The CRO-AP/3 and CRO-AP/5 PEL cell lines were derived from two distinct HIV-infected patients, defined as case 1 and case 2, respectively. Selected clinical characteristics of both patients have been included in a previous report ( Spina et al, 1998 ).
Case 1 was a 42-year-old HIV-positive homosexual male without a previous history of Kaposi's sarcoma. At the clinical level, the lymphoma presented exclusively as a malignant peritoneal effusion.
Case 2 was a 35-year-old HIV-positive homosexual male with a previous history of Kaposi's sarcoma. At the clinical level, the lymphoma presented exclusively as a malignant pleural effusion.
The assignment of both cases to PEL was based on multiple clinico-pathological and biological criteria as reported previously (reviewed by Jaffe, 1996; Carbone & Gaidano, 1997).
Establishment of cell lines and cell culture procedures
Lymphoma cells obtained from ascitic fluid (case 1) and pleural effusion (case 2) were separated by Ficoll-Hypaque (Pharmacia, Uppsala, Sweden) and washed twice in sterile phosphate-buffered saline. Subsequently, cells were cultured in RPMI 1640 (GIBCO, Paisley, Scotland) supplemented with 20% heat-inactivated fetal bovine serum (FBS, ICN Biomedicals Inc., Costa Mesa, Calif.), 2 m ML-glutamine, 100 U/ml penicillin, and 100 μg/ml streptomycin (Irvine Scientific, Santa Ana, Calif.) at 37°C in the presence of 5% CO2. After 3 weeks (case 1; CRO-AP/3) or 6 weeks (case 2; CRO-AP/5) the cells started to grow spontaneously. Cultures were fed twice weekly with the above-mentioned medium. Once established (defined as in vitro growth for more than 40 passages), the cell lines were continuously cultured in RPMI 1640–10% FBS at 37°C in the presence of 5% CO2.
The immunophenotype of PEL primary samples and derived cell lines was determined either by immunocyto-histochemistry (on cytospins and cell blocks) or by direct and indirect immunofluorescent flow cytometry using the FACScan fluorescent activated cell sorter (Becton Dickinson, Mountain View, Calif.) as previously described ( Carbone et al, 1996, 1997b). The monoclonal antibodies (MoAb) used in this study (together with their sources) are listed in Table I. A phenotypic analysis of EBV latent gene expression (latent membrane protein 1 [LMP 1] and EBV-encoded nuclear antigen 2 [EBNA 2]) was also performed on cytospins and cell blocks, as previously reported ( Carbone et al, 1997b ).
Table 1. Table I. Immunophenotype of PEL cell lines and corresponding primary tumour samples. −, negative; +, positive < 25%; ++, positive 25–50%; +++, positive 50–75%; ++++, positive > 75%. The following B-cell and T-cell associated markers were consistently negative: k (A8B5/D), λ (TB28-2/BD), IgM (R1/69/D), IgD (IgD26/D), IgG (A57H/Ser), CD10 (W8E7/BD), CD19 (Leu12/BD), CD22 (HD39/BOE), CD40 (89/JB), CD45RA (MB1/CB), CD79a (JCB117/D), CD3 (Leu4/BD), CD5 (Leu1/BD).* D, Dakopatts A/S (Glostrup, Denmark); BD, Becton Dickinson (Mountain View, Calif.); Ser, Serotec (Serotec, Oxford, U.K.); CB, Clonab Biotest (Dreiech, Germany); BS, The Binding Site (Birmingham, U.K.); V WS, V Workshop on Leukocyte Typing; BOE, Boehringer (Mannheim, Germany); JB, Dr J. Banchereau (Centre de Reserche, Schering-Plough, Dardilly, France); IT, Immunotech (Marseille, France).
In order to better define the stage of B-cell maturation of PEL and to gain better insights into the biological basis of PEL growth pattern, the PEL-derived cell lines and primary samples were investigated by immunocytochemistry for the expression of molecules selectively associated with late stages of B-cell differentiation (see Table II) as well as molecules implicated in cell-to-cell and cell-to-extracellular matrix (ECM) interactions (see Table III).
Table 2. Table II. Plasma-cell associated antigens in PEL cell lines and corresponding primary tumour samples. −, negative; +, positive < 25%; ++, positive 25–50%; +++, positive 50–75%; ++++, positive > 75%.* Ser, Serotec (Oxford, U.K.); D, Dakopatts A/S (Glostrup, Denmark); VI WS, VI Workshop on Leukocyte Typing. Table 3. Table III. Adhesion molecules in PEL cell lines and the corresponding primary tumour samples. −, negative; +, positive < 25%; ++, positive 25–50%; +++, positive 50–75%; ++++, positive > 75%.* TCS, T-Cell Sciences Inc. (Cambridge, Mass.); D, Dakopatts A/S (Glostrup, Denmark); AS, A. Sonnenberg (NKI, Amsterdam, The Netherlands); BD, Becton Dickinson (Mountain View, Calif.); Ser, Serotec (Oxford, U.K.); IT, Immunotech (Marseille, France); Mon, Monosan (Uden, The Netherlands); JTA, J. Thomas August (The Johns Hopkins University School of Medicine, Baltimore, Md.).
Chromosomal studies were performed on both cell lines (CRO-AP/3 and CRO-AP/5) by the trypsin-Giemsa banding method. Well-spread metaphases were photographed and arranged according to the recommendation of an International System for Human Cytogenetic Nomenclature ( ISCN, 1995) for cancer cytogenetics. Seven to 15 karyotypes per cell line were prepared.
As a first step, conventional cytogenetic analysis was performed on both cell lines. Subsequently, to assess doubtful findings, FISH analysis of metaphase spreads was performed using biotinylated DNA-painting probes according to the method of Pinkel et al (1986 ).
Cytogenetic analysis of PEL parental samples could not be performed due to the lack of sufficient numbers of metaphase spreads.
Analysis of viral infection
The presence of HHV-8 infection was assessed by multiple approaches, including PCR and Southern blot analysis of genomic DNA. PCR was performed with primers KS330233-F (5′-AGCCGAAAGGATTCCACCAT-3′) and KS330233-R (5′-TCCGTGTTGTCTACGTCCAG-3′) as previously reported ( Gaidano et al, 1996 ). Briefly, PCR was performed with 100 ng genomic DNA, 25 pmol of each primer, 200 μmol/l dNTPs, 10 mmol/l Tris-HCl (pH 8.8), 50 mmol/l KCl, 1 mmol/l MgCl2, 0.01% gelatin, 1 U Taq polymerase (Perkin-Elmer, Norwalk, Ct., U.S.A.), in a final volume of 25 μl. 35 cycles of denaturation (94°C), annealing (58°C) and extension (72°C) were performed in a Hybaid Omn-E thermocycler. For confirmatory purposes, samples scored positive by PCR were further tested by Southern blot hybridization of BamHI digested genomic DNA to the radiolabelled probe KS631Bam ( Chang et al, 1994 ). The HHV-8 viral load was assessed by semiquantitative PCR analysis performed according to a previously reported method ( Carbone et al, 1996 ).
Infection by EBV was investigated by multiple approaches, including PCR and Southern blot analysis of genomic DNA ( Gaidano et al, 1996 ). Analysis of EBV by PCR was performed with primers SL-1 (5′-GGACCTCAAAGAAGAGGGGG-3′) and SL-3 (5′-GCTCCTGGTCTTCCGCCTCC-3′), representative of the EBV nuclear antigen (EBNA)-1 gene. 35 cycles of PCR were performed as described above (annealing temperature 55°C). Southern blot analysis was performed with a probe representative of the EBV genomic termini (5.2 kb BamHI-EcoRI fragment isolated from the fused BamHI terminal fragment NJ-het) which, upon BamHI digestion, enables the definition of the clonality status of EBV infection ( Weiss et al, 1987 ).
The presence of HIV infection in the PEL cell lines and in the parental tumour samples was assessed by PCR analysis using primers derived from a highly conserved DNA sequence of the HIV clone HXB2, as previously reported ( Simmonds et al, 1990 ). Primers amplified a portion of the gp120 gene and their sequence was as follows: V3(a), 5′-TACAATGTACACATGGAATT-3′, and V3(d), 5′-ATTACAGTAGAAAAATTCCCC-3′. 30 cycles of PCR were performed as described above (annealing temperature 50°C).
Analysis of genetic lesions
Gross rearrangements of BCL-2, BCL-6 and c-MYC, as well as mutations of p53, c-MYC first exon–first intron border, and BCL-6 5′ noncoding regions were investigated by a combination of molecular approaches as previously reported ( Ballerini et al, 1993 ; Volpe et al, 1996 ; Gaidano et al, 1997a ).
Mutations of p53, c-MYC first exon–first intron border, and BCL-6 5′ non-coding regions were investigated by a two step approach, including PCR–Single Strand Conformation Polymorphism (PCR-SSCP) followed by DNA direct sequencing analysis of positive samples ( Gaidano et al, 1991 , 1997a; Ballerini et al, 1993 ). Briefly, PCR-SSCP was performed with 100 ng genomic DNA, 10 pmol of each primer, 2.5 μmol/l dNTPs, 37 kBq of [α-32P]dCTP (Amersham, U.K.; specific activity, 111 TBq/mmol), 10 mmol/l Tris-HCl (pH 8.8), 50 mmol/l KCl, 1 mmol/l MgCl2, 0.01% gelatin, 0.5 U Taq polymerase (Perkin-Elmer, Norwalk, Ct., U.S.A.), in a final volume of 10 μl. 30 cycles of denaturation (94°C), annealing (annealing temperatures were optimized for each pair of primers), and extension (72°C) were performed in a Hybaid Omn-E thermocycler. Samples were heated at 95°C for 5 min, chilled on ice, and immediately loaded (3 μl) onto a 6% acrylamide-TBE gel containing 10% glycerol. The sequence of oligonucleotides corresponding to p53 exons 5–9 was as follows ( Gaidano et al, 1991 ): p5-5, 5′-TTCCTCTTCCTGCAGTACTC-3′, and P5-3, 5′-ACCCTGGGCAACCAGCCCTGT-3′ (for p53 exon 5); P6-5, 5′-ACAGGGCTGGTTGCCCAGGGT-3′, and P6-3, 5′-AGTTGCAAACCAGACCTCAG-3′ (for p53 exon 6); P7-5, 5′-GTGTTGTCTCCTAGGTTGGC-3′, and P7-3, 5′-GTCAGAGGCAAGCAGAGGCT-3′ (for p53 exon 7); P8-5, 5′-TATCCTGAGTAGTGGTAATC-3′, and P8-3, 5′-AAGTGAATCTGAGGCATAAC-3′ (for p53 exon 8); P9-5, 5′-GCAGTTATGCCTCAGATTCAC-3′, and P9-3, 5′-AAGACTTAGTACC-
TGAAGGGT-3′ (for p53 exon 9). The sequence of oligonucleotides corresponding to c-MYC first exon-first intron border (fragment F and fragment G) was as follows ( Ballerini et al, 1993 ): F5′, 5′-GCACTGGAACTTACAACACC-3′, and F3′, 5′-GGTGCTTACCTGGTTTTCCA-3′ (for fragment F); G5′, 5′-CTGCCAGGACCCGCTTCTCT-3′, and G3′, 5′-TTTACCCCGATCCAGTTCTG-3′ (for fragment G). The sequence of oligonucleotides corresponding to BCL-6 5′ non-coding regions (fragments E1.10, E1.11 and E1.12) was as follows ( Migliazza et al, 1995 ): E1.21B, 5′-CTCTTGCCAAATGCTTTG-3′, and E1.24, 5′-TAATTCCCCTCCTTCCTC-3′ (for fragment E1.10); E1.23, 5′-AGGAAGGAGGGGAATTAG-3′, and IP1.6, 5′-AAGCAGTTTGCAAGCGAG-3′ (for fragment E1.11); IP1.7, 5′-TTCTCGCTTGCAAACTGC-3′, and E1.26, 5′-CACGATACTTCATCTCATC-3′ (for fragment E1.12).
For DNA sequencing of BCL-6 5′ non-coding regions, a unique PCR product encompassing fragments E1.10, E1.11 and E1.12 (nucleotides +404 to +1142) was amplified by primers E1.21B and E1.26. DNA direct sequencing of the amplified PCR fragment was performed with appropriate primers using a commercially available kit (ThermoSequenase, Amersham Life Sciences, Amersham, U.K.). α-33P-labelled terminator dideoxynucleotides were included in the sequencing mixture. Both strands were sequenced for each DNA fragment analysed.