, G800A, 4800 Sand Point Way N.E., Seattle, WA 98105, USA. Tel.: +1-206-987-2088; fax: +1-206-987-2793. 3 Tel.: +254-11-474-055; fax +254-11-474-055. 4 Present address: 325 Ninth Ave., Box 359909, Seattle, WA 98104-2499, USA. Tel.: +1-206-731-2425; fax: +1-206-731-2427. 5 Tel.: +254-11-313-379/228-827; fax: +254-11-316-458. 6 Tel.: +254-2-724-194; fax: +254-2-712-007. 7 Present address: Box 359931, 325 Ninth Ave., Seattle, WA 98104-2499, USA. Tel.: +1-206-731-8496; fax: +1-206-731-6005.

1386-6532/$ – see front matter © 2003 Elsevier B.V. All rights reserved. doi:10.1016/j.jcv.2003.08.017

138

B.H. Chohan et al. / Journal of Clinical Virology 30 (2004) 137–144

1. Introduction The human oncogenic virus, Kaposi’s sarcoma associated herpesvirus (KSHV) or human herpesvirus 8 (HHV-8) was discovered in biopsies from Kaposi’s sarcoma (KS) lesions (Chang et al., 1994). The diseases linked to the presence of antibodies to HHV-8 include KS, primary effusion lymphoma and a subtype of multicentric Castleman’s disease (Boshoff et al., 1995; Moore and Chang, 1995). In Africa, KS was quite common even before the appearance of AIDS but, with the current AIDS epidemic, KS has become the most commonly diagnosed cancer in some countries of Africa (Mayama et al., 1998; Sitas et al., 1999). HIV-infected people in Africa are 100 times more likely to develop KS than are HIV-uninfected people (Sitas et al., 1999). Among HIV-infected persons, those who seroconvert to HHV-8 are at higher risk of developing KS than those whose HHV-8 infections predate their HIV-1 infection (Goudsmit et al., 2000). A number of serologic assays have been developed and described for the detection of HHV-8 antibodies. These include enzyme-linked immunosorbent assays (ELISA), based on whole viral lysate (Chatlynne et al., 1998), synthetic peptides (Davis et al., 1997; Pau et al., 1998) or recombinant HHV-8 peptides (Andre et al., 1997; Smith et al., 1997). Immunofluorescence assays (IFA) are also well established for antibodies to uninduced (latent) or structural (lytic) proteins in HHV-8 infected cell-lines (Gao et al., 1996; Lennette et al., 1996; Simpson et al., 1996). Immunoblot techniques using viral proteins from infected cells or semi-purified virus have also been reported (Gao et al., 1996; Miller et al., 1996). To date, there is no “gold standard” serology against which others can be compared. Many of the HHV-8 seroprevalence studies have been based on IFA, which tests for antibodies to either lytic or latent HHV-8 antigens. These tests appear to identify not only most HIV-infected individuals diagnosed with KS but also those at increased risk to develop KS (Gao et al., 1996; Kedes et al., 1997; Simpson et al., 1996). However, the procedure for IFA testing is time-consuming, requires expensive microscopy, and is vulnerable to subjective reading to differentiate positive from negative results. These characteristics limit the utility of IFA in large-scale studies to assess HHV-8 seroprevalence, especially in countries with a high incidence of KS (Topino et al., 2001). This study was designed to compare the performance of an ELISA method and IFA in sequential sera of African women at high risk of HHV-8 infection. In addition, we also studied the impact of HIV-1 on the detection of HHV-8 antibodies.

1993 to January 2000, were offered HIV-1 serologic testing. Women who were HIV-1 seronegative (N = 1225) were enrolled in a prospective cohort study and their serum samples were stored (Martin et al., 1998). Of the 1225 women, 1025 (83%) had at least one follow-up visit to the clinic with median follow-up time of 18.6 months (range 0.7–82.1 months) and 741 women had at least 3 months of follow-up. At follow-up visits, STD screening and HIV-1 serologic testing was performed and plasma samples were collected and stored. Of the 741 subjects with follow-up, 305 (41%) were positive for HHV-8 antibodies by ELISA in their enrollment sera. Samples from the subjects negative for HHV-8 antibodies at enrollment were tested for this seroconversion study. 2.2. HIV-1 testing The blood samples were tested for HIV-1 antibodies using a commercial enzyme immunoassay kit (Detect HIV-1/2, Biochem Immunosystems Inc., Montreal, Canada). Samples that appeared reactive were retested with another commercial ELISA kit (Recombigen, HIV-1/HIV-2 EIA, Cambridge Biotech, Worcester, MA). Only women who were HIV-1 seronegative at enrollment were included in this HHV-8 study. During follow-up, if a woman had positive results for HIV-1 antibodies by both ELISAs, HIV-1 seroconversion was confirmed by Western blot (Cambridge Biotech, Worcester, MA). 2.3. Samples for HHV-8 testing Serologic testing for HHV-8-specific antibodies was done on paired archived specimens of women with at least three follow-up visits after the first screening visit. Enrollment serum and the most recent available plasma from each woman were tested as pairs by HHV-8 ELISA. Women who tested negative for HHV-8 antibodies in the first sample and positive in the last sample were analyzed for this study. Stored plasma samples from all clinic visits of these women were then selected, including the first and the last sample, and shipped to Seattle on dry ice for further testing. 2.4. Control sera for HHV-8 ELISA and IFA HHV-8 positive control serum was obtained from an HIV-1 seronegative subject who had clinically and virologically documented KS as defined by the presence of HHV-8 DNA in the plasma by PCR (Koelle et al., 1997). The negative control serum was from a low risk individual who was HHV-8 PCR negative as well as IFA-latent-negative and IFA-lytic-negative.

2. Materials and methods

2.5. Immunofluorescence assay

2.1. Study Subjects

BCBL-1 cells were used for IFA to measure HHV-8 antibodies to both the replicative (lytic) and latent nuclear antigen (LANA). The BCBL-1 cell-line was derived from an HHV-8 positive but EBV-negative B cell lymphoma (AIDS

Female sex workers in Mombasa, Kenya, attending a municipal clinic for regular STD checkups from February

B.H. Chohan et al. / Journal of Clinical Virology 30 (2004) 137–144

Research and Reference Reagents Program, National Institutes of Health, Bethesda, MD). BJAB cells, an EBV-negative and HHV-8 negative Burkitt’s lymphoma B cell line, (Menezes et al., 1975) were used for detection of antinuclear or anti-lymphocyte antibodies that may give false positive fluorescence. Both BCBL-1 and BJAB cells were grown and maintained at 37 ◦ C under CO2 in RPMI 1640 medium with 25 mM Hepes buffer supplemented with heat inactivated 10% fetal bovine serum (Gibco), 1% l-glutamine, 1% penicillin-streptomycin (Gibco), and 1% sodium pyruvate. For the IFA-LANA test, BCBL-1 cells were centrifuged, washed twice with phosphate buffered saline (PBS) and resuspended in PBS at 7.5 × 105 cells/ml. A volume of 10 ␮l of resuspended cells was dripped onto each well of 10-well glass-slides. Slides were dried and then fixed for 10 min in cold acetone. After fixation, the slides were again allowed to air-dry and stored at −70 ◦ C. For the IFA-lytic assay, BCBL-1 cells at a density of 106 cells/ml were treated with 20 ng/ml of a tumorpromoting agent, Phorbol 12-Myristate 13-acetate (TPA, Sigma, St Louis, MO). Slides were then prepared from the TPA-induced BCBL-1 cells as described above for the IFA-LANA test. The control slides with BJAB cells were prepared and stored as for uninduced BCBL-1 slides. IFA-lytic and IFA-LANA assays were performed according to the procedure described elsewhere (Koelle et al., 1997; Renne et al., 1996) with some modifications. In brief, test sera and negative control sera were diluted to 1:20 and 1:40 in 1× PBS containing 4% goat serum. The positive control serum was used at a dilution standardized previously to give maximum optimal positive fluorescence. A volume of 10 ␮l of each dilution was added to each of the three slides (latent, lytic, and BJAB). The slides were incubated in a moist chamber at 37 ◦ C for 45 min. Slides were washed for 10 min in 1× PBS, immersed in distilled water, and aspirated between wells to remove excess moisture. Slides were incubated for 30 min at 37 ◦ C in a moist chamber with a 1:80 dilution in 1× PBS of fluorescein isothiocyanate-conjugated goat anti-human IgG Fc region (Cappel–ICN Pharmaceuticals Inc., Durham, NC) containing a counterstain of 1:20,000 dilution of Evan’s blue (Sigma, St.Louis, MO). After washing and aspirating the slides as described above, the cover-slip was mounted using 90% glycerol in PBS. Cells were observed using a Zeiss Axiolab immunofluorescence microscope. Sera were considered HHV-8 IFA positive for latent antibody (IFA-LANA) if, at a dilution of 1:40, an apple-green punctate nuclear fluorescence was observed in the uninduced BCBL-1 cell that was not seen at a sample dilution of 1:20 in the nuclei of the BJAB control cells. If any sample showed fluorescence in the nucleus of both the BJAB and uninduced BCBL-1 cells, the sample was reported as presence of antinuclear antibody (ANA). IFA-LANA tests were scored based on the intensity of the fluorescence and reported as 1 = positive, 0.5 = weak positive, and 0 = negative.

139

Sera were scored HHV-8 IFA positive for lytic antibody if bright apple-green cytoplasmic fluorescence was observed in the TPA induced BCBL-1 cells at a dilution of 1:40. IFA-lytic results considered negative if no staining was seen and indeterminate (+/−) if cells stained green but did not fluoresce. Positive scores ranged from 1 indicating a barely visible cytoplasmic fluorescence to 4 indicating a very bright cytoplasmic fluorescence. 2.6. HHV-8 whole viral lysate enzyme immunoassay Sucrose gradient purified HHV-8 whole viral lysate (ABI; Columbia, MD) at 4 ␮g/ml in 0.05 M carbonate/bicarbonate coating buffer pH 9.6 was added to 96 well plates (Immulon 2 Highbinder Flat Microtiter Plates, Nunc, Chantilly, VA) using 100 ␮l per well. After incubating overnight at 4 ◦ C, plates were washed three times with PBS-Tween 20 (0.05% Tween 20 in PBS) and blocked by using 125 ␮l per well of ABI’s “Immunoassay Stabilizer” for 30 min. The plates were dried thoroughly and stored at 4 ◦ C after aspirating the immunoassay stabilizer. Samples were diluted 1:100 in 4% goat serum in PBS and 100 ␮l were placed in each of two wells. HHV-8 antibody positive and negative controls were included in duplicate in each ELISA run as well as one blank well containing all the reagents except serum. Plates were incubated for 30 min at 37 ◦ C. The wells were washed twice with PBS 0.05% Tween 20 and once with PBS. One hundred microlitres of goat anti-human IgG-conjugated to horseradish peroxidase (Kirkegaard & Perry Laboratories, Gaithersburg, MD) was added at a dilution of 1:2000 in the sample diluent (4% goat serum in PBS) and incubated for 30 min at 37 ◦ C. The plates were washed again three times as described above and 100 ␮l of freshly prepared substrate (equal parts of Solution A (TMB) and Solution B (H2 O2 ) (both from Boehinger-Mannheim, Indianapolis, IN), were added to each well and incubated for eight minutes at room temperature. The reaction was stopped using 100 ␮l 1 M phosphoric acid solution. The optical densities of the wells were read at 450 nm using an automated EIA plate reader. All samples were run in duplicate and the OD readings averaged. If there was a variation in the two OD of the same sample by more than 20%, the sample was retested. Acceptable ELISA runs required a mean negative control OD reading <0.2 and a mean positive control value >0.8. Samples giving mean OD readings of ≥0.35 were considered positive and measurements below 0.2 were considered negative; mean OD readings of 0.2–0.35 were scored as equivocal.

3. Results 3.1. Subjects with possible HHV8 seroconversion Of the 436 women who were negative for HHV-8 antibodies by ELISA at entry, subsequent plasma samples

140

B.H. Chohan et al. / Journal of Clinical Virology 30 (2004) 137–144

Table 1 HHV-8 IFA-LANA and IFA-lytic Results in 16 Women with Evidence of HHV-8 Seroconversion by ELISA Subject

ELISA OD

IFA-lytic score

IFA-LANA score

Sample

First

First

Last

First

Last

Seroconversion to HHV-8 Q138 0.01 1.2 Q205 0.08 0.85 Q264 0.09 0.60 Q219 0.04 1.15 Q274 0.10 0.91 Q323 0.17 1.28 Q375 0.19 1.11 Q498 0.19 0.72

0 0 0 0 4 0 0 0

2 3 2 4 4 4 4 4

0 0 0 0 0 0 0 0

1 0 1 1 1 0 1 ANAa

Past infection Q280b Q336b Q325 Q212b Q332

0 0 0 ± 2

1 1 3 1 3

1 1 1 0.5 1

1 1 1 0.5 1

ELISA antibody development only 0.05 1.08 0 Q165b 0.07 0.63 0 Q349b Q233 0.12 1.08 0

0 0 0

1 0 0.5

0 0 0

Last

with HHV-8 0.10 0.60 0.12 0.47 1.04 1.04 0.18 0.60 0.15 0.92

ANA = anti-nuclear antibody. HIV-1 seroconversion occurred coincident with HHV-8 ELISA OD increase. a

b

were available for 354 (81%). Sixteen (4.5%) of the 354 follow-up plasma samples were positive for HHV-8 antibodies by ELISA. The median follow-up period for these 16 women was 23.5 months (range 3–79) and a total of 124 post-enrollment plasma samples were available for these 16 women with possible HHV-8 seroconversion. These post-enrollment as well as the enrollment samples were thawed, divided, coded in sets, and tested for HHV-8 antibodies using ELISA, IFA-lytic, and IFA-LANA tests. Eight women had evidence of new HHV-8 infections by seroconversion in at least two of three tests and, in addition, lack of LANA antibodies in their enrollment sera (Table 1). The serologic test results of one of these eight women are shown in Fig. 1. Reactivity in all three assays increased over time in this subject. Of interest, the IFA-lytic test appeared to be more sensitive for detecting incident HHV-8 infection. Seroconversion by IFA-lytic occurred 1–2 months before increases in OD by ELISA in three of the eight women identified as HHV-8 seroconvertors. The IFA-LANA did not become positive in two of the eight women who appeared to have incident HHV-8 infection (Q205, Q323; Table 1) despite follow-up times of 22 months and 41 months, respectively. An additional five women had apparent seroconversion to HHV-8 by ELISA and IFA-lytic tests. However, all five had LANA antibodies in the enrollment sample and may have been HHV-8 infected at entry to the study. The remaining three women (Q165, Q233, Q349) had no samples that were definitely positive for HHV-8 antibodies

Fig. 1. Changes in HHV-8 test results over time in an initially seronegative subject (Q264). The top figure shows ELISA optical densities and IFA-lytic scores, while the lower figure shows ELISA optical densities and IFA-latent scores from the same subject.

by IFA-lytic or IFA-LANA despite having a clear increase in the ELISA OD values late in follow-up. Two of these three women (Q165, Q349) as well as three of the five women (Q280, Q336 and Q212) with presumed latent past HHV-8 infection seroconverted to HIV-1 coincident with the increase in HHV-8 ELISA OD values (Fig. 2). Unfortunately, we did not have mucosal or peripheral blood mononuclear cell (PBMC) samples to test for HHV-8 DNA by PCR for confirmation of infection in these women. Thus, we cannot rule out the possibility that HIV-1 infection may result in non-specific increases in HHV-8-ELISA reactivity. 3.2. Performance of serum versus plasma in ELISA Since the enrollment sample was serum and the subsequent serial samples were plasma, we wished to determine the relative performance of the ELISA with both sample types. Thirty-five individuals had both plasma and sera drawn at a single visit. These pairs were then tested together by ELISA. The mean OD for the plasma samples was 0.568 (95% CI: 0.401, 0.734), while the mean OD for paired serum samples was 0.683 (95% CI: 0.502, 0.864). The mean difference between the paired samples (plasma OD minus serum OD) was −0.116 (95% CI: −0.174, −0.057; P < 0.001 by paired t-test). Plasma and serum OD levels were correlated (r = 0.95; Pearson’s correlation coefficient). These results

B.H. Chohan et al. / Journal of Clinical Virology 30 (2004) 137–144

141

indicate that while the OD levels of serum and plasma samples were linearly related, the plasma OD levels were significantly lower than serum OD values (Fig. 3). 3.3. Duplicate versus single well determinations One of the reasons for our study on HHV-8 ELISA was to search for a cost-effective alternative to IFA for detection of HHV-8 antibodies in large field research studies. Hence, to be economical, we determined the correlation between performing an ELISA for detection of HHV-8 antibodies in single wells versus testing in duplicate wells. HHV-8 ELISA was performed in duplicate on 1820 plasma samples. The overall correlation between the OD levels in the two wells was 0.899 (Pearson’s correlation coefficient). Of 1820 samples, 1192 were negative (OD < 0.2) in the first well and of these 1185 (99.4%) were also negative in the second well (95% CI: 99.0, 99.8). An additional 520 samples were positive in the first well (OD > 0.35), and 515 (99.0%) of these were positive also in the second well (95% CI: 98.2–99.9). Finally, of the 1820 samples, 108 were equivocal in the first well and of these, 100 (92.6%) were equivocal in the second testing (95% CI: 87.7–97.5). Thus, a single well could be used for future studies to save costs. Fig. 2. HHV-8 antibody test results at the time of HIV-1 seroconversion in woman Q349 (the arrow indicates the time of HIV-1 seroconversion). Note the dramatic increase in ELISA OD readings for HHV-8 antibodies in conjunction with HIV-1 seroconversion. Neither IFA became positive.

4. Discussion In this study of female commercial sex workers in Kenya, we demonstrated an HHV-8 seroprevalence of 41% using a whole-antigen-based ELISA. A similar prevalence of

Fig. 3. The correlation between the HHV-8 ELISA OD values of the 35 paired serum/plasma samples.

142

B.H. Chohan et al. / Journal of Clinical Virology 30 (2004) 137–144

43% using the whole-antigen based ELISA was reported in a recent study conducted among heterosexual men in Kenya (Baeten et al., 2002). High HHV-8 seroprevalence has also been detected among populations in Uganda, Zimbabwe, Cameroon, and South Africa (Gessain et al., 1999; Lampinen et al., 2000; Mayama et al., 1998; Sitas et al., 1999). The HHV-8 seroprevalence among prostitutes in Cameroon has been reported as 51.9% (Bestetti et al., 1998). Archived samples from a long-term prospective cohort study of female commercial sex workers in Kenya gave us the opportunity to test HHV-8 serostatus over time. We found evidence of eight incident cases of HHV-8 among the 354 evaluated subjects. Hence, the estimated incidence rate for HHV-8 amongst this cohort was 1.2 per 100 person years (8/669 person years). Currently, little is known about the incidence of HHV-8 or seroconversion rates for HHV-8 in women in KS endemic areas. The few published studies reporting HHV-8 seroconversion have been performed in groups of intravenous drug users and men who have sex with men (Goudsmit et al., 2000). When we screened paired samples from the women for HHV-8 antibodies by a whole virus ELISA, we categorized 16 women as possibly having seroconverted to HHV-8, based on a rise in OD readings from negative to positive values over time. Seroconversion to HHV-8 in eight of these sixteen women was considered to be highly likely based on the ELISA results taken together with those of an IFA for antibodies to lytic antigens (IFA-lytic), and an IFA for antibodies to latent HHV8 antigens (IFA-LANA). IFA-lytic detected antibodies slightly earlier than did the whole virus ELISA. The IFA-LANA remained negative for two of the eight HHV-8 women with putative seroconversion. This could be due to relative lack of sensitivity in the IFA-LANA test (Andreoni et al., 2002). Alternatively, there could have been no expression or a very poor antibody response to HHV-8 latency associated antigens in the women. The other possibility is that the HHV-8 LANA antibodies, like those against Epstein Barr virus nuclear antibodies (EBNA), develop months after primary infection (Rea et al., 2002). However, samples from these two women were obtained at intervals of 22 months and 41 months, respectively, making the explanation of late rise in antibody less plausible. Five women had evidence of previously acquired HHV-8 in that their enrollment sera were positive by IFA-LANA. Seroconversion by ELISA in these women was accompanied by seroconversion or increased in reactivity by the IFA-lytic test. These five women may provide evidence of truly latent HHV-8 infection in which replicating virus and structural antigens are not being produced to stimulate the immune system. Further virological studies to characterize individuals with this serologic pattern are needed.

HIV-1 seroconversion was associated with de novo detection of HHV-8 antibodies by ELISA in five women. Three of these five women had presumed past HHV-8 infection as was evident by detection of HHV-8 LANA antibodies. One explanation for this finding is that latent HHV-8 virus is reactivated during the primary HIV-1 infection period. Alternatively, these women may have simultaneously acquired infection with the two viruses. As a practical application, until further data are accrued on HHV-8 natural history following HIV-1 infection, elevations in ELISA titers for HHV-8 antibodies at the time of HIV-1 seroconversion should be interpreted with caution. Studies in which samples are taken for HHV-8 PCR testing may help to explain these observations (Pauk et al., 2000). Currently there is no accepted ‘gold standard’ serological assay for the detection of the HHV-8 antibodies and studies have shown that a combination of assays is an effective strategy (Engles et al., 2000; Spira et al., 2000). Variability in detection of HHV-8 antibodies by different methods has been shown in many studies (Casper et al., 2002; Chatlynne et al., 1998; Engles et al., 2000; Spira et al., 2000; Topino et al., 2001). Casper, et al. have suggested a testing algorithm for HHV-8 antibodies that uses the whole virus ELISA in combination with IFA to yield a better sensitivity and specificity (Casper et al., 2002). Our data also suggest that using more than one HHV-8 specific antibody assay may be most accurate for determination of incident HHV-8 infection. In particular, the IFA-lytic test may improve sensitivity for seroconversion while the IFA-LANA may improve sensitivity for past latent infection. We found that plasma samples give slightly lower OD values compared to serum samples drawn from the same patient at the same time. Because of this greater sensitivity and since this test has been optimized using sera (Casper et al., 2002), serum is the preferred specimen and plasma and sera should not be considered interchangeable specimens for ELISA. In our study, initial enrollment samples were sera, while subsequent samples were plasma. Using the test cut-off values established with sera, it is unlikely that we mistakenly identified seroconverters (test results changing from negative to positive over time) due to use of plasma at later time points. However, seroconversion events might have been missed. Although a combination of ELISA and IFA tests are likely to be the most sensitive approach to detect HHV-8 seroconversion, the whole virus ELISA is a reasonable option for HHV-8 seroprevalence studies, especially in developing countries that lack the resources for expensive fluorescent microscopy. The ELISA is much less subjective in reporting than an IFA based test and requires less time and skill to perform and interpret (Casper et al., 2002). We found that samples could be reliably tested in single wells rather than in duplicate, thereby reducing the cost of testing even further. Thus, ELISA is a cost-effective method for seroprevalence or large-scale screening studies in developing countries and

B.H. Chohan et al. / Journal of Clinical Virology 30 (2004) 137–144

may prove useful in prospective studies to define clinical correlates of HHV-8 seroconversion.

Acknowledgements We are grateful to Dr. Hal Martin and Dr. Denis Jackson, who originally established this cohort in Mombasa, Kenya as well as the nurses and laboratory staff at Ganjoni Municipal Clinic and HIV-1 Research Laboratory in Coast General Hospital, respectively, for the assistance in collection and processing of the samples. Our greatest appreciation goes to the women who participated in this research study. Our sincere thanks to Dr. Lawrence Corey for his critical review of the manuscript. Ethical Approval: The study received ethical approval from the University of Washington’s Human Subjects Review Committee and the Ethical Review Committee of Kenyatta National Hospital in Nairobi. All subjects gave informed consent to participate in the study. This study was supported in part by the US National Institutes of Health through grants CA-86795 and D43-TW00007.

References Andre S, Schatz O, Bogner JR, Zeichhardt H, Stoffler-Meilicke M, Jahn HU, Ullrich R, Sonntag AK, Kehm R, Haas J. Detection of antibodies against viral capsid proteins of human herpesvirus 8 in AIDS-associated Kaposi’s sarcoma. J Mol Med 1997;75:145–52. Andreoni M, Sarmati L, Nicastri E, El Sawaf G, El Zalabani M, Uccella I, Bugarini R, Parisi SG, Rezza G. Primary human herpesvirus 8 infection in immunocompetent children. JAMA 2002;287:1295–300. Baeten JM, Chohan BH, Lavreys L, Rakwar JP, Ashley R, Richardson BA, Mandaliya KM, Bwayo JJ, Kreiss JK. Correlates of human herpesvirus 8 seropositivity among heterosexual men in Kenya. AIDS 2002;16:2073–8. Bestetti G, Renon G, Mauclere P, Ruffie A, Mbopi-Keou FX, Eme D, Parravicini C, Corbellino M, de The G, Gessain A. High seroprevalence of human herpesvirus-8 in pregnant women and prostitutes from Cameroon. AIDS 1998;12:541–3. Boshoff C, Whitby D, Hatziioannou T, Fisher C, van der Walt J, Hatzakis A, Weiss R, Schulz T. Kaposi’s-sarcoma-associated herpesvirus in HIV-negative Kaposi’s sarcoma. Lancet 1995;345:1043–4. Casper C, Krantz E, Taylor H, Dalessio J, Carrell D, Wald A, Corey L, Ashley R. Assessment of a combined testing strategy for detection of antibodies to human herpesvirus 8 (HHV-8) in persons with Kaposi’s sarcoma, persons with asymptomatic HHV-8 infection, and persons at low risk for HHV-8 infection. J Clin Microbiol 2002;40:3822–5. Chang Y, Cesarman E, Pessin MS, Lee F, Culpepper J, Knowles DM, Moore PS. Identification of herpesvirus-like DNA sequences in AIDS-associated Kaposi’s sarcoma. Science 1994;266:1865–9. Chatlynne LG, Lapps W, Handy M, Huang YQ, Masood R, Hamilton AS, Said JW, Koeffler HP, Kaplan MH, Friedman-Kien A, Gill PS, Whitman JE, Ablashi DV. Detection and titration of human herpesvirus-8-specific antibodies in sera from blood donors, acquired immunodeficiency syndrome patients, and Kaposi’s sarcoma patients using a whole virus enzyme-linked immunosorbent assay. Blood 1998;92:53–8. Davis DA, Humphrey RW, Newcomb FM, O’Brien TR, Goedert JJ, Straus SE, Yarchoan R. Detection of serum antibodies to a Kaposi’s sarcoma-associated herpesvirus-specific peptide. J Infect Dis 1997;175:1071–9.

143

Engles EA, Whitby D, Goebel PB, Stossel A, Waters D, Pintus A, Contu L, Biggar RJ, Goedert JJ. Identifying human herpesvirus 8 infection: performance characteristics of serologic assays. J Acquir Immune Defic Syndr 2000;23:346–54. Gao SJ, Kingsley L, Hoover DR, Spira TJ, Rinaldo CR, Saah A, Phair J, Detels R, Parry P, Chang Y, Moore PS. Seroconversion to antibodies against Kaposi’s sarcoma-associated herpesvirus-related latent nuclear antigens before the development of Kaposi’s sarcoma. N Engl J Med 1996;335:233–41. Gessain A, Mauclere P, van Beveren M, Plancoulaine S, Ayouba A, Essame-Oyono JL, Martin PM, DE THE G. Human herpesvirus 8 primary infection occurs during childhood in Cameroon, Central Africa. Int J Cancer 1999;81:189–92. Goudsmit J, Renwick N, Dukers NH, Coutinho RA, Heisterkamp S, Bakker M, Schulz TF, Cornelissen M, Weverling GJ. Human herpesvirus 8 infections in the Amsterdam Cohort Studies (1984–1997): analysis of seroconversions to ORF65 and ORF73. Proc Natl Acad Sci 2000;97:4838–43. Kedes DH, Ganem D, Ameli N, Bacchetti P, Greenblatt R. The prevalence of serum antibody to human herpesvirus 8 (Kaposi sarcoma-associated herpesvirus) among HIV-seropositive and high-risk HIV-seronegative women. JAMA 1997;277:478–81. Koelle DM, Huang ML, Chandran B, Vieira J, Piepkorn M, Corey L. Frequent detection of Kaposi’s sarcoma-associated herpesvirus (human herpesvirus 8) DNA in saliva of human immunodeficiency virus-infected men: clinical and immunologic correlates. J Infect Dis 1997;176:94–102. Lampinen TM, Kulasingam S, Min J, Borok M, Gwanzura L, Lamb J, Mahomed K, et al. Detection of Kaposi’s sarcoma-associated herpesvirus in oral and genital secretions of Zimbabwean women. J Infect Dis 2000;181:1785–90. Lennette ET, Blackbourn DJ, Levy JA. Antibodies to human herpesvirus type 8 in the general population and in Kaposi’s sarcoma patients. Lancet 1996;348:858–61. Martin HL, Nyange PM, Richardson BA, Lavreys L, Mandaliya K, Jackson DJ, Ndinya-Achola JO, Kreiss J. Hormonal contraception, sexually transmitted diseases and risk of heterosexual transmission of HIV-1. J Infect Dis 1998;178:1053–9. Mayama S, Cuevas LE, Sheldon J, Omar OH, Smith DH, Okong P, Silvel B, Hart CA, Schulz TF. Prevalence and transmission of Kaposi’s sarcoma-associated herpesvirus (human herpesvirus-8) in Ugandan children and adolescents. Int J Cancer 1998;11:817–20. Menezes J, Leibold W, Klein G, Clements G. Establishment and characterization of an Epstein-Barr virus (EBC)-negative lymphoblastoid B cell line (BJA-B) from an exceptional, EBV-genome-negative African Burkitt’s lymphoma. Biomedicine 1975;22:276–84. Miller G, Rigsby MO, Heston L, Grogan E, Sun R, Metroka C, Levy JA, Gao SJ, Chang Y, Moore P. Antibodies to butyrate-inducible antigens of Kaposi’s sarcoma-associated herpesvirus in patients with HIV-1 infection. N Engl J Med 1996;334:1292–7. Moore PS, Chang Y. Detection of herpesvirus-like DNA sequences in Kaposi’s sarcoma in patients with and without HIV infection. N Engl J Med 1995;332:1181–5. Pau CP, Lam LL, Spira TJ, Black JB, Stewart JA, Pellett PE, Respess RA. Mapping and serodiagnostic application of a dominant epitope within the human herpesvirus 8 ORF 65-encoded protein. J Clin Microbiol 1998;36:1574–7. Pauk J, Huang ML, Brodie SJ, Wald A, Koelle DM, Schacker T, Celum C, Selke S, Corey L. Mucosal shedding of human herpesvirus 8 in men. N Engl J Med 2000;343:1369–77. Rea TD, Ashley RL, Russo JE, Buchwald DS. A systematic study of Epstein-Barr virus serologic assays following acute infection. Am J Clin Pathol 2002;117:56–61. Renne R, Zhong W, Herndier B, McGrath M, Abbey N, Kedes D, Ganem D. Lytic growth of Kaposi’s sarcoma-associated herpesvirus (human herpesvirus 8) in culture. Nat Med 1996;2:342–6.

144

B.H. Chohan et al. / Journal of Clinical Virology 30 (2004) 137–144

Simpson GR, Schulz TF, Whitby D, Cook PM, Boshoff C, Rainbow L, Howard MR, Gao SJ, et al. Prevalence of Kaposi’s sarcoma associated herpesvirus infection measured by antibodies to recombinant capsid protein and latent immunofluorescence antigen. Lancet 1996;348:1133–8. Sitas F, Carrara H, Beral V, Newton R, Reeves G, Bull D, Jentsch U, Pacella-Norman R, Bourboulia D, Whitby D, Boshoff C, Weiss

R. Antibodies against human herpesvirus 8 in black South African patients with cancer. N Engl J Med 1999;340:1863–71. Topino S, Vincenzi L, Mezzaroma I, Nicastri E, Andreoni M, Sirianni MC. Correlation between enzyme-linked immunosorbent assay and immunofluorescence assay with lytic antigens for detection of antibodies to human herpesvirus 8a. Clin Diagn Lab Immunol 2001;8: 203–5.