| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
Regular Articles |
Division of Hematology and Oncology, Cedars-Sinai Medical Center [H. J. M., N. N. S-S., R. A. V., M. K., A. M., K. P., D. B., J. R. B.], Division of Hematology and Oncology, Veterans Affairs Greater Los Angeles Health Care System [M. P., C. H. W., J. Z., L. Z., G. C.], Division of Pathology [J. W. S.], University of California Los Angeles School of Medicine, and Jonsson Comprehensive Cancer Center [H. J. M., N. N. S-S., R. A. V., M. K., A. M., M. P., K. P., D. B., C. H. W., J. Z., L. Z., G. C., J. R. B.], Los Angeles, California 90048; Department of Hematology, Ibni Sina Hospital, Sihhiye Ankara, Turkey [M. B.]; and BIS Laboratories, Reseda, California [T. J. M.]
ABSTRACT
Human herpesvirus 8 (HHV-8), also known as Kaposi’s sarcoma-associated herpesvirus, has been implicated in the pathogenesis of Kaposi’s sarcoma (KS), primary effusion lymphoma (PEL), multicentric Castleman’s disease, and recently multiple myeloma (MM). DNA sequence analyses of HHV-8 suggest that multiple HHV-8 strains exist. We extracted DNA from 24 patients with MM and 3 patients with monoclonal gammopathy of undetermined significance and compared HHV-8 open reading frames (ORFs) 26 and 65 sequences with those derived from patients with KS, PEL, and two HHV-8-positive PEL cell lines KS-1 and BC-1. ORF26 sequence data suggest that MM patients are consistently carriers of HHV-8 strain subtype C3. All MM patients also consistently revealed either a single bp deletion or substitution at position 112197 in ORF65. This unique alteration is not present in patients with KS or PEL or in PEL cell lines. It occurs in the portion of ORF65 that is known to be responsible for a serological response to HHV-8.
INTRODUCTION
MM3
is the
second most common hematological malignancy in the United States. It is
characterized by an accumulation of abnormal plasma cells in the bone
marrow and a monoclonal gammopathy in serum, urine, or both.
Clinically, patients with MM commonly develop lytic bone lesions,
anemia, recurrent bacterial infections, and renal insufficiency
(1)
and have a median survival of 
3 years
(2)
.
Much of the pathogenesis of MM remains unknown. Recently, DNA sequences from a new herpesvirus, HHV-8, were found in the bone marrow dendritic cells from patients with MM, suggesting a role for this virus in the pathogenesis of this disease (3, 4, 5) . HHV-8 has been implicated in the pathogenesis of other malignant disorders including KS, PEL, and multicentric Castleman’s disease (6, 7, 8) . Whereas in the latter three diseases the virus infects the malignant cells (6, 7, 8) , in MM HHV-8 is found in nonmalignant dendritic cells within the bone marrow stroma (3) .
However, some investigators have been unable to detect HHV-8 in MM (9, 10, 11) . Moreover, serological assays have largely failed to detect HHV-8 expression in patients with MM (12 , 13) , although Gao et al. (14) recently successfully identified antibodies against the HHV-8 recombinant minor capsid antigen in 81% of patients with MM. Previous studies have shown the presence of heterogeneity of HHV-8 sequences among specimens from different individuals harboring this virus (15 , 16) . We explored the possibility that nucleotide variations in the HHV-8 strain may be unique to myeloma patients. Such a finding could offer insights into the controversy regarding the association between MM and HHV-8 and possibly explain some of the difficulties in detecting serological responses in these patients.
In our study, two HHV-8 ORFs derived from KS, MM, PEL, and MGUS patients and from PEL cell lines were sequenced and compared with sequences published previously. ORF26 that encodes a minor capsid protein has been most frequently used to establish the presence of HHV-8. In addition, interpatient differences in this sequence have been identified (17) , and this enables investigators to both rule out PCR contamination and determine the viral strain of each HHV-8-infected sample. ORF65 encodes the small viral capsid antigen. The expressed protein is known to be immunogenic in humans and serves as a specific serological marker for the presence of HHV-8 in KS and PEL-infected individuals (18 , 19) . Thus, consistent changes in the sequence from ORF65 may produce alterations in the protein that may prevent serological detection of HHV-8 infection in myeloma patients.
MATERIALS AND METHODS
Twenty-four patients with active MM (19 American and 5 Turkish), 3 patients with MGUS, 3 patients with AIDS-associated KS, 2 patients with PEL, and 2 PEL cell lines were investigated. Bone marrow or peripheral blood specimens were collected in heparinized tubes after informed consent was obtained in accordance with the Human Subjects Protection Committee. Whole bone marrow or blood specimens were diluted 1:1 with Leibovitz L-15 medium (Life Technologies, Inc., Grand Island, NY) supplemented with fetal bovine serum (L-15/FBS; Life Technologies, Inc.). The mononuclear cell fraction was obtained by density sedimentation on Ficoll-Hypaque (Pharmacia, Uppsala, Sweden) at 600 x g for 20 min and washed twice in L-15/FBS media. KS-1 and BC-1 are both HHV-8-containing lymphoma cell lines derived from patients with PEL (20 , 21) . Genomic DNA from the KS-1 cell line served as a positive control. Informed consent was obtained from patients. Human experimentation guidelines of the United States Department of Health and Human Services and those of the authors’ institutions were followed in the conduct of clinical research.
BMSC Cultures.
Bone marrow mononuclear cells were used to establish long-term bone
marrow stromal cell cultures as follows. Bone marrow mononuclear cells
were suspended in 10 ml of Iscove’s modified Dulbecco’s medium
(Irvine Scientific, Santa Ana, CA), supplemented with 10% FCS (Gemini
Bio-Products, Calabasas, CA), 10% horse serum (Gemini Bio-Products),
200 mM L-glutamine (Gemini Bio-Products), 100
units/ml penicillin (Irvine Scientific), and 100 µg/ml streptomycin
(Irvine Scientific) in 75-cm flasks (Costar, Cambridge, MA). Cells were
incubated at 37°C in a CO2 incubator around
4–8 weeks until confluence.
Dendritic Cell Enrichment from PBMCs.
Dendritic cells were enriched from PBMCs based upon their expression of
CD68 and CD83. Our laboratory has demonstrated previously that MM
bone marrow stromal cells infected with HHV-8 express CD68 and CD83.
PBMCs were placed in PBS (Sigma Chemical Co., St. Louis, MO) containing
1% FBS and 0.2% sodium citrate (PBS-FBS) medium and incubated at
4°C for 30 min with anti-CD68 and anti-CD83 antibodies (Dako,
Carpenteria, CA and Coulter, Miami, FL, respectively) at a
concentration of 25 µg/ml. The cells were washed twice with PBS-FBS
to remove unbound antibody and incubated with Dynal
immunoglobulin-coated beads (Dynal, Oslo, Norway) at a ratio of 1
bead/100 cells at 4°C for 30 min with gentle agitation. The cells
were again washed twice with PBS-FBS to remove unbound beads and
centrifuged at 600 x g for 10 min. The cells were
transferred to 15-ml tubes and then placed on an electromagnet (Dynal)
for 2 min at room temperature to bind cells to the magnet. Unbound
cells were removed by aspiration. Bound cells were released from the
electromagnet and resuspended in PBS-FBS medium. This immunomagnetic
enrichment procedure increases the proportion of CD68 and/or
CD83-expressing cells in PBMCs by two logs as reported previously
(22)
.
MM Bone Marrow Core Biopsy and KS and PEL Tissue Specimens.
KS and PEL tissue specimens and bone marrow core biopsies obtained from
patients with active MM were frozen at -80°C for 4 h and then
crushed. DNA extraction was completed for each specimen using the Easy
DNA kit (Invitrogen, San Diego, CA).
PCR.
PCR was performed using 200 ng (30,000 cells) of genomic DNA
extracted with the Easy DNA kit (Invitrogen) from bone marrow core
biopsies, BMSCs, PBDCs isolated from patients with MM, tumor biopsies
from patients with KS or PEL, as well as from the KS-1 and BC-1
cell-lines. Forty-five cycles of PCR amplification were performed on
each sample in triplicate using three pairs of nonoverlapping, newly
synthesized HHV-8 primers derived from ORF26 and ORF65. PCR with
ß-actin primers was also performed on all specimens to evaluate
adequate isolation of genomic DNA. Genomic DNA from the KS-1 cell line
served as a positive control. Placental DNA (Sigma) served as a
negative control. The ORF26 primers, which produced a 233-bp product,
were 5'-AGCCGAAAGGATTCCACCAT-3' and 5'-TCCGTGTTGTCTACGTCCAG-3'. One set
of primers was used to isolate and sequence ORF65. This ORF65 primer,
5'-TCCACGGTTGTCCAATCGTT-3' and 5'-TGTCCAACTTTAAGGTGAGA-3', was used to
generate a 533-bp product. To further eliminate bias in our results,
the PCR assays were performed randomly and in a blinded fashion. PCR
products were electrophoresed in agarose gels impregnated with ethidium
bromide and photographed.
DNA Cloning and Sequencing.
Appropriately sized PCR products were cloned into the PCR2.1 vector
(Invitrogen) and sequenced with the Sequenase version 2.0 sequencing
kit (Amersham Life Science, Cleveland, OH). All variant bp positions
were confirmed by bidirectional sequencing. The nucleotide position
nomenclature used conforms to that introduced by Neipel et
al. (23)
, despite the fact that ORF65 is oriented in
the opposite direction compared with ORF26 in the viral genome.
Restriction Enzyme Digestion.
Appropriately sized PCR products were cloned into the PCR2.1 vector
(Invitrogen). PCR2.1 vector DNA containing a 233-bp segment of ORF26 or
a 533-bp segment of ORF65 was subjected to restriction enzyme digestion
using XbaI (5'-CCCGGG-3') and NarI
(5'-GGCGCC-3'), respectively (New England BioLabs, Beverly, MA) to
confirm the observed sequence variability.
RESULTS
ORF26.
Amplified products using HHV-8 ORF26 primers were obtained from DNA
extracted from 9 fresh bone marrow biopsies and 15 BMSCs from American
(n = 19) and Turkish (n = 5) patients
with active MM and 3 patients with MGUS. HHV-8 ORF26 DNA segments was
also amplified from the tumor of three KS patients with HIV, two PEL
patients, as well as from the cell lines KS-1 and BC-1. Representative
results of DNA PCR amplification of a 233-bp segment of ORF26 in
7 myeloma patients (even-numbered Lanes 2–14) and in the
KS-1 cell-line (Lane 17) are shown in Fig. 1
. Appropriately sized ß-actin
amplified product was also present in all samples (odd-numbered
Lanes 3–15). We have shown previously the lack of HHV-8 in bone
marrow biopsies in normal subjects.
|
summarizes the nucleotide sequence
variation among the different ORF26 fragments isolated from BC-1 and
KS-1, as well as tissue from patients with KS, MM, and MGUS. For
comparison, the ORF26 sequences in KS tissues and the BBG-1 PEL cell
line reported to GenBank are also included. All MGUS and MM-ORF26
segments (except 2 MM patients), regardless of geographic origin, had
identical polymorphisms that had been identified previously (15
, 16)
at nucleotides 47631, 47632, 47654, 47731, and 47738, when
compared with the original HHV-8 sequence described by Chang et
al. (7)
. These latter two MM sequences (m26-01 and
m26-02) were unchanged at position 47731 but contained the same
polymorphisms as the other MM cases at the other four respective
nucleotides. As a result, ORF26 sequences consistently encode for
isoleucine (47631, 47632) and glycine (47731) in MM patients, whereas
in KS, these positions encode for either proline or isoleucine (47631,
47632) and asparagine or glycine (47731). As documented for KS, a small
number of patient-specific, single-bp substitutions were noted among
individual myeloma ORF26 segments.
|
, consistent with previous studies showing the high
frequency of this strain among patients with AIDS-associated KS in the
United States (15
, 16)
. Thus, the ORF26 sequences in
myeloma patients that contain interpatient differences demonstrated a
consistent C3 subtype that has been observed in only a minority of
patients with KS or PEL. Although the thymine at 47632 has been
reported rarely in GenBank sequences (Fig. 2)
, it is commonly found in
sequences published previously from this ORF26 (15
, 16)
.
To confirm the frequent presence of the adenine to guanine substitution
at 47731 in the HHV-8 derived from MM patients, 7 myeloma bone marrow
samples were digested with the restriction enzyme XbaI (Fig. 3)
. The adenine-to-guanine substitution
creates a novel XbaI cleavage domain in ORF26, resulting in
a single DNA fragment upon digestion of the cloning vector with this
restriction endonuclease. The undigested ORF26-containing vectors
with an adenine at 47731 remain mostly in supercoiled form.
|
summarizes the nucleotide sequence data
of the 233-bp ORF26 segments isolated from these three tissues in four
patients with active MM. Despite interpatient differences, the ORF26
sequences were identical among the different tissues from each patient
except for one patient (patient m26-04) in which a single-bp
substitution was observed at position 47719 in a single sample (BMSCs).
Our ability to consistently detect virtually identical HHV-8 ORF26 DNA
sequences from separate tissue sources in each patient strongly argues
against mere culture artifact or PCR contamination as an explanation
for viral presence in these patients.
|
summarizes the
nucleotide sequence data of the ORF65 segments isolated from bone
marrow specimens from 16 patients with MM (12 American and 4 Turkish),
3 patients with AIDS-associated KS, 2 patients with lymphoma, and
from the KS-1 and BC-1 cell lines. Interestingly, most (14 of
16) myeloma sequences, regardless of geographic origin, consistently
revealed a single-bp deletion at position 112197. In contrast, none of
our PEL or KS samples or those reported previously in GenBank
(n = 10) have shown this deletion (Fig. 5)
. The
observed deletion was confirmed using two separate pairs of ORF65
primers and by restriction enzyme digestion using NarI. The
deletion at position 112197 creates an additional NarI
cleavage domain in ORF65, resulting in smaller DNA fragments upon
digestion with this restriction endonuclease (data not shown). Thus,
this deletion is present in 14 myeloma patients examined but is absent
in the lymphoma cell lines and fresh samples from patients with KS and
PEL. This myeloma-specific deletion in ORF65 results in a reading frame
shift that translates to the expression of a myeloma viral protein that
is putatively 32 amino acids longer than the protein encoded by the
viral sequences found in KS-1, BC-1, or KS patients (Fig. 6)
. The two other MM patients that lacked
this deletion instead revealed a single-bp substitution changing
cytosine to adenosine at position 112197, which has not been found in
other KS or PEL cases or cell lines. This substitution results in the
expression of the amino acid threonine instead of alanine. The large
structural change at this position is likely to lead to alterations in
the structure of the ORF65 protein.
|
|
DISCUSSION
Using direct nonnested PCR analysis, we have detected ORF26 and
ORF65 HHV-8 DNA fragments in patients with active MM, KS, PEL, and in
the PEL cell lines KS-1 and BC-1. As reported previously, HHV-8 DNA is
rarely detected in the bone marrow or blood of normal individuals (2 of
72 normal subjects screened to date harbor HHV-8) or patients with
malignancies other than PEL or KS (5
, 22)
. However, it was
found in the vast majority of patients with MM; and, moreover, 25%
of patients with MGUS also show viral presence. The latter observation
suggests the possibility that HHV-8 plays a causative role in the
transformation of MGUS to MM, which occurs in approximately one-fourth
of patients with MGUS (24)
.
HHV-8 is a gamma class herpesvirus, which also includes EBV and herpesvirus saimiri. After analyzing tissue samples from 12 patients with HHV-8-positive KS, Zong et al. (16) suggested that the HHV-8 genome could be divided into four distinct groups (A–D) based on polymorphic nucleotide positions within ORF26 and several other ORFs. These HHV-8 strains differed from each other at approximately 15–26 genomic positions, whereas within each subgroup only one to two positions vary. In herpesvirus saimiri, three different major viral types have also been identified (25) . Recent studies from Poole et al. (15) have revised the earlier classification system and further subtyped HHV-8. In the United States, the A1, A4, or C3 subtypes of HHV-8 predominate in AIDS-related KS, whereas the C2 variant predominates in classic KS. In the Middle East, the C2 subtype is also found among classic KS patients. In contrast, all of the myeloma patients whether from the United States or Turkey showed a C3 HHV-8 subtype. Although early studies suggested that specific strains (B and C) were more prevalent among patients with PEL (26) , more recent results have not confirmed these findings and suggest that the variability is attributable to geographic location. However, recent observations suggest that the three HHV-8 strains may, in fact, each possess different biological properties and cell tropism. For example, Friborg et al. (27) demonstrated that HHV-8 isolates from KS lesions possess different cytotoxic properties compared with lymphoma-derived HHV-8. Furthermore, Boralevi et al. (28) analysis of clinical data regarding the evolution of KS and HIV suggests that there is a clinical correlation between HHV-8 variants and aggressiveness of the tumor. Other parts of the HHV-8 genome, which to date have been less well characterized with regard to sequence analysis, may eventually explain these differences in the biology among different types of HHV-8. Certainly, strong evidence exists suggesting that certain EBV strains are preferentially associated with malignant disease (29) . Zong and Poole’s (15) proposed classification systems for HHV-8 are based solely on DNA variability. It remains to be seen whether this classification system translates into differential expression of distinctive latency genes as it does in EBV (30) or into differences in the ability to transform T cells, as it does in herpesvirus saimiri.
In the current report, we confirmed the presence of HHV-8 DNA
among patients with MM as reported previously by our group and others
(3, 4, 5)
. Of note, the HHV-8 strain that was identified in
13 MM patients, subtype C3, was the same strain as was detected in 3
patients with MGUS (Fig. 2)
. Sequence analysis of HHV-8 in KS suggested
a correlation between HHV-8 strain and geographic location (15
, 31)
. However, our evaluation of American and Turkish MM patients
suggests infection with mainly strain C3, regardless of geographic
origin. Certainly, further sequence analysis of these samples at
additional ORFs and evaluation of additional MM patients from other
geographic regions is necessary to fully evaluate which HHV-8 strains
are associated with MM.
Of interest, the HHV-8 sequences from patients with MM contained either a unique single-bp deletion or substitution at the 3'-end of the ORF65 coding region. This alteration has not been identified in other HHV-8-containing non-MM samples. Although random Taq incorporation errors can result in false sequence differences, it is highly unlikely that these random errors would result in a reproducible deletion or substitution at a single site as seen in ORF65 or in a disease-specific conserved sequence pattern as found in ORF26. In addition, the sequence we obtained for the BC-1 cell line completely matched the results published previously (6) . Furthermore, all PCR assays were performed in a random and blinded fashion, using three newly synthesized, nonoverlapping HHV-8 primers and at least two independently amplified PCR products were sequenced. Key sequences were also confirmed by restriction enzyme digestion. Last, despite interpatient differences, virtually identical ORF26 and ORF65 sequence patterns were identified in three different tissues from the same myeloma patient, making artifact unlikely.
The observed DNA variations in ORF26 and ORF65 may also be present in other segments of the HHV-8 genome. Conserved deletions can potentially lead to reading frameshifts as occurred in MM HHV-8 sequences at position 112197 in ORF65. Reading frameshifts invariably result in the expression of significantly altered proteins. The reading frameshift or the change from cytosine to adenosine at 112197 are both likely to change the tertiary structure of the expressed ORF65 protein. Our observations, therefore, lead us to speculate that available serological assays directed against epitopes expressed by the HHV-8 strain in KS may not recognize altered MM viral proteins. In support of this, two recent studies (32 , 33) have demonstrated that the epitope defined by HHV-8 serological tests is localized to the COOH-terminal end of the ORF65 protein. In fact, this epitope lies specifically in the portion of the molecule that is encoded by 112197, the site of the myeloma-specific alteration. In addition to ORF65, some serological HHV-8 assays are also directed against the latent nuclear antigen encoded by ORF73 (34) . Currently, part of our research effort is directed toward sequence analysis of the ORF73 from MM patients to assess whether myeloma-specific sequence variability is also present there. We are also expressing the putatively altered MM-specific ORF65 protein to develop antibodies selectively directed against the MM HHV-8 viral strain as well as synthesizing myeloma-specific ORF65 peptides for use in serological testing.
Although the consistent association between MM and HHV-8 implies a causal role in the pathogenesis of this disease, no cause-and-effect relationship has yet been demonstrated. Such evidence may be obtained more directly by using an animal model and indirectly through therapeutic interventions with antiviral agents or through extensive epidemiological studies. Such epidemiological studies would be greatly facilitated by the development of serological tests capable of consistently demonstrating exposure to HHV-8 among patients with MM. A direct or indirect causal effect of HHV-8 in the pathogenesis of MM has enormous implications for trying to use a therapeutic approach with antiviral agents and preventative strategies using vaccines. There was preliminary evidence that antiviral therapy in HIV-infected patients reduced the risk of development of KS (35) . More recently, a recent large randomized clinical trial study shows that the use of ganciclovir significantly reduces the development of KS among HIV-infected individuals with cytomegalovirus retinitis (36) . Clinical improvement for patients with KS treated with antiviral agents has also been reported (37) .
In conclusion, we showed a consistent HHV-8 subtype, C3, present in MM patients and demonstrated consistent myeloma-specific sequence variations at position 11227 in ORF65. These latter changes lead to alterations in the final expressed ORF65 in the part of the protein that is known to be responsible for a specific serological response to HHV-8 in patients with KS.
FOOTNOTES
The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked advertisement in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.
1 These studies were conducted with support from
the Lymphoma Research Foundation. 
2 To whom requests for reprints should be
addressed, at Myeloma and Bone Metastasis Programs, Cedars-Sinai
Medical Center, Beverly Modular-1, Room 120, 8700 Beverly Boulevard,
Los Angeles, CA 90048. Phone: (310) 423-5093; Fax: (310) 423-1977;
E-mail: berensonj{at}cshs.org
3 The abbreviations used are: MM, multiple
myeloma; HHV, human herpesvirus; KS, Kaposi’s sarcoma; PEL, primary
effusion lymphoma; ORF, open reading frame; BMSC, bone marrow stromal
cell; PBMC, peripheral blood mononuclear cell; PBDC, peripheral blood
dendritic cell; MGUS, monoclonal gammopathy of undetermined
significance.
Received 6/12/00; revised 9/ 1/00; accepted 9/12/00.
REFERENCES
This article has been cited by other articles:
|
|
 |
|
  A. M. Polstra, R. van den Burg, J. Goudsmit, and M. Cornelissen Human Herpesvirus 8 Load in Matched Serum and Plasma Samples of Patients with AIDS-Associated Kaposi's Sarcoma J. Clin. Microbiol., December 1, 2003; 41(12): 5488 - 5491. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
T. Endo, T. Miura, T. Koibuchi, H. Nakamura, T. Takahashi, T. Odawara, M. Goto, A. Ajisawa, A. Iwamoto, and T. Nakamura Molecular Analysis of Human Herpesvirus 8 by Using Single Nucleotide Polymorphisms in Open Reading Frame 26 J. Clin. Microbiol., June 1, 2003; 41(6): 2492 - 2497. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 C. Brander, N. Raje, P. G. O'Connor, F. Davies, J. Davis, D. Chauhan, T. Hideshima, J. Martin, D. Osmond, D. H. Kedes, et al. Absence of biologically important Kaposi sarcoma-associated herpesvirus gene products and virus-specific cellular immune responses in multiple myeloma Blood, June 28, 2002; 100(2): 698 - 700. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| Cancer Research | Clinical Cancer Research |
| Cancer Epidemiology Biomarkers & Prevention | Molecular Cancer Therapeutics |
| Molecular Cancer Research | Cancer Prevention Research |
| Cancer Prevention Journals Portal | Cancer Reviews Online |
| Annual Meeting Education Book | Meeting Abstracts Online |
