点击显示 收起
Division of Infectious Diseases, Department of Medicine, University of Colorado Health Sciences Center, Denver
University of Minnesota Medical School, Minneapolis
Inguinal lymph nodes from 24 human immunodeficiency virus (HIV) type 1infected subjects without Kaposi sarcomaassociated herpesvirus (KSHV)associated diseases were examined for KSHV infection. KSHV-infected cells were detected at a very low frequency in the lymph nodes of 7 subjects (median frequency, 2 infected cells/107 lymph node cells). Latent, but not lytic, KSHV gene expression was detected and KSHV-infected cells were located in B cellrich areas of lymph node follicles. These findings provide evidence that, in the absence of KSHV-associated diseases, latent infection of lymph node cells provides a mechanism for the persistence of KSHV in KSHV/HIV-1coinfected persons.
Coinfection with Kaposi sarcomaassociated herpesvirus (KSHV; also called "human herpesvirus 8") and HIV-1 is associated with increased risk of Kaposi sarcoma (KS), primary effusion lymphoma, and some forms of Castleman disease [13]. When KSHV-associated diseases are present, KSHV-infected cells are abundant in diseased tissue. However, in the absence of these diseases, KSHV-infected cells are uncommon in infected persons. Although B cells are host cells for KSHV infection [4], <20% of KSHV/HIV-1coinfected persons without KSHV-associated diseases have detectable KSHV DNA in peripheral-blood mononuclear cells (PBMCs) [5].
Little is known about the characteristics of the persistence of KSHV in KSHV/HIV-1coinfected persons. Because B cells are host cells for KSHV, we hypothesized that lymphoid tissue is an important reservoir in which KSHV-infected cells persist. Furthermore, because persons without KSHV-associated diseases usually do not display evidence of active KSHV replication, we hypothesized that KSHV gene expression in lymph node cells is restricted. To test these hypotheses, the lymph nodes of HIV-1infected persons were examined for evidence of KSHV infection, and KSHV gene expression was characterized in those lymph nodes that contained KSHV-infected cells.
Subjects, materials, and methods.
Inguinal lymph nodes, PBMCs, and plasma samples that had been obtained from HIV-1infected persons as part of a study of immune responses in HIV-1infected lymph node cells [6] were analyzed. Antibodies to KSHV latencyassociated nuclear antigen (LANA) were detected by use of a latent BCP-1 cell immunofluorescent assay, as described elsewhere [7]. The present study adhered to the US Department of Health and Human Services guidelines for human experimentation and was approved by the Colorado Multiple Institutional Review Board.
For detection and quantification of KSHV DNA, DNA was extracted from frozen lymph nodes, PBMCs, and plasma samples by use of QIAamp reagents (Qiagen). All DNA samples gave a strong signal when human -globin DNA was amplified [5]. For qualitative detection of KSHV in lymph nodes, KSHV DNA was amplified by nested polymerase chain reaction (PCR) with primers to open-reading frame (ORF) 25 (major capsid antigen) and ORF26 (minor capsid antigen) [8]. All PCR assays included, as a negative control, DNA from cells that were not infected with KSHV and, as a positive control, DNA that contained 10 copies of ORF26 DNA [5]. KSHV DNA in lymph nodes, PBMCs, and plasma samples was quantified by real-time PCR amplification of ORF26 [9]. Cellular DNA was quantified by real-time PCR amplification of human -actin DNA, with Taqman Actin reagents (PE Applied Biosystems).
All assays for the detection and quantification of KSHV infection by in situ hybridization and immunochemistry were performed and interpreted by persons blinded to sample identity, subject demographics, and the results of other assays. KSHV RNA was detected in sections of paraffin-embedded lymph node tissue by hybridization with [35S]-labeled antisense riboprobes for the latent T0.7 transcript (kaposin), lytic polyadenylated nuclear RNA (T1.1), and the lytic transcript that encodes viral interleukin 6 homologue (vIL-6) [10, 11]. Lymph nodes were hybridized with a sense riboprobe for T0.7 RNA as a negative control. A KS tumor and a lymph node from a patient with multicentric Castleman disease were used as positive controls.
Immunohistochemical staining for LANA-1 (ORF73) and viral envelope protein (K8.1) was performed with monoclonal antibodies to each viral protein (Advanced Biotechnologies). Paraffin-embedded sections were deparaffinized and rehydrated, and high-temperature antigen retrieval was performed in 10 mmol/L sodium citrate buffer (pH 6.0) for 20 min. Frozen sections were fixed for 20 min in 2% formaldehyde. Primary antibody was diluted 1 : 1500 and was incubated with sections for 1 h at room temperature. The Vectastain Elite ABC Kit (Vector Laboratories) was used in accordance with the manufacturer's instructions. Chromagen development was performed by use of 3,3-diaminobenzidine; Mayer's hematoxylin was used for counterstaining. Positive controls for LANA-1 and K8.1 were sections of a paraffin-embedded cutaneous KS tumor and a lymph node from a patient with multicentric Castleman disease, as well as paraffin-embedded BCBL-1 cells treated with phorbol myristic acid. Rat IgG (Zymed Laboratories) and mouse ascites (Sigma) were used as isotype controls for LANA-1 and K8.1 immunohistochemistry, respectively. Immunofluorescent staining for LANA-1 was detected by use of AlexaFluor488-labeled anti-rat antibody (Molecular Probes) and double stained with CD20 (DAKO), which was detected by use of AlexaFluor594-labeled anti-mouse antibody (Molecular Probes).
Cells that were positive by immunohistochemistry or in situ hybridization were counted in a complete section, and the total area of the section was measured by use Qwin Image Analysis software (version 2.2; Leica Image Systems). The frequency of KSHV-infected cells was estimated as the mean of the number of cells positive by immunohistochemistry or in situ hybridization in 5 separate 100-m2 areas.
All statistical analyses were performed by use of Statview software (version 4.53; Abacus Concepts); a 2-sided significance level of .05 was assumed. Between-group characteristics were compared by use of the Mann-Whitney U test for continuous variables and Fisher's exact test for categorical variables. Relationships of continuous variables were tested by use of Spearman's rank correlation.
Results and discussion.
Twenty-four HIV-1infected subjects were studied: 15 whites, 4 blacks, and 5 Hispanics. Twenty-two were men and 2 were women. The median age was 37 years (range, 2450 years). The median CD4+ lymphocyte count was 303 cells/L (range, 108689 cells/L). The median plasma HIV-1 RNA load was 29,367 copies/mL (range, 453464,000 copies/mL). Six subjects were receiving antiretroviral therapy at the time their lymph nodes were obtained; in all instances, therapy consisted of 12 nucleoside analog reverse-transcriptase inhibitors. Four other subjects had received antiretroviral therapy in the past, but not within the preceding 6 weeks. The duration of HIV-1 infection ranged from <6 months (4 subjects) to >10 years.
KSHV DNA was detected in lymph nodes from 7 of the subjects by qualitative PCR amplification of either ORF25 or ORF26 (table 1). KSHV DNA was detected in 4 lymph nodes by PCR amplification of both ORF25 and ORF26, in 2 lymph nodes by PCR amplification of ORF25 only, and in 1 lymph node by PCR amplification of ORF25 only. In a univariate analysis, detection of KSHV DNA in lymph nodes was not associated with age, sex, race, risk factors for HIV-1, prior or current receipt of antiretroviral therapy, CD4+ lymphocyte count, or plasma HIV-1 RNA load (P .3, for each comparison). Quantitative PCR confirmed the presence of KSHV DNA in lymph nodes (median, 20 copies/105 cells; range, 250 copies/105 cells) from all 7 subjects who had a positive qualitative PCR result. Detection of KSHV DNA in lymph nodes was not correlated with either plasma HIV-1 RNA load or CD4+ lymphocyte count (P > .1, for both comparisons). KSHV DNA was not detected in PBMCs or plasma samples from the 7 subjects with KSHV DNA in their lymph nodes, and none of these subjects subsequently developed KS (median follow-up duration, 40 months; range of follow-up duration, 072 months).
Antibodies to LANA were detected in 9 (38%) of the 24 subjects, a finding that is similar to the estimated 46% seroprevalence of KSHV in HIV-1infected populations in Denver, CO [5]. The concordance between the results of LANA antibody assays and PCR assays for KSHV DNA was 75% (P = .06, Fisher's exact test). Five of the 7 subjects with KSHV DNA in their lymph nodes had detectable antibody to LANA and, thus, also had serologic evidence of KSHV infection. The inability to detect antibodies to LANA in 2 subjects with KSHV DNA in their lymph nodes was likely due to false negative antibody tests, because the presence of KSHV in the lymph nodes from these subjects was subsequently confirmed by non-PCR methods (in situ hybridization for KSHV RNA and immunohistochemistry for KSHV proteins; see below). The discordance between the results of LANA antibody assays and PCR assays for KSHV DNA in the present study is consistent with the observed 20%30% frequency of negative LANA antibody assays for persons with both AIDS and KS [7].
The expression of LANA-1 was detected by immunohistochemistry in all 7 of the lymph nodes in which KSHV DNA was detected by qualitative PCR; however, it was not detected in the lymph nodes from 3 KSHV/HIV-1uninfected control subjects. In contrast to the low frequency of KSHV-infected cells in the lymph nodes from the 7 coinfected subjects (table 1), the frequency of KSHV-infected cells in the lymph node from the patient with multicentric Castleman disease (the positive control) was 105 infected cells/107 lymph node cells. The intracellular pattern of LANA-1 staining in the lymph nodes from the 7 coinfected subjects was similar to that in the lymph node from the patient with multicentric Castleman disease (figure 1A and 1B). Latent KSHV gene expression in lymph node cells was confirmed in 6 of the 6 lymph nodes that were examined by in situ hybridization with an antisense riboprobe for T0.7 RNA (figure 1C). Subject 22 was not evaluated by in situ hybridization, because adequate lymph node sample was not available. No foci of silver grains were detected when sections from each lymph node were hybridized with a sense riboprobe for T0.7 RNA as a negative control. Likewise, expression of T0.7 RNA was not detected when sections from 2 lymph nodes that were negative for KSHV DNA by PCR were hybridized with antisense riboprobes. There was a correlation between the frequency of KSHV-infected lymph node cells and plasma HIV-1 RNA load (r = 0.8; P = .06) but not between the frequency of KSHV-infected lymph node cells and CD4+ lymphocyte count (r = -0.6; P = .2).
Previous studies have demonstrated the localization of KSHV-infected cells in B cellrich areas of the mantle zone of lymph node follicles during multicentric Castleman disease [11, 12] and AIDS-related lymphadenopathy [13]. The localization of KSHV-infected cells in lymph nodes was investigated by dual immunohistochemistry for LANA-1 and CD20 expression in lymph node sections from subjects 10, 12, 22, and 23 (figure 1D and 1E). Eighty-two percent of LANA-1+ cells colocalized with CD20+ cells in lymph node follicles, and 4% were in extrafollicular areas; the CD20 stain was indeterminate for 14% of LANA-1+ cells.
Although the expression of vIL-6 in KS tumors is uncommon, vIL-6 is consistently expressed in KSHV-infected lymph node cells during primary effusion lymphoma and multicentric Castleman disease, and it has been hypothesized that vIL-6 plays a role in the pathogenesis of these KSHV-associated diseases via its effects on B cell proliferation and differentiation [11, 14, 15]. Therefore, we sought to determine whether vIL-6 is expressed in KSHV-infected lymph node cells in the absence of KSHV-associated diseases. As expected, KSHV-infected cells in the lymph node from the patient with multicentric Castleman disease expressed high levels of vIL-6 RNA. Although (1) the sensitivity of the T0.7 and vIL-6 riboprobes for the detection of KSHV-infected cells in lymph nodes from patients with multicentric Castleman disease is similar and (2) both probes can detect 1030 copies of viral RNA per cell [11], vIL-6 was not detected by in situ hybridization of lymph node sections from 6 KSHV/HIV-1coinfected subjects with detectable T0.7 RNA and LANA-1 expression. Thus, in the absence of KSHV-associated diseases, there were both fewer infected cells in lymph nodes and qualitative differences in the patterns of viral gene expression in infected lymph node cells.
The lytic T1.1 RNA was detected in the KS tumor positive control by in situ hybridization but not in any of the lymph nodes from the 7 coinfected subjects; likewise, K8.1 was not detected by immunohistochemistry in the lymph nodes from the 7 coinfected subjects. The apparent lack of lytic KSHV gene expression in infected lymph nodes is consistent with the absence of KSHV DNA in peripheral blood from any of the 7 subjects with infected lymph node cells. It is important to note that all 7 of these subjects also had CD4+ lymphocyte counts that were higher than the AIDS-defining range. Indeed, 1 subject had been infected with HIV-1 4 months before lymph nodes were obtained. Thus, the low frequency of KSHV-infected lymph node cells and the lack of detectable lytic KSHV gene expression may have been a consequence of relatively intact antiviral immune responses in the subjects we studied.
The present study provides the first systematic analysis of KSHV lymph node infection in HIV-1infected persons without KSHV-associated diseases. A previous study of KSHV lymph node infection in the setting of AIDS-related lymphadenopathy concluded that the occurrence of KSHV-infected cells in lymph nodes represents the seeding of lymph nodes by KSHV-infected peripheral-blood B cells, although peripheral-blood KSHV DNA loads and patterns of viral gene expression in infected lymph nodes were not reported [13]. Our finding of a low frequency of latent KSHV infection of lymph node cells from subjects without detectable KSHV DNA in PBMCsand in some cases without detectable titers of antibody to KSHVsuggests a different conclusion. Our finding suggests that KSHV-infected B cells in lymph node follicles are a reservoir of persistent viral infection in HIV-1infected persons without KSHV-associated diseases.
Acknowledgments
Real-time polymerase chain reaction (PCR) assays were performed by Uma Pugazenthi, at the Quantitative PCR Core Laboratory of the University of Colorado Cancer Center. We are also grateful to Carlyne Cool, for assistance with immunohistochemistry; M. Graham Ray and David Cohn, for assistance with the follow-up of subjects; and Leica Image Systems, for the loan of a microscope and quantitative image-analysis equipment.
References
1. Chang Y, Cesarman E, Pessin MS, et al. Identification of herpesvirus-like DNA sequences in AIDS-associated Kaposi's sarcoma. Science 1994; 266:18659. First citation in article
2. Cesarman E, Chang Y, Moore PS, Said JW, Knowles DM. Kaposi's sarcomaassociated herpesvirus-like DNA sequences in AIDS-related body-cavitybased lymphomas. N Engl J Med 1995; 332:118691. First citation in article
3. Soulier J, Grollet L, Oksenhendler E, et al. Kaposi's sarcomaassociated herpesvirus-like DNA sequences in multicentric Castleman's disease. Blood 1995; 86:127680. First citation in article
4. Mesri EA, Cesarman E, Arvanitakis L, et al. Human herpesvirus8/Kaposi's sarcomaassociated herpesvirus is a new transmissible virus that infects B cells. J Exp Med 1996; 183:238590. First citation in article
5. Campbell TB, Fitzpatrick L, MaWhinney S, Zhang XQ, Schooley RT. Human herpesvirus 8 (Kaposi's sarcomaassociated herpesvirus) infection in men receiving treatment for HIV-1 infection. J Acquir Immune Defic Syndr 1999; 22:33340. First citation in article
6. Folkvord JM, Anderson DM, Arya J, MaWhinney S, Connick E. Microanatomic relationships between CD8+ cells and HIV-1producing cells in human lymphoid tissue in vivo. J Acquir Immune Defic Syndr 2003; 32:46976. First citation in article
7. Simpson GR, Schulz TF, Whitby D, et al. Prevalence of Kaposi's sarcoma associated herpesvirus infection measured by antibodies to recombinant capsid protein and latent immunofluorescence antigen. Lancet 1996; 348:11338. First citation in article
8. Moore PS, Kingsley LA, Holmberg SD, et al. Kaposi's sarcomaassociated herpesvirus infection prior to onset of Kaposi's sarcoma. AIDS 1996; 10:17580. First citation in article
9. White IE, Campbell TB. Quantitation of cell-free and cell-associated Kaposi's sarcomaassociated herpesvirus DNA by real-time PCR. J Clin Microbiol 2000; 38:19925. First citation in article
10. Staskus KA, Zhong W, Gebhard K, et al. Kaposi's sarcomaassociated herpesvirus gene expression in endothelial (spindle) tumor cells. J Virol 1997; 71:7159. First citation in article
11. Staskus KA, Sun R, Miller G, et al. Cellular tropism and viral interleukin-6 expression distinguish human herpesvirus 8 involvement in Kaposi's sarcoma, primary effusion lymphoma, and multicentric Castleman's disease. J Virol 1999; 73:41817. First citation in article
12. Dupin N, Fisher C, Kellam P, et al. Distribution of human herpesvirus8 latently infected cells in Kaposi's sarcoma, multicentric Castleman's disease, and primary effusion lymphoma. Proc Nat Acad Sci USA 1999; 96:454651. First citation in article
13. O'Leary JJ, Kennedy M, Luttich K, et al. Localisation of HHV-8 in AIDS related lymphamdenopaty. J Clin Pathol: Mol Pathol 2000; 53:437. First citation in article
14. Parravicini C, Chandran B, Corbellino M, et al. Differential viral protein expression in Kaposi's sarcomaassociated herpesvirus-infected diseases: Kaposi's sarcoma, primary effusion lymphoma, and multicentric Castleman's disease. Am J Pathol 2000; 156:7439. First citation in article
15. Cannon JS, Nicholas J, Orenstein JM, et al. Heterogeneity of viral IL-6 expression in HHV-8associated diseases. J Infect Dis 1999; 180:8248. First citation in article