sign in sign up
<<
Home , Plasma cell

4181 JClin Pathol 1994;47:418-422 VS38: a new monoclonal for detecting differentiation in routine sections J Clin Pathol: first published as 10.1136/jcp.47.5.418 on 1 May 1994. Downloaded from

H Turley, M Jones, W Erber, K Mayne, M de Waele, K Gatter

Abstract mouse , VS38, that is Aims-To characterise a new mouse reactive with an expressed strongly on monoclonal antibody, VS38, which recog- normal and neoplastic plasma cells but is nises an intracytoplasmic antigen of 64 absent from haemopoietic cells of other lin- kilodaltons present in normal and neo- eages. This antigen is also expressed to some plastic plasma cells; and to establish its extent in both normal and neoplastic epithe- value as a diagnostic reagent for routine lial cells. VS38 was tested on a wide range of pathological practice. normal and neoplastic tissues including solid Methods-A range of normal and neo- tumours as well as and plastic tissue sections, both frozen and specimens, with the aim of establishing its routinely fixed, were immunostained, value as a diagnostic reagent for pathological using the microwave method of antigen practice. retrieval for routinely fixed specimens. The antibody was also tested on blood and bone marrow specimens and a range Methods of human cell lines. The molecular PRODUCTION OF ANTIBODY VS38 weight of the antigen recognised by the The source of antigen was a membrane prepa- antibody was obtained by western blot ration of MCF-7 AdR cells. About 4 x 1O' analysis. FACS analysis was used to cells were injected intraperitoneally into a demonstrate the cellular location of the Balb/c mouse, together with Freund's com- antigen and its presence on tonsil cell plete adjuvant. Eleven days later, the mouse suspensions and myeloma cases. was given a booster injection of 1 6 x 106 cells, Results-VS38 recognised normal and with non-complete adjuvant. The booster was neoplastic plasma cells in all of the repeated again seven days later. Three days all tissues, including routinely fixed after the last injection the mouse was killed http://jcp.bmj.com/ plasma cell neoplasms tested. The anti- and the removed. Cell suspensions body also weakly stained epithelial ele- were prepared from the spleen and were fused ments within the tissue but was absent with cells from the myeloma cell line NSIs, from haemopoietic cells of other using polyethylene glycol to induce cell lineages. fusion, as described by Mason and Conclusion-Antibody VS38 is of poten- colleagues.7

tial value in identifying myeloma or on September 27, 2021 by guest. Protected copyright. in bone marrow or other BIOCHEMICAL ANALYSIS OF THE VS38 tissues. It differentiates lymphoplasma- Cell cultures were washed twice with phos- cytoid from lymphocytic and phate buffered saline (PBS) and then har- . It also subdivides vested with 0 25% trypsin and 0 02% EDTA. large cell into two groups The cell suspension was washed in culture University which may be a more reliable method medium and centrifuged for five minutes at Department of of separating these tumours than mor- 1000 and The cells were Cellular Science, John rpm 4°C. washed Radcliffe Hospital, phology alone. twice in PBS and centrifuged for a final spin Headington, Oxford at 2000 rpm for 10 minutes at 4°C. The cell OX3 9DU (J Clin Pathol 1994;47:418-422) pellet was solubilised with an equal volume of H Turley M Jones 1% (v/v) Triton X-100 (Sigma) in 10 mM K Gatter TRIS-HCL (pH 8) for 20 minutes on ice. K Mayne The terminal stage of differentiation is This was then spun at 2000 rpm for 15 min- Haematology the plasma cell for which no unique utes at 4°C and the supernatant fluid col- Department, Royal have been Perth Hospital, Perth, yet described. Apart from their lected. This was either used immediately for Australia morphology, plasma cells are usually distin- analysis by polyacrylamide gel electrophoresis W Erber guished from other B cells by their lack of sur- (PAGE), or stored as a membrane extract at Laboratorium face HLA class I and class II antigens, surface - 700C. Hematologie et immunoglobulin, Fc and C3 receptors, and Immunologie, Academisch the presence of intracytoplasmic immuno- SDS-PAGE Ziekenhuis, VUB globulin.' Most which stain plasma Sodium dodecyl sulphate (SDS) polyacry- Brussels, Belgium cells label other B cells as well as cells of other lamide gel electrophoresis was performed on M de Waele lineages.2 6 Most do not work on routinely the extract, according to the method of Correspondence to: Dr Kevin Gatter fixed material, thus further limiting their Laemmli.8 Samples were suspended in buffer Accepted for publication usefulness to pathologists. containing DL-dithiothreitol (Sigma) and 25 November 1993 We describe the characterisation of a boiled for two minutes, allowed to cool, and Plasma cell detection in routine sections 419

loaded on to a gel (5 5% stacking gel, 6-5% were then fixed in acetone/methanol (1 in 1) running gel). Molecular weight markers for 60 seconds at room temperature, placed (Sigma UK) were run at the same time. into TBS (05M TRIS, pH7-6, diluted 1 in 10 were transferred electrophoreti- in 0- 15M saline) and either stained using the J Clin Pathol: first published as 10.1136/jcp.47.5.418 on 1 May 1994. Downloaded from cally using a Transblot (Biorad UK) appara- alkaline phosphatase-antialkaline phosphatase tus to nitrocellulose membranes9 and the (APAAP) technique, or stored wrapped in detected by applying the antibody, aluminium foil at - 20°C until required. They followed by a rabbit-anti-mouse peroxidase were then allowed to reach room temperature conjugate (Dako UK) and visualised using before fixation, as described. diaminobenzidine with metal ion enhance- ment.'0 IMMUNOCYTOCHEMICAL STAINING The APAAP staining technique was per- TISSUES formed, as described before.'2 The enzyme A representative range of normal and neo- reaction was developed using napthol-AS- plastic tissues, including bone marrow biopsy MX-phosphate and fast red (TR-Salt, Sigma specimens, was obtained from the Histo- Chemical Co) as substrate. For immunoper- pathology Department at the John Radcliffe oxidase labelling tissue sections were incu- Hospital. Tissues were immediately snap bated first with the monoclonal antibody for frozen in liquid nitrogen and stored at 30 minutes, then with peroxidase conjugated - 70°C. Cryostat sections (5-8 ,um) were rabbit-anti-mouse (Dakopatts), and finally then cut and collected on to glass multiwell with peroxidase conjugated swine-anti-rabbit slides and dried overnight at room tempera- immunoglobulin."3 The peroxidase reaction ture. The slides were then fixed in acetone at was developed using diaminobenzidine room temperature. The sections were either (Sigma Chemical Co). The slides were then stained immediately or stored wrapped in alu- washed and mounted in aqueous mountant minium foil at - 20°C until staining. (Apathy's; BDH Poole, Dorset). Formalin fixed, paraffin wax embedded sec- tions were cut at approximately 5 pm and FACS ANALYSIS floated on to silane coated glass slides. These White cells and mononuclear cells were iso- were dewaxed, rehydrated, and microwaved lated from peripheral blood and bone marrow in a 700 Watt microwave for 8 minutes in aspirates by red cell lysis using ammonium 0-01 M sodium citrate, then into buffer before chloride'4 or density centrifugation on staining as normal." Lymphoprep (Nygaard, Oslo). The peripheral blood cells were permeabilised using buffered CELL LINES formol acetone (BFA), as described by The following cell lines were obtained from Slaper-Cortenbach et al.'5 After washing in

A B C the Sir William Dunn School of Pathology, phosphate buffered saline (PBS) containing http://jcp.bmj.com/ Oxford: K562 (erythroleukaemia); HeLa 0-1% bovine serum albumin (BSA), perme- (cervical epithelial carcinoma); U937 (malig- abilised cells were incubated for 30 minutes at 205 - nant histiocytosis); HL60 (promyelocytic 4°C with the antibodies. Cells were then leukaemia); RVH421 (melanoma); A431 washed and incubated for 30 minutes with (vulval carcinoma); Daudi and Raji (Burkitt's fluorescein isothiocyanate (FITC) conjugated 116 - lymphoma); Nalm-1 (pre-B cell leukaemia); rabbit-anti-mouse F(ab')2 immunoglobulin 97

and Jurkat ( lymphoma). The cell lines (Dakopatts a/s). After a final wash the cells on September 27, 2021 by guest. Protected copyright. L428 (Hodgkin's disease) and Thiel (plasma were fixed in 1-5% formaldehyde in washing 68 - cell) were obtained from Dr V Diehl; SU- buffer and analysed in a FACScan flow DHL-1 (T cell leukaemia) from Dr ML cytometer (Becton Dickinson) using Consort Cleary; Karpas 299 (T cell lymphoma with a 30 and Paint-a-Gate software. 2:5 translocation) and Karpas M (plasma cell) from Dr A Karpas; and HT29 (colon carci- noma), MCF 7 ADR (breast carcinoma) from Results the Imperial Research Fund, London. BIOCHEMICAL ANALYSIS OF EPITOPE The plasma cell lines L363, JJN3, Karpas A strong band of molecular weight 64 kilodal- 29 707, OPM-1, and EJM were obtained from tons (reduced) was detected by VS38 from Professor K Nilsson, University of Uppsala, extracts of the cell lines Thiel (plasma cell Sweden. LB 84-4 (plasma cell), ARH 77 derived), MCF-7, and MCF-7 ADR (breast Figure 1 Western blot (plasma cell), and 8226 R10 (plasma cell) carcinoma derived) (fig 1). Occasionally a analysis of lysatesfrom were from Professor Brian GM Durie. All cell higher molecular weight band of 73 kilodal- three cell lin Lane A M(Ces. (breast lines were cultured in RPMI containing 10% tons was detected in the MCF-7 ADR cell carcinoma) ;F-7 fetal calf serum (Gibco Biocult Ltd) at 37°C. line (fig 1). No bands could be detected with- Lane B M(CF-7ADR Cytocentrifuge preparations were made using a out reducing conditions. (breast carczinoma) multidnug resistant; Shandon cytocentrifuge. These were allowed Lane C Thiiel (plasma cell to air dry for at least two hours, then fixed for NORMAL TISSUE REACTIVITY derived). The antibody 10 minutes in acetone. Frozenlfresh material recognises the same 64 kilodalton band in all three VS38 was tested on a wide range of normal cases with an extra band of BLOOD AND BONE MARROW SAMPLES tissues (table 1). The most striking result is 73 kilodaltons present in Samples were obtained from the Haematology the intense cytoplasmic labelling of all plasma the drug resistant cell line. Molecular weight Department of the John Radcliffe Hospital cells (fig 2). All other. lymphoid cells were standards are indicated to and cytospins were prepared and allowed to unlabelled. Weak staining of epithelium was the left. dry at room temperature overnight. The slides detectable in most tissues, and in some cases 420 Turley, J7ones, Erber, Mayne, de Waele, Gatter

Table 1 some tissues but were unable to identify these Nonnal tissue staining ofroutinelyfixed specimens cells. Tonsil Weak epithelium Gut APUD cells FACS J Clin Pathol: first published as 10.1136/jcp.47.5.418 on 1 May 1994. Downloaded from Pancreas Exocrine positive, endocrine negative Analysis of the tonsil cell suspensions, Lung Plasma cells Brain Neurones, glial cells myeloma cases, and the plasma cell lines with Heart No staining and without permeabilisation showed that the Cerebellum Purkinje cells Adrenal Glomerulosa cells antigen is intracytoplasmic being undetectable Kidney Distal tubules, Bowman's capsule without prior permeabilisation (fig 3). Spleen No staining APUD: Amine precursor uptake decarboxylase. REACTIVITY OF VS38 WITH HUMAN CELL LINES The antibody reacted with all the plasma cell lines tested. This reactivity was strong and weak endothelial staining was also present. present on all of the cells in the sample. All of Peripheral blood were unstained, the lymphocytic lines were unreactive with although some weak staining of monocytes VS38 apart from Nalm-1 which showed a and neutrophils could occasionally be weak positivity. Of the myeloid lines tested, detected. only U937 was reactive. The carcinoma lines were also weakly positive in keeping with the Fixed material tissue staining. The single melanoma cell line Without microwave pre-treatment no reactivity examined RVH42 1 was also positively could be detected on routinely fixed sections labelled. whether or not enzyme predigestion was utilised. However, after heating in a sodium REACTIVITY WITH LYMPHOMAS AND citrate solution intense cytoplasmic labelling LEUKAEMLAS of plasma cells was seen (fig 2). Other lym- The results of staining lymphoma specimens phoid cells were negative. Of note was the fact with VS38 are summarised in table 2. All that the epithelial cell labelling seen in most specimens of myeloma and plasmacytoma tissues on frozen section examination was were strongly labelled (fig 2) as were the lym- considerably reduced or absent in these speci- phoplasmacytoid elements in all of the lym- mens (fig 2). We also observed strongly phoplasmacytic lymphomas. Only in frozen labelled spindle cells within the stroma of sections was there staining of follicular lym- phoma and common lymphocytic leukaemia (CLL). In these cases staining was generally focal and weak (fig 2). There was no staining ofT cell lymphomas but 10 ofthe fixed large B

cell lymphomas were positive, albeit much http://jcp.bmj.com/ more weakly than the myelomas (fig 2).

Discussion No plasma cell specific antigens have been described to date, with all of the previously

reported antibodies also labelling other cell on September 27, 2021 by guest. Protected copyright. types, most notably other B cells.3-5 16-1 Two antibodies well known to haematologists, PCA1 and PCA2,16 were reported as being expressed on plasma cells but were also pre- sent on and monocytes. Other examples more familiar to histopathologists might include antibodies against EMA (epithelial membrane antigen) and CD30 (notably antibody Ber-H2).1>" Most impor- tantly none of these antibodies has been shown to detect all (or virtually alt) plasma cells in formalin fixed material, the main interest of the current study to practising histopathologists. The only antibody other

Table 2 VS38 immunostaining on lymphoma Paraffin Diagnosis Frozen wax CLL 7*/13 0/12 Centroblastic centrocytic lymphoma 2/7 0/6 Hairy cell leukaemia 0/8 0/4 Figure 2 (A) Strongly labelled plasma cells surrounded by a weakly positive squamous Lymphoplasmacytoid lymphoma 2/2 7/7 cell carcinoma of the lung. (B) Positive labelling of myeloma cells. (C) Large cell B B cell large cell non-Hodgkin's lymphoma 0/8 12/38 lymphoma showing positive staining of the malignant cells contrasted with (D) a large cell T cell non-Hodgldn's lymphoma NT 0/9 B lymphoma in which the lymphoma cells are unstained (note afew.scattered positive Myeloma 42/42 7/7 plasma cells). (E) Small intestine showing strong positivity ofAPUD cells in the glandtelar Plasmacytoma NT 4/4 crypts. (F) Low grade lymphoplasmacytoid Iymphoma uniformly stained though not as intensely as the myeloma (B). Key: NT = not tested. *weak/focal staining. Plasma cell detection in routine sections 421

UNPERMEABILISED PERMEABILI'BED only disagreed with each other but also with A B themselves when they re-examined individual cases. This difficulty is reflected in the Kiel

classification which makes substantial efforts J Clin Pathol: first published as 10.1136/jcp.47.5.418 on 1 May 1994. Downloaded from to define reliably the distinction between immunoblastic and centroblastic lymphoma. In early studies this group believed that immunoblastic was the commonest large cell non-Hodgkin's lymphoma. This is what the general reader will glean from many text- books-for example, The Oxford Textbook of Patholog93 and the first edition of Biopsy Interpretation by Stansfeld.24 In the sec- ond edition of Stansfeld's book the revised Kiel classification is used and the following , explanation is given for centroblastic now being the most common large cell non- FL1 FL1 Hodgkin's lymphoma. Bu2O Bu2O "Formerly believed to be the commonest C D type of high grade large cell lymphoma it is now clear that ML-IB-B (immunoblastic) is less common than centroblastic lymphoma. This reversal is largely due to the inclusion in the ML centroblastic-polymorphic category of tumour as containing a large number of when it can be shown that the tumour is basically of follicular centre cell S, origin."25 This undoubtedly reflects difficulty in making the diagnostic subdivision rather than any change in the incidence of the different subtypes of disease. Separation of these large cell cases on the grounds of reproducible immunostaining, such as that achieved well be a Ina lot Ina 1B3 104 log lot Ia2 by VS38, may FL1 FL1 better way of subdividing large cell lym- VS38 VS38 phomas than by morphological examination

Figure 3 FACS analysis of a bone marrow samplefrom a case ofmultiple nyeloma. alone. It is to be hoped that lymphoma groups http://jcp.bmj.com/ Antibody VS38 shows no surface staining ofthe neoplastic cells (C). However, following will incorporate such findings into their study BFA permeabilisation (D), the myeloma ceUls are labelled by antibody VS38. o(A) and of non-Hodgkin's lymphomas so that in time (B) show the absence oflabelling obtained with antibody Bu2O (negative control) before a arise as to or not it and after permeabilisation. (SSC = side scatter) consensus might whether is useful to subdivide large cell lymphomas and if so on what grounds.

to This work was in part the Cancer than anti-Ig antibodies describedI date, supported by Imperial on September 27, 2021 by guest. Protected copyright. which stains neoplastic plasma cel[Is in fixed Research Fund and the Leukaemia Research Fund. sections, is antibody MM4.5 A drawback of this antibody is that only a subset of the myeloma cells is positive, and this could be a I Harris NL. Lymphomas. In: Colvin RB, Bhan AK, potential problem in small or lightly infiltrated York:McCluskeyRavenRT,Press,eds.1988:271-300.Diagnostic immunopathology. New biopsy specimens. 2 Delmastro-Galfre P, Calabi F, Arno J, Karpas A, Hayhoe In the current study we have shovwn that the myeloma/plasmaFGJ. Monoclonalcells.antibodiesIn: McMichaelspecificAJ, ed.for Leucocytehuman recently produced antibody VS38 i:s of poten- typing III. Oxford: Oxford University Press, 1987: 510-11. tial value in clinical practice. It can be helpful 3 Nathan PD, Walker L, Hardie D, et al. An antigenic study in identifying myeloma or plasmacytoma in of human plasma cells in normal tissue and in myeloma: bone-marrow and other tissues. It also helps identificationClin Exp Immunolof a 1986;65:112-9.novel plasma cell associated antigen. to distinguish cases of lymphoplaismacytoid 4 Tazzari PL, Gobbi M, Dinota A, et al. Normal and neo- plastic plasma cell membrane phenotype: studies with lymphoma from lymphocytic andI follicular new monoclonal antibodies. Clin Exp Immunol 1987; lymphoma. The only other lymphomas show- 70:192-200. mono- ing positivity were about one third cAf the cases 5 TongclonalAW,antibodyLee JC,havingStone MJ.selectiveCharacterizationreactivity withof anormal of large cell non-Hodgkin's lym]phoma. It and neoplastic plasma cells. Blood 1987;69:238-45. 6 Uckun FM. of human B-cell Blood be that these woiLuld be the Regulation ontogeny. might predicted stic 1990;76:1908-23. immunoblastic rather than the ceuntrobldntroblastic 7 Mason DY, Cordell JL, Pulford KAF. Production of mon- oclonal antibodies for immunocytochemical use. In: or other cell types. However, this iis a notori- Bullock GR, Petrusz P, eds. Techniques in immunocyto- ously difficult area of morphology with very chemistry. Vol 2. New York: Academic Press, 1983:175. 8 Laemmli UK. Cleavage of structural proteins during the little consensus among patholog,istsnists about assembly of the head of bacteriphage T4. Nature their identification. 1970;227:680-5. rnke et al 22 9 Towbin H, Staehlin T, Gordon J. Electrophoretic transfer Indeed, a study undertaken by Wa of proteins from polyacrylamide gels to nitrocellulose more than 10 years ago showed that even sheets. Proc Nad Acad Sci USA 1979;76:4350-4. 10 Harlow E, Lane D. Antibodies. A laboratory manual. Cold expert haematopathologists disagre-ed on the Spring Harbor: Cold Spring Harbor Laboratory, 1988. subdivision of large cell lymphomas They not 1 1 Cattoretti G, Becker MH, Key G, et al. Monoclonal anti- 422 Turley, Jones, Erber, Mayne, de Waele, Gatter

bodies against recombinant parts of the Ki-67 antigen biology and clinical course of the disease. Castellano- (MIB 1 and MIB 3) detect proliferating cells in Leones (Spain) Cooperative Group for the Study of microwave-processed formalin-fixed paraffin sections. Monoclonal Gammopathies. Br Haematol 1991;77: Pathol 1992;168:357-63. 185-90. 12 Cordell JL, Falini B, Erber WN, et al. Immunoenzymatic 19 Cordell J, Richardson TC, Pulford KA, et al. Production

labelling of monoclonal antibodies using immune of monoclonal antibodies against human epithelial mem- J Clin Pathol: first published as 10.1136/jcp.47.5.418 on 1 May 1994. Downloaded from complexes of alkaline phosphatase and monoclonal anti- brane antigen for use in diagnostic immunocytochem- alkaline phosphatase (APAAP complexes). Histochem istry. BrJGCancer 1985;52:347-54. Cytochem 1984;32:219-29. 20 Schwarting R, Gerdes J, Durkop H, Falini B, Pileri S, 13 Mason DY, Abdulaziz Z, Falini B, Stein H. Single and Stein H. BER-H2: a new anti-Ki-l (CD30) monoclonal double immunoenzymatic techniques for labelling tissue antibody directed at a formol-resistant epitope. Blood section with monoclonal antibodies. Ann NY Acad Sci 1989;74: 1678-89. 1984;420: 1127-33. 21 Heyderman E, Strudley I, Powell G, Richardson TC, 14 Carter N. Introduction to the principles offlow cytometry. Cordell JL, Mason DY. A new monoclonal antibody to In: Omerod MG, ed. . A practical epithelial membrane antigen (EMA)-E29. A comparison approach. Oxford: Oxford University Press, 1989:1-28. of its immunocytochemical reactivity with polyclonal 15 Slaper-Cortenbach ICM, Admiraal LG, Kerr JM, anti-EMA antibodies and with another monoclonal anti- Leeuwen EF, von dem Borne A, Tetteroo PAT. Flow body, HMFG-2. BrJ7 Cancer 1985;52:355-6 1. cytometric detection of terminal deoxynucleotidyl trans- 22 Warnke RA, Strauchen JA, Burke JS, Hoppe RT, ferase and other intracellular antigens in combination Campbell BA, Dorfman RF. Morphologic types of with membrane antigens in acute lymphatic leukaemia. diffuse large cell lymphoma. Cancer 1982;50:690-5. Blood 1988;72:1639. 23 Isaacson PG. The non-Hodgkin's lymphomas. In: McGee 16 Anderson KC, Park EK, Bates MP, et al. Antigens on JO, Isaacson PG, Wright NA, eds. Oxford textbook of human plasma cells identified by monoclonal antibodies. pathology. Oxford: Oxford University Press, 1992: 9 Immunol 1983;130:1132-8. 1775-87. 17 Linden MD, Fishleder AJ, Tubbs RR, Park H. Immuno- 24 Stansfeld AG. Lymph node biopsy interpretation. Edinburgh: phenotypic spectrum of . Cancer Churchill Livingstone, 1985. 1989;63:859-62. 25 Stansfeld AG, d'Ardenne AJ, eds. Lymph node biopsy 18 San MJ, Gonzalez M, Gascon A, et al. Immunophenotypic interpretation. Second edn. Edinburgh: Churchill heterogeneity of : influence on the Livingstone, 1992. http://jcp.bmj.com/ on September 27, 2021 by guest. Protected copyright.