- Email: [email protected]
The Kaposi's sarcoma-associated herpesvirus (human herpesvirus-8) in Kaposi's sarcoma, malignant lymphoma, and other diseases
Annals of Oncology 8 (Suppl. 2): S123-S129, 1997. © 1997 Kluwer Academic Publishers. Printed in the Netherlands.
Original article The Kaposi's sarcoma-associated herpesvirus (human herpesvirus-8) in Kaposi's sarcoma, malignant lymphoma, and other diseases D. M. Knowles & E. Cesarman Department of Pathology, The New York Hospital- Cornell Medical Center, New York, NY, USA
Summary
Kaposi's sarcoma Kaposi's sarcoma (KS) has been largely considered an enigmatic clinicopathologic entity since it was discovered by Moritz Kaposi in 1872 [1]. Four clinical-epidemiologic forms of the disease have been described: classicalMediterranean KS, which occurs very rarely in elderly individuals, predominantly men, of Mediterranean and Eastern European descent, in whom it usually exhibits indolent clinical behavior; endemic-African KS, which occurs rather commonly in HIV-negative individuals in equatorial Africa, in whom it often exhibits aggressive clinical behavior; iatrogenic KS, which occurs relatively frequently in solid organ transplant recipients, in whom it often regresses following reduction of immunosuppressive therapy; and AIDS-epidemic KS [2], which has emerged as the most common neoplasm in individuals who have AIDS [2] and is a denning condition for that disease [3]. Kaposi's sarcoma is composed of a variable mixture of ectatic, irregularly shaped, round capillary and slit-like endothelial-lined vascular spaces and spindle-shaped cells accompanied by a variable inflammatory mononuclear cell infiltrate. Red blood cells and hemosiderin pigment are usually present, often extravasated between the spindle cells. The earliest patch and plaque stage lesions sometimes may be difficult to distinguish from granulation tissue. The spindle cells increasingly become the predom-
Key words: body cavity-based lymphoma, human herpesvirus-8, Kaposi's sarcoma, Kaposi's sarcoma-associated herpesvirus, multicentric Castleman's disease
inant cell population, forming fascicles that compress the vascular slits, and the lesions become progressively nodular [4]. The histopathologic features of the patch, plaque, and nodular lesions of KS are similar among the different clinical-epidemiologic categories of the disease [4]. The histogenesis of the spindle cell component, believed to be the 'tumor cell' of KS, has remained controversial, although most investigators favor an endothelial cell origin [5]A considerable body of clinical observations and epidemiologic data suggests that KS may have an infectious etiology and be transmitted sexually [6, 7]. For example, among AIDS patients, KS occurs almost exclusively among homosexual men, in whom it may be associated with specific sexual practices [6, 7]. In contrast, KS is very uncommon among adult AIDS patients infected through heterosexual or parenteral HIV transmission or among pediatric AIDS patients infected through vertical HIV transmission [5]. Numerous organisms, including cytomegalovirus, hepatitis B virus, human herpesvirus-6, human immunodeficiency virus (HIV), mycoplasma penetrans, and human papillomavirus, have been suggested as the etiologic agent of KS [5]. However, a strong etiologic association between any of these infectious organisms and KS has never been demonstrated [5]. Consequently, none of them have been accepted as a causative agent in the pathogenesis of KS.
Downloaded from http://annonc.oxfordjournals.org/ at UNIVERSITY OF PITTSBURGH on March 21, 2015
Background: Two novel nonhuman DNA fragments were discovered in an AIDS-related Kaposi's sarcoma (KS) lesion using representational difference analysis. Design: These sequences belong to a previously unidentified gamma-2-herpesvirus exhibiting homology with Herpesvirus saimiri and Epstein-Barr virus. This virus was named Kaposi's sarcoma-associated herpesvirus (KSHV) and provisionally designated human herpesvirus-8 (HHV-8). Results: KSHV is detectable in more than 90% of classicalMediterranean, iatrogenic, endemic-African, and AIDS-epidemic KS lesions. In situ PCR studies have demonstrated KSHV in the spindle cells and endothelial cells of KS lesions. KSHV appears to be a transmissible B-lymphotropic herpes-
virus. It is detectable in circulating B cells in some HIV-infected patients, and this finding appears to predict the future development of KS among these individuals. KSHV has been identified in a rare and distinct subset of malignant lymphoma referred to as body cavity-based lymphoma but not in other lymphoid neoplasms. KSHV is absent from most other HIV- and non-HIVassociated lymphadenopathies. Conclusions: Further studies should lead to a better understanding of the role of KSHV in the pathogenesis of these disorders and may eventually show that KSHV represents the long sought-after etiologic agent of Kaposi's sarcoma.
124 Discovery of Kaposi's sarcoma-associated herpesvirus
Characterization of KSHV Evidence supporting the notion that these DNA sequences are part of a new human herpesvirus has accumulated rapidly. Two lymphoma cell lines derived from AIDSrelated body cavity-based lymphomas (discussed below) that retain these sequences were established in our laboratory. These cell lines, designated BC-1 and BC-2, have been used extensively to characterize this viral agent [12]. We discovered that these sequences represent a portion of a much larger DNA molecule that is located in the nucleus of the infected cells. Fluorescence in situ hybridization studies of metaphase spreads prepared from both cell lines demonstrate these sequences in episomal structures [12]. Their presence as large nuclear episomes in these cells is consistent with a latent herpesvirus genome and, in conjunction with the homology data, strongly suggested that this agent belonged to the herpesvirus family. Sequence analysis of separate cloned fragments obtained from genomic DNA libraries by Moore et al. [13] and Cesarman et al. [14] further demonstrated that these sequences are part of a large herpesvirus genome. Moore and colleagues [13] used one of the originally described fragments, KS330Bam, to clone a 20.7 kb DNA fragment from a KS lesion. This fragment was entirely sequenced, revealing 17 partial or complete open reading frames (ORF), 16 of which have sequence and positional homology to known gamma-2 herpesvirus genes, including major capsid protein and thymidine kinase genes [13]. Phylogenetic analysis demonstrated that KSHV belongs to the gamma-2 sublineage (genus Rhadionovirus) of the Gammaherpesvirinae subfamily, making it the first member of this genus known to infect humans [13]. Cesarman et al. [14] used the second orginally described fragment, KS631Bam, to clone a 13 kb DNA fragment from an AIDS-related body cavity-based lymphoma. Sequence analysis of this fragment revealed four ORFs colinear with similar genes in HVS and present in the same tran-
Association of KSHV with Kaposi's sarcoma To determine the specificity of the two DNA fragments for AIDS-KS, Chang et al. hybridized them by Southern blotting to DNAs extracted from a variety of AIDS and non-AIDS tissues. They found that 78% of AIDS-KS lesions hybridized with variable intensity to KS330Bam and/or KS631Bam. They subsequently designed primers from KS33OBam that amplify a 233 bp fragment (KS330233) and found that 93% of the same AIDS-KS tissues were positive by polymerase chain reaction (PCR) amplification for KS330233, clearly demonstrating that KS33OBam is present in some KS lesions at levels below the threshold for detection by Southern blotting. They also identified the PCR-amplified 233 bp fragment in 3 of 12 AIDS lymph nodes and in 3 of 27 AIDS lymphomas
Downloaded from http://annonc.oxfordjournals.org/ at UNIVERSITY OF PITTSBURGH on March 21, 2015
Chang, Moore, Cesarman, Knowles, and coworkers utilized representational difference analysis [8] to identify two novel nonhuman DNA sequences, designated KS330Bam and KS631Bam, in KS tissue obtained from a homosexual man who had died from AIDS [9]. They showed that these DNA sequencs exhibit partial homology with genes encoding tegument and capsid proteins of Herpesvirus saimiri (HVS) and Epstein-Barr virus (EBV) [9], two herpesviruses belonging to the Gammaherpesvirinae subfamily [10]. The degree of homology suggested that these sequences belonged to a novel, previously unidentified member of the same herpesvirus family. For this reason, and because of its association with KS, this agent was initially named descriptively as Kaposi's sarcoma-associated herpesvirus (KSHV) [9]. It has been provisionally designated human herpesvirus-8 (HHV-8) [11].
scriptional orientation. Three of these four ORFs have homologies to known viral and/or cellular genes. These include homologies to (1) EBV and HVS membrane antigens pl40 and pl60, respectively; (2) HVS and cellular cyclines, predominantly the cyclin D family; and (3) HVS and cellular G-protein coupled receptors [14]. These findings demonstrate that KSHV is colinear with HVS, another gamma-2 herpesvirus, which appears to represent its closest relative. Subsequently, Mesri et al. [15] demonstrated that KSHV is encapsidated and distinct from EBV and that the KSHV-containing BC-1 lymphoma cell line produces infective viral particles capable of specifically transmitting KSHV to umbilical cord blood CD19-positive B cells but not to CD19-negative cells. This transmission requires a biologically active and replicating herpesvirus, i.e., KSHV, since it is blocked by ultraviolet irradiation, which is known to inactivate herpesviral DNA replication and transcription, and also by foscarnet, a viral DNA polymerase inhibitor. These findings strongly suggest that KSHV is a transmissible B-lymphotropic herpesvirus. Recently, we established a KSHV-containing, EBVnegative lymphoma cell line, designated BC-3, from a body cavity-based lymphoma arising in an HIV-negative patient [16]. Pulsed field gel electrophoresis of intact cells and viral pellets demonstrate KSHV genomes of 170 kb molecular weight, consistent with herpesviral genomes. Sonication and DNase treatment studies provide evidence for the encapsidation of these 170 kb viral genomes [16]. We further showed that the BC-3 cells exhibiting viral cytopathic effect contain typical 100 to 115 nm intranuclear herpesvirus nucleocapsids and complete cytoplasmic viral particles, i.e., KSHV [17]. Renne and colleagues [18] demonstrated that the in vitro production of large numbers of KSHV virions can be activated in KSHV-positive, EBV-negative body cavity-based lymphoma cell line cells by phorbol esters. These cell lines represent an invaluable source of KSHV for the evaluation of its pathogenic potential as well as characterization of its mechanistic role in the development of KS and malignant lymphoma.
125
Association of KSHV with malignant lymphoma Non-Hodgkin's lymphoma (NHL) is the second most common neoplasm occurring among persons who have AIDS [25]. The majority of AIDS-related NHLs are diffuse aggressive B-cell lymphomas exhibiting Burkitt, large-cell, or immunoblastic morphology [25]. In addition, Knowles et al. [26] and later other investigators [27, 28] have described an unusual and uncommonly occurring subset of AIDS-related NHLs that grow exclusively or mainly in the pleural, pericardial, or abdominal cavities as lymphomatous effusions, usually in the absence of a tumor mass. These body cavity-based lymphomas appear to exhibit other distinctive clinical as well as biological features, which include immunoblastic morphology, an indeterminate phenotype, clonal immunoglobulin gene rearrangements indicating a B-cell lineage derivation, and EBV, in the absence of c-myc gene rearrangements [29]. In their initial studies, Chang et al. [9] identified KSHV in three malignant lymphomas, prompting tts to investigate a large clinically and pathologically diverse panel of cases of AIDS and non-AIDS-related lymphoid neoplasia for KSHV by Southern blotting, PCR, or both. We found that all eight AIDS-related body cavity-based lymphomas, but none of the remaining 185 cases of AIDS and non-AIDS-related NHL, Hodgkin's disease, and lymphoid leukemia contained KSHV [30]. We subsequently demonstrated that body cavity-based lympho-
mas rarely occur in HIV-negative individuals also, and these cases similarly contain KSHV [31]. We also failed to identify KSHV in the pyothorax-associated lymphomas (PALs) [32], an uncommon subset of malignant lymphomas, most often reported from Japan, that are associated with long-standing pyothorax resulting from artificial pneumothorax for the treatment of pulmonary tuberculosis [33]. The PALs are body cavity-based but develop as a tumor mass in or near the pleura rather than as a lymphomatous effusion. They similarly are high-grade B-cell lymphomas, exhibit immunoblastic morphology, and contain EBV [33]. These findings strongly suggest a specific link between KSHV and the subset of body cavity-based lymphomas associated with an effusion. These observations have since been confirmed by other investigators [34, 35]. KSHV apparently has been rarely detected in other lymphomas [36, 37], but the significance of these anecdotal reports remains to be determined. The consistent presence of KSHV in this uncommon and unusual subset of NHLs that grow mainly in the body cavities as lymphomatous effusions without an identifiable tumor mass strongly suggests that they represent a distinct entity. We tested that hypothesis by investigating 19 malignant lymphomatous effusions occurring in the absence of a contiguous tumor mass for their clinical, morphologic, immunophenotypic, viral, and molecular characteristics [38]. We detected KSHV in 15 of the 19 lymphomas. The four KSHV-negative lymphomatous effusions exhibited Burkitt or Burkitt-like morphology and c-myc gene rearrangements and thus appeared to be Burkitt-type lymphomas occurring in the body cavities. The 15 KSHV-positive lymphomatous effusions exhibited a distinctive morphology bridging large-cell immunoblastic lymphoma and anaplastic large-cell lymphoma, and all 12 cases studied lacked c-myc gene rearrangements. In addition, these lymphomas occurred in men (15/15), frequently but not exclusively in association with HIV infection (13/15), in whom homosexuality was a risk factor (13/13), presented initially as a lymphomatous effusion in the absence of a contiguous tumor mass (14/15), remained localized to the body cavity of origin (13/15), expressed CD45 (15/15) and one or more activation-associated antigens (9/10) in the frequent absence of B-cell-associated antigens (11/15), exhibited clonal immunoglobulin gene rearrangements (13/13), contained EBV (14/15) and lacked bcl-2, bcl-6, ras and p53 gene alterations (13/15) [38]. These findings suggest that KSHV infection and c-myc gene inactivation may be mutually exclusive events and potentially represent alternative molecular mechanisms in the development of malignant lymphomas presenting as lymphomatous effusions. Based on these findings, we have proposed replacing the term 'body cavity-based lymphoma' with the term 'primary effusion lymphoma' since the latter term describes these lymphomas more accurately and avoids confusion with other lymphomas that occur in the body cavities. These findings further suggest that the KSHV-containing primary effusion lymphomas represent a distinct clinicopathologic and biologic entity.
Downloaded from http://annonc.oxfordjournals.org/ at UNIVERSITY OF PITTSBURGH on March 21, 2015
but not in any of 29 non-AIDS lymphomas, 7 non-AIDS benign lymph nodes, 5 vascular tumors, 13 biopsy specimens containing AIDS-associated opportunistic infections, or 49 consecutive surgical biopsies [9]. In summary, they showed that these novel sequences are highly associated with AIDS-KS. Since then, many investigators have confirmed our original observations and identified this PCR-amplified 233 bp fragment in virtually all KS lesions belonging to all four forms of the disease: classical-Mediterranean, endemic-African, iatrogenic, and AIDS-epidemic [19-23]. Moreover, these numerous samples have exhibited a high degree of homology with the published sequence, indicating that the DNA sequence of this virus is highly conserved and that similar viral strains are present in KS lesions worldwide. These molecular studies cannot demonstrate the cell population within the KS lesions that actually harbors the virus, however. This leaves open the possibility that KSHV resides in an inflammatory cell population associated with KS lesions and not in the KS cells themselves. However, Boshoff et al. [24] have demonstated by in situ PCR that KSHV is present in the nuclei of the majority of the spindle cells and the flat vascular lining cells in AIDS and non-AIDS cutaneous nodular KS lesions. Thus, KSHV appears to be present within the KS cells generally considered to represent the 'tumor cells' of KS and not in innocent bystander inflammatory cells.
126 Association of KSHV with other diseases
Figure 1. Electron photomicrographs of KSHV-positive, EBV-negative BC-3 primary effusion lymphoma cell line cells. Top: A BC-3 cell displays lysis of nuclear chromatin because of KSHV. Bottom: A BC-3 cell exhibiting intranuclear KSHV nucleocapsids (arrow heads) and a complete extracellular KSHV particle (') (uranyl acetate, lead citrate, x 15,000, x 50,000). (Photomicrographs courtesy of Dr. Jonathan Said, Cedars-Sinai Medical Center.)
ders. However, further studies are clearly needed to delineate the complete clinical spectrum of KSHV infection in both immunocompetent and in immunodeficient individuals.
Pathogenetic significance of KSHV A fierce debate concerning the pathogenetic significance of KSHV, especially in KS, has already begun [47, 48]. The presence of KSHV in virtually all KS lesions arising in all anatomic sites and belonging to all clinical-epidemiologic forms of the disease occurring worldwide suggests an incontrovertible link between KSHV and KS. However, KSHV is not detectable by PCR in the endo-
Downloaded from http://annonc.oxfordjournals.org/ at UNIVERSITY OF PITTSBURGH on March 21, 2015
We and most other investigators have failed to identify KSHV in vascular and lymphatic endothelial-cell proliferations other than KS, including reparative granulation tissue, hemangiomas, lymphangiomas, and angiosarcomas [9, 20, 24, 39, 40]. Therefore, KSHV appears to be specifically associated with the endothelial and spindle cells comprising KS lesions, rather than merely exhibiting a tropism for endothelium. Jin et al. [40] have even recommended that the PCR detection of KSHV serve as a reliable diagnostic marker for distinguishing KS from other vascular lesions. We and others have identified KSHV in the majority of cases of AIDS- and non-AIDS-associated multicentric Castleman's disease (MCD), whether accompanied by KS or not [41, 42], as well as in their peripheral blood mononuclear cells [42]. Multicentric Castleman's disease is an atypical lymphoproliferative disorder believed to be an immune dysregulation [43]. It occurs in older individuals, predominantly men, in association with a senescent immune system. The patients present with multiple lymphadenopathies and a variety of constitutional and other symptoms. They develop multisystem disease, including autoimmune phenomena, cytopenias, skin rashes, and intercurrent infections [43]. Also, they frequently develop malignancies, most commonly KS and NHL [43]. Kaposi's sarcoma develops in a particularly high proportion of HIV-positive patients who have MCD [42]. The identification of KSHV in AIDS and non-AIDS-associated MCD supports an even closer relationship between MCD and KS than previously hypothesized and suggests a role for KSHV in the pathogenesis of MCD as well. Luppi et al. [35] identified KSHV in 3 of 15 cases of non-HIV-related angioimmunoblastic lymphadenopathy and in 4 of 5 cases of a peculiar form of non-HIV-related lymphadenopathy characterized by florid follicular hyperplasia with increased vascularity that resembles MCD and certain stages of HIV-relaicd lymphadenopathy. KSHV has been identified only rarely in other nonAIDS-related benign lymphoid proliferations [35, 41, 42, 44] and thus does not appear to be widely distributed among the non-neoplastic lymphoproliferative disorders occurring in the general population. Last, Rady et al. [45] reported the detection of KSHV in a variety of benign and malignant skin lesions in four HIV-negative organ transplant recipients receiving immunosuppressive therapy. The population from which these patients were selected is not evident from the report, and control patient populations were not studied. Moreover, these findings have been challenged by other investigators who have failed to identify KSHV in such cutaneous proliferations [46]. In summary, KSHV appears to be preferentially and highly associated with KS, primary effusion lymphomas, and MCD occurring in both HIV-infected and non-HIVinfected individuals and to be present only very infrequently in other vascular and lymphoproliferative disor-
127
The strong association between KSHV and AIDS- and non-AIDS-related primary effusion lymphomas strongly suggests that this virus plays an important role in their pathogenesis. KSHV may represent a transforming virus. Support for this notion comes from the homology data. KSHV belongs to the Gammaherpesvirinae subfamily of herpesvirus, which is characterized by the ability to replicate in lymphoblastoid cells [10]. Herpesvirus saimiri and EBV, the two herpesviruses bearing the most structural homology to KSHV, are capable of inducing latent infection of peripheral blood lymphocytes of their natural hosts, immortalizing lymphocytes in vitro, and leading to the development of malignant lymphomas [10]. It is well known that EBV immortalizes B cells in vitro and is associated with various malignant lymphomas, including those associated with immune deficiency [50]. Herpesvirus saimiri is a squirrel monkey virus that can be isolated from lymphomas in New World primates other than its natural host [51]. Herpesvirus saimiri can transform T
cells to grow continuously in vitro [52]. As already discussed, KSHV has been identified in circulating benign peripheral blood CD19-positive B cells of patients with AIDS and non-AIDS-associated KS [21]. Also, as discussed above, we have demonstrated that KSHV contains homologues to known viral and/or cellular genes, including HVS and cellular G-protein-coupled receptors and HVS and cellular cyclins, predominantly the cyclin D family [14]. Many G-protein-coupled receptors are involved in cell growth and differentiation and some are involved in malignant transformation [53]. Cyclins are required for cellular division and thus play a key role in cellular proliferation [54]. Some of these homologues are expressed in primary effusion lymphomas and in KS lesions [14], suggesting that KSHV is playing an active biological role in these diseases. It is also remarkable that the majority of primary effusion lymphomas contain EBV in addition to KSHV [30, 38]. The presence of these two viruses in combination appears to be unique to these malignant lymphomas. Perhaps KSHV acts in conjunction with EBV to induce full transformation. It is well known that EBV is capable of immortalizing B cell in vitro but alone may be insufficient for tumor development, as in the case of the complementary role of EBV and an activated c-myc gene in Burkitt's lymphoma [55]. Genetic complementation can occur in vitro with dual viral infections, an example being activation of the EBV replicative cycle by infection with HHV-6 [56]. Alternatively, KSHV may be merely associated with their growth as an effusion. Obviously, as in the case of KS, the precise pathogenic role of KSHV in the development of these primary effusion lymphomas remains to be determined. The presence of KSHV in virtually all cases of MCD occurring in HIV-positive patients and many cases of MCD occurring in HIV-negative patients [41,42] strongly suggests a role for KSHV in the pathogenesis of this disease as well. However, several points require clarification. For example, given the very high frequency of KS in patients with MCD, sometimes even in the very same lymph nodes involved by MCD, one cannot entirely rule out the possibility that undetected microscopic foci of KS are responsible for some of the cases of apparent KSH Vcontaining MCD. Alternatively, some of the KSHVcontaining MCD lymph nodes may simply represent disseminated KSHV infection in those hosts. Furthermore, in situ hybridization studies have not yet been performed to determine whether KSHV resides in the lymphoid, the vascular, or both components of MCD. Satisfactory resolution of these issues will help clarify the role, if any, of KSHV in the pathogenesis of MCD. Finally, whether KSHV enjoys a restricted or a ubiquitous distribution has been hotly debated. However, Monini et al. [57] recently reported finding KSHV in more than 90% of ejaculates obtained from HIV-negative immunocompetent men undergoing surgery for varicocoele. If confirmed, these findings suggest that KSHV actually infects a large proportion of the healthy adult general population and that the spermatic organs serve
Downloaded from http://annonc.oxfordjournals.org/ at UNIVERSITY OF PITTSBURGH on March 21, 2015
thelial cell lines established from KS lesions, including those capable of inducing KS-like lesions of murine origin in nude mice and those exhibiting characteristics of malignant tumor cell lines [21, 22, 48]. KSHV appears to be lost from short-term cultures of KS lesions after a few culture passages, suggesting a cytopathic effect of the virus toward recipient cells and disappearance of the virus reservoir or, alternatively, the necessity of specific cofactors for the maintenance of KSHV-infected cells [22]. For this and other reasons, it could be argued that KSHV is simply an innocent bystander in KS lesions. It has been suggested that KSHV may not contribute to the formation of KS lesions but merely replicates in the endothelial and spindle cells of KS because its replication is favored by one or another of the cy tokines expressed in the lesions [24]. If this were true, however, the vascular proliferation of KS would occur prior to KSHV infection. Ambroziak et al. [21] demonstrated that KSHVinfected peripheral blood mononuclear cells correlate with the presence of KS. Whitby et al. [49] extended these observations. They detected KSHV in 52% of HIVpositive patients with KS but in only 8% of those without KS and not in any of 160 HIV-negative patients. Moreover, 6 of 11 (55%) HIV-positive patients without KS whose blood contained KSHV developed KS, compared with only 12 of 132 (9%) similar patients whose blood was KSHV-negative. Thus, KSHV infection appears to precede the occurrence of KS and predict the subsequent development of KS lesions as well, at least in many individuals with AIDS-related KS. These findings are incompatible with KSHV merely representing an innocent bystander and strongly suggest that KSHV is the long sought-after etiologic agent of KS. It is still conceivable, however, that KSHV infection of the KS spindle cells merely facilitates their growth factor-dependent proliferation without causing a completely transformed phenotype [24]. The pathogenesis of KS may involve a complex interplay of genetic factors, immunocompromise, KSHV infection, and additional cofactors including, for example, the HIV Tat protein.
128 as the site of that persistent latent infection. This would make KSHV similar to other known human herpesviruses. One obvious implication is that KSHV is transmissible by sexual contact. Another implication is that the mechanisms operational in the pathogenesis of the KSHV-associated diseases are related to viral reactivation rather than primary infection. Obviously, as in the case of the other human herpesviruses, many conditions, including for example, immunosupression, immune stimulation, and multiple genetic, environmental, and behavioral factors may cooperate in variable combinations in the pathogenesis of the KSHV-associated diseases. Significant additional investigation will be necessary to elucidate these mechanisms.
17.
18.
19.
20.
21.
22.
Supported in part by National Institutes for Health grants EY06337 to D. M. Knowles and CA68939 to E. Cesarman.
23.
References
25.
1. Kaposi M. Idiopathisches multiples Pigment sarcoma der haut. Arch Derm Syph 1872; 4: 265. 2. Beral V. Epidemiology of Kaposi's sarcoma. Cancer Surv 1991; 10: 5-22. 3. Centers for Disease Control. Revision of the CDC surveillance case definition for acquired immune deficiency syndrome. Morb Mort Wkly Rep 1987; 36 (Suppl): Sl-5. 4. Cockerell CJ. Histopathological features of Kaposi's sarcoma in HIV-infected individuals. Cancer Surv 1991; 10: 73-90. 5. Peterman TA, Jaflfe HW, Friedman-Kein AE et al. The etiology of Kaposi's sarcoma. Cancer Surv 1991; 10: 23-7. 6. Beral V, Peterman TA, Berkelman RL et al. Kaposi's sarcoma among persons with AIDS: A sexually transmitted infection? Lancet 1990; 335:123-8. 7. Beral V, Bull D, Darby S et al. Risk of Kaposi's sarcoma and sexual practices associated with fecal contact in homosexual or bisexual men with AIDS. Lancet 1992; 339: 632-5. 8. Lisitsyn N, Lisitsyn N, Wigler M. Cloning the differences between two complex genomes. Science 1993; 259: 946-51. 9. Chang Y, Cesarman E, Pessin MS et al. Identification of herpesvirus-like DNA sequences in AIDS-associated Kaposi's sarcoma. Science 1994; 266:1865-9. 10. Roizman B. Herpesviridae: A brief introduction. In Fields BN, Knipe DM (eds): Fields Virology, 2nd ed. New York: Raven Press 1990; 2: 1787-93. 11. Roizman B. New viral footprints in Kaposi's sarcoma. N Engl J Med 1995; 332: 1227. 12. Cesarman E, Moore PS, Inghirami G et al. In vitro establishment and characterization of two AIDS-related body cavity-based lymphoma cell lines containing the Kaposi's sarcoma-associated herpesvirus-like (KSHV) DNA sequences. Blood 1995; 86: 2708-14. 13. Moore PS, Gao S, Dominguez G et al. Primary characterization of a herpesvirus agent associated with Kaposi's sarcoma. J Virol 1996: 70: 549-58. 14. Cesarman E, Nador RG, Bai F et al. KSHV/HHV-8 contains G protein-coupled receptor and cyclin D homologues which are expressed in Kaposi's sarcoma and malignant lymphoma. J Virol 1996; 70: 8218-23. 15. Mesri EA, Cesarman E, Arvanitakis L et al. Human herpesvirus-8 (HHV-8/KSHV) is a new transmissable virus that infects B-cells. J Exp Med 1996; 183: 2385-90. 16. Arvanitakis L. Mesri EA. Nador RG et al. Establishment
24.
26.
27.
28.
29.
30.
31.
32.
33.
34.
35.
36.
37.
Downloaded from http://annonc.oxfordjournals.org/ at UNIVERSITY OF PITTSBURGH on March 21, 2015
Acknowledgements
and characterization of a body cavity-based lymphoma cell line (BC-3) harboring Kaposi's sarcoma-associated herpesvirus (KSHV/ HHV-8) in the absence of EBV. Blood 1996; 88: 2648-54. Said JW, Chien K, Tasaka T et al. Kaposi's sarcoma associated herpesvirus (KSHV or HHV-8) in primary effusion lymphoma: Ultrastructural demonstration of herpesvirus in lymphoma cell lines. Blood 1996; 87:4937^13. Renne R, Zhong W, Herndier B et al. Lytic growth of Kaposi's sarcoma-associated herpesvirus (human herpesvirus 8) in culture. Nature Med 1996; 2: 342-6. Dupin N, Grandadam M, Calvez V et al. Herpesvirus-like DNA sequences in patients with Mediterranean Kaposi's sarcoma. Lancet 1995; 345: 761-2. Boshoff C, Whitby D, Hatziionnou T et al. Kaposi's sarcomaassociated herpesvirus in HIV-negative Kaposi's sarcoma. Lancet 1995; 345: 1043-4. Ambroziak JA, Blackbourn DJ, Herndier BG et al. Herpes-like sequences in HIV-infected and uninfected Kaposi's sarcoma patients. Science 1995; 268: 582-3. Lebbe C, de Cremoux P, Rybojad M et al. Kaposi's sarcoma and new herpesvirus. Lancet 1995; 345:1180. Schalling M, Ekman M, Kaaya EE et al. A role for a new herpes virus (KSHV) in different forms of Kaposi's sarcoma. Nature Med 1995; 1: 707-8. Boshoff C, Schulz TF, Kennedy MM et al. Kaposi's sarcomaassociated herpesvirus infects endothelial and spindle cells. Nature Med 1995; 1:1274-8. Knowles DM, Chadburn A. Lymphadenopathy and the lymphoid neoplasms associated with the acquired immune deficiency syndrome (AIDS). In Knowles DM (ed): Neoplastic Hematopathology. Baltimore: Williams and Wilkins 1992; 773-835. Knowles DM, Inghirami G, Ubriaco A et al. Molecular genetic analysis of three AIDS-associated neoplasms of uncertain lineage demonstrates their B-cell derivation and the possible pathogenetic role of the Epstein-Barr virus. Blood 1989; 73: 792-9. Walts AE, Shintaku IP, Said JW. Diagnosis of malignant lymphoma in effusions from patients with AIDS by gene rearrangement. Am J Clin Path 1990; 94:170-5. Green I, Espiritu E, Ladanyi M et al. Primary lymphomatous effusions in AIDS: A morphologic, immunophenotypic, and molecular study. Modern Pathol 1995; 8: 39-45. Knowles DM, Dalla-Favera R. AIDS-associated malignant lymphoma. In Broder S, Merrigan TC, Bolognesi D (eds): Textbook of AIDS Medicine. Baltimore: Williams and Wilkins 1994; 431-63. Cesarman E, Chang Y, Moore PS et al. Kaposi's sarcoma-associated herpesvirus-like DNA sequences are present in AIDS-related body cavity based lymphomas. N Engl J Med 1995; 332: 1186-91. Nador RG, Cesarman E, Knowles DM et al. Herpes-like DNA sequences in a body cavity-based lymphoma in an HIV-negative patient. N Engl J Med 1995; 333: 943. Cesarman E, Nador RG, Aozasa K et al. Kaposi's sarcomaassociated herpesvirus in non-AIDS-related lymphomas occurring in the body cavities. Am J Pathol 1996; 149: 53-7. Fukayama M, Ibuka T, Hayashi Y et al. Epstein-Barr virus in pyothorax-associated pleural lymphoma. Am J Pathol 1993; 143: 1044-9. Carbone A, Tirelli U, Gloghini A et al. Herpesvirus-like DNA sequences selectively cluster with body cavity-based lymphomas throughout the spectrum of AIDS-related lymphomatous effusions. Eur J Cancer 1996; 32A: 555-6. Luppi M, Barozzi P, Maiorana A et al. Human herpesvirus-8 DNA sequences in human immunodeficiency virus-negative angioimmunoblastic lymphadenopathy and benign lymphadenopathy with giant germinal center hyperplasia and increased vascularity. Blood 1996; 87:3903-9. Luppi M, Barozzi P, Marasca R et al. HHV-8-associated primary cerebral B-cell lymphoma in HIV-negative patients after long-term steroids. Lancet 1996; 347: 980. Robert C, Agbalika F, Blanc F et al. HIV-negative patient with HHV-8 DNA follicular B-cell lymphoma associated with Kaposi's sarcoma. Lancet 1996; 347:1042-3.
129 50. Liebowitz D, Kieff E. Epstein-Barr virus. In Roizman B, Whitley RJ, Lopez C (eds): The Human Herpesviruses. New York: Raven Press 1993; 107. 51. Fleckenstein B, Desrosiers RC. Herpesvirus saimiri and herpesvirus ateles. In Roizman B (ed): The Herpesviruses. New York: Plenum Press 1982; 1: 253-32. 52. Biesinger B, Muller-Fleckenstein I, Simmer B et al. Stable growth transformation of human T lymphocytes by herpesvirus saimiri. Proc Natl Acad Sci USA 1992; 89: 3116-9. 53. Allen LF, Lefkowitz RJ, Caron MG et al. G-protein-coupled receptor genes as protooncogenes: Constitutively activating mutation of the alpha-lB-adrenergic receptor enhances mitogenesis and tumorigenicity. Proc Natl Acad Sci USA 1991; 88: 11354-8. 54. Peters G. The D-type cyclins and their role in tumorigenesis. J Cell Science 1994; 18: 89-96. 55. Lombardi L, Newcomb EW, Dalla-Favera R. Pathogenesis of Burkitt lymphoma: Expression of an activated c-myc oncogene causes the tumorigenic conversion of EBV-infected human B lymphoblasts. Cell 1987; 49:161-70. 56. Flamand L, Stefanescu I, Ablashi DV et al. Activation of the Epstein-Barr virus replicative cycle by human herpesvirus 6. J Virol 1993; 67: 6768-77. 57. Monini P, de Lellis L, Fabris M et al. Kaposi's sarcoma-associated herpesvirus DNA sequences in prostate tissue and human semen. N Engl J Med 1996; 334: 1168-72.
Correspondence to: Dr. Daniel M. Knowles Department of Pathology The New York Hospital - Cornell Medical Center 525 East 68th Street New York, NY 10021 USA
Downloaded from http://annonc.oxfordjournals.org/ at UNIVERSITY OF PITTSBURGH on March 21, 2015
38. Nador RG, Cesarman E, Chadburn A et al. Primary effusion lymphoma: A distinct clinicopathologic entity associated with the Kaposi's sarcoma-associated herpesvirus. Blood 1996; 88: 645-56. 39. Tomita Y, Naka N, Aozasa K et al. Absence of Kaposi's-sarcomaassociated herpesvirus-like DNA sequence (KSHV) in angiosarcomas developing in body-cavity and other sites. Int J Cancer 1996; 66:141-2. 40. Jin YT, Tsai ST, Yan JJ et al. Detection of Kaposi's sarcomaassociated herpesvirus-like DNA sequence in vascular lesions: A reliable diagnostic marker for Kaposi's sarcoma. Am J Clin Pathol 1996; 105: 360-3. 41. Cesarman E, Knowles DM. Herpes-like DNA sequences, AIDSrelated tumors, and Castleman's disease. N Engl J Med 1995; 333: 799. 42. Soulier J, Grollet L, Oksenhendler E et al. Kaposi's sarcomaassociated herpesvirus-like DNA sequences in multicentric Castleman's disease. Blood 1995; 86: 1276-80. 43. Frizzera G. Atypical lymphoproliferative disorders. In Knowles DM (ed): Neoplastic Hematopathology. Baltimore: Williams and Wilkins 1992; 459-96. 44. Chadburn A, Cesarman E, Nador RG et al. Kaposi's sarcomaassociated herpesvirus sequences in non-HIV-associated benign lymphoproliferative lesions. Lab Invest 1996; 74:109A. 45. Rady PL, Yen A, Rollefson JL et al. Herpesvirus-like DNA sequences in non-Kaposi's sarcoma skin lesions of transplant patients. Lancet 1995; 345: 1339-40. 46. Boshoff C, Talbot S, Kennedy M et al. HHV8 and skin cancers in immunosuppressed patients. Lancet 1996; 347: 338-9. 47. Cohen J. Is a new virus the cause of KS? Science 1994; 266:1803-4. 48. Cohen J. Controversy: Is KS really caused by new herpes virus? Science 1995; 268:1847-8. 49. Whitby D, Howard MR, Tenant-Flowers M et al. Detection of Kaposi sarcoma associated herpesvirus in peripheral blood of HIV-infected individuals and progression to Kaposi's sarcoma. Lancet 1995; 346: 799-802.
Original article The Kaposi's sarcoma-associated herpesvirus (human herpesvirus-8) in Kaposi's sarcoma, malignant lymphoma, and other diseases D. M. Knowles & E. Cesarman Department of Pathology, The New York Hospital- Cornell Medical Center, New York, NY, USA
Summary
Kaposi's sarcoma Kaposi's sarcoma (KS) has been largely considered an enigmatic clinicopathologic entity since it was discovered by Moritz Kaposi in 1872 [1]. Four clinical-epidemiologic forms of the disease have been described: classicalMediterranean KS, which occurs very rarely in elderly individuals, predominantly men, of Mediterranean and Eastern European descent, in whom it usually exhibits indolent clinical behavior; endemic-African KS, which occurs rather commonly in HIV-negative individuals in equatorial Africa, in whom it often exhibits aggressive clinical behavior; iatrogenic KS, which occurs relatively frequently in solid organ transplant recipients, in whom it often regresses following reduction of immunosuppressive therapy; and AIDS-epidemic KS [2], which has emerged as the most common neoplasm in individuals who have AIDS [2] and is a denning condition for that disease [3]. Kaposi's sarcoma is composed of a variable mixture of ectatic, irregularly shaped, round capillary and slit-like endothelial-lined vascular spaces and spindle-shaped cells accompanied by a variable inflammatory mononuclear cell infiltrate. Red blood cells and hemosiderin pigment are usually present, often extravasated between the spindle cells. The earliest patch and plaque stage lesions sometimes may be difficult to distinguish from granulation tissue. The spindle cells increasingly become the predom-
Key words: body cavity-based lymphoma, human herpesvirus-8, Kaposi's sarcoma, Kaposi's sarcoma-associated herpesvirus, multicentric Castleman's disease
inant cell population, forming fascicles that compress the vascular slits, and the lesions become progressively nodular [4]. The histopathologic features of the patch, plaque, and nodular lesions of KS are similar among the different clinical-epidemiologic categories of the disease [4]. The histogenesis of the spindle cell component, believed to be the 'tumor cell' of KS, has remained controversial, although most investigators favor an endothelial cell origin [5]A considerable body of clinical observations and epidemiologic data suggests that KS may have an infectious etiology and be transmitted sexually [6, 7]. For example, among AIDS patients, KS occurs almost exclusively among homosexual men, in whom it may be associated with specific sexual practices [6, 7]. In contrast, KS is very uncommon among adult AIDS patients infected through heterosexual or parenteral HIV transmission or among pediatric AIDS patients infected through vertical HIV transmission [5]. Numerous organisms, including cytomegalovirus, hepatitis B virus, human herpesvirus-6, human immunodeficiency virus (HIV), mycoplasma penetrans, and human papillomavirus, have been suggested as the etiologic agent of KS [5]. However, a strong etiologic association between any of these infectious organisms and KS has never been demonstrated [5]. Consequently, none of them have been accepted as a causative agent in the pathogenesis of KS.
Downloaded from http://annonc.oxfordjournals.org/ at UNIVERSITY OF PITTSBURGH on March 21, 2015
Background: Two novel nonhuman DNA fragments were discovered in an AIDS-related Kaposi's sarcoma (KS) lesion using representational difference analysis. Design: These sequences belong to a previously unidentified gamma-2-herpesvirus exhibiting homology with Herpesvirus saimiri and Epstein-Barr virus. This virus was named Kaposi's sarcoma-associated herpesvirus (KSHV) and provisionally designated human herpesvirus-8 (HHV-8). Results: KSHV is detectable in more than 90% of classicalMediterranean, iatrogenic, endemic-African, and AIDS-epidemic KS lesions. In situ PCR studies have demonstrated KSHV in the spindle cells and endothelial cells of KS lesions. KSHV appears to be a transmissible B-lymphotropic herpes-
virus. It is detectable in circulating B cells in some HIV-infected patients, and this finding appears to predict the future development of KS among these individuals. KSHV has been identified in a rare and distinct subset of malignant lymphoma referred to as body cavity-based lymphoma but not in other lymphoid neoplasms. KSHV is absent from most other HIV- and non-HIVassociated lymphadenopathies. Conclusions: Further studies should lead to a better understanding of the role of KSHV in the pathogenesis of these disorders and may eventually show that KSHV represents the long sought-after etiologic agent of Kaposi's sarcoma.
124 Discovery of Kaposi's sarcoma-associated herpesvirus
Characterization of KSHV Evidence supporting the notion that these DNA sequences are part of a new human herpesvirus has accumulated rapidly. Two lymphoma cell lines derived from AIDSrelated body cavity-based lymphomas (discussed below) that retain these sequences were established in our laboratory. These cell lines, designated BC-1 and BC-2, have been used extensively to characterize this viral agent [12]. We discovered that these sequences represent a portion of a much larger DNA molecule that is located in the nucleus of the infected cells. Fluorescence in situ hybridization studies of metaphase spreads prepared from both cell lines demonstrate these sequences in episomal structures [12]. Their presence as large nuclear episomes in these cells is consistent with a latent herpesvirus genome and, in conjunction with the homology data, strongly suggested that this agent belonged to the herpesvirus family. Sequence analysis of separate cloned fragments obtained from genomic DNA libraries by Moore et al. [13] and Cesarman et al. [14] further demonstrated that these sequences are part of a large herpesvirus genome. Moore and colleagues [13] used one of the originally described fragments, KS330Bam, to clone a 20.7 kb DNA fragment from a KS lesion. This fragment was entirely sequenced, revealing 17 partial or complete open reading frames (ORF), 16 of which have sequence and positional homology to known gamma-2 herpesvirus genes, including major capsid protein and thymidine kinase genes [13]. Phylogenetic analysis demonstrated that KSHV belongs to the gamma-2 sublineage (genus Rhadionovirus) of the Gammaherpesvirinae subfamily, making it the first member of this genus known to infect humans [13]. Cesarman et al. [14] used the second orginally described fragment, KS631Bam, to clone a 13 kb DNA fragment from an AIDS-related body cavity-based lymphoma. Sequence analysis of this fragment revealed four ORFs colinear with similar genes in HVS and present in the same tran-
Association of KSHV with Kaposi's sarcoma To determine the specificity of the two DNA fragments for AIDS-KS, Chang et al. hybridized them by Southern blotting to DNAs extracted from a variety of AIDS and non-AIDS tissues. They found that 78% of AIDS-KS lesions hybridized with variable intensity to KS330Bam and/or KS631Bam. They subsequently designed primers from KS33OBam that amplify a 233 bp fragment (KS330233) and found that 93% of the same AIDS-KS tissues were positive by polymerase chain reaction (PCR) amplification for KS330233, clearly demonstrating that KS33OBam is present in some KS lesions at levels below the threshold for detection by Southern blotting. They also identified the PCR-amplified 233 bp fragment in 3 of 12 AIDS lymph nodes and in 3 of 27 AIDS lymphomas
Downloaded from http://annonc.oxfordjournals.org/ at UNIVERSITY OF PITTSBURGH on March 21, 2015
Chang, Moore, Cesarman, Knowles, and coworkers utilized representational difference analysis [8] to identify two novel nonhuman DNA sequences, designated KS330Bam and KS631Bam, in KS tissue obtained from a homosexual man who had died from AIDS [9]. They showed that these DNA sequencs exhibit partial homology with genes encoding tegument and capsid proteins of Herpesvirus saimiri (HVS) and Epstein-Barr virus (EBV) [9], two herpesviruses belonging to the Gammaherpesvirinae subfamily [10]. The degree of homology suggested that these sequences belonged to a novel, previously unidentified member of the same herpesvirus family. For this reason, and because of its association with KS, this agent was initially named descriptively as Kaposi's sarcoma-associated herpesvirus (KSHV) [9]. It has been provisionally designated human herpesvirus-8 (HHV-8) [11].
scriptional orientation. Three of these four ORFs have homologies to known viral and/or cellular genes. These include homologies to (1) EBV and HVS membrane antigens pl40 and pl60, respectively; (2) HVS and cellular cyclines, predominantly the cyclin D family; and (3) HVS and cellular G-protein coupled receptors [14]. These findings demonstrate that KSHV is colinear with HVS, another gamma-2 herpesvirus, which appears to represent its closest relative. Subsequently, Mesri et al. [15] demonstrated that KSHV is encapsidated and distinct from EBV and that the KSHV-containing BC-1 lymphoma cell line produces infective viral particles capable of specifically transmitting KSHV to umbilical cord blood CD19-positive B cells but not to CD19-negative cells. This transmission requires a biologically active and replicating herpesvirus, i.e., KSHV, since it is blocked by ultraviolet irradiation, which is known to inactivate herpesviral DNA replication and transcription, and also by foscarnet, a viral DNA polymerase inhibitor. These findings strongly suggest that KSHV is a transmissible B-lymphotropic herpesvirus. Recently, we established a KSHV-containing, EBVnegative lymphoma cell line, designated BC-3, from a body cavity-based lymphoma arising in an HIV-negative patient [16]. Pulsed field gel electrophoresis of intact cells and viral pellets demonstrate KSHV genomes of 170 kb molecular weight, consistent with herpesviral genomes. Sonication and DNase treatment studies provide evidence for the encapsidation of these 170 kb viral genomes [16]. We further showed that the BC-3 cells exhibiting viral cytopathic effect contain typical 100 to 115 nm intranuclear herpesvirus nucleocapsids and complete cytoplasmic viral particles, i.e., KSHV [17]. Renne and colleagues [18] demonstrated that the in vitro production of large numbers of KSHV virions can be activated in KSHV-positive, EBV-negative body cavity-based lymphoma cell line cells by phorbol esters. These cell lines represent an invaluable source of KSHV for the evaluation of its pathogenic potential as well as characterization of its mechanistic role in the development of KS and malignant lymphoma.
125
Association of KSHV with malignant lymphoma Non-Hodgkin's lymphoma (NHL) is the second most common neoplasm occurring among persons who have AIDS [25]. The majority of AIDS-related NHLs are diffuse aggressive B-cell lymphomas exhibiting Burkitt, large-cell, or immunoblastic morphology [25]. In addition, Knowles et al. [26] and later other investigators [27, 28] have described an unusual and uncommonly occurring subset of AIDS-related NHLs that grow exclusively or mainly in the pleural, pericardial, or abdominal cavities as lymphomatous effusions, usually in the absence of a tumor mass. These body cavity-based lymphomas appear to exhibit other distinctive clinical as well as biological features, which include immunoblastic morphology, an indeterminate phenotype, clonal immunoglobulin gene rearrangements indicating a B-cell lineage derivation, and EBV, in the absence of c-myc gene rearrangements [29]. In their initial studies, Chang et al. [9] identified KSHV in three malignant lymphomas, prompting tts to investigate a large clinically and pathologically diverse panel of cases of AIDS and non-AIDS-related lymphoid neoplasia for KSHV by Southern blotting, PCR, or both. We found that all eight AIDS-related body cavity-based lymphomas, but none of the remaining 185 cases of AIDS and non-AIDS-related NHL, Hodgkin's disease, and lymphoid leukemia contained KSHV [30]. We subsequently demonstrated that body cavity-based lympho-
mas rarely occur in HIV-negative individuals also, and these cases similarly contain KSHV [31]. We also failed to identify KSHV in the pyothorax-associated lymphomas (PALs) [32], an uncommon subset of malignant lymphomas, most often reported from Japan, that are associated with long-standing pyothorax resulting from artificial pneumothorax for the treatment of pulmonary tuberculosis [33]. The PALs are body cavity-based but develop as a tumor mass in or near the pleura rather than as a lymphomatous effusion. They similarly are high-grade B-cell lymphomas, exhibit immunoblastic morphology, and contain EBV [33]. These findings strongly suggest a specific link between KSHV and the subset of body cavity-based lymphomas associated with an effusion. These observations have since been confirmed by other investigators [34, 35]. KSHV apparently has been rarely detected in other lymphomas [36, 37], but the significance of these anecdotal reports remains to be determined. The consistent presence of KSHV in this uncommon and unusual subset of NHLs that grow mainly in the body cavities as lymphomatous effusions without an identifiable tumor mass strongly suggests that they represent a distinct entity. We tested that hypothesis by investigating 19 malignant lymphomatous effusions occurring in the absence of a contiguous tumor mass for their clinical, morphologic, immunophenotypic, viral, and molecular characteristics [38]. We detected KSHV in 15 of the 19 lymphomas. The four KSHV-negative lymphomatous effusions exhibited Burkitt or Burkitt-like morphology and c-myc gene rearrangements and thus appeared to be Burkitt-type lymphomas occurring in the body cavities. The 15 KSHV-positive lymphomatous effusions exhibited a distinctive morphology bridging large-cell immunoblastic lymphoma and anaplastic large-cell lymphoma, and all 12 cases studied lacked c-myc gene rearrangements. In addition, these lymphomas occurred in men (15/15), frequently but not exclusively in association with HIV infection (13/15), in whom homosexuality was a risk factor (13/13), presented initially as a lymphomatous effusion in the absence of a contiguous tumor mass (14/15), remained localized to the body cavity of origin (13/15), expressed CD45 (15/15) and one or more activation-associated antigens (9/10) in the frequent absence of B-cell-associated antigens (11/15), exhibited clonal immunoglobulin gene rearrangements (13/13), contained EBV (14/15) and lacked bcl-2, bcl-6, ras and p53 gene alterations (13/15) [38]. These findings suggest that KSHV infection and c-myc gene inactivation may be mutually exclusive events and potentially represent alternative molecular mechanisms in the development of malignant lymphomas presenting as lymphomatous effusions. Based on these findings, we have proposed replacing the term 'body cavity-based lymphoma' with the term 'primary effusion lymphoma' since the latter term describes these lymphomas more accurately and avoids confusion with other lymphomas that occur in the body cavities. These findings further suggest that the KSHV-containing primary effusion lymphomas represent a distinct clinicopathologic and biologic entity.
Downloaded from http://annonc.oxfordjournals.org/ at UNIVERSITY OF PITTSBURGH on March 21, 2015
but not in any of 29 non-AIDS lymphomas, 7 non-AIDS benign lymph nodes, 5 vascular tumors, 13 biopsy specimens containing AIDS-associated opportunistic infections, or 49 consecutive surgical biopsies [9]. In summary, they showed that these novel sequences are highly associated with AIDS-KS. Since then, many investigators have confirmed our original observations and identified this PCR-amplified 233 bp fragment in virtually all KS lesions belonging to all four forms of the disease: classical-Mediterranean, endemic-African, iatrogenic, and AIDS-epidemic [19-23]. Moreover, these numerous samples have exhibited a high degree of homology with the published sequence, indicating that the DNA sequence of this virus is highly conserved and that similar viral strains are present in KS lesions worldwide. These molecular studies cannot demonstrate the cell population within the KS lesions that actually harbors the virus, however. This leaves open the possibility that KSHV resides in an inflammatory cell population associated with KS lesions and not in the KS cells themselves. However, Boshoff et al. [24] have demonstated by in situ PCR that KSHV is present in the nuclei of the majority of the spindle cells and the flat vascular lining cells in AIDS and non-AIDS cutaneous nodular KS lesions. Thus, KSHV appears to be present within the KS cells generally considered to represent the 'tumor cells' of KS and not in innocent bystander inflammatory cells.
126 Association of KSHV with other diseases
Figure 1. Electron photomicrographs of KSHV-positive, EBV-negative BC-3 primary effusion lymphoma cell line cells. Top: A BC-3 cell displays lysis of nuclear chromatin because of KSHV. Bottom: A BC-3 cell exhibiting intranuclear KSHV nucleocapsids (arrow heads) and a complete extracellular KSHV particle (') (uranyl acetate, lead citrate, x 15,000, x 50,000). (Photomicrographs courtesy of Dr. Jonathan Said, Cedars-Sinai Medical Center.)
ders. However, further studies are clearly needed to delineate the complete clinical spectrum of KSHV infection in both immunocompetent and in immunodeficient individuals.
Pathogenetic significance of KSHV A fierce debate concerning the pathogenetic significance of KSHV, especially in KS, has already begun [47, 48]. The presence of KSHV in virtually all KS lesions arising in all anatomic sites and belonging to all clinical-epidemiologic forms of the disease occurring worldwide suggests an incontrovertible link between KSHV and KS. However, KSHV is not detectable by PCR in the endo-
Downloaded from http://annonc.oxfordjournals.org/ at UNIVERSITY OF PITTSBURGH on March 21, 2015
We and most other investigators have failed to identify KSHV in vascular and lymphatic endothelial-cell proliferations other than KS, including reparative granulation tissue, hemangiomas, lymphangiomas, and angiosarcomas [9, 20, 24, 39, 40]. Therefore, KSHV appears to be specifically associated with the endothelial and spindle cells comprising KS lesions, rather than merely exhibiting a tropism for endothelium. Jin et al. [40] have even recommended that the PCR detection of KSHV serve as a reliable diagnostic marker for distinguishing KS from other vascular lesions. We and others have identified KSHV in the majority of cases of AIDS- and non-AIDS-associated multicentric Castleman's disease (MCD), whether accompanied by KS or not [41, 42], as well as in their peripheral blood mononuclear cells [42]. Multicentric Castleman's disease is an atypical lymphoproliferative disorder believed to be an immune dysregulation [43]. It occurs in older individuals, predominantly men, in association with a senescent immune system. The patients present with multiple lymphadenopathies and a variety of constitutional and other symptoms. They develop multisystem disease, including autoimmune phenomena, cytopenias, skin rashes, and intercurrent infections [43]. Also, they frequently develop malignancies, most commonly KS and NHL [43]. Kaposi's sarcoma develops in a particularly high proportion of HIV-positive patients who have MCD [42]. The identification of KSHV in AIDS and non-AIDS-associated MCD supports an even closer relationship between MCD and KS than previously hypothesized and suggests a role for KSHV in the pathogenesis of MCD as well. Luppi et al. [35] identified KSHV in 3 of 15 cases of non-HIV-related angioimmunoblastic lymphadenopathy and in 4 of 5 cases of a peculiar form of non-HIV-related lymphadenopathy characterized by florid follicular hyperplasia with increased vascularity that resembles MCD and certain stages of HIV-relaicd lymphadenopathy. KSHV has been identified only rarely in other nonAIDS-related benign lymphoid proliferations [35, 41, 42, 44] and thus does not appear to be widely distributed among the non-neoplastic lymphoproliferative disorders occurring in the general population. Last, Rady et al. [45] reported the detection of KSHV in a variety of benign and malignant skin lesions in four HIV-negative organ transplant recipients receiving immunosuppressive therapy. The population from which these patients were selected is not evident from the report, and control patient populations were not studied. Moreover, these findings have been challenged by other investigators who have failed to identify KSHV in such cutaneous proliferations [46]. In summary, KSHV appears to be preferentially and highly associated with KS, primary effusion lymphomas, and MCD occurring in both HIV-infected and non-HIVinfected individuals and to be present only very infrequently in other vascular and lymphoproliferative disor-
127
The strong association between KSHV and AIDS- and non-AIDS-related primary effusion lymphomas strongly suggests that this virus plays an important role in their pathogenesis. KSHV may represent a transforming virus. Support for this notion comes from the homology data. KSHV belongs to the Gammaherpesvirinae subfamily of herpesvirus, which is characterized by the ability to replicate in lymphoblastoid cells [10]. Herpesvirus saimiri and EBV, the two herpesviruses bearing the most structural homology to KSHV, are capable of inducing latent infection of peripheral blood lymphocytes of their natural hosts, immortalizing lymphocytes in vitro, and leading to the development of malignant lymphomas [10]. It is well known that EBV immortalizes B cells in vitro and is associated with various malignant lymphomas, including those associated with immune deficiency [50]. Herpesvirus saimiri is a squirrel monkey virus that can be isolated from lymphomas in New World primates other than its natural host [51]. Herpesvirus saimiri can transform T
cells to grow continuously in vitro [52]. As already discussed, KSHV has been identified in circulating benign peripheral blood CD19-positive B cells of patients with AIDS and non-AIDS-associated KS [21]. Also, as discussed above, we have demonstrated that KSHV contains homologues to known viral and/or cellular genes, including HVS and cellular G-protein-coupled receptors and HVS and cellular cyclins, predominantly the cyclin D family [14]. Many G-protein-coupled receptors are involved in cell growth and differentiation and some are involved in malignant transformation [53]. Cyclins are required for cellular division and thus play a key role in cellular proliferation [54]. Some of these homologues are expressed in primary effusion lymphomas and in KS lesions [14], suggesting that KSHV is playing an active biological role in these diseases. It is also remarkable that the majority of primary effusion lymphomas contain EBV in addition to KSHV [30, 38]. The presence of these two viruses in combination appears to be unique to these malignant lymphomas. Perhaps KSHV acts in conjunction with EBV to induce full transformation. It is well known that EBV is capable of immortalizing B cell in vitro but alone may be insufficient for tumor development, as in the case of the complementary role of EBV and an activated c-myc gene in Burkitt's lymphoma [55]. Genetic complementation can occur in vitro with dual viral infections, an example being activation of the EBV replicative cycle by infection with HHV-6 [56]. Alternatively, KSHV may be merely associated with their growth as an effusion. Obviously, as in the case of KS, the precise pathogenic role of KSHV in the development of these primary effusion lymphomas remains to be determined. The presence of KSHV in virtually all cases of MCD occurring in HIV-positive patients and many cases of MCD occurring in HIV-negative patients [41,42] strongly suggests a role for KSHV in the pathogenesis of this disease as well. However, several points require clarification. For example, given the very high frequency of KS in patients with MCD, sometimes even in the very same lymph nodes involved by MCD, one cannot entirely rule out the possibility that undetected microscopic foci of KS are responsible for some of the cases of apparent KSH Vcontaining MCD. Alternatively, some of the KSHVcontaining MCD lymph nodes may simply represent disseminated KSHV infection in those hosts. Furthermore, in situ hybridization studies have not yet been performed to determine whether KSHV resides in the lymphoid, the vascular, or both components of MCD. Satisfactory resolution of these issues will help clarify the role, if any, of KSHV in the pathogenesis of MCD. Finally, whether KSHV enjoys a restricted or a ubiquitous distribution has been hotly debated. However, Monini et al. [57] recently reported finding KSHV in more than 90% of ejaculates obtained from HIV-negative immunocompetent men undergoing surgery for varicocoele. If confirmed, these findings suggest that KSHV actually infects a large proportion of the healthy adult general population and that the spermatic organs serve
Downloaded from http://annonc.oxfordjournals.org/ at UNIVERSITY OF PITTSBURGH on March 21, 2015
thelial cell lines established from KS lesions, including those capable of inducing KS-like lesions of murine origin in nude mice and those exhibiting characteristics of malignant tumor cell lines [21, 22, 48]. KSHV appears to be lost from short-term cultures of KS lesions after a few culture passages, suggesting a cytopathic effect of the virus toward recipient cells and disappearance of the virus reservoir or, alternatively, the necessity of specific cofactors for the maintenance of KSHV-infected cells [22]. For this and other reasons, it could be argued that KSHV is simply an innocent bystander in KS lesions. It has been suggested that KSHV may not contribute to the formation of KS lesions but merely replicates in the endothelial and spindle cells of KS because its replication is favored by one or another of the cy tokines expressed in the lesions [24]. If this were true, however, the vascular proliferation of KS would occur prior to KSHV infection. Ambroziak et al. [21] demonstrated that KSHVinfected peripheral blood mononuclear cells correlate with the presence of KS. Whitby et al. [49] extended these observations. They detected KSHV in 52% of HIVpositive patients with KS but in only 8% of those without KS and not in any of 160 HIV-negative patients. Moreover, 6 of 11 (55%) HIV-positive patients without KS whose blood contained KSHV developed KS, compared with only 12 of 132 (9%) similar patients whose blood was KSHV-negative. Thus, KSHV infection appears to precede the occurrence of KS and predict the subsequent development of KS lesions as well, at least in many individuals with AIDS-related KS. These findings are incompatible with KSHV merely representing an innocent bystander and strongly suggest that KSHV is the long sought-after etiologic agent of KS. It is still conceivable, however, that KSHV infection of the KS spindle cells merely facilitates their growth factor-dependent proliferation without causing a completely transformed phenotype [24]. The pathogenesis of KS may involve a complex interplay of genetic factors, immunocompromise, KSHV infection, and additional cofactors including, for example, the HIV Tat protein.
128 as the site of that persistent latent infection. This would make KSHV similar to other known human herpesviruses. One obvious implication is that KSHV is transmissible by sexual contact. Another implication is that the mechanisms operational in the pathogenesis of the KSHV-associated diseases are related to viral reactivation rather than primary infection. Obviously, as in the case of the other human herpesviruses, many conditions, including for example, immunosupression, immune stimulation, and multiple genetic, environmental, and behavioral factors may cooperate in variable combinations in the pathogenesis of the KSHV-associated diseases. Significant additional investigation will be necessary to elucidate these mechanisms.
17.
18.
19.
20.
21.
22.
Supported in part by National Institutes for Health grants EY06337 to D. M. Knowles and CA68939 to E. Cesarman.
23.
References
25.
1. Kaposi M. Idiopathisches multiples Pigment sarcoma der haut. Arch Derm Syph 1872; 4: 265. 2. Beral V. Epidemiology of Kaposi's sarcoma. Cancer Surv 1991; 10: 5-22. 3. Centers for Disease Control. Revision of the CDC surveillance case definition for acquired immune deficiency syndrome. Morb Mort Wkly Rep 1987; 36 (Suppl): Sl-5. 4. Cockerell CJ. Histopathological features of Kaposi's sarcoma in HIV-infected individuals. Cancer Surv 1991; 10: 73-90. 5. Peterman TA, Jaflfe HW, Friedman-Kein AE et al. The etiology of Kaposi's sarcoma. Cancer Surv 1991; 10: 23-7. 6. Beral V, Peterman TA, Berkelman RL et al. Kaposi's sarcoma among persons with AIDS: A sexually transmitted infection? Lancet 1990; 335:123-8. 7. Beral V, Bull D, Darby S et al. Risk of Kaposi's sarcoma and sexual practices associated with fecal contact in homosexual or bisexual men with AIDS. Lancet 1992; 339: 632-5. 8. Lisitsyn N, Lisitsyn N, Wigler M. Cloning the differences between two complex genomes. Science 1993; 259: 946-51. 9. Chang Y, Cesarman E, Pessin MS et al. Identification of herpesvirus-like DNA sequences in AIDS-associated Kaposi's sarcoma. Science 1994; 266:1865-9. 10. Roizman B. Herpesviridae: A brief introduction. In Fields BN, Knipe DM (eds): Fields Virology, 2nd ed. New York: Raven Press 1990; 2: 1787-93. 11. Roizman B. New viral footprints in Kaposi's sarcoma. N Engl J Med 1995; 332: 1227. 12. Cesarman E, Moore PS, Inghirami G et al. In vitro establishment and characterization of two AIDS-related body cavity-based lymphoma cell lines containing the Kaposi's sarcoma-associated herpesvirus-like (KSHV) DNA sequences. Blood 1995; 86: 2708-14. 13. Moore PS, Gao S, Dominguez G et al. Primary characterization of a herpesvirus agent associated with Kaposi's sarcoma. J Virol 1996: 70: 549-58. 14. Cesarman E, Nador RG, Bai F et al. KSHV/HHV-8 contains G protein-coupled receptor and cyclin D homologues which are expressed in Kaposi's sarcoma and malignant lymphoma. J Virol 1996; 70: 8218-23. 15. Mesri EA, Cesarman E, Arvanitakis L et al. Human herpesvirus-8 (HHV-8/KSHV) is a new transmissable virus that infects B-cells. J Exp Med 1996; 183: 2385-90. 16. Arvanitakis L. Mesri EA. Nador RG et al. Establishment
24.
26.
27.
28.
29.
30.
31.
32.
33.
34.
35.
36.
37.
Downloaded from http://annonc.oxfordjournals.org/ at UNIVERSITY OF PITTSBURGH on March 21, 2015
Acknowledgements
and characterization of a body cavity-based lymphoma cell line (BC-3) harboring Kaposi's sarcoma-associated herpesvirus (KSHV/ HHV-8) in the absence of EBV. Blood 1996; 88: 2648-54. Said JW, Chien K, Tasaka T et al. Kaposi's sarcoma associated herpesvirus (KSHV or HHV-8) in primary effusion lymphoma: Ultrastructural demonstration of herpesvirus in lymphoma cell lines. Blood 1996; 87:4937^13. Renne R, Zhong W, Herndier B et al. Lytic growth of Kaposi's sarcoma-associated herpesvirus (human herpesvirus 8) in culture. Nature Med 1996; 2: 342-6. Dupin N, Grandadam M, Calvez V et al. Herpesvirus-like DNA sequences in patients with Mediterranean Kaposi's sarcoma. Lancet 1995; 345: 761-2. Boshoff C, Whitby D, Hatziionnou T et al. Kaposi's sarcomaassociated herpesvirus in HIV-negative Kaposi's sarcoma. Lancet 1995; 345: 1043-4. Ambroziak JA, Blackbourn DJ, Herndier BG et al. Herpes-like sequences in HIV-infected and uninfected Kaposi's sarcoma patients. Science 1995; 268: 582-3. Lebbe C, de Cremoux P, Rybojad M et al. Kaposi's sarcoma and new herpesvirus. Lancet 1995; 345:1180. Schalling M, Ekman M, Kaaya EE et al. A role for a new herpes virus (KSHV) in different forms of Kaposi's sarcoma. Nature Med 1995; 1: 707-8. Boshoff C, Schulz TF, Kennedy MM et al. Kaposi's sarcomaassociated herpesvirus infects endothelial and spindle cells. Nature Med 1995; 1:1274-8. Knowles DM, Chadburn A. Lymphadenopathy and the lymphoid neoplasms associated with the acquired immune deficiency syndrome (AIDS). In Knowles DM (ed): Neoplastic Hematopathology. Baltimore: Williams and Wilkins 1992; 773-835. Knowles DM, Inghirami G, Ubriaco A et al. Molecular genetic analysis of three AIDS-associated neoplasms of uncertain lineage demonstrates their B-cell derivation and the possible pathogenetic role of the Epstein-Barr virus. Blood 1989; 73: 792-9. Walts AE, Shintaku IP, Said JW. Diagnosis of malignant lymphoma in effusions from patients with AIDS by gene rearrangement. Am J Clin Path 1990; 94:170-5. Green I, Espiritu E, Ladanyi M et al. Primary lymphomatous effusions in AIDS: A morphologic, immunophenotypic, and molecular study. Modern Pathol 1995; 8: 39-45. Knowles DM, Dalla-Favera R. AIDS-associated malignant lymphoma. In Broder S, Merrigan TC, Bolognesi D (eds): Textbook of AIDS Medicine. Baltimore: Williams and Wilkins 1994; 431-63. Cesarman E, Chang Y, Moore PS et al. Kaposi's sarcoma-associated herpesvirus-like DNA sequences are present in AIDS-related body cavity based lymphomas. N Engl J Med 1995; 332: 1186-91. Nador RG, Cesarman E, Knowles DM et al. Herpes-like DNA sequences in a body cavity-based lymphoma in an HIV-negative patient. N Engl J Med 1995; 333: 943. Cesarman E, Nador RG, Aozasa K et al. Kaposi's sarcomaassociated herpesvirus in non-AIDS-related lymphomas occurring in the body cavities. Am J Pathol 1996; 149: 53-7. Fukayama M, Ibuka T, Hayashi Y et al. Epstein-Barr virus in pyothorax-associated pleural lymphoma. Am J Pathol 1993; 143: 1044-9. Carbone A, Tirelli U, Gloghini A et al. Herpesvirus-like DNA sequences selectively cluster with body cavity-based lymphomas throughout the spectrum of AIDS-related lymphomatous effusions. Eur J Cancer 1996; 32A: 555-6. Luppi M, Barozzi P, Maiorana A et al. Human herpesvirus-8 DNA sequences in human immunodeficiency virus-negative angioimmunoblastic lymphadenopathy and benign lymphadenopathy with giant germinal center hyperplasia and increased vascularity. Blood 1996; 87:3903-9. Luppi M, Barozzi P, Marasca R et al. HHV-8-associated primary cerebral B-cell lymphoma in HIV-negative patients after long-term steroids. Lancet 1996; 347: 980. Robert C, Agbalika F, Blanc F et al. HIV-negative patient with HHV-8 DNA follicular B-cell lymphoma associated with Kaposi's sarcoma. Lancet 1996; 347:1042-3.
129 50. Liebowitz D, Kieff E. Epstein-Barr virus. In Roizman B, Whitley RJ, Lopez C (eds): The Human Herpesviruses. New York: Raven Press 1993; 107. 51. Fleckenstein B, Desrosiers RC. Herpesvirus saimiri and herpesvirus ateles. In Roizman B (ed): The Herpesviruses. New York: Plenum Press 1982; 1: 253-32. 52. Biesinger B, Muller-Fleckenstein I, Simmer B et al. Stable growth transformation of human T lymphocytes by herpesvirus saimiri. Proc Natl Acad Sci USA 1992; 89: 3116-9. 53. Allen LF, Lefkowitz RJ, Caron MG et al. G-protein-coupled receptor genes as protooncogenes: Constitutively activating mutation of the alpha-lB-adrenergic receptor enhances mitogenesis and tumorigenicity. Proc Natl Acad Sci USA 1991; 88: 11354-8. 54. Peters G. The D-type cyclins and their role in tumorigenesis. J Cell Science 1994; 18: 89-96. 55. Lombardi L, Newcomb EW, Dalla-Favera R. Pathogenesis of Burkitt lymphoma: Expression of an activated c-myc oncogene causes the tumorigenic conversion of EBV-infected human B lymphoblasts. Cell 1987; 49:161-70. 56. Flamand L, Stefanescu I, Ablashi DV et al. Activation of the Epstein-Barr virus replicative cycle by human herpesvirus 6. J Virol 1993; 67: 6768-77. 57. Monini P, de Lellis L, Fabris M et al. Kaposi's sarcoma-associated herpesvirus DNA sequences in prostate tissue and human semen. N Engl J Med 1996; 334: 1168-72.
Correspondence to: Dr. Daniel M. Knowles Department of Pathology The New York Hospital - Cornell Medical Center 525 East 68th Street New York, NY 10021 USA
Downloaded from http://annonc.oxfordjournals.org/ at UNIVERSITY OF PITTSBURGH on March 21, 2015
38. Nador RG, Cesarman E, Chadburn A et al. Primary effusion lymphoma: A distinct clinicopathologic entity associated with the Kaposi's sarcoma-associated herpesvirus. Blood 1996; 88: 645-56. 39. Tomita Y, Naka N, Aozasa K et al. Absence of Kaposi's-sarcomaassociated herpesvirus-like DNA sequence (KSHV) in angiosarcomas developing in body-cavity and other sites. Int J Cancer 1996; 66:141-2. 40. Jin YT, Tsai ST, Yan JJ et al. Detection of Kaposi's sarcomaassociated herpesvirus-like DNA sequence in vascular lesions: A reliable diagnostic marker for Kaposi's sarcoma. Am J Clin Pathol 1996; 105: 360-3. 41. Cesarman E, Knowles DM. Herpes-like DNA sequences, AIDSrelated tumors, and Castleman's disease. N Engl J Med 1995; 333: 799. 42. Soulier J, Grollet L, Oksenhendler E et al. Kaposi's sarcomaassociated herpesvirus-like DNA sequences in multicentric Castleman's disease. Blood 1995; 86: 1276-80. 43. Frizzera G. Atypical lymphoproliferative disorders. In Knowles DM (ed): Neoplastic Hematopathology. Baltimore: Williams and Wilkins 1992; 459-96. 44. Chadburn A, Cesarman E, Nador RG et al. Kaposi's sarcomaassociated herpesvirus sequences in non-HIV-associated benign lymphoproliferative lesions. Lab Invest 1996; 74:109A. 45. Rady PL, Yen A, Rollefson JL et al. Herpesvirus-like DNA sequences in non-Kaposi's sarcoma skin lesions of transplant patients. Lancet 1995; 345: 1339-40. 46. Boshoff C, Talbot S, Kennedy M et al. HHV8 and skin cancers in immunosuppressed patients. Lancet 1996; 347: 338-9. 47. Cohen J. Is a new virus the cause of KS? Science 1994; 266:1803-4. 48. Cohen J. Controversy: Is KS really caused by new herpes virus? Science 1995; 268:1847-8. 49. Whitby D, Howard MR, Tenant-Flowers M et al. Detection of Kaposi sarcoma associated herpesvirus in peripheral blood of HIV-infected individuals and progression to Kaposi's sarcoma. Lancet 1995; 346: 799-802.