80 Clin Pathol 1995;48:840-844

Immunohistochemical staining of plasticembedded bone marrow trephine biopsyspecimens after microwave heating

W G McCluggage, S Roddy, C Whiteside, J Burton, H McBride, P Maxwell,H Bharucha

AbstractAims-To investigate (1) whether ad-equate immunohistochemical staining canbe achieved on sections cut from plasticembedded bone marrow trephine biopsyspecimens after microwave heating in cit-rate buffer; and (2) whether this immuno-histochemical staining is comparable withthat achieved on routine sections cut fromparaffin wax embedded trephine biopsyspecimens after decalcification pro-cedures.Methods-Sixty five consecutive bonemarrow trephine biopsy specimens ofmore than 1 cm in length were dividedtransversely into two equal parts. One partwas processed in paraffin wax foliowed bydecalcification. The other part was em-bedded in the epoxyresin Polarbed 812 fol-lowed by the cutting of 1 pm sections. Bothparts underwent immunohistochemicalstaining by an identical panel of anti-bodies. With Polarbed 812 plastic em-bedded sections, microwave heating incitrate buffer was undertaken before theapplication of antisera.Results-On sections cut from plastic em-bedded material, immunohistochemicalstaining was generally satisfactory, easyto interpret and comparable with thatachieved with paraffin wax embeddedmaterial. Exceptions were antibodiesto neutrophil elastase and CD61 whereimmunostaining was consistently negativeon plastic embedded sections. Immuno-histochemical staining for CD20 wasconsistently more reliable on plasticembedded sections.Conclusions-The results provide evi-dence that, with few exceptions, satisfactoryimmunohistochemical staining is possibleon plastic embedded bone marrow tre-phine biopsy specimens after microwaveheating in citrate buffer. This, combinedwith the advantage of superior cellularmorphology with semi-thin (1 pm) sec-tions of plastic embedded material, makesuch embedding procedures the preferredmethod for the processing ofbone marrowtrephine biopsy specimens.(J Clin Pathol 1995;48:840-844)

Keywords: Bone marrow trephine biopsy specimen,immunohistochemistry, plastic embedding, microwaveheating.

Bone marrow trephine biopsies are routinelyperformed, in conjunction with marrow as-pirates, in the investigation of patients with avariety of haemopoietic disorders, both neo-plastic and non-neoplastic. Trephine biopsyspecimens are particularly useful where thereis significant bone marrow fibrosis which fre-quently results in an inadequately aspiratedspecimen or "dry tap". Trephine biopsy speci-mens are also useful in the assessment of mar-row cellularity and in the determination of theextent of marrow involvement by neoplastic orother infiltrates. The relation between cellularelements and normal marrow structures, suchas bony trabeculae and blood vessels, can beassessed accurately only on bone marrow tre-phine biopsy specimens. In addition focal le-sions, such as granulomas, are more readilyidentifiable on trephine biopsy specimens thanon marrow aspirates.Two main methods are used for the routine

histological examination of bone marrow tre-phine biopsy specimens, namely paraffin waxembedding followed by decalcification andplastic embedding followed by the cutting of1 ,um sections. A third approach to processingsuch specimens is the use of cryostat sections ofundecalcified bone marrow biopsy specimens.This method, although allowing for optimalantigen preservation, results in very poor cellu-lar morphology, a serious obstacle in evaluatingthese specimens, and thus has not gained wide-spread acceptance. Problems arise in cuttingsections ofbone marrow trephine biopsy speci-mens because of the intimate mixture of hardtissue (bone) and soft tissue (marrow and fat).To cut adequate intact sections, one can eithermake the tissue uniformly soft by paraffin waxembedding followed by decalcification, ormakethe tissue uniformly hard by using a plasticembedding procedure. There has been muchdebate in the literature regarding the relativemerits of these two techniques.`'7 Those whoadvocate paraffin wax embedding of trephinebiopsy specimens stress the advantages of thefamiliarity of most pathologists with sectionscut from such material combined with theirunfamiliarity with plastic embedded sectionsand the fact that widespread antigen pre-servation allows a wide range of immuno-histochemical reactions to be performed.'2Furthermore, plastic embedding represents asignificant departure from normal routine forthe diagnostic histopathology laboratory as spe-cial equipment, different preparation protocols

Departmentof Pathology,Royal Group ofHospitals Trust,Grosvenor Road,Belfast BT12 6BLW G McCluggageS RoddyC WhitesideJ BurtonH McBrideP Maxwell

The Queen'sUniversity ofBelfastH Bharucha

Correspondence to:Dr W G McCluggage.Accepted for publication21 February 1995

840

on April 8, 2021 by guest. P

rotected by copyright.http://jcp.bm

j.com/

J Clin P

athol: first published as 10.1136/jcp.48.9.840 on 1 Septem

ber 1995. Dow

nloaded from

Processing of bone marrow trephine biopsy specimens

and a higher degree of technical expertise arerequired. The protagonists of plastic em-bedding ofbone marrow trephine biopsy speci-mens point to the advantages of the excellentcellular morphology due to the thin sectionsobtained by this technique and to the lack oftissue distortion which is inevitable followingdecalcification procedures.36 In addition, plas-tic embedding procedures allow a wide rangeof histochemical stains to be performed.4 Amajor disadvantage with plastic embedded ma-terial is the loss of immunoreactivity which hashampered the routine application of im-munohistochemical reactions in the evaluationof such trephine biopsy specimens.Two main types of plastic resin are available

for embedding tissue, namely epoxy and acrylicresins.8 Acrylic resins include methyl me-thacrylate and glycol methacrylate. In recentyears glycol methacrylate has become the resinof choice for embedding bone marrow trephinebiopsy specimens in plastic.45 Several authorshave described methods of glycol methacrylateembedding which combine the advantage ofexcellent cellular morphology with optimalantigen preservation, permitting the immuno-histochemical detection of a wide range ofcellular antigens.57 We routinely use the epoxy-resin Polarbed 812 for processing bone mar-row trephine biopsy specimens. This resin isalso used for the preparation of semi-thin tissuesections prior to the cutting ofultrathin sectionsfor routine electron microscopy examination.In utilising this resin for the preparation ofbone marrow trephine biopsy specimens wehave avoided the need for specialised pro-cedures and equipment. Previously we havehad disappointing results when attempting toperform immunohistochemical reactions on tis-sue embedded in this resin, with little or nospecific staining together with excessive back-ground staining. Several articles have recentlyappeared in the literature advocating the useof prior microwave heating to enhance im-munohistochemical staining on a variety oftissue types.9'-" These reports stimulated us toundertake this study to determine whether priormicrowave heating would allow a range of an-tigens to be detected, using routine im-munohistochemical techniques, in sections cutfrom plastic embedded tissue. We have com-pared results of immunohistochemical stainingbetween routine paraffin wax embedded ma-terial without microwave heating and plasticembedded material following microwave heat-ing in citrate buffer.

MethodsSixty five consecutive bone marrow trephinebiopsy specimens more than 1 cm in lengthwere studied. All were obtained from the pos-terior, superior iliac crest of patients seen by ahaematologist at the Royal Group of Hospitals,Belfast, or the Mater Infirmorum Hospital,Belfast. Trephine biopsy specimens were re-ceived fresh and processed in the Departmentof Pathology, Royal Group of Hospitals,Belfast.

FIXATION AND PROCESSING OF TREPHINEBIOPSY SPECIMENSAll specimens were divided transversely intotwo equal parts. One part was fixed in 3%acetic acid in 10% unbuffered formol saline fora minimum of one and a half hours and amaximum of three hours. It was then washedin running water, treated with EDTA in buffer(pH 7 0) for 24 hours and processed to paraffinwax. Sections (5 gm) were cut and stained withhaematoxylin and eosin, and Giemsa.The second part of each trephine biopsy

was fixed overnight in 10% neutral bufferedformalin, followed by washing in phosphatebuffer prior to processing. Processing was car-ried out on a rotator as follows: 50% alcohol(30 minutes), 70% alcohol ( 30 minutes), 90%alcohol (30 minutes), 100% alcohol (2 x 30minutes), dried acetone (2 x 30 minutes), 1acetone: 3 resin (one hour), neat resin at 37°C( 2 x 1-5 hours). The trephine biopsy specimenwas then embedded in a rubber mould or plasticcapsule and polymerised at 60°C overnight.Blocks were trimmed on a Leica Ultracut EUltramicrotome, set at 1 jim, using a 450 glassknife. Sections (1 gim) were then cut using aHisto-Diatome diamond knife. Each sectionwas lifted onto a slide containing a drop ofwater, flamed to remove any wrinkles, dried ona hot plate and reflamed to ensure the sectiondid not wash off during staining. Etching ofsections was undertaken prior to staining.Slides were placed in a jar containing sodiumethoxide for 25 minutes, washed in 50% alcoholfor 30 minutes and then in running waterfor 15 minutes. Sections were stained withhaemotoxylin and eosin, and Giemsa.The resin used for this plastic embedding

procedure was the epoxyresin Polarbed 812(Fison's Scientific Equipment, Loughborough,UK). This was made up as follows: Polarbed812 (50g), DDSA (32g), MNA (21g), andDMP-30 (2 g).

IMMUNOHISTOCHEMICAL METHODSA number of primary antisera was applied toserial sections cut from both paraffin wax andplastic embedded tissue. An identical panel ofantibodies was used to stain the two halvesof each trephine biopsy specimen to facilitatecomparison of immunohistochemical stainingbetween sections cut from paraffin wax andplastic embedded tissue.

Sections cut from plastic embedded tissuewere transferred onto APES coated slides(Sigma, Poole, Dorset, UK), etched in sodiumethoxide (25 minutes), rinsed in alcohol, en-dogenous peroxidase activity blocked in 3%alcoholic hydrogen peroxide (15 minutes), andwashed well in water. Slides were placed in aplastic Coplin jar filled with citrate buffer (2 1 gcitric acid monohydrate in 1 litre of distilledwater, pH 6-0) and placed for 30 minutesin an 850W domestic microwave oven. TheCoplin jar was topped up at intervals withdistilled water to avoid complete evaporation.After microwave heating slides were allowed tocool to room temperature. Incubation with theprimary antibody was performed and antigens

841

on April 8, 2021 by guest. P

rotected by copyright.http://jcp.bm

j.com/

J Clin P

athol: first published as 10.1136/jcp.48.9.840 on 1 Septem

ber 1995. Dow

nloaded from

McCluggage, Roddy, Whiteside, Burton, McBride, Maxwell, et al

Antibodies used in the study together with their source and major specificities

Antibody Source Main specificity

Leukocyte common antigen (LCA) (CD45) Dako, Copenhagen, Denmark LeucocytesL26 (CD20) Dako B lymphoid cellsUCHL1 (CD45 RO) Dako T lymphoid cellsMT1 (CD43) Biotest, Dreieich, Germany T lymphoid cells, myeloid cells, megakaryocytes,

erythroid cellsCD3 Dako T lymphoid cellsCD61 Dako MegakaryocytesCD68 Dako Histiocytic cells, myeloid cellsFactor VIII related antigen Signet, St Louis, MO, USA Megakaryocytes, endotheliumNeutrophil elastase Dako NeutrophilsGlycophorin A Mason DY, Nuffield Dept. of Haematology, John Erythroid cells

Radcliffe Hospital, Oxford, UKBer-H2 (CD30) Dako Reed-Stemnberg cells in Hodgkin's disease, activated

lymphoid cells, some high grade non-Hodgkin'slymphomas

Leu Ml (CD 15) Dako Reed-Steinberg cells in Hodgkin's disease,myelomonocytic cells, some adenocarcinomas

Ki67 (MIB1) The Binding Site, Birmingham, UK Ki67 antigenNeurone specific enolase (NSE) IncStar, Wokingham, UK Neuroendocrine cells and tumours, neural cells and

tumoursCam 5-2 Becton Dickinson, Sunnyvale, CA, USA CytokeratinsAE1/AE3 ICN Biomedicals Ltd, Thame, UK CytokeratinsCarcinoembryonic antigen (CEA) Dako Adenocarcinomas, other carcinomasEpithelial membrane antigen (EMA) Dako Epithelial cells, plasma cells, some lymphomasChromogranin A Dako Neuroendocrine cells and tumoursS100 protein Diagnostic Products Ltd, Abingdon, UK S100 protein (melanocytic lesions, dendritic cells,

others)K light chains Dako K light chainsX light chains Dako k light chains

Figure 1 Immunohistochemistry of a plastic embedded bone marrow trephine specfor CD20 (L26). (Case of low grade non-Hodgkin s lymphoma.)

4. .L.

t .

'... B,4

Figure 2 Immunohistochemistry of a plastic embedded bone marrow trephine specfor CD68. (Normal bone marrow showing positive staining of myeloid cells.)

were localised using a standard Streptavidin-biotin peroxidase method (Dako, Copenhagen,Denmark).

Sections from paraffin wax embedded tissuefor immunohistochemistry were cut onto APEScoated slides and dried at 560C overnight. Im-munohistochemical staining was as outlinedfor plastic embedded material, except sectionswere not subjected to prior microwave heating.Several antisera required pretreatment withtrypsin (ICN, Amsterdam, Netherlands),namely CD3, CD68, K and k light chains,

. *; carcinoembryonic antigen (CEA), Cam 5-2,and AE1/AE3.

ANTIBODIES USED IN STUDY>-'- The table shows the primary antisera used in

cimen the study together with their main specificities.These antibodies were chosen, some becausethey stain normal haemopoietic elements pres-ent in bone marrow and others because theyare of use in the diagnosis of a variety ofhaemopoietic and non-haemopoietic disordersaffecting the bone marrow.

ResultsIn all cases cellular morphology was superiorwith sections cut from plastic embedded tissuecompared with those cut from paraffin waxembedded tissue. Microwave heating resultedin specific, easily interpretable staining ofplastic embedded material with 20 out of 22antibodies used in the study. Staining was com-parable with that achieved with paraffinwax embedded sections (in which specific,easily interpretable staining was achieved with-out microwave heating with all 22 antibodies),although generally of lower intensity. Excessivebackground staining was not present. The twoexceptions were antibodies against neutrophil

cimen elastase and CD6 1 where immunohisto-chemical staining was consistently negative on

842

on April 8, 2021 by guest. P

rotected by copyright.http://jcp.bm

j.com/

J Clin P

athol: first published as 10.1136/jcp.48.9.840 on 1 Septem

ber 1995. Dow

nloaded from

Processing of bone marrow trephine biopsy specimens

..... ...~

X,;tSlS ~ q3 '

A br.

0Wf'

Figure 3 Immunohistochemistry of a plastic embedded bone marrow trephine biopsyspecimen for Cam 5 2. (Case of metastatic breast carcinoma.)

sections cut from plastic embedded tissue butpositive on sections cut from paraffin wax em-

bedded tissue. Immunostaining with the CD20antibody was more satisfactory on plastic em-

bedded material. With all other antibodiesused, immunostaining was comparable on sec-tions cut from plastic and paraffin wax em-

bedded material.With sections cut from plastic embedded

tissue and stained with neurone specific enolase(NSE), strong cytoplasmic staining of matureplasma cells was identified in a number ofcases.This phenomenon was not seen in sections ofparaffin wax embedded material stained withNSE.

Figures 1 to 3 show sections cut from plasticembedded biopsy specimens and stained withthe antibodies CD20, CD68 and Cam 5-2.

DiscussionImmunohistochemical staining of tissue sec-

tions has become a valuable diagnostic tool forthe histopathologist. With plastic embeddedbone marrow trephine biopsy specimens, it hasgenerally been difficult to achieve adequateimmunohistochemical staining and, in mostpathology laboratories, trephine biopsy speci-mens are routinely processed in paraffin wax.This is in spite of the superior cellular mor-phology obtainable with semi-thin sections cutfrom plastic embedded material. For plasticembedding of bone marrow trephine biopsyspecimens, the water miscible plastic glycolmethacrylate has become the resin of choiceand, in recent years, there have been reportsin the literature of modified glycol methacylateembedding techniques for bone marrow tre-phine biopsy specimens which permit highquality visualisation ofmorphological detail to-gether with optimal antigen preservation.57Such embedding techniques also allow a widerange of detailed enzyme histochemical re-

actions to be performed.'5-'7In our laboratory we routinely use a plastic

embedding technique for the histological ex-

amination of bone marrow trephine biopsy

specimens. We use the epoxyresin Polarbed812, which was chosen primarily because it isalso routinely used for electron microscopy.All trephine biopsy specimens are routinelyprocessed in the Electron Microscopy Unit.This alleviates the need for the separate speci-men preparation protocols and different equip-ment necessary if a different method of plasticembedding were to be used. Excellent cy-tological detail is achieved and the applicationof a range of enzyme histochemical reactionsis possible. Ultrathin sections from the sameblock can also subsequently be examined byelectron microscopy in cases where this is in-dicated.

In the past we have attempted immuno-histochemical staining of such plastic em-bedded trephine biopsy specimens with littlesuccess. We found a lack of specific, readilyinterpretable staining together with excessivebackground, non-specific staining. With therecent interest in microwave heating,9-14 as op-posed to the predigestion of tissue sections byproteolytic enzymes, as an easily applicablemethod for re-exposing antigens for im-munohistochemical reactions, we undertookthis study to investigate whether microwaveheating would result in enhanced im-munohistochemical staining on sections cutfrom plastic embedded bone marrow trephinebiopsy specimens. The results provide evidencethat prior microwave heating does indeed resultin enhanced immunoreactivity with specific,readily interpretable staining and abolition ofexcessive background staining with most ofantibodies used in the study. Staining was com-parable with that achieved on paraffin waxembedded material without prior microwaveheating. Only with antibodies against neu-trophil elastase and CD61 was positive im-munostaining not obtained. Although stainingintensity was generally less with plastic thanwith paraffin wax embedded material, stainingwas specific; the less intense staining can beexplained on the basis of the thinness (1 pm)of sections obtained with the plastic embeddingtechnique.The exact mechanisms of antigen retrieval

achieved following microwave heating are notestablished but it is possible that microwavesdisrupt the cross-linking of proteins in a similarmanner to enzymatic predigestion.'8 An im-portant consideration when undertaking mi-crowave heating is whether normally maskedantigens may be exposed, resulting in falsepositive results with previously trusted anti-bodies.'9 The only unexpected binding patternwe have identified in the present study is pos-itive cytoplasmic staining of plasma cells byNSE on plastic embedded material. Positivestaining of plasma cells for NSE was not seenin sections cut from paraffin wax embeddedmaterial. This unexpected immunohisto-chemical stainingreaction was potentially mis-leading in bone marrow trephine biopsy speci-mens from two patients with neuroblastoma.A diagnosis of bone marrow involvement bymetastatic neuroblastoma could have beenrendered and inappropriate treatment in-stigated had it not been appreciated that the

843

on April 8, 2021 by guest. P

rotected by copyright.http://jcp.bm

j.com/

J Clin P

athol: first published as 10.1136/jcp.48.9.840 on 1 Septem

ber 1995. Dow

nloaded from

McCluggage, Roddy, Whiteside, Burton, McBride, Maxwell, et al

small numbers of cells exhibiting positive stain-ing for NSE were, in fact, mature plasma cells.

In conclusion, the results of the study con-firm that prior microwave heating results inunmasking of antigens and permits adequateimmunohistochemical staining of plastic em-bedded material with a range of antibodies.Only with occasional antibodies did microwaveheating not result in positive staining. Priormicrowave heating may be of use in exposingantigens for immunohistochemical staining ofbone marrow trephine biopsy specimens em-bedded in a variety of plastic media. Only byundertaking a study such as ours can the use

of microwave antigen retrieval be assessed. Webelieve that the ability to perform immuno-histochemical reactions together with the un-

doubted superior cytological detail make plasticembedding the method of choice for processingbone marrow trephine biopsy specimens. Thestudy adds to the growing list of tissue typeson which prior microwave heating results inimproved immunohistochemical staining.

1 Gatter KC, Heryet A, Brown DC, Mason DY. Is it necessaryto embed bone marrow biopsies in plastic for haema-tological diagnosis? Histopatholog 1987;11:1-7.

2 Vincic L, Weston S, Ridell RH. Bone core biopsies. Plasticor paraffin? Am JT Surg Pathol 1989;13:329-34.

3 Islam A, Frisch B. Plastic embedding in routine histology I:Preparation of semi-thin sections of undecalcified marrowcores. Histopathology 1985;9:1263-74.

4 Islam A, Henderson ES. Glycol methacrylate embeddingfor light microscopy I: Enzyme histochemistry of semi-thin sections of undecalcified marrow cores. Clin Pathol1987;40: 1194-200.

5 Islam A, Archimbaud E, Henderson ES, Han T. Glycolmethacrylate embedding for light microscopy II: Im-

munohistochemical analysis of semi-thin sections of un-decalcified marrow cores. J Clin Pathol 1988;41:892-6.

6 Casey TT, Cousar JB, Collins RD. A simplified plasticembedding and immunohistologic technique for im-munophenotypic analysis of human haematopoietic andlymphoid tissues. Am J Pathol 1988;31:183-9.

7 Burgio VL, Pignoloni P, Baroni CD. Immunohistology ofbone marrow: a modified method of glycol-methacrylateembedding. Histopathology 1991;18:37-43.

8 Murray GI. Is wax on the wane? J Pathol 1988;56:187-8.9 Cattoretti G, Pileri S, Parravicini C, Becker MHG, Poggi

S, Bifulco C, et al. Antigen unmasking on formalin-fixed,paraffin-embedded tissue sections. J Pathol 1993;171:83-98.

10 Cuevas EC, Bateman AC, Wilkins BS, Johnson PA, WilliamsJH, Lee AHS, et al. Microwave antigen retrieval in im-munocytochemistry: a study of 80 antibodies. J Clin Pathol1994;47:448-52.

11 Norton AJ. Microwave oven heating for antigen unmaskingin routinely processed tissue sections. J Pathol 1993;171:79-80.

12 Charalambous C, Singh N, Isaacson PG. Immuno-histo-chemical analysis of Hodgkin's disease using microwaveheating. J. Clin Pathol 1993;46:1085-8.

13 Reynolds GM, Rowlands DC, Young JA. The use of micro-wave antigen retrieval for immunostaining in cyto-pathology [abstract]. J Pathol 1994;173:192.

14 Gerdes J, Becker MHG, Key G, Cattoretti G. Im-munohistological detection oftumour growth fraction (Ki-67 antigen) in formalin-fixed and routinely processedtissues. J Pathol 1992;168:85-7.

15 Burgio VL, Morra E, Ascari E. Histochemistry on glycol-methacrylate embedded human bone marrow biopsies. Anew method and first evaluations. Haematologica 1982;67:177-87.

16 Westen H, Bainton DF. Association ofalkaline-phosphatase-positive reticulum cells in bone marrow with granulocyticprecursors. J Exp Med 1979;150:919-37.

17 Beckstead JH, Halverson PS, Ries CA, Bainton DF. Enzymehistochemistry and immunohistochemistry on biopsyspecimens of pathologic bone marrow. Blood 1981;57:1088-98.

18 Shi SR, Key ME, Kalra KL. Antigen retrieval in formalin-fixed, paraffin-embedded tissues: An enhancementmethod for immunohistochemical staining based on mi-crowave oven heating of tissue sections. J Histochem Cy-tochem 1991;39:741-8.

19 McKee PH, Hobbs C, Hall PA. Antigen retrieval by mi-crowave irradiation lowers immunohistological detectionthresholds. Histopathology 1993;23:377-9.

844

on April 8, 2021 by guest. P

rotected by copyright.http://jcp.bm

j.com/

J Clin P

athol: first published as 10.1136/jcp.48.9.840 on 1 Septem

ber 1995. Dow

nloaded from