Immunohistochemical analysis of murine CD95/Fas/Apo-1 receptor and its ultrastructural distribution in the thymus

AP.WIS 105: 5 9 7 4 0 2 . 1997 Priritcrl ir i Divirri(rrk . All ri'qlits r e s e r w d

C o p y r i ~ h t 0 A P M I S 1997

WJUUg ISSN 0903-4641

Immunohistochemical analvsis of murine CD95/Fas/A~o-l receDtor and its ultrastructural distribution in the thvmus

PAULA RAMOS,' MARYANN D. GANGI,' ANN BAREN,2 DANIEL FILIPPA2 and KEITH B. ELKON'

'Hospital for Special Surgery-Cornell Univ Med Center, New York, 2Memorial Sloan-Kettering Cancer Center, New York, U.S.A.

Ramos, F?, Gangi, M. D., Baren, A., Filippa, D. &. Elkon, K. B. Immunohistochemical analysis of murine CD95/Fas/Apo-l receptor and its ultrastructural distribution in the thymus. APMIS 105: 597- 602, 1997.

CD95/Fas/Apo-l is a cell surface receptor that, upon contact with its ligand, induces cells to die by apoptosis. In view of the importance of Fas receptor (FasR) in immunologic tolerance, an immunohistochemical analysis of FasR expression was performed in the lymphoid and certain parenchymal tissues of normal and mutant MRL/lpr mice using a rabbit polyclonal anti-Fas receptor antibody. FasR was expressed by immunoperoxidase (IP) in the cortex and at the corticomedullary junction of the thymus of normal mice. By immunoelectron microscopy FasR was detected on the cell membrane of normal thymocytes. In MRL/lpr mice, FasR protein expression could not be clearly detected. FasR protein expression was not detected in the heart, liver or ovary by IP, presumably reflecting the low number of receptors in these tissues.

Key words: FasR/CD95; immunoperoxidase; immunoelectron microscopy.

Paula Ramos, Department of Rheumatology, Hospital Ntra. Sra. Candelaria, Carretera del Rosario s/n, Santa Cruz de Tenerife 38010, Espaiia.

CD95/Fas/Apo is a cell surface receptor belong- ing to the nerve growth factor/tumor necrosis factor receptor family that signals apoptosis in lymphocytes (reviewed in 13). In mice, North- ern blot analysis revealed high levels of Fas receptor (FasR) mRNA expression in the thymus and lower levels of expression in the heart, liver, and ovary (20). Flow cytometry analysis con- firmed a high level of FasR protein expression on thymocytes, with a lower level of expression on mature splenic CD4+ T lymphocytes (2, 6). A spontaneous mutation in the gene encoding the FasR has occurred in the MRL strain of

Received July 10, 1995. Accepted August 9, 1996.

mice and results in a lupus-like disease associated with massive lymphadenopathy (19). The FasR mutation in MRL/lpr mice is caused by the insertion of a retrotransposon, ETn, into the second intron of the FasR gene resulting in reduced transcription of wild type FasR mRNA as well as in frame chimeric FasR-ETn transcripts (1, 4, 21). Negative selection appears to be normal in MRUlpr mice (10) and wild type FasR mRNA is reduced but not absent (1, 4, 21). Detailed evaluation of FasR by immunoperoxidase (IP) has recently been reported in humans using the monoclonal antibody anti- Apo-1 (11). FasR expression was not detected in the thymus, but was detected in peripheral lymphoid tissues. FasR expression was also detected in non-lymphoid organs (lung, kidney,

597

RAMOS et al.

and intestine) (11). By flow cytometry, FasR protein has been detected on the majority of human immature thymocytes and down-regulated at a mature stage of thymocyte development (5) . The anatomic distribution of FasR within the thymus and murine tissues has not, however, previously been reported. In view of the importance of FasR in immunologic tolerance, in the present study we examined FasR protein expression by IP technique in lymphoid and non- lymphoid organs and by immunoelectron microscopy (IEM) technique in thymus from normal Balb/c and MRL/lpr mice.

MATERIALS AND METHODS

Mice All studies were performed with normal. 6-8-

week-old Balb/cJ and MRL/MpJ-lpr (MRL/lpr) mice, 4 mice per strain (2 male and 2 female), obtained from the Jackson Laboratory (Bar Harbor, ME) and subsequently bred at our Animal Facility in a non-pathogen-free environment.

Tissue processing The organs were immediately quick-frozen in li-

quid nitrogen after removal. Frozen sections of about 1 cm2 and 3 pm thickness were placed on pretreated slides (Superfrost Plus glass slides, Fisher Scientific, Pittsburgh, PA), air-dried for 6 8 h and stored at - 70°C. Before immunostaining, sections were air- dried once again and fixed in a 3: 1 proportion of cold acetone and methanol for 1 min at 4"C, followed by three changes of PBS. Lymph nodes were obtained from the axilla and mesentery.

Immunoperoxidase ( IP) Tissue sections were blocked with 2% BSA/PBS for

1 h at room temperature (RT) followed by pre-in- cubation with non-immune suppressor 5% goat

serum in 2% BSA/PBS for 30 rnin at RT. Liver and kidney sections were also blocked with avidin followed by biotin (Vector Laboratories, Burlingame, CA) for 15 rnin each at RT. A polyclonal monospec- ific affinity-purified rabbit anti-mouse IgG F(ab), FasR antibody (6) at a concentration of 25 pg/ml in 2% BSNPBS was then applied overnight. This anti- FasR antibody binds to the aminoacids 5-23 of the FasR protein. After washing in PBS, the sections were incubated with biotinylated goat anti-rabbit IgG (H+L) affinity purified with< 1% of cross-reactivity to mouse IgG and IgM (Vector) diluted 1:2000 in 0.5% BSNPBS for 10 rnin at RT, quenched in 1% hydrogen peroxide in PBS for 5 rnin at RT, and placed in peroxidase-conjugated streptavidin (Boehringer- Mannheim, Germany) diluted 1:lOOO in 0.5% BSA/ PBS for 5 rnin at RT. Diaminobenzidine (Sigma, St Louis, MO) was used as the chromogen for the en- zyme for 5-10 rnin at a final concentration of 0.06%,/ 0.005% hydrogen peroxide in 0.5% TritoniPBS.

The peroxidase reaction resulted in an intense brown precipitate. Slides were lightly counterstained with modified Harris hematoxylin (Fisher), blued in ammonia water, dehydrated in graded ethanols, cleared with xylene and mounted in Permount (Fisher) (9). Routine staining with hematoxylin and eosin was also performed. Negative controls, either omitting the primary antibody or replacing it by normal rabbit IgG F(ab),, were run in parallel.

Immunoelectron microscopy ( IEM) A pre-embedding method was used. Cell suspen-

sions were obtained from the thymus by mashing the tissue with a sterile rubber and washed three times with RPMI medium. 5X lo6 cells were pre-incubated in 5% goat serum in PBS for 30 rnin at RT. Polyclonal rabbit anti-mouse IgG F(ab), FasR antibody was added in a final concentration of 50 pgiml in 5%) BSA/PBS for 1 h at RT. Cells were washed three times with 1% BSA/l% sodium azideiPBS and incubated with goat anti-rabbit IgG lOnm gold conjugate (Ted- Pella Inc, Redding, CA) diluted 1 : 10 in Tris buffer (20 mM Tris, 20 mM NaN,, 225 mM NaCl, 0.1% BSA, pH 8.2) for 1 h at RT. Following three washes, the

Fig. 1. Distribution of FasR expression in the thymus. Thymi were obtained from 6-week-old Balbic (A, C) or MRL/lpr (B) mice and processed by IP hematoxylin as described in Methods. A rabbit polyclonal anti-FasR antibody (6) was used as the primary antibody in A and B, whereas normal rabbit IgG was used as a control in C. All fields shown are of the thymic cortex except (A) which shows the corticomedullary junction. A, B and C, X 1000. Fig. 2. FasR expression in lymph nodes. Lymph nodes from Balb/c mice were examined for FasR expression by IP hematoxylin as described in Fig. 1. The field shows a lymphoid follicle with lymphoid cellular elements in the germinal center immunostained positively for FasR. X 1000. Fig. 3. Ultrastructural expression of FasR in the thymus. Thymocytes were obtained from Balbic (A) or MRLi lpr (B) mice and processed for immunoelectron microscopy (IEM) as described in Methods. Immunogold labeling was performed with the anti-FasR antibody. The arrows indicate gold particles on the membrane surface. Magnification: A, X46800; B, X48000.

598

MURINE CD95iFasiApo-1 RECEPTOR IN THE THYMUS

599

RAMOS et al.

cells were fixed in a pellet with 3% glutaraldehydei formaldehyde in 0.1 M sodium phosphate buffer (Electron Microscopy Sciences, PA) overnight at 4°C. The fixative was removed and replaced with 0.09 M phosphate buffer (X2, 15 min each). They were post- fixed with 1% osmium tetroxide (Stevens Metallurgi- cal Corp., New York) in 0.09 M phosphate buffer for 45 min at RT. Following dehydration in graded ethanols, the pellets were embedded in a 1:l proportion of propylene oxide and epoxy mixture (72% Maraglas 655, 16% epi-res 502, 10% dibutyl phthal- ate, 2% benzyldimethylamine) for 45 min at RT with two changes of epoxy mixture for 1 h each at RT. Then, they were polymerized at 58°C overnight (8). Sections were cut on a Reichert microtome, picked up on copper grids, and stained with uranyl acetate and lead citrate in an Ultrostainer Carlsberg System. They were examined with a Zeiss EM 109 electron microscope. Negative controls were performed as for II?

RT-PCR RNA was extracted by the phenol-acid-guanidium

method (3) and RT-PCR for FasR expression performed as described previously (4). Briefly, cDNA was synthesized from total RNA with Moloney murine leukemia virus RT and oligo(dT) as described (7). cDNA was amplified by PCR on a thermal cycler (480; Perkin Elmer Corp., Norwalk, CT) with Taq polymerase (Bethesda Research Laboratories). The primers used for amplification were: sense ATGCTGTGGATCTGGGCTGT, antisense GTTTCCTGCAGTTTGTATTGCT. The conditions used for amplification were: denaturation at 94°C for 1 min, annealing at 5°C below the melting temperature for 1 min, and extension at 72°C for 1 min.

RESULTS

FasR is expressed in the thymic cortex and at the corticomedullary junction

In Balblc mice, FasR staining was detected only in lymphoid organs by IF? Staining was observed on the membrane of cortical thymocytes. In addition, staining was detected at the corticomedullary junction of the normal thymus with weak expression in the medulla (Fig. 1A). Examination of the thymus obtained from MRL/lpr mice revealed no staining (Fig. 1B). Similarly, staining was absent when normal rabbit IgG was used instead of rabbit anti-FasR antibody (Fig. 1C). When normal Balb/c spleen was analyzed by I€', FasR protein was detected at low density around the periarteriolar sheath; the follicles and the red pulp were devoid of

FasR expression (not shown). In lymph nodes, FasR was detected in the germinal centers spar- ing the mantle (Fig.2). IP staining of the non- lymphoid organs (brain, heart, lung, liver, kidney, small bowel, colon or ovary) did not reveal specific staining for FasR when compared with the irrelevant rabbit IgG F(ab)z (not shown).

FasR expression by IEM To further examine FasR distribution on

thymocytes at the subcellular level, we performed EM immunogold studies using cells obtained from Balb/c and MRL/lpr mice. As shown in Fig. 3, specific FasR labeling was found sparsely situated on the cell membrane of Balb/c thymocytes (Fig 3A). FasR labeling was detected on almost all cells (19/20) obtained from Balbk thymocytes, but only on 1/20 thymocytes from MRL/lpr mice (Fig.3B). Inter- estingly, some degree of chromatin condensation (a marker of apoptosis) was detected on almost all thymocytes.

Analysis of FasR mRNA in non-lymphoid tissues

To examine whether failure to detect FasR protein expression by IP in Balb/c organs could be explained by alternative splicing and removal of the epitope recognized by the anti-peptide antibody, the size of 5' mRNA (between nt 50 and 539) obtained from heart and liver was examined by RT-PCR. No differences were detected in the size of cDNA obtained from Balb/ c spleen, heart and liver when analyzed on a 2% agarose gel (not shown).

DISCUSSION

Using flow cytometry analysis, we recently reported that FasR is expressed at high levels on CD4+CD8 + (double-positive) normal murine thymocytes (2, 6). Using the same polyclonal rabbit anti-FasR antibody, we now report membrane staining of cells in the cortex and at the corticomedullary junction of the thymus consistent with the location of double-positive thymocytes (14). Lower intensity and incidence of FasR+ cells were observed in the medulla by IP, consistent with lower levels of FasR expression on CD4+ or CD8+ (single-positive) cells detected by flow cytometry (2, 6). At least

600

MURINE CD95iFasiApo-1 RECEPTOR IN THE THYMUS

95%) of thymocytes are thought to die by apoptosis in the thymus. Since negative and positive selection of thymocytes probably occur at the double-positive stage of thymic matu- ration (14) and FasR signals apoptosis in double-positive thymocytes (1 5) , these findings suggest a role for FasR in thymic selection. To date, however, no convincing evidence for a defect in negative selection has been observed in FasR-deficient (lpr) mice (10). Although some degree of chromatin condensation was fre- quently observed in thymocytes by IEM, the very high rate of apoptosis (17) and the pres- ence of chromatin condensation in MRL/lpr thymocytes do not prove a functional relation- ship between FasR and apoptosis in the thymus.

The insertion of the retrotransposon into an intron of the FasR gene in MRL mice induces a marked (at least 10-fold) reduction in the transcription of wild type FasR mRNA as well as abnormal chimeric FasR-ETn transcripts (1, 4, 21). One possible explanation for the appar- ently normal negative selection in MRL/lpr mice could be that these mice express some FasR protein. To address this question, we examined FasR protein expression by IP and IEM. No clear staining of MRL/lpr thymocytes, compared to Balb/c thymocytes, was observed by IP or IEM. Recent attempts to induce apoptosis of MRL/lpr double-positive thymocytes with an agonist monoclonal anti-FasR antibody have been unsuccessful (1 5) . Together, these findings support the assertion (6) that no functional FasR is present in the thymus of MRL/lpr mice.

Flow cytometry analysis of normal murine splenocytes (6) and lymph nodes (18) revealed that FasR was expressed at lower incidence on peripheral lymphocytes compared to thymocytes and that FasR was predominantly expressed on CD4+ T cells. IP analysis of Balb/c spleen demonstrated FasR staining in cells of the periarteriolar sheaths but not in the follicles, consistent with the known distribution of T and B cells respectively in the spleen. By IP, FasR staining was observed in the germinal centers of Balb/c lymph nodes. Germinal centers are com- posed of activated B cells that die by apoptosis unless they undergo positive antigen-driven selection (12). This finding suggests a role for FasR in peripheral B selection.

Northern blot analysis of RNA obtained

from murine tissues has previously shown FasR mRNA expression in the heart, liver and ovary (20). Furthermore, Leithauser et al. (11) reported FasR detection by IP in human lung, liver, kidney, small bowel, colon,and ovary. We considered the possibility that our failure to detect FasR expression in non-lymphoid organs could be explained by alternative splicing of FasR mRNA as observed in a B-cell line 2PK3 (1 6). RT-PCR analysis, meanwhile, revealed no differences between the size of FasR cDNA de- rived from Balb/c heart and liver compared to spleen. Based on the lower levels of expression of FasR mRNA in non-lymphoid organs (20), we assume that the failure to detect FasR by IP in the present study is explained by the relatively low number of receptors in these organs. It is, however, surprising that FasR was not detected in human thymus by IP (1 1) whereas FasR was readily detected in the thymus by IP in the present study. Whether these are species-related differences or due to technical factors remains to be determined. Further studies of the distribution and function of FasR in non-lymphoid organs will be of interest.

We thank Drs. Robert A . Erlandson (Memorial Sloan Kettering Cancer Center) and Drs. E. F. Di Carlo and S. Doty (Hospital for Special Surgery) for help and advice. We also thank Kin Chung Kong for photo- graph assistance. Paula Ramos was awarded a grant .from Fondo Investigaciones Sanitarias (Ministerio de Sanidad, Espafia).

REFERENCES

1. Adachi, M. , Watanabe-Fukunaga, R. & Nagata, S. : Aberrant transcription caused by the insertion of an early transposable element in an intron of the Fas antigen gene of Ipr mice. Proc. Natl. Acad. Sci. USA 90: 1756-1760, 1993.

2. Andjelic, S . , Drappa, J . , Lacy, L. , Elkon, K. B. & Nikolic-Zugic, J. : The onset of Fas expression parallels the acquisition of CD8 and CD4 in fetal and adult a/O thymocytes. Int. Immunol. 6; 73- 79, 1994.

3. Chomczynski, P. & Sacchi, N . : Single-step method of RNA isolation by acid guanidinium thiocyanate-phenol-chloroform extraction. Anal. Biochem. 162: 156-1 59, 1987.

4. Chu, J. L. , Drappa, J . , Parnassa, A. P. & Elkon, K. B.: The defect in Fas mRNA expression in MRL/lpr mice is associated with insertion of the

60 1

RAMOS et al.

retrotransposon, ETn. J. Exp. Med. 178: 723- 730, 1993.

5 . Debatin, K. M . , Suss, D. & Krammer, P. H.: Dif- ferential expression of APO-1 on human thymocytes: implications for negative selection? Eur. J. Immunol. 24: 753-158, 1994.

6. Drappa, J., Brot, N . & Elkon, K. B.: The Fas protein is expressed at high levels on CD4+CD8+ thymocytes and activated mature lymphocytes in normal mice but not in the lupus-prone strain, MRL 1prApr. Proc. Natl. Acad. Sci. USA 90: 10340-10344, 1993.

7. Elkon, K. B., Hines, J. J., Chu, J. L. & Parnassa, A. P.: Epitope mapping of recombinant Hela SmB and B’ peptides obtained by the polymerase chain reaction. J. Immunol. 145: 636-643, 1990.

8. Erlandson, R. A.: A new Maraglas, DER 732, embedment for electron microscopy. J. Cell Biol. 22: 704-709, 1964.

9. Erlandson, R. A,: Diagnostic Transmission Elec- tron Microscopy of Tumors, Raven Press, New York 1994, I : 1-68.

10. Herron, L. R . , Eisenberg, R. A , , Roper, E., Kak- kanaiah, V., Cohen. P. L. & Kotzin, B. L.: Selec- tion of the T cell repertoire in lpr mice. J. Immu- nol. 151: 3450-3459, 1993.

11. Leithauser, F., Dhein, J . , Mechtersheimer, G. et al.: Constitutive and induced expression of APO- 1, a new member of the nerve growth factorh- mor necrosis factor receptor superfamily, in normal and neoplastic cells. Lab. Invest. 69: 415- 429, 1993.

12. Liu, Y . J., Joshua, D. E., Williams, G. T., Smith, C. A , , Gordon, J. & MacLennan, I. C. M.: Mech- anism of antigen-driven selection in germinal centers. Nature 342: 929-931, 1989.

13. Nagata, S. & Suda, T.: Fas and Fas ligand: lpr

and gld mutations. Immunol Today 16: 3943, 1995.

14. Nikolic-Zugic, J . : Phenotypic and functional stages in the intrathymic development of ci/B T cells. Immunol. Today 12: 65-70, 199 1.

15. Ogasawara, J., Suda, T. & Nagata, S.: Selective apoptosis of CD4+CD8+ thymocytes by the anti-Fas antibody. J. Exp. Med. 181: 485491, 1995.

16. Onel, K . , Ashany, D . , Nikolic-Zugic. J . , Lacy. E. & Elkon, K. B.: Expression and function of the murine CD95/FasR/APO-I receptor in relation to B cell ontogeny. Eur. J. Immunol. 25:

17. Surh, C. D. & Sprent, J.: T cell apoptosis detected in situ during positive and negative selection in the thymus. Nature 372: 100-103, 1994.

18. Tucek-Szabo, C. L. , Andjelic, S . , Lacy, L., Elkon, K. B. & Nikolic-Zugic, J.: Surface T cell Fas re- ceptorKD95 regulation, in vivo activation, and apoptosis. Activation-induced death can occur without Fas receptor. J. Immunol. 156: 192-200, 1996.

19. Watanabe-Fukunaga, R. , Brannan, C. I . , Cope- land, N. G., Jenkins, N. & Nagata, s.: Lympho- proliferation disorder in mice explained by de- fects in Fas antigen that mediates apoptosis. Na- ture 356: 314-317, 1992.

20. Watanabe-Fukunaga, R., Brannan, C. I . , Itoh, N . et al.: The cDNA structure, expression, and chro- mosomal assignment of the mouse Fas antigen. J. Immunol. 148: 12761279, 1992.

21. Wu, J., Zhou, T., He, J. & Mountz, J. D.: Auto- immune disease in mice due to integration of an endogenous retrovirus in an apoptosis gene. J. Exp. Med. 178: 461468, 1993.

2940-2947, 1995.

602

Documents

Immunohistochemical analysis of murine CD95/Fas/Apo-1 receptor and its ultrastructural distribution in the thymus