Thymic lymphocytes

Thymic lymphocytes

CELLULAR IMMUNOLOGY 97,23-33 (1986) Thymic Lymphocytes III. Cooperative Phenomenon in the Proliferation of Thymocytes under Con A Stimulation’ MAR...

845KB Sizes 0 Downloads 28 Views





Thymic Lymphocytes III. Cooperative Phenomenon in the Proliferation of Thymocytes under Con A Stimulation’ MARTINE INSERM


U 25, Hcfpital Necker, 161, rue de St?vres, 75730 Paris Cedex 15, France Received February 25, 1985; accepted August 6, 1985

In the present paper, the responseof thymocytes to Con A is analyzed in terms of a cooperative phenomenon between medullary thymocytes, cortical thymocytes, thymic accessorycells, and interleukin 2. Medullary thymocytes respond spontaneously to Con A and produce IL-2. The addition of exogenously produced IL-2 enhances their proliferation. Small numbers of cortical (PNA+) thymocytes do not respond to Con A, even in the presenceof IL2-containing supematant. By increasing the number of PNA+ cells per well, sensitivity to Con A and IL-2 appears. This responsemay be linked either to the increase in a minor PNA’-responding population and/or to the enhancedcontamination by medullary thymocytes and macrophagesin non-responding PNA+ thymocyte population. In this hypothesis, either the contaminating cells respond by themselves and/or cooperate with PNA+ cells to induce their proliferation. &culture of non-responding low numbers of PNA+ thymocytes with Con A- and IL-2-containing supematant in the presence of PNA- cells containing thymic medullary thymocytes and macrophagesalways produces a higher responsethan that ofeach individual population. Theseresultsshowthat a cooperative phenomenon occurs in the cocultures of PNA+ and PNA- thymic cells. We can show using PNA+ and PNAthymocytes with different Thy 1 alleles, that indeed both PNA+ and PNA- populations participate in the generation of proliferating cells. We can demonstrate, by lysis experiments with monoclonal antibodies and complement that at the end of coculture, most of the proliferating cells are Lyt l+, and part are Lyt 2+ or L3T4+. We discuss the fact that the phenotype of the cells after activation does not allow us to deduce the phenotype of their precursors. Lysis of Ia+ cells prior to coculture, reduces the level of the proliferative response but does not modify the percentage of cooperation produced by the coculture. Cooperation with medullary mature thymocytes or the presenceof active Ia- accessorycells possibly able to convert to Ia expression during coculture CXpHhentS may aCCOUnt for these reSUkS. 0 1986 Academic Pnss, IIIC.

INTRODUCTION Intrathymic T-lymphocyte differentiation is a ghost story, where the majority of the cells are said to be produced just to die locally ( 1,2), to be functionally immature (3), and even to be devoid of any functional potentialities (4, 5). The fact is that the origin, fate, and functions of thymic lymphocytes have been highly controversial subjects for years. On the basis of their membrane antigen composition, corticosteroid sensitivities, and immune reactivities, two major thymocyte populations have been described ’ This work was supported by INSERM (CRL 811044) and MRI 83C 085. 23 0008-8749186 $3.00 Copyright @ 1986 by Academic Press, Inc. All rigbts of reproduction in any form nrserwd.



(3-6): one minor population (5 to 10%) is essentially located in the thymic medulla, and is a mature population with antigenic characters similar to those of peripheral T cells; the bulk of thymocytes (85 to 90%) is located in the cortex, has an immature antigenic profile and no immune reactivity. The relationship between these two populations is far from clear. It was previously acceptedthat the mature population was derived from the immature cortical one, and was the source of the migrating cells intended to give rise to peripheral T cells (7-8,6). On the other hand, other data suggestthat cortical and medullary thymocytes are two different cell lineages (9-l 1). In this light, what might be the relationship between, and fate of these two populations? The development of studies using interleukin 2 (IL-2)2 produced by the mature T cells ( 12, 13), opens a new field of discussion concerning the role of mature thymic medullary lymphocytes and their relationship with immature cortical ones. By analogy with what is known about the IL-2 requirement for peripheral T-cell proliferation in the presence of antigen or mitogen, it was proposed that cortical thymocytes need IL-2 to proliferate, and that mature thymocytes were able to provide this helper factor (14). Two questions were then asked: (1) were the mature thymic medullary cells able to secrete IL-2? and (2) were the immature cortical cells able to respond to mitogen or antigen in the presence of IL-2? The production of IL-2 by thymocytes was not demonstrated by some authors, for example, Hayward, in man ( 15) or Palacios, in mice ( 16); while others found an IL-2 production by the mature, PNA-, corticoresistant thymocytes (4, 17, 18, our unpublished data). More controversial, however, is the relationship between IL-2 and immature cortical thymocytes. Several papers dealing with the reactivity of immature cortical thymocytes toward mitogens or alloantigens, demonstrate that the addition of exogenously produced IL-2 to culture medium leads to cortical thymocyte proliferation or generation of cytotoxic T cells ( 19-22, 18). Hence, it was said that cortical immature thymocytes have receptors for IL-2 and could respond to mitogen and antigen in the presence of exogenously produced IL-2 which they are unable to produce by themselves. Some data show that among cortical thymocytes, only restricted subpopulations were responsive to IL2 (23-25). However, Bijdeker et al. (4) and more recently Chen et al. (5) and Ceredig et al. ( 17) were unable to activate immature cortical thymocytes with Con A or to generate cytotoxic lymphocytes. Previous results were interpreted by Chen and colleagues (5) as contaminations of immature PNA+ cortical thymocytes by mature medullary ones which are supposedto be the only ones to respond. Another alternative may be that cortical thymocytes (or a subpopulation of them) need not only IL-2 to proliferate, but also cell to cell cooperation with other types, such as, mature thymocytes, macrophages, and/or dendritic cells which segregratetogether during the separation procedure by peanut agglutination or density gradient centrifugation (26). The present paper explores this hypothesis of a cell to cell cooperation. MATERIALS


Mice. DBA/2 (H-zd) C3H (H-2k) and AKR (H-2k) 4- to 6-week-old female mice were used (Iffa Credo, Lyon, France, and CSEAL, Orleans, France). C3H and AKR * Abbreviations used: C, complement; S-CM, spleen cell conditioned medium; PNA, peanut agglutinin; PNA+, PNA agglutinated thymocytes; fPNA+, irradiated PNA+; PNA-, PNA non-agglutinated thymocytes; fPNA-, irradiated PNA-; IL-2, interleukin 2; IL-I, interleukin-1; FCS, fetal calf serum; P-TR, phagocytic cells of the thymic reticulum.






mice were used only in cooperation experiments with anti-Thy 1.1 and anti-Thy 1.2 treatments. Thymocyte separation. Thymocytes were separated by the differential peanut agglutination technique (27). Thymocytes were prepared aseptically, and diluted to 8 X lO’/ml in RPM1 1640 (GIBCO laboratories, Paisley, Scotland, U.K.). Thymocytes (250 ~1) were diluted in the same volume of a 1 mg/ml solution of peanut agglutinin (PNA) solution (I.B.F. Clichy, France), and incubated at room temperature for 20 min. The cell suspension was then layered onto the top of 10 ml of undiluted heat-inactivated fetal calf serum (GIBCO) and allowed to sediment at room temperature for 45 min. Cells above the medium/FCS interface (PNA-) were recovered and washed in RPMI. Cells of the pellet (PNA+) were sedimented twice, at room temperature for 45 min by passage through 10 ml of RPMI. The final pellet was then resuspended in 5 ml of 0.15 M Dgalactose, incubated for 30 min at 37°C and washed three times. In some experiments, enrichment in medullary and cortical type thymocytes was carried out on a Ficoll density gradient (28-29). Since the type of separation and the results were the same, results were pooled with those performed with PNA agglutination. Preparation of IL-2-containing supernatant and IL-2 testing. Spleen cells were stimulated with Con A coupled to biotin. This technique, which was established by H. J. Garchon and J. Altman in our laboratory is a method readily adaptable for the subsequent removal of Con A from the supernatant, by taking advantage of the affinity of avidin-Ultrogel for the biotin-conjugated Con A. Spleencells were diluted (5 X 106/ ml) in RPM1 1640 containing 1% L-glutamine, 100 U/ml penicillin-streptomycin, 1% Hepes, 5 X lo-* A4 2-mercaptoethanol, 2% heat-inactivated fetal calf serum (GIBCO), and 5 pg/ml of Con A-biotin (Sigma, St. Louis, MO. Ref. C-2272). Cells were cultivated for 24 hr at 37°C in a humidified atmosphere of 5% CO;! in air. Cultures were set up in Falcon plastic flasks (Oxnard, Calif.) containing 10 ml/small flask (Ref. 30 13 F) or 30 ml/large flask (Ref. 3024 F). After 24 hr, cultures were centrifuged (3000 rpm 5 min) and the supernatants were collected. AvidinUltrogel containing 2 mg of avidin/ml of Ultrogel (IBF-Villeneuve la Garenne, France) was washed three times in RPM1 containing 100 U/ml of penicillin and streptomycin. Two hundred and fifty microliters of avidin-Ultrogel were used for 16-20 ml of supematant. Avidin-Ultrogel and supematant containing biotin-conjugated Con A were incubated at 4°C for 20 min, with intermittent shaking of the mixture which was then centrifuged ( 1500 rpm for 7 min) to eliminate the avidin-Ultrogel coupled to Con A biotin. The resultant supematants were sterilized by filtration through a 20-pm filter (Schleicher and Schiill, Dassel, Germany), one aliquot being tested for IL-2 activity, the rest being stored at -20°C until use.The presenceof IL-2 in spleencell supematants was tested by their ability to sustain the growth of an IL-2-addicted cell line, CTLL2 which was initially described by Gillis and Smith (3 1). IL-2 absorption. Spleen cell conditioned medium (S-CM) which was shown to contain IL-2 activity was absorbed onto washed CTLL2. Absorption was performed by mixing 30 X 1O6CTLL2 with 1 ml of undiluted supemamnt for 4 hr with intermittent shaking in a 5% CO2 humidified atmosphere at 37°C. Supematants were then centrifuged, and the pellet was discarded. IL-2 absorption was verified by the inability of absorbed supematant to sustain the growth of CTLL2. The effectsof absorbed and non-absorbed supematants were com-



pared on the Con A-induced proliferation of PNA- and PNA+ thymocytes as described below. Thymocyteculture. Total thymocytes, PNA- and PNA+ separatedthymocytes were cultivated for 72 hr in flat-bottomed microplates (Falcon microtest II) in a 5% CO2 and humidified atmosphere at 37°C. Culture medium was the same as that described for IL-2 preparation, but contained 10% FCS. Con A (5 pg, Miles Yeda, Rehovot, Israel) was added to each well. The number of cells per well was 1.6 X 105,3.2 X 105, 6.4 X 105,or 10 X lo5 in 200 ~1 of medium. For each cell dilution, serial dilutions of S-CM were used (from 50 to 6%). Control cultures with medium or Con A alone were performed for each cell dilution. In coculture experiments, PNA+ thymocytes were associated with PNA- ones, to explore the cooperation between mature and immature subpopulations. PNA+ cells (1.6 X lo5 per well) were used in all the experiments and 0.32 X lo5 or 0.16 X 10’ PNA-. Both were added at the onset of culture with 5 pg Con A/ml. Four experimental groups were formed, PNA+ plus PNA-, irradiated PNA+ plus PNA-, PNA+ plus PNAirradiated, both PNAf and PNA- irradiated (2500 R). Cultures were incubated for 72 hr and proliferation was measured after 18 hr of [3H]thymidine incorporation.

Antibody Treatments Anti-la antibody treatment. To assessthe role of Ia+ cells in the cooperative phenomenon between PNA+ and PNA- cells, PNA- cells were pretreated with anti-Ia antibody (ATH anti-ATL antibody gift from Dr. [email protected] CNRS Orleans la Source, France) in the presenceof complement (C’) (Low tox M rabbit complement Cedarlane laboratories, Hornby, Ontario, Canada). PNA- cells were treated with anti-Ia antibody ( l/ 10 final dilution) for 30 min at 37°C. Complement was then added for an additional 45 min (l/20 final dilution). The control group was treated with C’ alone. Cells were washed in fresh medium and used in the coculture experiments described above. Anti-Lyt I, and Lyt 2 and anti-L3T4. To determine the phenotype of cells proliferating in coculture experiments, PNA+/PNA- associations were cultivated for 3 days, and treated 18 hr before the end of cultures with anti-Lyt 1 (rat IgG*, clone 53.7.3), anti-Lyt 2 (rat IgG*, clone 53.6.7), or anti-L3T4 (rat IgGZb,clone GK-1.5) (32) antibody plus C’, or C’ alone and then pulsed with [3H]thymidine for the remaining 18 hr. AntiLyt 1 and anti-Lyt 2 were purchased from Becton-Dickinson (Sunnyvale, Calif.). Cocultures were centrifuged at 1500 rpm for 5 min. The supernatant was discarded and 100 ~1 of antibody (l/ 100 dilution) was added to the cells for 30 min at 37°C. RPM1 or C (100 ~1)was added (final dilution 1:20)for an additional 45 min incubation. Microplates were then centrifuged and the supernatant was replaced by fresh conditioned medium plus 1 &i of [3H]thymidine. Anti-Thy 1.1 anti-Thy 1.2 antibody treatments.To analyze the origin of the proliferating cells, cocultures were established between 1.6 X lo5 PNA+ cells from C3H mice (H-2k; Thy 1.2) and 0.32 X lo5 PNA- from AKR mice (H-2k; Thy 1.1). Using the same protocol as that described above, cocultures were treated either with antiThy 1.1 or anti-Thy 1.2 mouse antibody (l/l00 dilution) (NEN, Boston, Mass.) plus C’. Immunofluorescence.For immunofluorescence, cells were washed at the end of the culture period and incubated with each of the antibodies for an additional 30 min with a goat anti-rat antibody (GARat-FITC Cappel) or a goat anti-mouse antibody





(GAM-FITC Nordic). To analyze the Lyt subsets,cells were labeled with either antiLyt 2 or anti-L3T4 or with both antibodies. Percentagesof L3T4-Lyt 2+, L3T4’Lyt 2- subsets were determined by subtracting the percentage of cells labeled with one antibody to the percentage of cells labeled with both of them. L3T4’Lyt 2+ cells were then calculated from the known percentages of single antigen-positive cells. To determine the origin of the cells present at the end of the culture period, cells from Day 3 cocultures between PNA+, Thy 1.2 and PNA-, Thy 1.1 thymocytes were subjected to either anti-Thy 1.1 (mouse antibody clone 19 X E5) or to anti-Thy 1.2 (rat IgG2b clone 30-H 12) and revealed by a labeled antibody as described above. Before counting the results of the staining procedure, cells were incubated for a short period with propidium iodide (2 pg/ml) to exclude dead cells. Five experiments were evaluated with a Leitz Orthoplan microscope equipped for epiluminescence. Results were confirmed in one experiment using a cell analyzer (Ortho System). RESULTS

Efect of S-CM on ThymocyteSubpopulations Increasing numbers of total, PNA+ or PNA- thymocytes were subjected to Con A stimulation in the presence of increasing percentagesof S-CM (Fig. 1). Increasing the number of total thymocytes per well augmented the Con A-induced proliferation in the absence of S-CM, which was null when 1.6 X lo5 cells per well were used (0.3 I20




l~ly-/yA,A, 4020ot

Nb of cells


12 16.105



12 32.105

50 3% Supmmn1

12 64.lO5


3120 10.105

FIG. 1. Effect of S-CM and Con A on total thymocytes and on PNA+ and PNA- separatedones.Increasing numbers of total thymocytes, PNA+ and PNA- ones (1.6 X lo5 to 10 X 10’ per well) were subjected to Con A and increasing percentages of S-CM containing IL-2. (A) Proliferative response to Con A in the absenceof conditioned medium.



X lo3 cpm) and which rose to 36.0 X lo3 cpm for 10 X lo5 cells per well. In all cases, the addition of S-CM with IL-2 activity increased the level of the total thymocyte response.To further analyze these results, thymocytes were separatedinto PNA+ and PNA- subsets. PNA- cells subjected to Con A stimulation proliferated strongly with a plateau of 62.9 X lo3 cpm for 6.4 X lo5 cells per well. Addition of S-CM containing IL-2 activity increased this response. These results show that PNA- cells which are able to produce IL-2 in the presence of Con A, may enhance their proliferative response to Con A if extra IL-2 is added to the culture medium. Increasing the number of cells from 6.4 X lo5 to 10 X lo5 per well did not enhance the level of the response. Nutritional reasons linked to the high number of dividing cells per well may account for these results. When PNAf cells were used in these experiments (Fig. I), a spontaneous response to Con A was obtained for the highest cell dilution (6.2 X 10’ cpm for lo6 cells per well). If S-CM containing IL-2 was added to the culture medium, increasing proliferative responseswere obtained by increasing the number of PNA+ cells per well. Cells, 1.6 X lo5 per well, did not significantly respond to Con A plus S-CM, but they began to respond at 3.2 X lo5 cells per well and the response reached a plateau at 10 X lo5 cells per well. The role of IL-2 contained in crude S-CM in the response of PNA+ and PNA- cells to Con A was confirmed by experiments showing that the absorption of IL-2 activity onto CTLL2 totally abrogated the effect of S-CM (results not shown). The augmented response of PNA+ thymocytes obtained by increasing the number of cells per well may be linked either to the enhancement of PNA+ responding cells, or to the enhancement of mature medullary type of cells which may contaminate PNA+ cortical cells. If this is the case,two hypotheses may be proposed. (1) Contaminating PNA-, mature thymocytes are the only ones to proliferate within the PNA+ separated population, as may be concluded from the experiments of Chen and colleagues (5). (2) Mature thymocytes, and perhaps other cell types such as macrophages which migrate with mature lymphocytes during the separation procedures, are necessary contaminants for an immature cortical PNA+ cell response to Con A plus S-CM containing IL-2. This second hypothesis was tested, as reported in the next paragraph. Cooperative Functions betweenPNA’ and PNA- Thymic Cells in the Responseto Con A and S-CM PNA+ cells were associated with PNA- cells which contain mature thymic lymphocytes as well as macrophage/dendritic cells. PNA+ cells were used ( 1.6 X lo5 per well), as it is a cell dilution unresponsive to Con A plus SCM. To these non-responding PNA’ cells, PNA- cells (17 or 9%) were added at the same time as Con A and S-CM. The proliferative responseswere measuredand the experimental results were compared to the theoretical responses calculated by adding the individual responsesof PNA+ (plus irradiated PNA-) and PNA- (plus irradiated PNAf cells). Table 1 shows that in each case,the calculated results are lower than the experimental ones. These results suggestthat the presence of both cortical and medullary types of cells in the culture system potentiates the resultant proliferative response to Con A plus S-CM containing IG2. This cooperation phenomenon is dependent upon the percentage of PNA- cells. PNA- cells ( 17%) give the best cooperative effect.





TABLE I Cooperation between PNA+ Cells and PNA- Cells in the Proliferative Response to Con A and S-CM Containing IL-2 PNA+ + PNAPNA+ + PNA-.$




17 17 17 17

25 12 6 3

2.1 + 0.5 1.2 Zk0.2 0.4 + 0.09 0.2 + 0.03

2.8 k 2.5 t 0.6 + 0.2 +

9 9

25 12

2.1 * 0.5 1.2 k 0.2

2.2 + 0.4 1.8 f 0.3

0.4 0.5 0.1 0.06

PNA+ 3 + PNA6.2 -t 5.4 + 2.0 + 1.0 +


1.4 1.2 0.2 0.2

14.1 k1.4 12.6 f 1.3 5.9 + 0.9 1.7 + 0.6

2.3 k 0.3 1.5 f 0.4

5.7 + 1.1 3.7 ?I 0.5



9.0 f 7.9 + 2.7 + 1.3 f

36.8 + 36.8 k 51.5 Ii 10.6 +

1.3 1.4 0.2 0.2

4.5 f 0.5 3.6 k 0.3

2.9 5.5 5.1 5.3

15.7 + 7.2 6.5 + 3.7

1.6 X 10’ PNA+ cells were cultivated in the presence of S-CM plus Con A either alone or with PNAcells (17 or 9%). Response of PNA+ + PNA- is compared to the calculated response (PNA+ + PNA-) ) + PNA+ 2 + PNA-). The experimental response is higher than the calculated one, showing a cooperative phenomenon between PNA+ and PNA- cells.Cooperation is expressedasthe %potentialization: (experimental - calculated)/experimental X 100. Cell proliferation is expressed as cpm X IO-‘. Mean value f SEM for five different experiments.

To analyze the roles of macrophage-like cells in this cooperative phenomenon, lysis experiments were performed using anti-Ia antibody. PNA- cells were treated with anti-la antibody plus C’ before the onset of association experiments with PNA+ cells. Table 2 gives the results for two different experiments comparing the proliferation of PNA+ plus PNA- cells, with anti-Ia treated and non-treated PNA- cells. The depletion of Ia+ cells was verified by immunofluorescence (data not shown). In each case, the depletion of Ia+ cells reduces the level of the response to Con A plus S-CM containing IL-2.

TABLE 2 Role of la+ Cells in the Cooperation between PNA+ and PNA- Ceils during Con A Activation in the Presenceof S-CM Containing IL-2 Treatment



PNA+ + PNA- )

PNA+$ + PNA-


Expt I

None Anti-la + C’

1.6 -

39.5 20.0

4.3 3.0

19.8 10.1

39 35

Expt 2

None Anti-la + C’ C’ alone

1.8 -

37.3 19.6 32.1

4.5 3.0 3.0

17.8 11.0 19.0

40 29 31

Note. PNA- cells were treated by anti-la antibody plus C’ prior to association with PNA+ cells. Cultures were then performed between PNA+ and treated or non-treated PNA- cells in the presenceof Con A and 25% S-CM. Treatment with anti-Ia antibody consistently reducesthe proliferative responsein each group, but does not modify the potentialization of the responsewhich is obtained when PNA+ and PNA- cells are cultivated together instead of separately. Results are expressedin cpm X 10e3.%Potentialization is calculated as described for Table I.



However, when looking at Table 2, it can be seenthat the elimination of Ia+ cells, reduces the level of the response but does not inhibit the cooperation phenomenon between PNAf and PNA- cells: the samepercentage of potentiation is obtained when PNA+/PNA- coculture proliferation is compared to the calculated, expected result. To determine the type and the origin of the cells which are proliferating at the end of the coculture experiments two different techniques were used: in the first series of experiments the phenotypes of the cells were studied by immunofluorescence in DBA2/DBA2 and in C3H/AKR cocultures which revealed the Lyt phenotype and the origin of the cells. Figure 2B showsthat after 3 days in culture, all the Lyt phenotypes are represented: 95.3 & 1.7% of the cells are Lyt l+, 37.6 f 3.8% are L3T4+ Lyt 2-, 26.2 f 4.5% are L3T4-Lyt 2+, 18.6 & 4% are L3T4+Lyt 2+, and 21.4 + 2.2% are L3T4-Lyt 2-. Experiments using PNA+ and PNA- cells of respectively the Thy 1.2 and Thy 1.1 phenotypes demonstrated that after 3 days half of the recovered cells are of PNA- and half are of PNA+ origin (Fig. 2A). Since the cells with a given phenotype may be surviving cells and not dividing ones, the second type of experiment was performed: cells were treated with anti-Thy 1.1, anti-Thy 1.2, or anti-Lyt antibody plus C’ and then pulsed for 18 hr with [‘Hlthymidine. The contribution of each cell type to the proliferating pool is expressedas the percentage of proliferation inhibited by each antibody. Figure 3A shows that half of the dividing cells are killed by an anti-Thy 1.1 plus C’ and half by an anti-Thy 1.2 plus C’. These findings confirm and extend the results obtained by immunofluorescence, which showed that proliferating cells arise from both PNA+ and PNA- subsets. Figure 3B shows that 94 * 1% of proliferation is inhibited by anti-Lyt 1 plus C’ and that part of the proliferation is blocked by either anti-Lyt 2 (59.6 f 6.9%) or anti-L3T4 (36 + 8.4%) antibody plus C’. DISCUSSION DO all the cortical functionally immature thymocytes represent end-line, deathdestined cells, or are they able to mount functional reactivity under some experimental conditions? This is one of the questions which may be raised, subsequent to the results obtained by Chen and colleagues (5) showing that purified PNA+ cells do not react with Con A in the presence of IL-2 by proliferating and generating cytotoxic ‘I cells.


100 + s 8



r" ZR .aL FIG. 2. Cells were cultivated for 3 days and then tested for membrane antigen expression. In (A), the

PNA+, Thy I .2+, C3H cells were mixed with PNA-, Thy 1. l+ AKR cells and tested for expression of Thy 1.1 or Thy 1.2 antigen. In (B), PNA+ and PNA-, DBA/Z cells were. mixed and tested for Lyt antigen expression. Mean values + SEM for, respectively, four (A) and five (B) experiments.






!I 100


E m a

4 S %



FIG. 3. PNA+ and PNA- ceils were mixed under the same conditions as those described in Fig. 2. Eighteen hours before the end of the culture, cells were subjected to monoclonal antibodies and complement. Results are expressedas percent inhibition of [‘Hlthymidine incorporation by each antibody. Mean values + SEM for three different experiments.

In that perspective, the described reactivity of cortical thymocytes may be linked to the reactivity of mature medullary lymphocytes which contaminate cortical thymocyte preparations. However, as we are now accustomed to considering immune reactivity as cooperative events, one may ask if cortical thymocytes do not require, in addition to IL-2, the physical presenceof other cell types such as medullary mature thymocytes or macrophages in order to react. These cooperation phenomena were analyzed in the present study. Medullary thymocytes (PNA-) react spontaneously with Con A, and the addition of exogenous IL-2 enhances this response. Conversely, cortical thymocytes (PNA+) do not react to Con A, even in the presence of IL-2, when small numbers of cells are used. Increasing the number of cells per well clearly leads to the responseof PNA+ cells to Con A in the presence of IG2-containing supernatants. To analyze the role in that response of cells which may contaminate the PNA+ cell population, low numbers of PNA+ cells, which do not respond to Con A and IL-2-contaming supematant, were associatedwith PNA- cells containing both mature medullary thymocytes and thymic macrophages (26). In that case, it can be shown that the associatedcells do react to a greater extent than could be predicted from the individual responsesof both isolated subsets(Table 1). These results indicate a cooperative phenomenon in the response of thymocytes to Con A and IL-2. Since it has been shown that accessorycells intervene not only in IL-2 production, but also in the induction of IL-2 responsiveness of peripheral T cells (33, 34), we have explored the role of thymic accessorycells in the cooperative phenomenon. Lysis of Ia+ cells which contaminate the thymic medullary lymphocyte-enriched population, leads to a net reduction of Con A responsiveness in the presence of S-CM containing IL-2. These results are consistent with data reported by Hunig et al. (33) showing that Ia+ cells participate in IL2 responsivenessaswell asin IL2 production. However, our technique of depletion does not totally inhibit the Con A plus S-CM response as does that of Hunig and colleagues performed with spleen and lymph node cells. Furthermore, when Iaf cells are depleted from PNA- cells prior to PNA-/PNA+ cell association, the cooperation is not modified: although the level of the responseis lower compared to non depleted cocultures the response of Ia-depleted PNA-/PNA+ cell associations remains higher, in the same proportion as non-treated populations, than the response which would be predicted from Ia-depleted PNA- cells and PNA+ cells cultivated separately (Table 2).



This may be explained either by the fact that very small levels of Ia+ cells (not detectable by immunofluorescence) are sufficient to induce S-CM and Con A responsiveness,or that Ia- macrophage-like cells in the thymus are also responsible for the response to S-CM and Con A. Indeed, we have shown that phagocytic cells of the thymic reticulum (P-TR), both Ia- and Ia+ have a close relationship with both mature and immature thymocytes (35, 36). Another explanation is that Ia antigen expression may be induced after anti-Ia treatment on Ia- cells during the activation process (possibly through interferon production). A final explanation is that, in the absenceof Ia+ accessorycells, thymic meduky lymphocytes, and thymic cortical ones, cooperate in the proliferative response to Con A in the presence of IL-2. That proliferation in the cooperation experiments was not restricted to only one subset was assessedby immunofluorescence and lysis experiments using PNA+, Thy 1.2+and PNA-, Thy 1. l+ thymocytes. These experiments showed that proliferating cells arise from both PNA+ and PNA- subsets.A cooperative phenomenon between cortical and medullary thymocytes has already been shown by Eisenthal and colleagues(37) who demonstrated that both PNA+ and PNA- thymocytes were required for the generation of suppressor T cells in culture. The Lyt phenotypes of the cells at the end of the culture period were also analyzed by immunofluorescence and lysis experiments, prior to [3H]thymidine incorporation. These experiments demonstrate that all Lyt phenotype combinations are represented at the end of the culture period in the coculture experiments. Unfortunately, this type of experiment which reveals the phenotype of the proliferating cells, does not allow us to deduce the phenotype of the originally activated cells. Indeed, it hasbeen clearly shown that cell phenotypes may be profoundly modified during culture in the presence of spleen-conditioned medium; for example: the appearanceof PNA receptors on PNA- thymocytes (38), the appearanceof Thy 1 antigen on bone marrow cells (39), or the loss of Lyt 1 antigen on the surface of Lyt 1+2+ thymocytes leading to Lyt 1 (low) 2+ cells (40). In this latter study, we showed that the early reduction of Lyt 1 antigen on Lyt 1+2+cells is followed by the proliferation of both Lyt l-2+ and Lyt 1+2- cells. In any case, the present data show that when a non-Con A-responding subset of PNA+ thymocytes is cocultured with a PNA--responding one, this association leads to an enhanced response in which the PNA+ subset participates. This proliferation occurs in a complex environment, in which IL2 is essential, in which L3T4+ and Lyt 2+ cells coproliferate, and in which, thymic accessorycells are amplifiers of this response. REFERENCES I. Metcalf, D., in “Thymus: Experimental and Clinical Studies (Wolstenholme and Porter, Eds.), p. 242. Ciba Foundation Symposium. Churchill, London, 1966. 2. McPhee, D., Pye, J., and Shortman, G., Thymus 1, 151, 1970. 3. Shortman, K., Von Boehmer, H., and Hoppner, K., Transplant. Rev. 25, 163, 1975. 4. Biideker, B. G. D., Van Eijk, R. V. K., and Mtlhlradt, P. F., Eur. J. Immunol. 10,702, 1980. 5. Chen, W. F., Scollay, R., and Shortman, K., J. Immunol. 129, 18, 1982. 6. Cantor, H., and Weissman, I. L., Prog. Allergy 20, 1, 1976. 7. Weissman, I. L., J. Exp. Med. 137, 504,~1973. 8. Weissman, I. L., Mass&a, J., Olive, C., and Friedberg, S. H., J. Med. Sci. 11, 1267, 1975. 9. Shortman, R., and Jackson, H., Cell. Immunol. 12,230, 1974. 10. Goldschneider, J., Shortman, K., McPhee, D., Linthicum, S., Mitchell, G., Battye, F., and Bollum, F. J., Cell. Immunol. 69, 59, 1982.






11. Jotereau, F. V., and Le Douarin, N. M., I. Immunol. 129, 1869, 1982. 12. Smith, K. A., Immunol. Rev. 51, 337, 1980. 13. Farrar, J. J., Benjamin, W. R., Hilfrker, M. L., Howard, M., Farrar, W. L., and Fuller-Farrar, J., Immunol. Rev. 63, 129, 1982. 14. Gillis, S., Union, N. A., Baker, P. E., and Smith, K. A., .I. Exp. Med. 149, 1460, 1978. 15. Hayward, A. R., Kumick, J. T., and Clarke, D. R., I. Immunol. 127,2079, 198 1. 16. Palacios, R., Immunol. Rev. 63, 73, 1982. 17. Ceredig, R., Glasebrook, A. L., and MacDonald, H. R., .I. Exp. Med. 155, 358, 1982. 18. Conlon, P. J., Henney, C. S., and Gillis, S., .I. Immunol. 128, 797, 1982. 19. IrlC, C., Piguet, P. F., and Vassali, P., .I. Exp. Med. 148, 32, 1978. 20. Wagner, H., Rollinghoff, M., Shawaller, R., Hardt, C., and Plizenmaier, K., Nufure (London) 280,405, 1979. 21. Wagner, H., Hardt, C., Bartlett, R., Rollinghoff, M., and Phzenmaier, K., .I. Immunol. 125,3532, 1980. 22. Hard& C., Pfizenmaier, K., Rollinghoff, M., Klein, J., and Wagner, H. .I Exp. Med. 152, 1413, 1980. 23. Piantelli, M., Lauriola, L., Maggiano, N., Ranelletti, F. O., and Musiani, P., Cell. Immunol. 64, 337, 1981. 24. Draber, P., and Kisielow, P., Eur. J. Immunol. 11, 1, 198 1. 25. Umiel, T., Daley, J. F., Bhan, A. K., Levey, R. H., Schlossman, S. F., and Reinherz, E. L., .I. Immunol. 129, 1054, 1982. 26. Nabarra, B., and Papiemik, M., Cell. Immunol. 50, 72, 1980. 27. Reisner, Y., Linker-Israeli, M., and Sharon, M., Cell. Immunol. 26, 1290, 1976. 28. Papiemik, M., Laroche, L., and Bach, J. F., Eur. J. Immunol. 7, 796, 1977. 29. Papiemik, M., and Bach, J. F., Eur. J. Immunol. 7, 800, 1977. 30. London, J., Berrih, S., and Papiemik, M., Dev. Comp. Immunol. 3, 349, 1979. 31. Gillis, S., and Smith, K. A., Nature (London) 268, 154, 1978. 32. Dialynas, D. P., Wilde, D. B., Marrack, P., Pierres, A., Wall, K. A., Havran, W., Otten, G., Loken, R. R., Pierres, M., Kappler, J., and Fitch, F. W., Immunol. Rev. 74, 29, 1983. 33. Hunig, T., Loos, M., and Schimpl, A., Eur. J. Immunol. 13, 1, 1983. 34. Austyn, J. M., Steinman, R. M., Weinstein, D. E., Granellil-Pipemo, A., and Palladino, M. A., J. Exp. Med. 157, 1101, 1983. 35. Papiemik, M., Nabarra, B., Savino, W., Pontoux, C., and Barbey, S., Eur. J. Immunol. 13, 147, 1983. 36. El Rouby, S., Praz, F., Halbwachs-Mecarelli, L., and Papiemik, M., Eur. J. Immunol. 134, 3625, 1985. 37, Eisenthal, A., Nachtigal, D., and Feldman, M., Immunologv46,697, 1982. 38. Schrader, J. W., Chen, W. F., and Scollay, R., J. Immunol. 129, 545, 1982. 39. Schrader, J. W., Battye, F., and Scollay, R., Proc. Natl. Acad. Sci. USA 79,416 1, 1982. 40. Vatteroni, M. L., and Papiemik, M., Cell. Immunol. 83, 124, 1984.