272 G. Brown, P. Biberfeld, B. Christensson and D. Y. Mason Eur. J. Immunol. 1979. 9: 272-275

This can also be suggested by the very early description of the null or K cell by Allison [ 111. In this report, the null cell has been described as adherent but nonphagocytic. With our bone marrow cultures, w e find that the development of adherence is an earlier step in maturation, whereas phagocytic properties are acquired later (unpublished observation).

A connection between macrophages and null cells has also been described by Koren [12]. He characterized the cell line P388D which was first reported as being a null cell line by Shevach [13]. When Koren handled the cell line, it possessed all characteristics of mature macrophages, and he maintained that macrophages and null cells may have the same origin.

The assumption that promonocytes may be identical to K lym- phocytes was further supported by the finding that mouse spleen cells which in our experiments under normal conditions have no lytic capacity in the ADCC system, become cytotoxic for antibody-coated tumor target cells when the mice are preactivated with CP. By measuring against L5 178 Y cells we could exclude NK effects [7] . Also pretreatment of the mice with thioglycollate induces a PE population with ADCC activi- ty against tumor targets (Table 6). Both treatments are known to induce proliferation of macrophages and their precursors in the animal. Thus, it may be deduced that the positive effects obtained with CP or thioglycollate-induced cells are per- formed by effector cells of macrophage origin.

Further studies on the nature of the effector cell under the described in vivo conditions are in progress.

Geoffrey Brown', Peter Biberfeld', Birger Christenssono and David Y. Mason0

Nuffield Department of Medicine+, Department of Pathologyo, Radcliffe Infirmary, Oxford and Department of Pathology, Karolinska Hospitalo and Department of Immunology, National Bacteriological Laboratory, Stockholm

The excellent technical assistance of Mss. I . Hiilsmann and U. Hauss- mann is gratefully acknowledged.

Received August 28, 1978; in final revised form November 23, 1978.

5 References

1

2

3 4

5

6

7

8 9

10

11 12

13

Lohmann-Matthes, M.-L., Domzig, W. and Taskov, H., Eur. J . Immunol. 1979. 9: 261. Meerpohl, H.-G., Lohmann-Matthes, M.-L., Fischer, H., Eur. J . Immunol. 1976. 6: 213. Wekerle, H., J. Exp. Med. 1978. 147: 233. Peter, H. H., Pane-Fischer, J., Friedman, W. H., Aubert, C., Cesarini, J.-P., Roubin, R. and Kourilsky, F. M., J . Immunol. 1975. 11.5: S39. Koski, I., in Bloom, B. R. and David, J. R. (Eds.), In vitro Methods in Cell-mediated and l'umor lmmunily, Academic Pre\\ New York 1976, p. 359. Archer, J. R., Smith, D. A,, Davie.;, D. A. L. and Staines, N. A,, J . Immunogenetics 1974. I : 113. Roder, J., Lohmann-Matthes, M.-L. and Domzig, W. 1978, in press. van Furth R., Agents Actions 1976. 6: 91. Goud, T. J. L. M., Schotte, C. and van Furth, R., J. Exp. Med. 1975.142: 1180. Oehl, S., Schaefer, U. W., Boecker, W. R., Kalden, J. R. and Peter, H. H., Z. Imrnunitaetsforsch. 1971. 152: 423. Allison, A. C., Ann. Imrnunol. Pans 1972. I23: 585. Koren, H. S., Handwerger, B. S. and Wunderlich, H., J . Immunol. 1975.114: 894. Shevach, E. M., Stobo, J. D. and Green, J., J. Imrnunol. 1972. 108: 1146.

The distribution of HLA on human lymphoid, bone marrow and peripheral blood cells*

The cellular distribution and the differential expression of HLA on cell suspensions and tissue sections has been investigated using the monoclonal antibody W6-32, which reacts with the high molecular weight chain of the major histocompatibility antigen. Lymphocytes and platelets, as assessed by autoradiographic and immuno- peroxidase labeling, were the most densely labeled cells. Myeloid precur5ors showed more labeling than mature neutrophils. Electron microscopic immunoperoxidase labeling showed a continuous distribution of HLA antigen on lymphoid and myeloid cell membranes. Erythroid precursors (including reticulocytes), although very weakly labeled, were clearly positive, in comparison with mature erythrocytes. In the thymus, HLA-negative, thymocyte antigen-positive cells (85 %) can be distinguished from HLA-positive, thymocyte antigen-negative cells (15 %). By using immunofluores- cence techniques on tissue sections, the former cells were shown to be cortical thymo- cytes and the latter medullary cells.

[I 22.581 1 Introduction * This study was supported by the Leukaemia Research Fund and the

Swedish Cancer Society.

Correspondence: David Y. Mason, Department of Pathology, Rad- cliffe Infirmary, Oxford, GB

Abbreviations: BBSS: Balanced buffered saline solution saline

HBSA: Hanks' solution plus bovine serum albumin P B S Phosphate-buffered

Although it has been recognized for several years that the serologically defined antigens coded for by the HLA gene complex are widely distributed on human tissues [ 11, there have been few reports on the relative density of these antigens on different cell populations [2-4]. This lack of information may, in part, be due to the technical difficulties of quantitating HLA antigens on individual cells using human alloantisera. The recent production of monoclonal mouse antibody

0 Verlag Chemie, GmbH, D-6940 Weinheim, 1979 0014-2980/79/0404-0272$02.50/0

Eur. J. Immunol. 1979. 9: 272-275 Cellular distribution of HLA 273

(W6-32) to the high molecular weight chain of the major histocompatibility antigen [ 5 ] , by the plasma cell hybridization technique [6], provides a valuable new reagent for studying the relative distribution of HLA. This antibody appears to recognize an antigenic determinant shared between products of the HLA-A, B and C loci [5] . In the present report, we describe its application to the analysis of HLA antigens on human lymphoid, bone marrow and peripheral blood cells.

2 Materials and methods

2.1 Antibodies

The W 6-32 antibody used was a 10-fold concentrated culture supernatant from the myeloma-spleen cell hybrid secreting antibody against HLA [ S ] . Monoclonal anti-thymocyte anti- body was kindly donated Dr. A. McMichael and F(ab'), rabbit anti-mouse IgG kindly provided by Dr. A. F. Williams.

2.2 Cell preparations

Cell preparations used in this investigation consisted either of whole peripheral blood or aspirated bone marrow, or of Ficoll- Triosil-separated mononuclear cells from blood, bone marrow, tonsil or thymus [7]. These cell suspensions were washed in Hanks' solution containing 5 % bovine serum albumin (HBSA).

2.3 Surface labeling techniques

The binding of W 6 3 2 to various cell types was demonstrated using immunofluorescence, immunoenzymatic and auto- radiographic techniques. Cell preparations were incubated at 4 ° C for 1 h with the culture supernatant from the W6-32 hybrid (concentrated 10-fold) and then washed twice in HBSA. For the immunoenzymatic technique, cells were smeared on clean microscope slides, fixed in buffered formol/ acetone for 30 s [8], rinsed in distilled water and Tris-buffered saline, pH 7.6, and then incubated with rabbit anti-mouse IgG (purified by elution from a mouse IgG immunosorbent) conjugated with horseradish peroxidase [9] . The slides were then washed thoroughly with Tris-saline, incubated in peroxi- dase substrate solution (diaminobenzidine 0.6 mg/ml plus 0.01% H,O, in Tris-saline, pH 7.6), counterstained with hematoxylin and mounted in DPX.

For autoradiography, cells which had been incubated with the culture supernatant were washed twice with a balanced buf- fered saline solution (BBSS [lo]) and then incubated with lz5I-1abeled F(ab'), rabbit anti-mouse IgG ( 5 pCi/pg = 185 KBq) at 30 pg/ml for 1 h at 4°C. After washing the cells twice with BBSS, smears were made on albumin-coated slides, fixed overnight in methanol and rinsed in distilled water. Autoradiographs were prepared with Ilford KS emulsion and exposed for 6 to 72 h before developing and counterstaining with Leishmann's stain. In some experiments on bone marrow, endogenous peroxidase activity was developed with di- aminobenzidine and H202, before coating with emulsion, in order to aid identification of myeloid and erythroid precursors.

Immunohistological labeling of antigens in tissue sections was performed on cryostat sections which had been air-dried and

fixed for 15 s in acetone. After rehydration in phosphate- buffered saline (PBS, 0.15 M, pH 7.2), the sections were incubated with monoclonal antibodies, diluted in PBS, for 30 min, washed and then incubated with fluorescein isothio- cyanate-labeled sheep anti-mouse IgG (Nat. Bacteriological Laboratory, Stockholm). After a further wash, sections were mounted in PBS: glycerol (SO: SO, v: v) for fluorescence microscopy.

In order to assess the surface membrane distribution of HLA, tonsil and bone marrow cells were examined by immuno- electron microscopy. For this purpose, cells were fixed in 1 % paraformaldehyde and 0.05% glutaraldehyde in 0.15 M PBS for 30 min, washed (in PBS) and then incubated with the monoclonal antibody to HLA. After further washing in PBS, the cells were incubated with the peroxidase-conjugated anti- mouse IgG as above, fixed in 1% glutaraldehyde and then incubated in the diaminobenzidine/H,O, substrate before processing for electron microscopy [ 121.

3 Results

3.1 Distribution of HLA on lymphoid, bone marrow and peripheral blood cells

The results of the labeling experiments are summarized in Table 1. The strongest binding of anti-HLA antibody, as assessed by immunoperoxidase and/or autoradiography, was to peripheral and tonsillar lymphoid cells, myeloid precursors and platelets (Fig. 1 a-c).

Table 1. Distribution of HLA on human lymphoid, bone marrow and peripheral blood cellsa)

Tissue

Peripheral hloob'

Lymphocytes Erythrocytes Platelets

Bone marrowd) Lymphocytes Myeloid precurson Erythroid precursors

Lymphoid cells

Pol ymOrph!

Thy m use)

Tonsil lymphoid cells"

Reactivityb)

+ + + + + + ++++

-

+ + + + +++ f

- (SSOr,) + + + ( 1 5 8 ) + + + +

a) Results from three experiments. b) Autoradiographs were exposed for 71 h. c) Peripheral blood cells were prepared by mixing heparinized blood

with an equal volume of 3% gelatin in normal saline and incubating for 40 min at 37 "C. The leukocyte-rich upper fraction was care- fully removed and the cells washed three times in BBSS before use.

d) Normal bone marrow was obtained from patients undergoing routine hematological investigation. Bone marrow cells were tested either prior to any separation procedure or after gelatin sedi- mentation as above.

e) Thymuses were obtained from patients undergoing cardiovascular surgery. Thymus and tonsil cells were prepared as described previously [7].

273 G. Brown, P. Biberfeld, B. Christensson and D. Y. Mason Eur. J. Immunol. 1979. Y: 272-275

Figure 3. Immunofluorescence staining of thymus cryohtdt section with anti-HLA. There is strong labeling of medullary cells around a Hassal’s corpuscle (H) hut the cortex (C) is unstained.

Figure 1 . Reactivity of monoclonal antibody to HLA with bone marrow cells. (a) lymphocyte; (b) myelocyte and normoblast; (c) platelet; (d) polymorphonuclear leukocyte. (a), (b) and (d) are auto- radiographs, and the binding to platelets (c) is demonstrated using an immunoperoxidase procedure.

r

Groin count

0-10 11-20 21-30 31-40 41-50 >50

pEiq Groin count

J Figure 4. Cryostat section of the thymus labeled for the thymocyte- specific antigen. The cortex shows strong labeling, but the medulla (M) is negative except for scattered cells. A Hassal’s corpuscle is marked with an asterisk.

Figure 2. Comparative di\tribution of HLA on Iwnc marrow auto- radiograph\. ( a ) Autorndiographs were exposed for 72 h. Reactivity o f W h-32 \\it11 nomiohlasts 0 and erythroc! tc\ 3 ; (b) Autoradio- graphs wcrc c \ p ~ . d Ii)r 6 h. Reactivity of WO-32 \\ith normoblasts M. I>niphoc\ ti‘\ c:. polq’morphonuclear leukocytes m0, promyelocytes and rnyeloc!, tL,,

Mature polymorphonuclear leukocytes showed weaker label- ing than their precursors (Fig, 1 b and d) while normoblasts were very weakly labeled after exposure periods (6 h) suf- ficient to give strong lymphoid and myeloid labeling (Fig.

1 a and b). Longer exposure revealed clearly that normoblasts had bound anti-HLA antibody, while the labeling intensity on the majority of red cells was not above background. A minor- ity of bone marrow red cells (ll%), however, showed com- parable labeling intensity to that of normoblasts and were presumed to be reticulocytes.

These findings were confirmed by comparative grain counts on bone marrow autoradiographs. A short exposure time (6 h) was used to compare the reactivity of lymphocytes, myelocytes and polymorphs, and a longer exposure time (72 h) was necessary to visualize the normoblast activity. The results are shown in Fig. 2.

3.2 Distribution of HLA in the thymus

In the case of thymus, approximately 15% of thymocytes were labeled with the HLA antibody by immunoperoxidase or autoradiography. In the same experiments, approximately 85% of thymocytes were labeled with a monoclonal anti- thymocyte antibody prepared by McMichael e t al. [ 111, and

Eur. J . Immunol. 1979. 9: 272-275 Cellular distribution of HLA 275

a combination of both antibodies labeled 99% of thymo- cytes. Immunohistological staining of thymus sections showed that HLA determinants were present on the majority of medullary cells (Fig. 3), but were absent from the cortex, with the exception of clusters of cells in the vicinity of vessels or septae. The thymocyte-specific antigen, in contrast, was present on the majority of cortical cells (Fig. 4) but only on a few scattered cells in the medulla.

3.3 Surface membrane distribution of HLA

Tonsil and bone marrow cells were examined by immuno- electron microscopy, and the results confirmed the autoradio- graphic and immunoperoxidase studies [ 121 in thar lymphoid cells were more densely labeled than polymorphs. The surface distribution of HLA was continuous without evidence of patches or clustering of antigen, as shown in Fig. 5 .

Figure 5. Surface mcmbrane distribution of HLA on tonsil lympho- cytes. For this purpose the cells were fixed in paraformaldehyde and glutaraldehyde before incubation with the monoclonal antibody to HLA and peroxidase-conjugated anti-mouse IgG. The bar represents 1 ym.

4 Discussion

These results are largely in accord with previous, less direct, data on the distribution and concentration of HLA antigens. The relatively higher density of HLA on lymphocytes and platelets compared to polymorphs had been inferred from the intensity of labeled human anti-HLA binding by these cell types [13-161. The stronger expression of HLA on the pre- cursors of polymorphs described above has not been reported. However, Kourilsky et al. [2] have tentatively suggested that myeloid precursors carry tess HLA than their progeny. The much weaker expression of HLA on monoblasts relative to other marrow elements has also not previously been described. This finding is in disagreement with Harris’ and Zervas’ report (based upon absorption studies) that blood reticulocytes carry

approximately as much HLA as do peripheral white cells [17]. However, our data do support the observation of Kourilsky et al. [2] that HLA antigens persist for a period on the erythroid cell after the loss of its nucleus.

In contrast to the ubiquitous expression of HLA on bone marrow cells and peripheral and tonsillar lymphoid cells, the antigen was confined to a minor subpopulation of thymocytes and exhibited a reciprocal relationship to a major popula- tion of cells in this tissue which carried a thymocyte-specific antigen. The immunohistological staining patterns (Figs. 3 and 4) showed that HLA-positive cells are localized predominantly in the medulla, while the thymocyte antigen is expressed almost exclusively on cortical cells. These findings are closely analogous to observations on the mouse thymus in which the majority of cells express little or no H-2 but carry Thy-1 and TL antigens, while a minor population shows the reverse phenotype [18].

A further observation of interest which was readily made with the monoclonal anti-HLA antibody, was that this antigen shows a continuous distribution on cell surfaces when studied by immunoelectron microscopy. The present data confirm suggestions [3] that the discontinuous labeling patterns reported in previous studies [2] can be explained as an artefact attributable to antibody-induced redistribution of antigenic determinants.

We are extremely grateful to Mss. Rita Sammons and Kaye Yuta for technical assistunce and to D. W. Jerome for electron microscopy.

Received September 1, 1978.

5 References

1 Albert, E. and Gotze, D., The Major Histocompatibility System in

2 Kourilsky, F. M., Silvestre, O., Levy, J. P., Dausset, J., Nicolai,

3 Davis, W. C., Alspaugh, M. A,, Stimpfling, J. H. and Walford,

4 Silvestre, D., Kourilsky, F. M., Nicolai, M. G. and Levy, J. P.,

5 Barnstable, C. J., Bodmer, W. F., Brown, G., Galfre, G., Milstein,

6 Kohler, G. and Milstein, C., Nature 1975.256: 495. 7 Brown, G. and Greaves, M. F., Eur. J . Immunol. 1974.4: 302. 8 Mason, D. Y., Farrell, C. and Taylor, C. R., Br. J . Haematol.

1975.31: 361. 9 Nakane, P. K. and Kawaoi, A., J. Histochem. Cytochem. 1974.22:

1084. 10 Shortman, K., Williams, N. and Adams, P., J . Immunol. Methods

1972.1: 273. 11 McMichael, A. J., Pilch, J. R., Galfre, G., Mason, D. Y., Fabre, J.

and Milstein, C., Eur. J . Immunol. 1979. 9: 205. 12 Reyes, F., Lejonc, J . L., Gourdin, M. F., Mannoni, P. and Dreyfus,

B., J . Exp. Med. 1975. 141: 392. 13 Giphart, M. J., Doyer, E., Wisse, E. and Bruning, J. W., in Kiss-

meyer-Nielsen, F. (Ed.), Histocompatibility Testing 197.5, Munks- gaard, Copenhagen 1075, p. 739.

14 Verheugt, F. W. A,, Von Der Borne, A. E. G. K., Decary, F. and Engelfriet, C. P., Br. J . Haematol. 1977. 36: 533.

15 Haynes, W. D. G., Jones, J. V. and Cummins, J. C. O., Transplan- tation 1974. 18: 81.

16 Thorsby, E., Scand. J . Haematol. 1969. 6: 119. 17 Harris, R. and Zervas, J. D., Nature 1969.22: 1062. 18 Konda, S., Stockert, E. and Smith, R. T., Cell. Immunol. 1973.

Man and Animals, Springer Verlag, Berlin 1977, p. 27.

M. G. andSenik, A,, J . Immunol. 1971. 106:454.

R. L., Tissue Antigens 1971. I : 89.

Nature 1970. 288: 67.

C., Williams, A. F. and Ziegler, A,, Cell 1978. 14: 9.

7; 275.