From www.bloodjournal.org by guest on June 17, 2017. For personal use only.
Establishment and Characterization of a Primary Effusion (Body Cavity Based) Lymphoma Cell Line (BC-3) Harboring Kaposi’s Sarcoma Associated Herpesvirus (KSHV/HHV-8) in the Absence
of Epstein-Barr Virus
By Leandros Arvanitakis, Enrique A. Mesri, Roland G. Nador, Jonathan W. Said, Adam S. Asch, Daniel M. Knowles,
and Ethel Cesarman
The recently identified Kaposi’s sarcoma-associated herpesvirus (KSHV), also known as human herpesvirus 8 (HHV-81,
has been found t o be consistently associated with an unusual subset of acquired immunodeficiency syndrome-related lymphomas, the so-called body cavity-based lymphomas (BCBL) or primary effusion lymphomas (PEL). These
lymphomas are characterized by a unique spectrum of morphologic and molecular characteristics, and grow as lymphomatous effusions without an identifiable contiguous tumor
mass. Until now, efforts t o delineate the role of KSHV in the
pathogenesis of PELs have been hampered by the lack of
appropriate model systems and the concomitant presence
of Epstein-Barr virus (EBV) in nearly all cases examined, and
in all previously established cell lines. We now report the
establishment and characterization of a novel PEL cell line,
BC-3, which is KSHV+ by polymerase chain reaction (PCR)
but EBV- as assessed by a variety of methods including PCR
for EBER, EBNA-2, and EBNA-3C. This cell line was established from a lymphomatous effusion obtained from an HIVpatient, and has immunophenotypic and molecular features
consistent with the diagnosis of PEL, including an indeterminate immunophenotype with a B-cell immunogenotype and
lack of c-myc proto-oncogene rearrangements. Pulsed-field
gel electrophoresis shows an intact KSHV genome of about
170 kb both in the cell line and in the viral isolate, whereas
herpesvirus-like capsids are visible by electron microscopy.
Consequently, the BC-3 cell line represents an invaluable
tool as a source of KSHV, for both the evaluation of the
pathogenic potential of this virus and the mechanistic characterization of its role in the development of Kaposi‘s sarcoma and malignant lymphoma.
Q 1996 by The American Society of Hematology.
R
restricted antigens. At the molecular level, they are characterized by a B-cell genotype as determined by clonal Ig gene
rearrangements, the presence of Epstein-Barr virus (EBV)
and the lack of c-myc gene rearrangement^.^,'
Extensive sequence analyses of fragments of KSHV have
shown considerable homology with Herpesvirus saimiri
(HVS) and EBV.’.’ Because these two viruses belong to the
gammaherpesvirus subfamily, which is characterized by its
propensity to infect and transform lymphoblastoid cells,8,’ it
would appear reasonable to assume that KSHV may also be
a transforming agent.
In an attempt to develop in vitro model system suitable for
the isolation and characterization of KSHV, our laboratory
recently established two PEL cell lines, namely BC-I and
BC-2.“’ Although very useful for several initial studies, these
cell lines, as well as nearly all KSHV’ lymphomas described
to date, fall short of the desired model system because of
their concomitant infection with EBV. Here we report the
establishment and characterization of a PEL cell line that
harbors KSHV in the absence of EBV and releases viral
particles containing the KSHV genome. We believe that this
cell line will serve as an invaluable reagent for the characterization of the properties and functions of this novel, infectious agent.
ECENT STUDIES have shown the presence of DNA
sequences belonging to a novel herpesvirus in
Kaposi’s sarcoma (KS) tissues.’ This virus has been called
KS-associated herpesvirus (KSHV) and is provisionally designated human herpesvirus 8 (HHV-8). Although KSHV
does not appear to be present in control tissues from nonacquired immunodeficiency syndrome (AIDS) patients’.2 or
in the peripheral blood of normal blood donors,’ it has also
been detected in two unique and distinctive categories of
lymphoid proliferations, namely the primary effusion (body
cavity -based) lymphomas (PEL)4 and multicentric Castleman’s disease.‘ The PELs are a distinct group of B-cell
non-Hodgkin lymphomas presenting with a unique spectrum
of clinical, morphologic, immunophenotypic, and molecular
genetic characteristics that distinguish them from most
AIDS-related lymphomas. More specifically, PELs tend to
present in the pleural, pericardial, and/or abdominal cavities
as lymphomatous effusions, usually in the absence of any
identifiable tumor mass throughout the clinical course. They
have an unusual immunophenotype in that they commonly
express CD4S in the absence of other B- or T-cell lineageFrom the Departments of Patholog? and Medicine. The New York
Hospitul-Cornell Medical Center, New York, NY; and the Department of Pathology, Cedurs-Sinai Medical Cenfer, Los Angeles. C A
Submitted Februap 28, 1996; accepted May 21, 1996.
Supported in purt by National Institutes of Health Grants No.
AI39192 to E.A.M.. CA42710 to J.W.S., EY06337 to D.M.K., und
CA68939 to E.C.
Address reprint requests to Ethel Cesarman, MD, PhD, Department ($Pathology. The New York Hospital-Cornel1 Medical Center,
-525 E 68th St, New York, NY 10021.
The publicution costs qf this article were defrayed in part by page
charge payment. This article must therefore be hereby marked
“advertisement” in accordance with 18 U.S.C. section 1734 solely to
indicate this fitct.
0 1996 by The American Sociev of Hematrilogy.
0006-497//96/8807-0022$3.00/0
2648
MATERIALS AND METHODS
Pathologic samples, cell line establishment, and cell culture.
The BC-3 cell line was established from a lymphomatous pleural
effusion sample collected from a human immunodeficiency virusnegative (HIV-) patient previously reported in the literature.”
Mononuclear cells were isolated from the effusion samples by FicollHypaque (Pharmacia Fine Chemicals, Piscataway, NJ) density gradient centrifugation. Viable cells were plated at a concentration of 2
x IO’/mL onto a layer of heterologous feeder cells as previously
described.“’ Cells were maintained in W M I 1640 supplemented with
2 mmol/L glutamine and 20% fetal bovine semm at 37°C in the
presence of 5% CO,. After approximately 4 weeks in culture, the
feeder cells progressively disappeared compared with control plates,
Blood, Vol 88, No 7 (October l ) , 1996: pp 2648-2654
From www.bloodjournal.org by guest on June 17, 2017. For personal use only.
KSHV’, EBV- LYMPHOMA CELL LINE
2649
CGCCACTCTATATGCAAACTG-3‘. A primer set specific for the
giving way to a proliferating cell population. The cells from this
human P-actin cDNA (Stratagene) was used as a quantitative control.
well were progressively expanded, passaged every 3 to 4 days, and
Because this primer set spans an intron, it can only amplify seseeded at 0.3 to 0.5 X 106/mL(cell viability was evaluated by Trypan
quences that have derived from reverse-transcribed RNA and not
blue exclusion). BC-1 and BC-2 cells, which are PEL cell lines
from contaminating DNA, thus serving as an ideal positive control.
previously established in our laboratory,” were used as controls for
The sequence of the P-actin internal oligonucleotide probe was: 5’the presence of KSHV and EBV genomes in the present studies.
GGATGTCCACGTCACACC-3‘. PCR reaction products (5-pL
lmnaunophenofypic analyses. The immunophenotype of the PEL
aliquots) were fractionated by electrophoresis and transferred to niand the BC-3 cell line were determined by direct and indirect immutrocellulose filters as described above for genomic DNA Southerns.
nofluorescent flow cytometry using a FACScan fluorescent activated
Filters were hybridized with oligonucleotide probes for KSHV and
cell sorter (Becton Dickinson Immunocytometry Systems, Mountain
EBV that were end-labeled with y-’*P(dATP) using T4 PolynucleoView, CA) as previously described.612.”The monoclonal antibodies
tide Kinase (Boehringer Mannheim). The similar melting tempera(MoAbs) used included BerH2 (CD30; Dako Corp, Santa Barbara,
tures of the EBV oligonucleotide probes allowed their simultaneous
CA), B4 (CD19), B1 (CD20), B2 (CD21), CALLA (CDlO; Coulter
hybridization to the nitrocellulose filter. Furthermore, direct sequencImmunology, Hialeah, EX), Leu1 (CD5), Leu14 (CD22), Leu20
ing of PCR products from the KS330 Bum region and flanking
(CD23), Leu9 (CD7), LeuM3 (CD14), Ia (HLA-DR), T4 (CD4), T8
sequences for the purpose of verification was performed as pre(CDS), My10 (CD34), T11 (CD2), IL2R (CD24), ICAM (CD54).
viously described. lo
G28-5 (CD40 Becton Dickinson), T3 (CD3; United Biomedical Inc,
Pulsed-field gel electrophoresis (PFGE). Cells (25 X 106/mL
Hauppauge, NY) and T9 (CD71), T10 (CD38; Ortho, Raritan, NJ).
final), viral pellets from culture supernatants or cellular fractions
Antisera to total Ig, kappa ( K ) and lambda (A) Ig light chains were
(see below) were molded into agarose plugs using 1% low-melting
obtained from Dako.
point (LMP) agarose (Bio-Rad, Richmond, CA) and 0.9% NaCI.
DNA extraction, Southern blot analyses. Genomic DNA was
Plugs were incubated 2 X 24 hours at 50°C in lysis buffer consisting
extracted by a salting-out procedure as previously described.I4Aliof 0.5 m o m EDTA pH 8.0, 1% Sarkosyl, 1 mg/mL Proteinase K
quots (5 pg) of genomic DNA were digested with appropriate
restriction endonucleases as per the manufacturer’s instructions
(GIBCO-BRL), washed, and stored in 0.5 mol/L EDTA at 4°C until
ready for use. PFGE was performed using a CHEF-DR I1 System
(Boehringer Mannheim, Indianapolis, IN), size-fractionated by electrophoresis in 0.8% agarose gels, denatured (1.5 mom NaCI, 0.5
(Bio-Rad) as per the manufacturer’s instructions. Samples were run
on 1% PFGE-certified agarose (Bio-Rad) gels in 0.5 X Tris borate
molL NaOH), neutralized (0.5 m o m Tris pH 7.4, 3 m o m NaCI),
EDTA (TBE) at 200 V, ramped from 3 to 30 seconds for 23 hours
and transferred to nitrocellulose filters as per Southem.” Filters were
at 1 4 T , with constant buffer recirculation using a peristaltic pump.
then hybridized appropriately with a-’*P(dCTP) random prime-laMolecular-weight marker plugs were obtained commercially (PFGE
beled (PrimeIt 11; Stratagene, La Jolla, CA) DNA probes for the Ig
Marker I-X Ladder; Boehringer Mannheim). Gels were transferred
heavy-chain joining region (JH), the T-cell receptor p-chain, the cto nitrocellulose filters and probed successively with radiolabeled
myc oncogene, KSHV (KS330Bam and KS631Bam fragments),’ or
DNA probes for KSHV and EBV as described above.
EBV (terminal repeat region)16 as described previously.”
Polymerase chain reaction (PCR) conditions, oligonucleotide primViral isolation and negative staining electron microscopy. Viral
pellets were prepared as previously described.*’” Briefly, BC-1 or
ers, and probes. Total RNA was isolated using the Tri-Reagent nucleic acid extraction method (Molecular Research Center Inc, CincinBC-3 cells were collected by centrifugation for IO minutes at 1,500
nati, OH) according to the manufacturer’s instructions. To eliminate
rpm, resuspended in 1:20 vol of culture medium, and snap-frozen
(cell virus extract). Supernatant virus was obtained by centrifugation
contaminating genomic DNA, the RNA samples were treated with 2
U RNase-& DNase1 (Boehringer Mannheim), with subsequent heatof a 1,400g-cleared conditioned culture medium for 2 hours at
inactivation of the enzyme. RNA-based PCR was performed on 1 pg
23,OOOg at 4°C. Pellets were then resuspended in 1:50 vol of culture
RNA using the Superscript Reverse Transcriptase System (GIBCOmedium and snap-frozen. Both preparations (cell or supernatant viBRL, Grand Island, NY) according to the manufacturer’s instructions.
rus) were then thawed, sonicated in a cup sonicator (Heat Systems,
PCR reactions were also performed on RNA samples without the reFramingdale, NY), cleared by centrifugation at 10,OOOg at 4”C,
verse transcription step to control for amplification due to genomic
loaded on a phosphate-buffered saline (PBS)-25% sucrose cushion,
DNA contamination. PCR conditions, and sequences of oligonucleoand pelleted by overnight centrifugation at 70,000g at 4°C. Finally,
tides used for the amplification and hybridization of both the KSHV
viral pellets were resuspended in PBS and either embedded in agarKS330233 E m fragment and its flanking sequences, were as originally
ose for pulsed-field electrophoresis or adsorbed onto parlodiodcarreported.‘*’’Primers and probes for EBV included sets for the EBNAbon-coated grids using the agar diffusion method, stained with 2%
2, EBNA-3C, and EBER regions, the first two allowing the accurate
phosphotungstic acid (pH 6.8) and examined with a JEOL electron
distinction of both types A and B EBV.” Oligo sequences and PCR
microscope (JEOL, Peabody, MA).
conditions for these EBV regions were as previously published.” All
oligonucleotides used in these studies as PCR primers or probes were
RESULTS
obtained Commercially (GIBCO-BRL). PCR primers and probes for
the KSHV open reading frames or 0RFs2O were as follows: (1)
Morphologic and imnunophenotypic characterization.
for ORF 75 (membrane antigen homolog), 5’ primer: 5’-AGGThe BC-3 cells were large and contained abundant basophilic
AGCGAGAGAGACGGGAT-3’, 3’ primer: S-CCAGGTGCCTGCcytoplasm, occasionally with a paranuclear hof, and large
CCACTTCC-3‘ and probe: 5’-CCTAGCTCTTGCAGCAGAAC-3’;
round to polygonal and sometimes pleomorphic nuclei with
(2) for ORF 74 (G-protein coupled receptor homolog), 5’ primer:
one or more prominent nucleoli. These morphologic features
5‘-CGGGGTGCCTTACACGTGG-3‘,
3’ primer: 5’-CAGTCTare consistent with their classification as a PEL. This classiGCAGTCATGTITCC-3’ and probe: 5’-TGTGTGCGTCAGTCTfication is further consolidated by immunophenotypic analyAGTGAG-3’; (3) for ORF 73, 5‘ primer: S-GCAGTCTCCAGAGTCTTCTC-3‘, 3’ primer: 5’-CGGAGCTAAAGAGTCTGGTG-3’ ses that showed that BC-3 cells express CD45, but do not
express B-cell lineage-restricted antigens (CD 19, CD20,
and probe: 5‘-TGGAGGTGTAGTCTGCTGCG-3’;
(4) for ORF 72
CD21, CD22, K or X Ig light chains) or T-cell lineage-re(cyclin D homolog), 5’ primer: 5‘-CACCCTGAAACTCCAGGC3’, 3’ primer: 5‘-GATCCGATCCTCACATAGCG-3’
and probe: 5’stricted antigens (CD2, CD3, CD4, CD5, CD8). There was
From www.bloodjournal.org by guest on June 17, 2017. For personal use only.
ARVANITAKIS ET AL
2650
17kb-
-
12.8 kb
11.6 kbs
12 kb
11 kb-
4.2 kb
ENZYME
PROBE
-- -EcoRi
Hindlll
EcoRl
JH
f3amHl
EcoRl
TP
Hindlll
c-myc
Fig 1. Southern blot hybridization analyses for lg heavy chain, T-cell receptor P-chain, and c-myc proto-oncogenerearrangements. (AI Five
micrograms of DNA from HL60 (germline control) and BC-3 cells was digested with EcoRl or Hidlll and hybridized with an lg heavy-chain
joining region (J,) probe. The HL60 lane shows a germline band at 17 kb for the EcoRl and at 11 kb for the Hindlll digests, respectively,
whereas the BC-3 lanes show clonal rearrangement bands. (B) Five micrograms of DNA from HL60 and BC-3 cells was digested with EcoRl or
6amHl and hybridized with a T-cell receptor P-chain probe. Under both digestion protocols, the BC-3 cells display a germline configuration
for the T,, chain, compared with the HL60 control. (C) Five micrograms of DNA from HL60 and BC-3 cells was digested with EcoRl or Hindlll
and hybridized with a probe for c-myc. Although HL60 cells have an amplified c-myc gene, both digestion protocols show that BC-3 cells
possess a germline configurationfor this proto-oncogene.
variable expression of activation antigens, with expression
of CD30. CD38. CD54, CD71, and HLA-DR.
Genoppic characterization. To determine the lineage
and clonality of these cells, we investigated several genomic
markers by Southern hybridization. DNA from HL60 cells
was used as a negative control for gene rearrangement, based
on previous knowledge that this cell line has a germline
configuration of the genes examined by the JH, To, and cm y probes, even though the c - m y gene is amplified and
gives a stronger hybridization signal. Restriction endonuclease digestion of genomic DNA and subsequent hybridization
to appropriate radiolabeled probes showed that BC-3 cells
have rearranged Ig heavy chain (Fig I), as well as K and A
light-chain genes (data not shown) and retain the germline
configuration for the To genes (Fig I), indicating a clonal
B-cell population. The c - m y proto-oncogene is also in its
germline configuration, further consolidating classification
as a PEL (Fig I). Direct comparison of Southern hybridization analyses for Ig genes between DNA from the original
tumor samples and DNA from the BC-3 cell line corroborated the tumor derivation of this cell line (data not shown).
Presence of virnl sequences in BC-3 cells. PCR analyses
on genomic DNA from the original tumor sample suggested
the presence of KSHV sequences in the absence of EBV and
HIV." Preliminary Southem blot hybridization analyses on genomic DNA from both the original tumor and the BC-3 cells
also suggested the presence of KSHV in the absence of EBV
sequences (data not shown). In addition, in situ hybridization
for EBER was negative." To further evaluate the viral genomes
present in the BC-3 cells, we examined the presence of KSHV
and EBV sequences in these cells by PCR. We used a panel
of three EBV genes, namely EBNA-X, EBNA-2, and EBER.
to detect both type A and B forms of EBV and to exclude the
possibility of false negatives arising as a result of the presence
of mutant EBV genomes with deleted sequences. As controls,
we used two other PEL cell lines previously established in our
laboratory, namely BC-I and BC-2, both of which are known
to be positive for KSHV, and EBV types B and A, respectively.'" PCR analyses showed that BC-3 cells, like BC-I and
BC-2, are positive for KSHV as judged by the presence of
sequences from the KSHV330 fragment. However, and in contrast to BC-I and BC-2 cells, BC-3 cells are negative for EBV
as shown by the absence of hybridization signals with the
EBNA-3C, EBNA-2. and EBER probes (Fig 2). To confirm
the KSHV origin of these amplification products, we sequenced
a I , I IO-bp region containing the originally described 233-bp
fragment as well as 492 bp upstream and 385 bp downstream
flanking sequences.'.'" The sequences of the amplified fragments were identical to the previously reported KSHV sequence derived from a KS specimen,' except for nine altered
bases at positions 683 (T C), 686 (T + C), 981 (T C),
1032 (C A), 1033 (C T), 1055 (G T), 1132 (A G),
1139 (A + C), and IS14 (G -+ A), which fall within the
variation expected among different isolates. BC-3 cells are also
negative for other herpesviruses such as herpes simplex virus
types I and 2 (HSV-I and HSV-2) and cytomegalovirus(CMV)
as determined by immunohistochemistry analyses.'3
Expression of KSHV sequences in BC-3 cells. In an attempt to characterize the biologic significance of the presence of KSHV sequences in the BC-3 cell line, we used an
RNA-based PCR assay to analyze the expression of four
KSHV ORFs previously identified in the region surrounding
the originally reported KS631 Bam fragment.' These ORFs
include a cyclin D homolog, a G-protein coupled receptor
homolog, a homolog of HVS ORF 75 which is thought to
encode a tegument protein, and the positional homolog of
+
+
+
+
+
+
From www.bloodjournal.org by guest on June 17, 2017. For personal use only.
2651
KSHV', EBV- LYMPHOMA CELL LINE
A KSHV
Fig 2. PCR analysesfor KSHV and EBV sequences
in the BC-3 cell line. (A) PCR was performed as described in the Materials and Methods, and amplified
fragments were transferred t o nitrocellulose filters
and hybridized with an internal oligonucleotide
probe for the 233-bp KSHV330 fragment. BC-3 cells
are shown t o be positive for the KSHV330 sequence
as judged by comparing t o the BC-1 and BC-2 cell
lines (positive controls). (BI PCR for the EBNA-3C.
EBER, and EBNA-2 genes of EBV detected their presence in the BC-1 and BC-2 cell lines but not in the
BC-3 cell line. Different primer sets allowed the discrimination between the EBNAJC genes in type B
EBV (246-bp fragment) and type A EBV (153-bp f r a g
ment) present in the BC-1 and BC-2 cell lines, respectiveIy
233bp-
-*
KS330233
B EBV
246 bp153 bp-
.
HVS ORF 73 whose function is unknown.20All four ORFs
were found to be expressed as determined by RNA-based
PCR analyses (Fig 3), suggesting that these KSHV sequences
are part of a functional viral genome rather than spurious
ORF 75
-217 bp
GCR
r---1 -475 bp
ORF 73
-169 bp
CYCLIN
- 106 bp
PACTIN
-661 bp
Fig 3. RNA-based PCR analyses for KSHV sequence transcripts in
the BC-3 cell line. One microgram of total RNA extracted from BC-1,
BC-2, and BC-3 cell lines was treated with RNase-free DNase I,reverse
transcribed, and amplified by PCR. All four KSHV ORFs tested were
found t o be transcribed in the BC-3 cells compared with the BC-1
and BC-2 cells which were used as positive controls. The KSHV- HL60
cell line did not exhibit any KSHV ORF transcription, despite the
presence of amplifiable RNA as determined by the presence of pactin signal.
DNA sequences. PCR on RNA samples lacking the reverse
transcription step failed to yield any amplification products,
thus confirming that the observed fragments did not arise
from genomic DNA contamination (data not shown).
Characterization of KSHV particles in BC-3 cells. To
show the presence of KSHV particles in the BC-3 cells, we
analyzed the genomic DNA content by PFGE. As before,
the BC-I and BC-2 cell lines provided useful controls for
the presence of KSHV and EBV genomes. As shown in Fig
4B, the BC-I cells when hybridized with a KSHV probe
In
show a band at about 270 kb, as previously
contrast, the BC-2 and BC-3 cells show bands of approximately 200 and 170 kb, respectively. This difference is
highly reproducible, and does not stem from technical or
experimental variables. Viral pellets obtained from BC-3
cellular or supernatant fractions show a band of 170 kb, at
the same position as that of the bands in the BC-3 cells (Fig
4C). As expected, a band at the appropriate molecular weight
was identified with an EBV probe in the P3HRI and Raji
control cell lines and in the BC-I and BC-2 cell lines, but
was absent from the BC-3 cell line (Fig 4A).
Previous attempts to directly visualize KSHV particles
have been obscured by the concomitant presence of EBV
in all biologic samples and established cell lines thus far
examined. However, the BC-3 cell line is ideally suited for
the examination of KSHV particles at the electron microscopic level. Negative staining electron microscopy showed
the presence of viral capsids of approximately 100 to 150
nm diameter (Fig 5). Although the two capsids shown are
collapsed, ring-shaped capsomers of approximately 9-nm diameter arranged in linear arrays are discernible. Some of
these capsomers clearly show up as hexons whereas others
in the periphery appear as hollow tubes giving a castellated
edge. These features are characteristic of herpesviruses and
herpesviral preparation^.^^ Taken together with the PFGE
analyses, these results indicate that BC-3 cells are infected
with KSHV and release KSHV particles.
DISCUSSION
In this report we describe the establishment and characterization of a cell line, BC-3, which is shown by immunophe-
From www.bloodjournal.org by guest on June 17, 2017. For personal use only.
ARVANITAKIS ET AL
2652
300-
m-
250-
300200150
100
-
50-
4
!
PROBE:
'
It
250
-
-
200-
100-
150-
50-
100-
KSHV
EBV
KSHV
Fig 4. PFGE analyses of the KSHV and EBV genomes in the BC-3 cell line. Intact cells (25 x 10e/mL final concentration) and viral pellets
from cellular or supernatant fractions were molded into agarose plugs, separated by PFGE, and probed for EBV and KSHV sequences. (A)
Southern hybridization analysis of a representative experiment showing the migration of the EBV genome as determined by probing with a
DNA insert for EBV terminal repeat sequences. The EBV genome is detectable in the P3HR1, Raji, BC-1, and BC-2 cells, but not in the BC-3
cells, as a band of 170-kb molecular weight. The presence of EBV in BC-2 cells was not as strong as in the other cell lines but was evident
after longer exposure. (E) The nitrocellulose filter depicted in (A) is shown after hybridization with the KS33013am fragment. KSHV particles
of variable yet reproducible sizes were detected in the BC-1, BC-2, and BC-3 cell lines. The KSHV probe does not cross-hybridize with EBV as
shown by the absence of signal in the P3HRl and Raji lanes. (C) Viral pellets collected from culture supernatants (sup), cellular fractions (cell),
and intact BC-3 cells all display a KSHV genome with a molecular weight of approximately 170 kb.
notypic and genotypic analyses to belong to a recently recognized rare group of malignancies termed PEL or primary
effusion lymphomas. BC-3 cells are of B-cell origin as
judged by their clonal Ig heavy- and light-chain gene rear-
'
O.lpm
'
Fig 5. Electron microscopy of the viral pellets from the BC-3 cell
line. Viral pellets were prepared as described in Materials and Methods, and visualized by negative staining using a JEOL electron microscope. In the upper particle, ring-shaped capsomers are visibly arranged in linear arrays. In the upper edge of the particle, they appear
as hollow tubes giving a castellated periphery.
rangements and germline T-cell receptor genes. In line with
all previously published cases of PEL, the BC-3 cells are
positive for KSHV sequences; however, the novelty of this
cell line lies in the absence of EBV coinfection as determined
by Southern hybridization, in situ hybridization, PCR, and
PFGE analyses. This cell line consequently provides us an
opportunity to isolate, characterize, and functionally dissect
the recently discovered KSHV.
The first report of the KSHV sequences as detected in
KS tissues' raised questions regarding the nature of these
sequences and the evidence for a bona fide viral particle.
Sequencing and formal phylogenetic analyses of large portions of the KSHV genome have demonstrated that it is a
7-2 herpesvirus (genus Rhadinovirus) and the first member
of this group to infect humans.' Furthermore, with the establishment of the BC-1 and BC-2 cell lines in our laboratory,"
it became possible to visualize the KSHV DNA by PFGE
analyses as well as to perform transmissibility studies to
prove that these sequences belong to a functional virus capable of infecting cells and replicating in them.'.'' According
to these studies, the KSHV genome in the BC-1 cell line
has a molecular weight of approximately 270 kb. The data
presented in this report confirm these findings, yet also show
KSHV particles of lower molecular weight in the BC-2 and
BC-3 cell lines (Fig 4). The molecular weight of the KSHV
genomes in the BC-3 cells and viral isolates is approximately
170 kb, and this is a highly reproducible finding not attributable to technical or experimental variations. Because the
viral preparations were extensively sonicated before PFGE
From www.bloodjournal.org by guest on June 17, 2017. For personal use only.
2653
KSHV+, EBV- LYMPHOMA CELL LINE
and Southem analyses, the fact that we obtained sharp bands
corresponding to intact genomes for BC-1” and BC-3 viral
isolates is an indication that these genomes are protected
from sonication, and therefore evidence for their encapsidation. Moreover, KSHV DNA fragments are protected from
DNase degradation, further suggesting that these viral isolates contain encapsidated genomesZZ(E.A.M., unpublished
data, 1995). One possible explanation for the observed differences could involve dimerization or other rearrangement
events of the viral genome. It is also tempting to speculate
that the KSHV genome present in BC-3 cells may be a
deletion mutant of the larger KSHV genomes found in BC1 cells, in a manner analogous to the various known deletion
mutants of EBV. It is interesting to note that functional
differences may accompany these differences in genomic
size. Viral isolates obtained from the BC-I cell lines are
capable of infecting umbilical cord blood B cells. Furthermore, KSHV transmission is blocked by UV-irradiation and
foscarnet (an inhibitor of viral DNA-polymerase), thus providing evidence for the presence of a biologically active
virus.’’ Similarly, viral isolates from BC-3 cells appear to
infect cell lines of the B-cell lineage (E.A.M., unpublished
data, 1995). Studies to characterize the molecular and functional differences between KSHV derived from the BC- 1
and BC-3 cell lines are currently under way in our laboratory.
Although the precise explanation and significance of the
differences in genomic size among various KSHV isolates
can only be addressed when a physical map of the virus
becomes available, these particles do correspond to KSHV
based on the presence and sequence analyses of the KS330
Bum fragment, which is thus far the single available criterion
for detecting the presence of this virus. Further evidence that
the sequences present in BC-3 cells correspond to KSHV is
provided by our RNA-based PCR assays, which detected
and amplified fragments from four different ORFs derived
from a region very distant from the KS330 Bum fragment.
Expression of these four ORFs further suggests that this
virus is functional and has an active biologic role in these
PEL cells. The homologies of these ORFs to genes involved
in cell cycle regulationm raise interesting questions as to
their potential role(s) in viral transformation and/or pathogenesis. For example, ORF 72, a cyclin D homologue, may
be involved in the subversion of the host cell cycle to the
advantage of viral replication. In this respect, HVS contains
an ORF that encodes a viral cyclin D homologue capable of
interacting with host-cell cyclin-dependent kinases and thus
potentially playing a role in the oncogenic transformation
pro~ess.’~
Although EBV lacks a cyclin D homologue, its
latent membrane protein-1 (LMP-1) has been shown to
upregulate the expression of cellular cyclin D2 leading to
the loss of transforming growth factor-pl -mediated growth
inhibition and perhaps contributing to cellular transformation.26We are currently investigating the functional characteristics of these O m s with a view to assessing their role
(if any) in KSHV-mediated transformation events.
Finally, electron microscopic examination of the viral
preparations showed the presence of viral particles having
the morphologic features typical of herpesviral capsids (Fig
5), and particularly those of gammaherpes~iruses,~~
sug-
gesting that they may correspond to KSHV. These results are
in agreement with transmission electron microscopy studies
performed on KSHV+ cells obtained from the same patient
as the BC-3 cells but after passage in BNX mice.23Besides
being an invaluable tool for the isolation and characterization
of KSHV, the BC-3 cell line provides us an opportunity to
investigate the pathogenesis of PEL. An expanded cytogenetic and immunophenotypic analysis of these cells may
yield clues as to the nature of the B lymphocyte which is
infected by KSHV and perhaps point to the cell surface
molecule(s) acting as the viral receptors. Studies involving
the administration of BC-3 cells to immunodeficient (nude)
mice are currently under way in our laboratory, in an effort
to determine whether KSHV dictates the development and
characteristics of these rare lymphomas. In summary, BC-3
cells will serve as a tool which will allow us to address the
more significant questions regarding the role(s) of KSHV in
the pathogenesis of PEL, KS, and multicentric Castleman’s
disease.
ACKNOWLEDGMENT
We thank Dr Amy Chadbum for assistance with the immunophenotypic characterization of the BC-3 cell line, Dr Yi-Fang Liu for
excellent technical assistance with the Southern blot analyses, and
Bianca Santomasso for the work with the viral preparations.
REFERENCES
1. Chang Y, Cesarman E, Pessin MS, Lee F, Culpepper J,
Knowles DM, Moore PS: Identification of herpesvirus-like DNA
sequences in AIDS-associated Kaposi’s sarcoma. Science 266: 1865,
1994
2. Moore PS, Chang Y: Detection of herpesvirus-like DNA sequences in Kaposi’s sarcoma in patients with and those without HIV
infection. N Engl J Med 332:1181, 1995
3. Whitby D, Howard MR, Tenant-Flowers M, Brink NS, Copas
A, Boshoff C , Hatzioannou T, Suggett FEA, Aldam DM, Denton
AS, Miller RF, Weller IVD, Weiss RA, Tedder RS, Schulz TF:
Detection of Kaposi sarcoma associated herpesvirus in peripheral
blood of HIV-infected individuals and progression to Kaposi’s sarcoma. Lancet 346:799, 1995
4. Cesarman E, Chang Y, Moore PS, Said JW, Knowles DM:
Kaposi’s sarcoma-associated herpesvirus-like DNA sequences in
AIDS-related body cavity-based lymphomas. N Engl J Med
332: 1186, 1995
5. Soulier J, Grollet L, Oksenhendler E, Cacoub P, Cazals-Hatem
D, Babinet P, D’Agay MF, Clauvel JP, Raphael M, Degos L, Sigaux
F: Kaposi’s sarcoma-associated herpesvirus-like DNA sequences in
multicentric Castleman’s disease. Blood 86: 1276, 1995
6. Knowles DM, Inghirami G, LJbriaco A, Dalla-Favera R: 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 73:792, 1989
7. Moore PS, Gao SJ, Dominguez G, Cesarman E, Lungu 0,
Knowles DM, Garber R, Pellett PE, McGeoch DJ, Chang Y: Primary
characterization of a herpesvirus agent associated with Kaposi’s sarcoma. J Virol 70549, 1996
8. Liebowitz D, Kieff E: Epstein-Barr virus, in Roizman B, Whitley RJ, Lopez C (eds): The Human Herpesviruses. New York, NY,
Raven, 1993, p 107
9. Fleckenstein B, Desrosiers RC: Herpesvirus saimiri and herpesvirus ateles, in Roizman B (ed): The Herpesviruses. New York,
NY, Plenum, 1982, p 253
From www.bloodjournal.org by guest on June 17, 2017. For personal use only.
2654
IO. Cesarman E, Moore PS, Rao PH, Inghirami G, Knowles DM,
Chang Y: In vitro establishment and characterization of two acquired
immunodeficiency syndrome-related lymphoma cell lines (BC- 1 and
BC-2) containing Kaposi’s sarcoma-associated herpesvirus-like
(KSHV) DNA sequences. Blood 862708, 1995
I I . Nador RG, Cesarman E, Knowles DM, Said JW: Herpeslike DNA sequences in a body cavity-based lymphoma in an HIVnegative patient. N Engl J Med 333:943, 1995
12. Walts AE, Shintaku IP, Said JW: Diagnosis of malignant
lymphoma in effusions from patients with AIDS by gene rearrangement. Am J Clin Pathol 94:170, 1990
13. Chadburn A, Cesarman E, Jagidar J , Subar M, Mir RN,
Knowles DM: CD30 (Ki-I) positive anaplastic large cell lymphomas
in individuals infected with the human immunodeficiency virus. Cancer 72:3078, 1993
14. Miller SA, Dykes DD, Polesky HF: A simple salting out
procedure for extracting DNA from human nucleated cells. Nucleic
Acids Res 16:I2 15, 1988
IS. Southem EM: Detection of specific sequences among DNA
fragments separated by gel electrophoresis. J Mol Biol 98503, 1975
16. Raab-Traub N, Flynn K: The structure of the termini of the
Epstein-Barr virus as a marker of clonal cellular proliferation. Cell
47:883, 1986
17. Pelicci PG, Knowles DM, Dalla-Fdvera R: Lymphoid tumors
displaying rearrangements of both immunoglobulin and T cell receptor genes. J Exp Med 162:1015, 1985
18. Lin JC, Lin SC, De BK, Chan WC, Evatt BL: Precision genotyping of Epstein-Barr virus by polymerase chain reaction using three gene
loci (EBNA-2, EBNA-3C, and EBER): Predominance of type A virus
associated with Hodgkin’s disease. Blood 8 1 :3372, 1993
ARVANITAKIS ET AL
19. Frank D, Cesarman E, Liu YF, Michler RE, Knowles DM:
Posttransplantation lymphoproliferative disorders frequently contain
type A and not type B Epstein-Barr virus. Blood 85: 1396, I995
20. Cesarman E, Nador RG, Bai F, Bohenzky RA, Russo JJ,
Moore PS, Chang Y, Knowles DM: Kaposi’s sarcoma-associated
herpesvirus (KSHVHHV 8) contains G protein-coupled receptor
and cyclin D homologs which are expressed in Kaposi’.; sarcoma
and malignant lymphoma. J Virol 1996 (in press)
21. Mesri EA, Federoff HJ, Brownlee M: Expression of vascular
endothelial growth factor from a defective herpes simplex virus type
I amplicon vector induces angiogenesis in mice. Circ Res 76:161,
1995
22. Mesri EA, Cesarman E, Arvanitakis L, Rafii S, Moore MAS,
Posnett DN, Knowles DM, Asch AS: Human herpesvirus4 /
Kaposi’s sarcoma associated herpesvirus is a new transmissible virus
that infects B-cells. J Exp Med 183:2385, 1996
23. Said JW, Chien K, Tasaka T, Takeuchi S, Asou H, Cho
SK, Cesarman E, Knowles DM, Koeffler HP: Kaposi’s sarcoma
associated herpesvirus (KSHV or HHVS) in primary effusion
lymphoma: Ultrastructural demonstration of herpesvirus in
lymphoma cells. Blood 87:4937, 1996
24. Madeley CR, Field AM: Virus Morphology. London, UK,
Churchill Livingstone, 1988, p 5 1
25. Jung JU, Stager M, Desrosiers RC: Virus-encoded cyclin.
Mol Cell Biol 14:7235, 1994
25. Arvanitakis L, Yaseen N, Sharma S: Latent membrane protein- I induces cyclin D2 expression, pRb hyperphosphorylation, and
loss of TGF-P I -mediated growth inhibition in EBV-positive B cells.
J Inimunol 155: 1047, 1995
From www.bloodjournal.org by guest on June 17, 2017. For personal use only.
1996 88: 2648-2654
Establishment and characterization of a primary effusion (body
cavity- based) lymphoma cell line (BC-3) harboring kaposi's
sarcoma-associated herpesvirus (KSHV/HHV-8) in the absence of
Epstein-Barr virus
L Arvanitakis, EA Mesri, RG Nador, JW Said, AS Asch, DM Knowles and E Cesarman
Updated information and services can be found at:
http://www.bloodjournal.org/content/88/7/2648.full.html
Articles on similar topics can be found in the following Blood collections
Information about reproducing this article in parts or in its entirety may be found online at:
http://www.bloodjournal.org/site/misc/rights.xhtml#repub_requests
Information about ordering reprints may be found online at:
http://www.bloodjournal.org/site/misc/rights.xhtml#reprints
Information about subscriptions and ASH membership may be found online at:
http://www.bloodjournal.org/site/subscriptions/index.xhtml
Blood (print ISSN 0006-4971, online ISSN 1528-0020), is published weekly by the American
Society of Hematology, 2021 L St, NW, Suite 900, Washington DC 20036.
Copyright 2011 by The American Society of Hematology; all rights reserved.