Eur. J. Immunol. 1990. 20: 2517-2520

Deletion of thymocytes


Short paper Eric J. Jenkinson, Rosetta Kingston and John J.T. Owen Department of Anatomy, Medical School, University of Birmingham, Bdgham

Newly generated thymocytes are not refractory to deletion when the a/P component of the T cell receptor is engaged by the superantigen staphylococcal enterotoxin B It has been reported that, following the initial expression of the Tcell receptor (TcR) dp, newly generated thymocytes pass through a developmental window characterized by ineffective coupling between the d p and CD3 components resulting in resistance to deletion (negative selection). However, we now provide evidence that the TcR d p on developing thymocytes is capable of delivering deletional signals in response to the superantigen staphylococcal enterotoxin B (SEB) as soon as the receptor is expressed.We also show that if TcR+ thymocytes are allowed to mature in organ cultures of embryonic thymus before SEB is added, they respond by proliferation giving rise to blast cells of CD4-CD8-, CD4+CD8- or CD4-CD8+ phenotypes.

1 Introduction The thymus is crucial for the production of T lymphocytes and supports the generation of large numbers of thymocytes with diverseTcR specificities (reviewed in [l]). Recent studies have shown that deletion (negative selection) of thymocytes reacting with self antigens [2-61 occurs by a process of apoptosis in which engagement of the TcR activates cell death by DNA degradation [7,8]. However, results have been reported that suggest that newly generated thymocytes pass through a developmental window following initial expression of the TcR d p , which is characterized by ineffective coupling between the d p and CD3 components of the TcR, resulting in impaired signaling and resistance to deletion when the T c R d p component is engaged [9, 101. Furthermore, it has been suggested that positive selection of thymocytes displaying receptors with specificity for thymic MHC antigens might only be possible in this window prior to the coupling of the TcR dp to CD3 [Ill. To investigate further whether, following the initial expression of the TcR, thymocytes are resistant to deletion, we have studied the effects of the superantigen Staphylococcus aureus enterotoxinB (SEB) on the generation of Vg8+ thymocytes in thymic organ cultures. SEB, when bound to class I1 MHC, reacts withTcR specificities containing either Vp3,Vg7 or Vg8 elements [12]. We have shown previously that when SEB is added for a final 18 h to established thymic cultures containingTcR+ cells a proportion of Vg8+ cells are deleted by apoptosis [8]. In this study, we have cultured 14-day thymus lobes (i.e. removed at least 2 days

prior to initial TcR expression) in the continuous presence of SEB added to the cultures at the outset. If newly formed thymocytes are resistant to deletion, they should be detected in cultures at all time points, since 14-day lobes contain stem cells from which new thymocytes continue to be generated for at least 2 weeks of culture [13]. However, we find an almost total elimination of Vg8+ cells arguing against the notion of an initial phase of resistance to deletion in immature thymocytes as a result of functional uncoupling of the a/p and CD3 components of the TcR. In line with our previous study [8], we find that if TcR+ thymocytes are allowed to develop in organ cultures before SEB is added, some Vg8+ cells resist deletion. We now provide evidence that at least some of these cells have reached a point of maturity where they respond to SEB by proliferation as described for peripheral Tcells [12]. These activated thymocytes become blast cells with the phenotypes CD4-CD8-, CD4+CD8- or CD4-CD8+.

2 Materials and methods 2.1 Embryonic material and organ culture BALB/c mouse embryos at day 14 of gestation were obtained from timed pregnancies (day of vaginal plug = day 0). Isolated embryonic thymus lobes were organ-cultured for periods varying from 5-19 days in DMEM containing 10% FCS as described previously [14]. SEB (Sigma, Poole, GB) was added to cultures to give a final concentration of 10 pg/ml[8] either at the outset of the culture or for the final 40 h of the culture period.

[I 86481 Correspondence: Eric J. Jenkinson, Department of Anatomy, Medical Schoo1,VincentDrive, Edgbaston, Birmingham B15 2TJ, GB Abbreviations: AP: Alkaline phosphatase BrdUrd: Bromodeoxyuridine SEB: Staphylococcal enterotoxin B 0 VCH Verlagsgesellschaft mbH, D-6940 Weinheim, 1990

2.2 Immunolabeling and FCM analyses

Cell suspensions were prepared from SEB-treated and from untreated cultures by gently teasing the thymus lobes apart with fine knives.Vg8 expression was detected by sequential incubation in F23.1 hybridoma SN [15], biotinylated antiOO14-2980/90/1111-2517$3.50 + .25/0


Eur. J. Immunol. 1990.20: 2517-2520

E. J. Jenkinson, R. Kingston and J. J. T. Owen

mouse IgG (Amersham International, Amersham, GB) and FITC-streptavidin (Amersham) with intervening washes over an FCS cushion. Vp6 was detected using hybridoma SN 44-22-1 [16] a gift from Dr. H. Hengartner. The SN was followed sequentially by biotinylated anti-rat Ig and FITC-streptavidin. After labeling , cell suspensions were fixed in PBS containing 0.5% paraformaldehyde and analyzed (20 000 events) in a FACS 440 (Becton Dickinson, Mountainview, CA) using logarithmic amplification. Control profiles for each treatment were prepared by analyzing cells stained with the second and third steps only. Double labeling for J l l d and Vp8 was carried out by sequential incubation in F23.1 SN, anti-mouse IgG alkaline phosphatase (AP), J l l d SN (containing 1% normal mouse serum) and biotinylated anti-rat IgG (containing 1% normal mouse serum). Cells were cytospun onto slides and incubated with AP substrate kit (Vector, Peterborough, GB) t o develop a red reaction product for identification of Vp8. They were then incubated in FITC-streptavidin to identify J l l d before mounting in an anti-fade reagent (diazobicyclo-octane; DABCO) and carrying out counts based on screening a minimum of 500 cells. We have found that the red AP substrate fluoresces brilliantly red under conditions for exciting either fluorescein or rhodamine fluorescence - the latter most strongly [17]. Thus, the red fluorescence identifying Vp8 staining can be viewed at the same time as the green fluorescence identifying J l l d .

anti-BrdUrd antibody (Becton Dickinson). The red reaction product generated by the AP substrate kit is unaffected by acid treatment and remains billiantly fluorescent. Thus, the red fluorescence of the AP product identifying Vg8 can be viewed simultaneously with green FITC fluorescence identifying BrdUrd incorporation using conditions for FTTC observation (Zeiss Filter set 9).

3 Results 3.1 The presence of SEB throughout the entire period of culture of thymus lobes results in the absence of thymocytes expressing surface Vp8 Cultures of 14-day mouse embryo thymus (with or without SEB) were examined at various time points for the presence of Vp8+ thymocytes. The results (Table 1) show that in control cultures (no SEB) there is an increasing proportion of Vp8+ cells with time (about 1% at 5 days of culture and up to 18%at 19 days of culture). By contrast, at most time points no Vp8+ cells are found in cultures exposed to the continuous presence of SEB (even after 19 days of culture). Table 1 also shows that Vp6+ cells are present in cultures irrespective of SEB treatment, suggesting that the effect of SEB onVg8+ cells is not due t o general toxicity. Fig. 1shows the FCM profiles obtained from one of these experiments.

To determine the CD4CD8 phenotype of Vp8+ cells in organ cultures treated with a final pulse of SEB,Vp8+ cells were obtained by immunomagnetic selection using Dynal beads (Dynal, Oslo, Norway). Cell suspensions were first depleted of M a using anti-rat IgG-conjugated beads precoated with a cocktail of rat anti-mouse M@ mAb F4/80, SER4 and 5C6 (a kind gift from Dr. S. Gordon).V$3+ cells were then isolated by incubation in F23.1 SN followed by rosetting with anti-mouse Ig-coated Dynal beads. The rosetted cells were separated magnetically and cultured for 2 h, during which time the majority of cells shed their beads, which were removed magnetically, t o leave a beadfree suspension of predominantly blast-like Vp8+ cells. These cells were collected and labeled with a mixture of PE-conjugated anti-CD4 and FITC-conjugated anti-CD8 (both Becton Dickinson) on ice for 40 min. Washed cells were fixed in 0.5% paraformaldehyde prior t o FCM analysis as before. 2.3 Analyses of DNA synthesis in Vb8+ cells Cell suspensions from organ cultured lobes were incubated for 1 h at 1 x 106-2 x lo6 cells/ml in medium containing bromodeoxyuridine (BrdUrd; 11500 dilution of BrdUrd, fluorodeoxyuridine-containing cell labeling reagent, Amersham Int.) and 5% heat-inactivated normal rabbit serum (to reduce Fc binding in subsequent surface labeling) in 96-well plates. Cells were recovered, washed and labeled for Vg8 by sequential exposure to F23.1 SN and anti-mouse Ig AP (Vector). Cytospin preparations were made onto glass slides and the cells were treated with the red AP substrate kit (Vector). Following exposure to 4 M HCI, newly incorporated BrdUrd within DNA was revealed with FITC-conjugated

Relative log fluorescence V P 8

Figure 1. FCM analysis of Vg8 expression by cells from 14-day thymus lobes cultured for 13 days in the absence (a) or continuous presence (b) of SEB. The hatched area in (a) between the profile after F23.1 labeling and the control profile (second and third labeling steps only) indicates the proportion of cells (17.5%) that are Vp8+. In (b) the F23.1 and control profiles are superimposed indicating an absence of Vp8+ cells.

Eur. J. Immunol. 1990.20: 2517-2520

Deletion of thymocytes


Table 1. Effect of the continuous presence of SEB on the proportion of cells expressing Vp8 or Vp6 in fetal thymus organ cultures

Vg element


Age of culture (days) 5 6 9 13 15 19


13 19


Untreated 1.2b) 1.7b) 13.1b) 9.6b) 11.5 17.5 14.7h) 12.ob) 18.4 4.0 5.4

v,+ SEB treatedP) ob) 0.2b)

ob) 0.9b)

ob) 0 0

ob) 0.2b) 0

6.5 10.3

a) Average cell yield per lobe (minimum of eight lobes per group) was either unaffected or slightly reduced by continuous culture in SEB. b) Cultures analyzed by scoring a minimum of 100 cells in the microscope following Vp8 staining and detection using APlabeled conjugates. All other cultures were analyzed by FCM .

3.2 Vp8 cells that resist deletion by SEB added to established cultures of embryonic thymus show DNA synthesis and most have mature phenotypes We have shown Dreviouslv that addition of SEB for a final 18 h to establishkd cultures of embryonic thymus results in partial depletion of Vp8+ cells accompanied by DNA degradation into oli~onucleosomal indicating that the apoptosis L81+In this we have investigated whether the resistant cells are activated into DNA synthesis as found when mature peripheral T cells are activated by SEB [12]. In two cultures of 6 days duration 34.7% and 33.3% of the residual Vp8+ cells were labeled by BrdUrd (Fig. 2) after a 1-h pulse of the latter. In culturesof 9 and 15 days duration 48.0% and 30.8% of Vp8+ cells were labeled under the same conditions.

Figure 2. Vg8+ blast cells from organ cultures pulsed with SEB for a final 40 h incorporate BrdUrd, indicating the presence of DNA synthesis. Surface Vg labeling is red, nuclear BrdUrd labeling is green. (a) Phase contrast, (b) Same field fluorescence.

Lc P)

1 U

The blast morphology of the Vp8+ cells remaining in cultures after a terminal 40-h pulse of SEB was confirmed by cytospin preparations (Fig. 2) and by forward light scatter in FCM analysis (Fig. 3). J l l d is a useful marker of maturity in the Tcell lineage distinguishing between immature cells up to and including the CD4+CD8+ stage, which have high levels of the antigen, and mature CD4+ or CD8+ Tcells, which have low levels [18]. To exclude the possibility that the blast cells found in organ cultures after a terminal pulse of SEB might be immature thymocytes, we examined the J l l d status of Vp8+ cells by double labeling using the fluorescence microscope. The results of two separate experiments showed the most (95% and 96%) of Vb8+ cells remaining after SEB treatment have only low levels of J l l d , thus confirming their mature status.

Forward light scatter Figure 3. Forward light scatter profiles giving relative cell size comparisons between newborn thymocytes (A) and Vg8+ cells isolated from 13-day cultures pulse with SEB for a final 40 h (B). It can be seen that the Vp8+ cells are predominantly large cells compared to the thymocyte profile.


E. J. Jenkinson. R . Kingston and J. J.T. Owen

To determine the CD4CD8 phenotype of these cells, Vp8+ cells were isolated by magnetic separation and analyzed for CD4 and CD8 expression after shedding their Dynal beads. In two separate FCM analyses, cells with the mature CD4+CD8- (10.8% and 32.6%) and CD4-CD8+ (64.4% and 40.5%) phenotypes characteristic of peripheral Tcells were observed. Of the remaining Vp8+ cells, most were CD4-CD8- (19.8% and 14.7%) and the rest were doublepositive CD4+CD8+cells.

4 Discussion

Eur. J. Immunol. 1990. 20: 2517-2520

continuous exposure to SEB argues that all these categories of cells, capable of a proliferative response to SEB, are derived from a precursor population which passes through a phase of sensitivity to deletion. Finally, our results support two main conclusions: first, that the TcRa/P on developing thymocytes is fully capable of delivering deletional signals as soon as it is expressed; and second, that, after a period of maturation in organ culture, thymocytes become resistant to deletion and respond to SEB by proliferation giving rise to blast cells. Thus, the conditions leading to either deletion or proliferation are present within the thymus lobes, suggesting that the decision as to which process is activated is more dependent on the maturational status of the thymocyte than the conditions of antigen presentation.

We have designed experiments to test the proposal that, following the initial expression of the TcR a@ on developing thymocytes, ineffective coupling between the a l p and CD3 components results in resistance to deletion and so perhaps permits positive selection following interaction of We are grateful to Alan Murdoch for skilled technical assistance, to the a@ component with thymic MHC [9-ll].We have used Alison Orchard for helping with photography and to Christine organ cultures of 14-day embryo thymus which is only Hepherd for typing this manuscript. beginning full TcR gene rearrangements but after 3 days of Received June 20, 1990. culture generates thymocytes with surface TcR [131.These lobes continue to generate new TcR+ cells from a stem cell pool over a prolonged period of culture. Thus, precursor 5 References cells which express cytoplasmic but not surfaceTcR p chains are still present in cultures of 31-day duration [13]. If up to 1 Adkins, B., Mueller, C., Craig,Y., Okada, R. A . , Weissman, 50% of TcR+ thymocytes are resistant to deletion as I. L. and Spangrude, G. F., Annu. Rev. Imrnunol. 1987. 5: previously suggested [9, lo], then even in the continuous 325. presence of SEB some Vp8+ cells should escape deletion 2 Kappler, J. W., Staerz, U. €!,White, J. and Marrack, P. C., Nature 1988. 332: 35. and so be readily detectable at all culture periods. However, 3 MacDonald, R., Schneider, R., Lees, R. K., Howe, R . C., we find that SEB prevents the appearance of Vp8+ cells. Archa-Orbea, H . , Festenstein, H., Zinkernagel, R. M. and Interestingly, precursor cells with cytoplasmic Vp8 chains Hengartner, H., Nature 1988. 332: 40. are present in these cultures, suggesting that the deletional 4 Kisielow, P., Bluthman, H . , Staerz, U. D., Steinmetz, M. and process occurs as soon as surface Vp8 is expressed (data not Von Boehmer, H . , Nature 1988. 333: 742. shown). Previously, we have provided evidence t o indicate 5 Pircher, H . , Burki, K., Lang, R., Hengartner, H. and Zinkerthat the deletional process occurs by apoptosis [8]. nagel, R. M., Nature 1989. 342: 599. Our results suggest that signals via the a@ component of the TcR are able to effect deletion in newly formed TcR+ thymocytes. This is in agreement with the finding of total deletion of TcR+ cells expressing an a/p transgene in male mice where the specificity of the transgene is for the H-Y antigen [19], but is in contrast with a previous report that SEB in thymus organ cultures produces only partial deletion of Vp8+ cells [9]. However, the latter cultures were initiated using 17-day embryo thymus when TcR+ cells are already present, providing a possible explanation for the difference if cells acquire resistance to deletion as they mature (see below). In contrast to the elimination of all Vp8+ if SEB is present continuously throughout culture, a terminal pulse of SEB in established cultures that already contain Vp8+ cells produces only partial deletion. In the present study, we show that at least a proportion of the resistant cells are activated into DNA synthesis and transform into blasts with low levels of J l l d , confirming their mature status. Although SEB is known to be presented by MHC class I1 [20], the blasts are heterogeneous in terms of phenotype and include either double-negative CD4-CD8- or single-positive CD4+ or CD8+ cells, in line with the observation that mature T cell clones can be stimulated by MHC class IIpresented staphylococcal enterotoxins irrespective of their CD4/CD8 phenotype [21]. In addition, the absence of activated blasts of any of these phenotypes in cultures with

6 Fry, A . M., Jones, L. A . , Kruisbeek, A . M. and Matis, L. A . , Science 1989. 246: 1044. 7 Smith, C. A., Williams, G. T., Kingston, R., Jenkinson, E. J. and Owen, J. J. T., Nature 1989. 337: 181. 8 Jenkinson, E. J., Kingston, R . , Smith, C. A . , Williams, G. T. and Owen, J. J. T., Eur. J. Imrnunol. 1989. 19: 2175. 9 Finkel, T. H., Marrack, I?, Kappler, J. W., Kubo, R. T. and Cambier, J. C., J. Immunol. 1989. 142: 3006. 10 Finkel, T. H., Cambier, J. C., Kubo, R. T., Born, W. K . . Marrack, €! and Kappler, J. W., Cell 1989. 58: 1047. 11 Kappler, J. W., Finkel, T. H. and Marrack, F!, Prog. Immunol. 1989. 7: 26.5. 12 White, J., Herman, A . , Pullen, A . M., Kubo, R., Kappler, J.W. and Marrack, €!, Cell 1989. 56: 27. 13 Owen, J. J.T., Kingston, R. and Jenkinson, E. J., Immunology 1986. 59: 23. 14 Jenkinson, E. J., Franchi, L. L., Kingston, R. and Owen, J. J. T., Eur. J. Immunol. 1982. 12: 583. 15 Staerz, U. D., Rammensee, H. G., Benedetto, J. and Bevan, M. J., Immunology 1985. 134: 3994. 16 Archa-Orbea, H . , Groscurth, P., Lang, R., Stilz, L. and Hengartner, H . , J. Immunol. 1983. 130: 2952. 17 Murdoch, A . , Jenkinson, E. J., Johnson, S. G. and Owen, J. J. T., J. Immunol. Methods, in press. 18 Crispe, I. N. and Bevan, M. J., J. Imrnunol. 1987. 137: 2013. 19 Kisielow, €!, Bluthman, H., Staerz, U. D., Steinmetz, M. and Von Boehmer, H . , Nature 1987. 333: 742. 20 Janeway, C. A. ,Yasi, J., Conrad, P., Katz, M. ,Vroegop, S. and Buxser, S., Immunol. Rev. 1989. 107: 61. 21 Fleischer, B. and Schrezenmeier, H., J. Exp. Med. 1988. 167: 1697.

beta component of the T cell receptor is engaged by the superantigen staphylococcal enterotoxin B.

It has been reported that, following the initial expression of the T cell receptor (TcR) alpha/beta, newly generated thymocytes pass through a develop...
465KB Sizes 0 Downloads 0 Views