Alpha-1-antitrypsin in human macrophages · Alpha-1-antitrypsin in humanmacrophages P ISAACSON, DBJONES, GHMILLWARD-SADLER, MAJUDD, S PAYNE ... nant lymphomas3 in the hopethat positive

J Clin Pathol 1981 ;34:982-990

Alpha-1-antitrypsin in human macrophagesP ISAACSON, DB JONES, GH MILLWARD-SADLER, MA JUDD, S PAYNE

From the Department ofPathology, Southampton University Medical School, Southampton S09 4XY

SUMMARY Preliminary studies have suggested that alpha-1-antitrypsin (AlAT) is a useful immuno-histochemical marker of histiocytes (monocytes/macrophages) and malignant tumours derived fromthem. To confirm the reliability ofthis marker a wide variety of benign and malignant lymphoreticularcells and tissues have been stained by the immunoperoxidase technique for AIAT and positivestaining was found to be confined to histiocytes. Immunodiffusion, isotope labelling, and isoelectricfocusing studies performed on cell lysates confirmed that the positive staining shown by monocytesand malignant histiocytes is due to the presence of AlAT identical with serum AlAT and that thismaterial is synthesised by these cells rather than taken up from their environment. Positive immuno-peroxidase staining for AlAT is thus a reliable marker oflymphoreticular neoplasms of true histiocyticorigin.

The development of immunologically definedmarkers of cell lineage has been largely responsiblefor recent advances in the understanding and classi-fication of lymphoreticular malignancies. Thus, bothin cell suspensions and, more importantly, in histo-logical sections, lymphocyte subsets can now beconfidently identified. Identification of macrophages(histiocytes) poses special problems however. Sus-pensions ofmacrophages are difficult to prepare fromtissue and there is no consistent immunologicalmarker for cells of the monocyte/macrophage seriesthat can be applied to histological sections. Lysozymecan be identified immunologically and, while it is agood marker ofbenign reactive macrophages in tissuesections, it has, contrary to early reports,1 proved tobe a poor marker of malignant macrophages.2 Con-sistent positive immunoperoxidase staining for alpha-1-antitrypsin (AIAT) in Kupffer cells was noted byone of us (GHMS) in the course of screening a largenumber of formalin-fixed paraffin embedded liverbiopsies for AlAT deficiency. The possibility thatthis might be a good immunological marker ofmacrophages was explored by staining a variety oftissues for AlAT including lung, spleen, and lymphnode; in each case tissue macrophages stained posi-tively. Antiserum to AlAT was therefore added intoa "panel" of antisera used to characterise malig-nant lymphomas3 in the hope that positive stainingforAlAT would distinguish true histiocytic (that is,monocyte/macrophage derived) tumours from thoseof lymphoid origin. Positive staining of malignant

Accepted for publication 18 February 1981

cells was found to be restricted to histiocytic tumoursand has been particularly useful in characterising agroup of intestinal lymphomas as a variant of malig-nant histiocytosis.4 This series of rather empiricalobservations raises several important questions.Firstly, are the routine specificity controls that wereused in the immunohistochemical procedure suffi-cient to ensure that the substance demonstrated inmacrophages is truly AIAT? Secondly, assuming thatthe material present is AlAT, does its presence inmacrophages reflect synthesis by the cells or uptakefrom their environment? Thirdly, how reliable isAIAT staining as a marker of the histiocytic origin ofa malignant lymphoma, and in particular do T-cells,which may resemble macrophages histochemicallyand immunologically,5 ever stain positively forAIAT? We have attempted to answer these questionsby extracting material from benign and malignantmacrophages and showing that it is identical withserum AIAT and, with the use of radiolabellingtechniques, showing that the cells synthesise thismaterial. Finally, we have stained a wide variety ofbenign and malignant lymphoreticular cells andtissues to show that positive staining for AlAT isconfined to macrophages and tumours derived fromthem.

Material and methods

PREPARATION OF BLOOD CELLS

Peripheral blood mononuclear cells were separatedfrom heparinised venous blood by the method de-scribed by Payne et al.6 Briefly, freshly drawn blood

982

copyright. on June 3, 2020 by guest. P

rotected byhttp://jcp.bm

j.com/

J Clin P

athol: first published as 10.1136/jcp.34.9.982 on 1 Septem

ber 1981. Dow

nloaded from

http://jcp.bmj.com/

Alpha-l-antitrypsin in human macrophages

was mixed with an equal volume of calcium andmagnesium-free Hanks' balanced salt solution(CMF-HBSS) and layered over ficoll/triosil.7 Mono-nuclear cells were removed from the interface aftercentrifugation at 400 g for 30 min, washed threetimes in CMF-HBSS and taken into medium RPMI1640 supplemented with 10% heat inactivated fetalbovine serum (SRPMI). All tissue culture reagentswere obtained from Gibco Biocult UK Ltd (Washing-ton Road, Paisley, Scotland).

T-ENRICHMENT OF LYMPHOCYTESEnrichment of T-lymphocytes was carried out by amodification of the method of Wahl et al.8 Mono-nuclear cells at 2 X 106/ml in RPMI 1640 wererosetted with 1% sheep red blood cells (SRBC) inthe presence of 2% fetal bovine serum. After incu-bating at 37°C for 5 min and further incubation onice overnight or for 4 h the cells were gently re-suspended. Rosetted and unrosetted lymphocyteswere separated by centrifugation through ficoll/trio-sil; erythrocytes were lysed with distilled water beforewashing.

PREPARATION OF BLOOD MONOCYTESTo prepare cell suspensions enriched in blood mono-cytes a maximum of 5 X 107 washed whole mono-nuclear cells were taken into 2 ml of isotonic Percol(PharmaciaUK Ltd, Hounslow, Middlesex) adjusted

to a density of 1-050 with 10 X CMF HBSS and1 M Hepes buffer, pH 7 0. The cells were then layeredover a discontinuous gradient of Percol of density1 -060 or 1 -070 and centrifuged at 400 g for 30 min atroom temperature. The top density fraction of thisgradient after centrifugation consisted almost entirelyof a-napthyl-acetate esterase positive monocytes(Fig. 1). This fraction was used after washing forimmunohistochemical and immunodiffusion studies.For isotope incorporation the monocyte fraction wasfurther purified by adherence to tissue culture gradeplastic surfaces.

MITOGEN STIMULATIONBlood mononuclear cells or T-lymphocytes werecultured for three days at a cell concentration of2 x 106/ml in the presence of Phytohaemagglutinin,(PHA, 1/50, Gibco); pokeweed mitogen (PWM;1/100, Gibco); or concanavalin A (Con-A; 10 jsg/ml,Miles Laboratories Ltd, Slough, England). Afterculture cells were washed once in SRPMI and pre-pared for immunocytochemical examination.

IMMUNODIFFUSION STUDIESApproximately 2 x 108blood monocytesorcells froma human histiocytic lymphoma cell line, U937,9 werefrozen and thawed three times in non-ionic detergent,0-1 % NP40 (BDH, Poole, England), centrifuged at2000 g for 20 min and concentrated twentyfold.

Fig. 1 Cytocentrifuge preparation ofperipheral blood monocytes prepared using a Percolgradient. This representative field demonstrates the homogeneity of the monocyte population.Wright-Giemsa x 1000.

983

.................F



j.com/

J Clin P


ber 1981. Dow

nloaded from

http://jcp.bmj.com/

984

Concentrated lysate was tested by immunodiffusionin gels of 0-6 % agarose containing 4%polyethyleneglycol 6000 against antibody to AlAT and lysozyme.

Culture medium containing fetal calf serum sup-

plement was unreactive to this system.

ISOTOPE INCORPORATIONPurified blood monocytes or U937 cells were culturedat a concentration of 2 x 106 cells/ml for 48 h inleucine-free medium containing 10% dialysed fetalcalf serum and 3H-leucine to a final concentration of25 jci/ml (TRK170; Radiochemical Centre, Amer-sham, England). After culture cells were lysed as

above and dialysed against a large volume ofphosphate-buffered saline (PBS). Isoelectric focus-ing10 (IEF) of the lysate was carried out for 45 minon a Pharmacia FBE300 flatbed apparatus in gels ofLKB Agarose-EF (0 3 g in 27 ml distilled water) with3-6g sorbitol and 2 ml ampholines, pH 3 5-10.Focused material was immunofixed into the gel witha 1/5 dilution of the antiserum to AlAT, the gelswere then washed in PBS and either stained in CBBR250 in acetic acid :methanol :water (1:4:4) or sliced,dissolved in NCS (Amersham Corporation, Arling-ton Heights, Illinois, USA) and 1-counted in scintil-lant cocktail-T (Hopkin and Williams, ChadwellHeath, Essex).

IMMUNOHISTOCHEMISTRY

Malignant lymphomasIn addition to 185 cases of malignant lymphomapreviously described3 sections of formalin-fixedparaffin-embedded lymph node tissue from six cases

of T-cell lymphoma were stained for AlAT. Theseincluded two cases of Sezary's syndrome, three largecell lymphomas shown to be of T-cell origin by Erosetting performed on cell suspensions and one case

of nodal involvement by T-cell chronic lymphocyticleukaemia (CLL).

Cytological preparationsThese consisted of cytocentrifuge slides and paraffinsections of pellets of formalin-fixed cells prepared as

previously described.3 Cells prepared in this way andstained for AIAT included peripheral blood mono-cytes, U937 cells, cells from a case of human T-cellCLL, three human T-cell lymphoma cell lines (Molt4, HSB-2 CCRF-CEM) and T-enriched peripheralblood lymphocytes transformed by culture withPHA, PWM, and Con-A.

Immunohistochemical stainingStaining for AlAT was carried out on all tissues,cell pellets and cytocentrifuge preparations by meansof the PAP immunoperoxidase technique. Sections

Isaacson, Jones, Millward-Sadler, Judd, Payne

of paraffin-embedded tissue and cell pellets were firsttreated with 01% trypsin as previously described."This step was omitted from cytocentrifuge prepara-tions. Staining then followed the standard PAPprocedure. Antiserum to AIAT was obtained fromtwo sources (Behringwerke AG, Germany, andDako-immunoglobulins AS, Denmark) which gavelines of complete identity when compared againstnormal human serum or AlAT standard (Sigma) byimmunodiffusion and both antisera were used at adilution of 1/1000. Controls were stained with anti-serum previously absorbed with partially purifiedAlAT (Sigma).

Results

These are summarised in the Table.

IMMUNOHISTOCHEMISTRYThe results of staining 185 cases of malignantlymphoma for AlAT have been previously described3and will be briefly summarised. With the exceptionof some Reed-Sternberg cells, positive staining forAlAT was found only in those tumours judged onmorphological, histochemical or immunologicalgrounds to be of true histiocytic derivation. Most ofthese were examples of malignant histiocytosis of theintestine.'2 The positive staining appeared in themalignant cells as cytoplasmic granules tucked intothe nucleus in the Golgi zone. There was considerablevariability in the amount of material present how-ever. In some cases there was an abundance ofcytoplasmic granules often extending into the cyto-plasm as a whole (Fig. 2) while in others a scatteringof cytoplasmic granules sometimes forming a peri-

AJAT in lymphoreticular cells and tissue

Stain for AlAT

Benign

Malignant

Peripheral blood monocytesTissue macrophages

liverspleenlymph nodelung

B-cellsreactive follicle centrestransformed with PWM

Transformed T-cells(PWM, PHA, CON-A)

U937Histiocytes

histiocytic lymphomaMHI

T-cell CLLT-cell lymphomaT-cell lymphoma

cell linesB-cell lymphomas

*

*Shown to synthesise alpha-l-antitrypsin (AlAT).



j.com/

J Clin P


ber 1981. Dow

nloaded from

http://jcp.bmj.com/


-W:

Ie:; t$F

4.. ji . A

*< rE. F

S.jjw S 9~~~~~~

*- 45 B..,:^i. *t'e,

$ o*;

... t.,.._ *se - ,^. s

5. .,f ,. *

::f;....o xei

S ss

.* '> w.w

@

*#~

.t

tt

4. *.

Fig. 2 (a) Section of intestinal Iymphoma from a case

ofMHI showing numerous large bizarre cells withabundant cytoplasm. Haematoxylin and eosin x 400.(b) stain for AJAT showing granules tucked into thenucleus which, in some cells, extend into the cytoplasmas a whole. Immunoperoxidase x 400.

Fig. 3 (a) A case ofMHI composed of smaller more

monomorphic histiocytes. Haematoxylin and eosin x 400.(b) stain for AJAT showing positively staining materialconcentrated close to the nucleus and a scattering ofgranules elsewhere in the cytoplasm. Immunoperoxidasex 400.

Aw.

1,:

0,

44;

985

IF,:0ik "

... .0. a aw?, .i, 4

*::

-.;M

:,1

imk 4S.-O,I-V

?' A&M1 .401EWA

.0 46



j.com/

J Clin P


ber 1981. Dow

nloaded from

http://jcp.bmj.com/

9saacson, Jones, Millward-Sadler, Judd, Payne

nuclear ring was observed (Fig. 3). Occasionally theamount of positively staining material was very smallindeed and the oil-immersion (x 1000) objective wasrequired to delineate single cytoplasmic granules(Fig. 4). In carrying out this study benign tissueadjacent to malignant lymphoma was frequentlystained. In this tissue which included lymph node,spleen, liver, and lung, reactive macrophages stainedfor AIAT. No staining was observed in otherlymphoreticular cells.

Since none of the 185 cases had been characterisedas a T-cell lymphoma, six cases of proven T-celllymphoma were stained for AlAT as part of thepresent study; all stained negatively, although accom-panying benign macrophages showed positive stain-ing (Fig. 5). Negative staining was also found in cellpellets and cytocentrifuge preparations from the caseof T-cell CLL, the human T-cell lymphoma cell linesand T-enriched human lymphocytes transformedwith PWM, PHA, and Con-A. Freshly isolatedperipheral blood monocytes and U937 cells stainedpositively for AlAT in both cell pellets and cyto-centrifuge preparations. In the monocytes positivestaining was found in approximately 50% of the cellsand was usually confined to an area close to thenucleus (Fig. 6). In U937 cells most of the cellsstained positively but with variable intensity, thefinely granular staining being unevenly distributedthroughout the cytoplasm (Fig. 7). In all cases posi-tive staining could be abolished by prior absorptionof the antiserum with partially purified AlAT.

Fig. 4 (a) Histiocytic lymphoma of spleen composed ofindistinctive immature lymphoreticular cells.Haematoxylin and eosin x 1000. (b) stain for AJAT;high magnification is required to distinguish isolatedgranules ofAIAT situated close to the nucleus (arrows).Immunoperoxidase x 1000.

IMMUNODIFFUSIONA single precipitin line was obtained when lysates ofU937 and peripheral blood monocytes were precipi-tated against the antiserum to AIAT and lines ofidentity were formed with serum AlAT and partiallypurified AIAT (Fig. 8).

ISOELECTRIC FOCUSINGIsoelectric focusing of leucine-labelled lysates ofU937 cells or purified blood monocytes producedlabelled proteins which focused at a pH equivalentto that observed with AlAT standards and whichwere fixed in the gel with antibody specific to AlAT(Fig. 9).

Discussion

Positive immunohistochemical staining with specificantiserum need not necessarily be due to the antigento which the antiserum was raised. Non-specificcross-reactions may occur and even abolition ofpositive staining by absorption of antiserum withspecific antigen, generally accepted as the best con-trol for specificity of immunohistochemical staining,

4q is not entirely reliable since, particularly with highmolecular weight antigens it is difficult to excludecross-reactions due to a few shared antigenic deter-

986

Ae

we

....:la.."4,.



j.com/

J Clin P


ber 1981. Dow

nloaded from

http://jcp.bmj.com/


.A

*

$ a4 .* .

.$ * w

4

#.++4 4t

Fig. 6 Cytocentrifuge preparation ofperipheral bloodmonocytes stainedfor AIAT. Approximately half of thecells stain positively with coarse granules tucked into thenuclear identation. Immunoperoxidase x 1000.

ft~ ~ ~ .. g

1#1 4k ; *

t:

Fig. 5 (a) Lymph node from a case of T-cell CLLshowing malignant infiltrate on left and residualcontaining macrophages on right. Haematoxylin andeosin x 320. (b) stain for AIAT showing positive stainingin sinusoidal macrophages while tumour infiltrate isnegative. Immunoperoxidase x 320.

Fig. 7 Cytocentrifuge preparation of malignanthistiocytes (U937) stainedfor AIAT. There is variablefinely granular staining of the cytoplasm.Immunoperoxidase x 1000.

b_

987

.6

* .. 0.-.

4 -i.. " O'A.0.f'.. .#

.A, 1. X,.40.t

t

m6e!.'i..

........

I:: W.I.A.I.-

.iw"



j.com/

J Clin P


ber 1981. Dow

nloaded from

http://jcp.bmj.com/


Fig. 8 Immunodiffusion plate prepared over 24 h at 4°C.Antibody to alpha-l-antitrypsin is present in the centrewell. The precipitin line shows complete identity betweenthe antigen extracts present. STD: alpha-l-antitrypsinstandard (Sigma); SER: human serum; MO: Percolpurified blood monocytes; U937: extracted U937histiocytic lymphoma cells.

minants. Strong evidence that our antisera are indeedrecognising AlAT in macrophages is obtained fromthe immunodiffusion studies which showed singleprecipitin lines of identity, without spur formation,between lysates of positively staining cells (benignmonocytes, and cultured malignant histiocytes(U937)), human serum and partially purified AlAT.Confirmation that the antiserum is recognising newlysynthesised AIAT in these cell preparations is ob-tained from the isoelectric focusing studies of 3H-labelled lysates of short-term cell cultures. In dialysedpreparations of both blood monocytes and U937cells 3H-leucine was incorporated into material withthe immunological and focusing characteristics ofAlAT.AIAT was first demonstrated in human alveolar

macrophages by Cohen in 197313 by means of animmunofluorescence technique. He noted that theintensity of fluorescence declined over 72 h and in-terpreted this finding as evidence that the macro-phages had taken up plasma AlAT which was thengradually lost from the cells. Gupta et al.14 foundAlAT in the pulmonary macrophages of smokersbut did not discuss their observations in any depth.Initially, we were also ofthe opinion that macrophageAlAT reflected uptake from the plasma, perhaps inthe form of complexes with proteases, but our pre-liminary studies on pellets of blood monocytes andU937 cells caused us to revise our views and suggestthat macrophages synthesised AlAT.'5 This sugges-tion has been confirmed by our present study andalso by the report of Wilson et al.16 who showedisotopically-labelled AIAT in the supernatants of

Fig. 9 Isoelectric focusing of detergent lysates ofperipheral blood monocytes and U937 cells. The shadedarea indicates the position of alpha-l-antitrypsin standardand vertical columns show the level of 1-counts,representing synthesised material fixed in each gel sliceby specific antibody to alpha-l-antitrypsin. Materialpresent in the cell lysate which is of too great a size toleave the loading trough of the gel is indicated by thearrowed thin column. The stained patterns showimmunofixed alpha-l-antitrypsin standard and below thattrichloroacetic acid precipitated whole human serumrun under the same conditions.

short-term cultures of the adherent (that is, mono-cyte) fraction of blood mononuclear cells.

Lipsky et al.17 showed positive staining for AlATon the surface, but not in the cytoplasm, of lympho-cytes transformed with Con-A and incubated withthe antiserum in suspension. Staining of cell surfaceis not successful in formalin-fixed tissues so that con-fusion between macrophages and transformed

6pol

988



j.com/

J Clin P


ber 1981. Dow

nloaded from

http://jcp.bmj.com/


lymphocytes in paraffin sections is unlikely on thisaccount. Our own studies of mitogen-transformedlymphocytes showed no positive staining either incell pellets or in cytocentrifuge preparations whichwere stained on the slide. In addition to monocytesBenitez-Bibiesca & Frere-Horta'8 demonstratedAIAT in granulocytes. In our study we too havenoted strong and apparently specific staining ingranulocytes. These cells are, however, sufficientlydistinctive and unlikely to be confused withmacrophages.When interpreting A1AT staining, care must be

taken not to confuse positive staining caused by non-specific uptake of the antigen from the plasma. Thisoccurs in dead or damaged cells in common withother plasma proteins such as IgG and albumin.3The nature of staining is, however, quite differentfrom thegranular staining of cells synthesising AlAT,being diffuse and non-granular and frequently stain-ing the nucleus as well as the cytoplasm.

Recent advances in the understanding and classi-fication of lymphoreticular tumours have restedheavily on the demonstration of chemical, immuno-logical, and morphological features in malignant cellsthat mimic those of their benign counterparts. To analmost uncanny extent, a normal cell type has beenfound to account for almost every type and subtypeof malignant lymphoreticular cell, and classificationof lymphoreticular malignancies has followedaccordingly. In normal tissues, while certain non-lymphoreticular cells such as granulocytes may con-tain AIAT the macrophage is the only lympho-reticular cell in which we have consistently been abletodemonstrateAlAT. Before being certain, however,that the presence of AIAT is indicative of macro-phage/monocyte origin, a thorough search forAlATin other benign and malignant lymphoreticular cellsis necessary. This applies particularly to T-cellswhich, in any case, may show certain resemblancesto macrophages.5 We have not observed AlAT inB-cell lymphomas of any type, in benign reactivefollicle centres (apart from macrophages), or in B-lymphocytes transformed with pokeweed mitogen.3Our inability to demonstrate AIAT in T-cell lymph-omas, T-cell lymphoma cell lines, T-cell CLL cellsand T-cells transformed withPHA and Con-A makesit highly unlikely that AlAT positive malignantlymphomas could be of T-cell origin. At present,therefore, it would seem justifiable to classify casesin which the cells stain positively for AlAT as of truehistiocytic (monocyte/macrophage) origin. Our re-cent report of A1AT in some Reed-Steinberg cells3is of interest in this respect. This finding, alsoreported by Poppema," suggests a histiocyticderivation of these cells.The role of macrophage A1AT is presumably one

of neutralisation of proteases either intracellularly orafter their release. The relation of macrophage AlATto the AlAT deficiencies needs to be explored as doesthe possibility of a relation to certain inflammatorydiseases. At present, however, we can state with con-fidence that human macrophages synthesise AIATand that the presence of A1AT in the cytoplasm ofmalignant lymphoreticular cells is a reliable indicatorof their histiocytic (monocyte/macrophage) deriva-tion.

We are grateful to Dr C Sundstrom of the Depart-ment of Pathology, University of Uppsala, Uppsala,Sweden, for the supply of U937 cells and to OliveHuber for preparation of the manuscript.

References

Taylor CR. An immunohistological study of follicularlymphoma, reticulum cell sarcoma and Hodgkin'sdisease. Eur J Cancer 1976:12:61-75.

2 Risdall RJ, Sibley RK, McKenna RW, Brunning RD,Dehner LP. Malignant histiocytosis: a light and electronmicroscopic and histochemical study. Am J Surg Pathol1980;4:439-50.

3 Isaacson R, Wright DH, Judd MA, Jones DB, Payne S.The nature of the immunoglobulin-containing cells inmalignant lymphoma: an immunoperoxidase study. JHistochem Cytochem 1980;28:761-70.

4Isaacson P, Wright DH, Judd MA, Mepham BL. Primarygastrointestinal lymphomas: a classification of 66 cases.Cancer 1979;43:1805-19.

5Rhernherz EL, Moretta L, Roper M, et al. Human T-lymphocyte subpopulations defined by Fc receptors andmonoclonal antibodies. J Exp Med 1980;151 :969-74.

Payne SV, Jones DB, Haegert DB, Smith JL, Wright DH.T and B lymphocytes and Reed-Stemnberg cells indiseased lymph nodes and spleens. Clin Exp Immunol1976 ;24 :280-6.

7Thorsby E, Bratlie A. A rapid method for preparation ofpure lymphocyte suspensions. In: Terasaki P, ed. Histo-comparability testing. Copenhagen: Munksgaard, 1970:655.

8 Wahl SM, Rosenstreich DL, Oppenheim JJ. Separationof human lymphocytes by E Rosette sedimentation. In:Bloom BR, David JR, ed. In Vitro methods in cellmediated and tumour immunity. New York: AcademicPress, 1976.

9 Sundstrom C, Nilsson K. Establishment and character-ization of a human histiocytic lymphoma cell line (U937).Int J Cancer 1976;17:565-77.

10 Wrigley CW. Analytical fractionation of plant and animalproteins by gel electrofocusing. J Chromatogr 1968;36:362-5.

Mepham BL, Frater W, Mitchell BS. The use of proteolyticenzymes to improve immunoglobulin staining by thePAP technique. HistochemJ 1979;11:345-57.

12Isaacson P, Wright DH. Malabsorption and intestinallymphomas. In: Wright R, ed. Recent advances ingastroenterology. Chapter 12. London: WB Saunders,1980.

13 Cohen AB. Interrelationships between the human alveolarmacrophage and alpha 1 -antitrypsin. J Clin Invest 1973:52:2793-9.

989



j.com/

J Clin P


ber 1981. Dow

nloaded from

http://jcp.bmj.com/


14 Gupta PK, Frost JK, Geddes S, Aracil B, Davidovski F.Morphological identification of alpha 1-antitrypsin inpulmonary macrophages. Hum Pathol 1979;10:345-7.

15 Isaacson P, Jones DB, Judd MA. Alpha l-antitrypsin inhuman macrophages. Lanicet 1979 ;ii :964.

16 Wilson GB, Walker JH, Wattanse JH, Wolgroch D.Determination of subpopulations of leukocytes involvedin the synthesis of al-antitrypsin in vitro. Proc Soc ExpBiol Med 1980;164:105-14.

Lipsky JJ, Berninger RW, Hyman LR, Talamo RC.Presence of alpha 1-antitrypsin on mitogen stimulatedhuman lymphocytes J Imminol 1979; 122 :24-6.

18 Benitez-Bibiesca L, Frere-Horta R. Immunofluorescentlocalisation of alpha- l antitrypsin in human polymorpho-nuclear leukocytes. Lift Sci 1978 ;21 :99-104.

Poppema S. The diversity of the immunohistologicalstaining pattern of Sternberg-Reed cells. J HistochenCYtochemn 1980;28:788-91.

Requests for reprints to: Dr P Isaacson, Department ofPathology, Level E, South Laboratory and PathologyBlock, General Hospital, Southampton S09 4XY,England.

990



j.com/

J Clin P


ber 1981. Dow

nloaded from

http://jcp.bmj.com/

Documents

Alpha-1-antitrypsin in human macrophages · Alpha-1-antitrypsin in humanmacrophages P ISAACSON, DBJONES, GHMILLWARD-SADLER, MAJUDD, S PAYNE ... nant lymphomas3 in the hopethat positive