Journal of Clinical Pathology, 1978, 31, 454460

Alkaline phosphatase and peroxidase for doubleimmunoenzymatic labelling of cellular constituentsD. Y. MASON AND RITA SAMMONS

From the University of Oxford, Department ofPathology, Radcliffe Infirmary, Oxford, UK

SUMMARY The use of alkaline phosphatase in an immunoenzymatic procedure for the localisationof antigens in paraffin sections or cell smears is described. The results of this method, when appliedto the detection of immunoglobulins, lysozyme, or lactoferrin, were comparable in intensity andclarity to those obtained with the PAP immunoperoxidase procedure. Furthermore, double im-munoenzymatic labelling (with alkaline phosphatase and peroxidase) of two cellular constituentsin a tissue section is possible, the brown peroxidase reaction product contrasting well with the bluealkaline phosphatase product. Since the two antibody 'sandwiches' are applied simultaneouslyrather than sequentially the total duration of this double immunostaining procedure is only a fewminutes longer than that required for detection of a single antigen.

It was also found that the unlabelled antibody immunohistological procedure (whether used inconjunction with alkaline phosphatase or peroxidase) can be shortened without loss of sensitivityby carrying out two of the incubation steps simultaneously.

Immunoperoxidase techniques for the detection ofantigens in human tissues have gained wide accep-tance in recent years because of their superiority,in terms of histological detail, over immunofluores-cent methods. Furthermore, it has become apparentthat a variety of antigens (eg immunoglobulin,lysozyme, hepatitis B antigen, CEA, and hormones)can be revealed by immunoperoxidase stainingof paraffin-embedded routine histological material(for references, see Mason and Taylor (1977)),thus greatly increasing the scope for retrospectiveimmunohistological studies of pathological tissuesamples.A number of techniques have been used for

linking peroxidase to specific antibody. The methodadopted in the earliest studies was directly tocouple the enzyme to antibody via covalent bonds(Avrameas and Uriel, 1966; Nakane and Pierce,1966). A subsequently developed approach (the'unlabelled antibody' method) involves the use ofan intermediate antibody acting bivalently to linkthe primary antibody to immune complexes ofperoxidase and anti-peroxidase (Mason et al.,1969; Sternberger and Cuculis, 1969). A modi-fication of this technique, the PAP method of

Received for publication 28 November 1977

Sternberger et al. (1970), is probably the mostsensitive immunoperoxidase method currently avail-able.

In addition to high sensitivity, unlabelled antibodytechniques offer the advantage that no chemicalmodification of either enzyme or antibody isnecessary in the preparation of reagents, thusminimising loss of peroxidase or antibody activity.The present paper describes an unlabelled

antibody procedure in which alkaline phosphataseis used to reveal antigens in paraffin sections orblood smears. By combining this procedure withimmunoperoxidase staining, two antigens in thesame section can be labelled simultaneously incontrasting colours. Since the two antisera 'sand-wiches' are applied at the same time rather thansequentially, the only addition to the length of theprocedure (compared to the single immunoenzy-matic technique) is the histochemical step for thesecond enzyme, lasting approximately five minutes.

Material and methods

BUFFERSTris buffered saline (05 M Tris-HCJ, pH 7-6diluted 1:10 with 0-15 M saline) was used throughoutas diluent for antisera, enzymes, and peroxidasesubstrate and for washing slides between incubation

454

group.bmj.com on July 6, 2017 - Published by http://jcp.bmj.com/Downloaded from

Alkaline phosphatase andperoxidase for double immunoenzymatic labelling of cellular constituents

steps. The alkaline phosphatase substrate wasmade up in Tris-HCI buffer (see below).

HISTOCHEMICAL REAGENTSDiaminobenzidine, naphthol AS phosphate, andnaphthol AS-MX were obtained from SigmaChemical Co, diazonium salts from Sigma orRaymond Lamb, and horseradish peroxidase(grade 1) from Boehringer.

ALKALINE PHOSPHATASECalf intestinal alkaline phosphatase (Sigma) wasobtained as a crude preparation (1-1 units/mgsolid, type 1) and as a highly purified grade (1075units/mg protein, type VII).

ANTISERAAntisera were obtained from Dakopatts A/s(rabbit anti-human kappa and lambda light chains,rabbit anti-human lactoferrin, rabbit anti-humanlysozyme, swine anti-rabbit IgG) or Miles Labora-tories (goat anti-human lysozyme, goat anti-humankappa and lambda light chains, donkey anti-goatIgG, sheep anti-rabbit IgG, goat anti-peroxidase:peroxidase complexes). The specificity of the Dakc-patts antisera has been established during extensiveuse in this laboratory for immunoperoxidasestaining (Taylor and Burns, 1974; Taylor and Mason,1974; Mason and Taylor, 1975; Mason et al.,1975). The Miles antisera gave staining reactionsidentical with those of corresponding Dakopattsantisera.

Rabbit anti-alkaline phosphatase antiserum wasobtained by immunising a rabbit subcutaneouslywith four injections of 230 ,tg of alkaline phos-phatase (type VII) in Freund's incomplete adjuvantat fortnightly intervals. Serum samples obtainedon bleeding six and 14 days after the last injectionwere pooled. A small amount of enzyme (2-5 mg)was coupled to 0(25 ml of CNBr activated Sepharose4B (Pharmacia), and this immunoabsorbent wasthen used to purify specific anti-alkaline phospha-tase antibodies from 0-5 ml aliquots of cruderabbit antiserum. The eluting buffer was pH 2'50'1 M glycine HCI, and the total yield (over sevenabsorption/elution cycles) was 1'3 mg of antibody/ml of antiserum. The antibody was concentrated to6 mg/ml and stored at - 20°C.A similar immunisation schedule was used to

raise an anti-horseradish peroxidase antiserum.

TISSUE SAMPLES AND BLOOD SMEARSStaining reactions were carried out on paraffinsections of aspirated human bone marrow fromcases of myeloma and other lymphoproliferativediseases (fixed in half-strength Zenker's solution),

on paraffin sections of rectal biopsies and tonsils(fixed in formol saline), and on paraffin sections of aHodgkin's disease lymph node (fixed in Carnoy'sfluid). Normal human blood smears were fixed inbuffered formol acetone, as previously described(Mason et al., 1975).

IMMUNOPEROXIDASE STAININGAntigens in tissue sections or cell smears werestained as described previously (Mason and Taylor,1975; Mason et al., 1975) by the peroxidase-anti-peroxidase (PAP) technique. In summary, thisprocedure involves applying to dewaxed paraffinsections or to cell smears the following antiserumsequence: goat anti-human light chains (or goatanti-human lysozyme), donkey anti-goat IgG, and,finally, goat PAP complexes. Each stage lasts 30minutes and is followed by a wash in bufferedsaline. Peroxidase activity is revealed using diamino-benzidine and H202 as substrate, and slides arecounterstained when necessary with haematoxylin.In control slides, normal rabbit serum was sub-situted for the primary antiserum, giving negativecellular staining reactions; see references citedunder Antisera for further details of specificitycontrols. In some experiments tissue sectionswere pretreated with pepsin (4 mg/ml in 0'01 MHCl at 370C for 15-20 minutes) before the antiserawere applied in order to diminish backgroundstaining (Reading, 1977) and to enhance plasmacell reactivity (Huang et al., 1976).

Table 1 lmmunoenzymatic technique for labelling cellconstituents with alkaline phosphatase

I Rabbit anti-immunoglobulin or lactoferrin2 Swine or sheep anti-rabbit IgG3 Rabbit anti-alkaline phosphatase4 Alkaline phosphatase5 Fast Blue or Fast Red plus naphthol AS phosphate in pH

9-0 Tris buffer

Sections are prepared for application of antisera by dewaxing inxylol and incubation for 10 minutes in normal swine serum (1/5).Blood smears are fixed in formol acetone.Incubation stages 1-4 each last 30 minutes and are followed by washesin Tris buffered saline pH 7-6. Stage 5 lasts approximately fourminutes, and slides are then washed in tap water and mounted inApathy's mountant.

IMMUNOENZYMATIC LABELLING WITHALKALINE PHOSPHATASE (Table 1)The first stages of this procedure were, in principle,identical with the immunoperoxidase technique,slides being treated with anti-human light chainantiserum (or rabbit anti-lactoferrin when stainingblood smears) followed by swine or sheep anti-rabbit IgG. Slides were then treated with anti-alkaline phosphatase antibody (either as whole

455

group.bmj.com on July 6, 2017 - Published by http://jcp.bmj.com/Downloaded from

D. Y. Mason and Rita Sammons

antiserum or as affinity purified specific antibody)followed by alkaline phosphatase. (In some ex-periments -see results -these two stages werecondensed into a single step, ie, anti-alkalinephosphatase and alkaline phosphatase were mixedand applied together to the slides.) As in the im-munoperoxidase procedure, all incubation stageslasted 30 minutes and were followed by washing inbuffered saline. A substrate solution was preparedby dissolving Fast Blue BBN or Fast Red TR(04 - 1.0 mg/ml) in Tris HCl buffer, 0'1 M pH9 0, containing 0 1 mg/ml naphthol AS phosphate,which was filtered directly onto the slides. Inaddition to the Fast Blue and Fast Red salts, thefollowing diazonium salts were also tested: FastViolet B, Fast Blue RR, Fast Black B, and FastRed Violet LB. In some experiments the sub-stituted naphthol AS-MX (Stage and Avrameas,1976) was compared to naphthol AS phosphate.After completion of the alkaline phosphatasereaction, slides were washed in tap water, andmounted, with or without prior counterstaining,in Apathy's aqueous mountant.

In some experiments Gomori's technique forrevealing alkaline phophatase was assessed as analternative to the naphthol AS phosphate method(Burstone, 1962). Slides were counterstained withhaematoxylin, dehydrated, and mounted in DPX.

DOUBLE IMMUNOENZYMATIC STAININGWhen staining two antigens simultaneously (usingboth peroxidase and alkaline phosphatase as labels)the two antisera or enzymes for each stage weremixed so as to give the final concentrations foundto be appropriate for the individual reagents inpreliminary single antigen staining experiments.Slides were then treated with DAB and H202 toreveal peroxidase labelling, rinsed, and incubatedin the alkaline phosphatase substrate. A repre-sentative experiment is outlined in Table 2.

Results

ALKALINE PHOSPHATASE IMMUNOCYTO-CHEMICAL STAININGThe unlabelled antibody procedure when evaluatedon peripheral blood smears (for the detection ofpolymorph lactoferrin) or on tissue sections (forthe detection of plasma cell immunoglobulin ormyeloid cell lysozyme) gave clear cytoplasmicstaining (Figs 1, 2, 5). This reaction closely re-sembled in intensity, cellular localisation, and levelof background the patterns obtained for theseconstituents by the PAP immunoperoxidase pro-cedure.

Table 2 Technique for staining two populations ofplasma cells in bone marrow sections

23

45

67

Details of a representative experiment:Dewaxed sections incubated in methanol plus 0.3 % H O,Normal swine serum 1/5Rabbit anti-lambda light chains 1/100 plus goat anti-kappa

light chains 1/30Swine anti-rabbit IgG 1/25 plus donkey anti-goat IgG 1/5Rabbit anti-alkaline phosphatase 1/500 plus alkaline phosphatase(5.5 U/ml) plus goat PAP 1/50Diaminobenzidine (0-6 mg/ml) plus 0-01% HOlNaphthol AS phosphate (0.1 mg/ml) plus Fast Blue BBN

(1-0 mg/ml)

Stages 1, 3, 4. and 5 last 30 minutes. Stage 2 lasts 15 minutes andstages 6 and 7 eight and four minutes respectively. Slides are washedin buffered saline after stages 1-6.Results: Lambda plasma cells: Blue

Kappa plasma cells: Brown

CHOICE OF SUBSTRATE CONDITIONSWhen using naphthol AS phosphate optimalstaining was obtained at pH 9-0 in combinationwith Fast Blue BBN, Fast Red TR, or Fast Red-Violet LB salts at 10 mg/ml. Fast Red TR was themost satisfactory stain for general use since it gavean intense red reaction product which contrastedwell with haematoxylin (Figs 1 and 5). For doubleimmunostaining, however (see below), Fast BlueBBN was preferable because of its clear contrastwith the brown DAB peroxidase reaction product(Figs 3, 4, 6). The substituted naphthol AS-MX wascomparable in staining intensity to naphthol ASphosphate.

Intense staining of alkaline phosphatase labelledcells was also obtained by Gomori's method,which yielded a black deposit of cobalt sulphide.However, this reaction, although of similar intensityto the Fast Blue and Fast Red reactions, did notstain cell cytoplasm as evenly as did the diazoniumsalts (or the immunoperoxidase procedure), althoughit had the advantages of being compatible withdehydration and mounting in DPX, and of contrast-ing well with haematoxylin counterstaining.

INCUBATION OF SECTIONS WITH ANTI-ALKALINE PHOSPHATASE AND ALKALINEPHOSPHATASEThe third and fourth stages of the immunoenzymaticsequence (application of anti-enzyme antibodyfollowed by incubation with enzyme) were initiallyperformed using affinity purified anti-alkalinephosphatase antibodies (see Methods) and highlypurified alkaline phosphatase (1075 units/mg).These reagents gave good staining when used athigh dilution (eg, less than 1 ,ug/ml). However,whole anti-alkaline phosphatase antiserum incombination with the crude enzyme preparation

456

group.bmj.com on July 6, 2017 - Published by http://jcp.bmj.com/Downloaded from

Alk-alinie phosphatase anidpero.xviase ,for double immunloenizy,}matic labellinig qf celluilar contstituienits

P ..

Q1

IDc). AS'U0

1IDy1 c

a L~.I

'S1

2..... ....* SW <ra s~~~

tJ* *A

f f fe . 4 ~. 4&,

5 ~ ~ -5

49

4rU

4

.3

0-

'5,

4Q0*, e4bA *

6Fig. I Alkaline phosphatase immncmostaining of normal periphewrl blood neutrophils for Iclttoferrin. Note negativelymiphtoc(tes. Suibstralte: Fast Red TR plus inphthol AS-AIX. C(ounterstainl: Haematoxlini.Fig. 2 AlA-lline phosphatase uninnumostaining ofplasma cells in human tonsil for ka-ppa light chains. Suhstralte: Fa.stRed TR plc/. naphtlol .4S-MX.. Counterstain: Haempiataxy,lini.Fig. 3 Double immuniuostaininig of sante amlple ofj'huimanr to/jAil /or kappa aulc lambda lighlt chains (blue atld bro witIrespectively). Subhstrate: Fast Blue BBN plus taphthol AS-AIX for alkaline phosphatase, dliaminioben:idine 1H202 forperoxidase. No counterstain.Fig. 4 Higher power view of santie section.Fig. 5 Alkaline phosphlatase staininlg of nleutrophil ly.sozyvmne in humltian tonsil. Suhbtralte and counlterstainl as for Fig. 1.Fig. 6 Doubhle staininig of same sample for neutrophil lysozymte (blue) anid plasina cell lambdalt lighlt chains(brown). Suihstrates as for Fig. 3.

457

r

. g.:

Mwru.S.

M

group.bmj.com on July 6, 2017 - Published by http://jcp.bmj.com/Downloaded from

D. Y. Mason and Rita Sammons

(1-1 units/mg) at appropriately higher concentrationsgave equally good staining.

After establishing the conditions for staining bythe four-stage method, the feasibility of adding thethird and fourth stages simultaneously (ratherthan sequentially) was explored. Staining by thisprocedure was identical with that observed in thefour-stage technique, using either purified anti-alkaline phosphatase and alkaline phosphataseor crude preparations of antibody and enzyme.When the three- and four-stage methods were com-pared in combination with a range of dilutions of theprimary anti-human light chain antiserum, the twotechniques were equally sensitive. However, incontrast to the four-stage procedure, it was necessaryin the three-stage technique to use the enzymesolution at a sufficient concentration (relativeto the anti-enzyme antiserum) to ensure that solublecomplexes (in antigen excess) were formed, ratherthan insoluble precipitates (at or near antigen:antibody equivalence). The same observation wasmade when anti-peroxidase antibody and purifiedhorseradish peroxidase were compared in a three-and four-stage procedure.

DOUBLE IMMUNOENZYMATIC STAININGDouble immunoenzymatic staining (Table 2) yieldedclear labelling in contrasting colours of plasma cellscontaining kappa and lambda chains (Figs 3 and 4)or of immunoglobulin positive plasma cells andlysozyme positive myeloid cells (Fig. 6). The colourof each enzyme reaction product was identical withthat obtained in single antigen staining, suggestingthat no cross reactions between the two antisera'sandwiches' were taking place.Although the ability of this procedure to detect

two antigens at the same site (rather than in twodifferent cell populations) was not extensivelyinvestigated, it was noted that whereas individualplasma cells in sections (stained for kappa andlambda chains) were either blue or brown, serumin blood vessels and the cytoplasm of atypicalHodgkin's reticulum cells gave an intermediatemuddy purple colour reaction, presumably repre-senting polyclonal immunoglobulin. A similarcolour was also seen in a proportion of plasma cellswhen sections were stained simultaneously forIgG and kappa or lambda light chains.

Discussion

Although the potential value of alkaline phosphataseas an antibody label was established some yearsago by Avrameas and his colleagues (Avrameas,1969; Ternynck and Avrameas, 1976), horseradishperoxidase has found widespread acceptance as the

enzyme of choice for immunocytochemical applica-tion. The only report in the literature on the useof alkaline phosphatase, apart from the work ofAvrameas' group, is the recent description byDruguet and Pepys (1977) on human lymphocytemembrane immunoglobulin labelling.The present report demonstrates that immuno-

enzymatic labelling of human tissue antigens withalkaline phophatase is similar in clarity and intensityto that obtained with horseradish peroxidasetechniques. The techniques used follow closelythe four-stage unlabelled antibody method firstdescribed in 1969 for use with horseradish peroxidase(Mason et al., 1969; Sternberger and Cuculis,1969). However, it was noted in the present experi-ments that this procedure can be simplified, withoutloss of sensitivity, by adding alkaline phosphataseand anti-alkaline phosphatase antibody in a singlestep (rather than sequentially). In these circumstancesit can be assumed that soluble complexes (in antigenexcess) of enzyme and antibody form and are boundby the bridging antibody. The technique thusresembles the three-stage PAP method (Sternbergeret al., 1970) and shares with the PAP procedurethe advantage that dissociation of antibody andenzyme during washing (which lowers the sensitivityof the four-stage unlabelled antibody methodrelative to the PAP technique (Sternberger, 1974))is minimised.The combination of alkaline phophatase with

peroxidase for double immunoenzymatic labellingof two constituents in a single section has notpreviously been reported. Several groups havereported labelling two cellular constituents by theuse of two contrasting substrates for peroxidase(Nakane, 1968; Vandesande and Dierickx, 1975;Erlandsen et al., 1976; Martin-Comin and Robyn,1976). However, this approach has the disadvantagethat the two immunohistological sandwiches mustbe applied sequentially so that double labelling bythe unlabelled antibody procedure would involvesix more incubation steps than are needed forvisualisation of a single antigen. Furthermore, thisprocedure entails the possibility of unwanted crossreactions and antigen denaturation (Campbell andBhatnagar, 1976).These considerations led Campbell and Bhatnagar

(1976) to develop a double immunoenzymaticstaining procedure in which two different enzymes(peroxidase and glucose oxidase) are covalentlycoupled to separate antisera. However, this pro-cedure also involved sequential rather than simul-taneous application of antibodies, thus doubling thelength of the procedure to a total of 48 hours.Furthermore, the necessity to prepare antibodiescovalently labelled with glucose oxidase, as well as

458

group.bmj.com on July 6, 2017 - Published by http://jcp.bmj.com/Downloaded from

Alkaline phosphatase and peroxidase for double immunoenzymatic labelling of cellular constituents 459

peroxidase coupled antibody, is an additionaltechnical disadvantage.The double labelling method described in the

present report offers several advantages overprevious techniques. The procedure involves onlyone extra incubation step (lasting less than fiveminutes) compared to the PAP immunoperoxidaseprocedure. All reagents, with the exception ofrabbit anti-alkaline phosphatase antiserum, arecommercially available. This antiserum is readilyprepared by a brief immunisation schedule (usinga total of less than 1 mg of enzyme) and, althoughraised against highly purified enzyme, gave identicalimmunocytochemical staining reactions whetherused in combination with purified or with crudeenzyme preparations. In the latter instance theanti-alkaline phosphatase antibody presumablyspecifically selects molecules of pure enzyme and'ignores' contaminating material which was absentfrom the preparation used for immunisation.The existence of a simple and reliable method for

the simultaneous immunoenzymatic demonstrationof two constituents in paraffin-embedded tissue sec-tions or peripheral blood films suggests a numberof applications in a clinicopathological context.Relative ratios of two cell populations and theirtopographical interrelationships can rapidly andaccurately be assessed by this procedure, for example,plasma cells of one immunochemical class versusthose of a second type, or lysozyme-positive cellsversus immunoglobulin-positive cells (of relevancein cases of leukaemia supervening on myeloma).In addition to these applications in the context

of the identification of different cell populations,it is of interest that the double immunoenzymaticprocedure appears to be capable of detecting twoantigens within a single cell. This feature is ofparticular interest in the context of studies oflymphoproliferative disorders such as Hodgkin'sdisease in which the presence of polyclonal immuno-globulin within proliferating cells has been sus-pected from staining of adjacent sections (Taylor,1976) but not directly demonstrated.

Our thanks are due to Mr J. Burns for advice onalkaline phosphate cytochemistry and for providinga number of histochemical reagents; to Dr R. I.Vanhegan for providing tissue samples; to Dr T.Parry for photographic assistance; and to MrsJ. Braidwood for typing the manuscript.We also gratefully acknowledge the generous

contributions of Dakopatts A/s and of the Oxford-shire Area Health Authority (Teaching) ResearchFund towards the cost of colour printing.

This work was partly supported by a grantfrom the Leukaemia Research Fund.

References

Avrameas, S. (1969). Coupling of enzymes to proteinswith glutaraldehyde. Use of the conjugates for thedetection of antigens and antibodies. Immunochemistry,6, 43-52.

Avrameas, S., and Uriel, J. (1966). Methode de marquaged'antigenes et d'anticorps avec des enzymes et sonapplication en immunodiffusion. Comptes Rendus del'Academie des Sciences (Paris) (D), 262, 2543-2545.

Burstone, M. S. (1962). Enzyme Histochemistry, p. 267.Academic Press, London.

Campbell, G. T., and Bhatnagar, A. S. (1976). Simul-taneous visualization by light microscopy of twopituitary hormones in a single tissue section using acombination of indirect immunohistochemical methods.Journal of Histochemistry and Cytochemistry, 24,448-452.

Druguet, M., and Pepys, M. B. (1977). Enumeration oflymphocyte populations in whole peripheral bloodwith alkaline phosphatase-labelled reagents. Clinicaland Experimental Immunology, 29, 162-167.

Erlandsen, S. L., Hegre, 0. D., Parsons, J. A., McEvoy,R. C., and Elde, R. P. (1976). Pancreatic islet cellhormones: distribution of cell types in the islet andevidence for the presence of somatostatin and gastrinwithin the D cell. Journal of Histochemistry andCytochemistry, 24, 883-897.

Huang, S. N., Minassian, H., and More, J. D. (1976).Application of immunofluorescent staining on paraffinsections improved by trypsin digestion. LaboratoryInvestigation, 35, 383-390.

Martin-Comin, J., and Robyn, C. (1976). Comparativeimmunoenzymatic localization of prolactin and growthhormone in human and rat pituitaries. Journal ofHistochemistry and Cytochemistry, 24, 1012-1016.

Mason, D. Y., Farrell, C., and Taylor, C. R. (1975). Thedetection of intracellular antigens in human leucocytesby immunoperoxidase staining. British Journal ofHaematology, 31, 361-370.

Mason, D. Y., and Taylor, C. R. (1975). The distributionof muramidase (lysozyme) in human tissues. Journalof Clinical Pathology, 28, 124-132.

Mason, D. Y., and Taylor, C. R. (1977). Distribution oftransferrin, ferritin, and lactoferrin in human tissues.Journal of Clinical Pathology, 31, 316-327.

Mason, T. E., Phifer, R. F., Spicer, S. S., Swallow, R. A.,and Dreskin, R. B. (1969). An immunoglobulin-enzyme bridge method for localising tissue antigens.Journal of Histochemistry and Cytochemistry, 17,563-569.

Nakane, P. K. (1968). Simultaneous localization ofmultiple tissue antigens using the peroxidase-labeledantibody method: a study on pituitary glands of therat. Journal of Histochemistry and Cytochemistry,16, 557-560.

Nakane, P. K., and Pierce, G. B., Jr (1966). Enzymelabeled antibodies: preparation and application forthe localization of antigens. Journal of Histochemistryand Cytochemistry, 14, 929-93 1.

Reading, M. (1977). A digestion technique for the reduc-tion of background staining in the immunoperoxidase

group.bmj.com on July 6, 2017 - Published by http://jcp.bmj.com/Downloaded from

460 D. Y. Mason and Rita Sammons

method. Journal of Clinical Pathology, 30, 88-90.Stage, D. E., and Avrameas, S. (1976). Detection ofAnti-TNP antibody forming cells (AFC) with TNP-enzyme and TNP-Fab anti-enzyme conjugates.Journal of Immunological Methods, 10, 105-118.

Sternberger, L. A. (1974). Immunochemistry, p. 182.Prentice Hall, New Jersey.

Stemberger, L. A., and Cuculis, J. J. (1969). Method forenzymatic intensification of the immunocytochemicalreaction without use of labeled antibodies (Abstract).Journal of Histochemistry and Cytochemistry, 17, 190.

Sternberger, L. A., Hardy, P. H. Jr., Cuculis, J. J., andMeyer, H. G. (1970). The unlabeled antibody methodof immunohistochemistry. Journal of Histochemistryand Cytochemistry, 18, 315-333.

Taylor, C. R. (1976). An immunohistological study offollicular lymphoma, reticulum cell sarcoma andHodgkin's disease. European Journal of Cancer, 12,61-75.

Taylor, C. R., and Burns, J. (1974). The demonstrationof plasma cells and other immunoglobulin-containingcells in formalin-fixed, paraffin-embedded tissues

using peroxidase-labeled antibody. Journal of ClinicalPathology, 27, 14-20.

Taylor, C. R., and Mason, D. Y. (1974). The immuno-histological detection of intracellular immuno-globulin in formalin-paraffin sections from multiplemyeloma and related conditions using the immuno-peroxidase technique. Clinical and ExperimentalImmunology, 18, 417-429.

Ternynck, T., and Avrameas, S. (1976). A new methodusing P-benzoquinone for coupling antigens andantibodies to marker substances. Annales d'Im-munologie, 127c, 197-208.

Vandesande, F., and Dierickx, K. (1975). Identificationof the vasopressin producing and of the oxytocinproducing neurones in-the hypothalamic magnocellularneurosecretory system of the rat. Cell and TissueResearch, 164, 153-162.

Requests for reprints to: Dr D. Y. Mason, HarknessLaboratory, The Radcliffe Infirmary, Oxford, OX2 6HE.

group.bmj.com on July 6, 2017 - Published by http://jcp.bmj.com/Downloaded from

cellular constituents.immunoenzymatic labelling ofperoxidase for double Alkaline phosphatase and

D Y Mason and R Sammons

doi: 10.1136/jcp.31.5.4541978 31: 454-460 J Clin Pathol 

http://jcp.bmj.com/content/31/5/454Updated information and services can be found at:

These include:

serviceEmail alerting

online article. article. Sign up in the box at the top right corner of the Receive free email alerts when new articles cite this

Notes

http://group.bmj.com/group/rights-licensing/permissionsTo request permissions go to:

http://journals.bmj.com/cgi/reprintformTo order reprints go to:

http://group.bmj.com/subscribe/To subscribe to BMJ go to:

group.bmj.com on July 6, 2017 - Published by http://jcp.bmj.com/Downloaded from