Eur. J. Immunol. 1991.21: 819-822

Biased Tcell receptor V gene usage in T lymphocytes

819

Short paper Johan Grunewald, Carl Harald Janson and Hans Wigzell Department of Immunology, Karolinska Institute, Stockholm

Biased expression of individual T cell receptor Vgene segments in CD4+ and CDS+ human peripheral blood T lymphocytes* The expression of seven different a and p gene segments of theTcell receptor on normal human CD4+ and CD8+ T lymphocytes, respectively,was examined using V gene-specific monoclonal antibodies. We found a statistically significant (p < 0.001) bias of the expression of four V gene products towards the CD4+ subpopulation. In every individual analyzed, the Vp5.1 gene segment was expressed to a higher degree among CD4+ compared to CD8+ cells, with a median value of 4.8% among CD4+ cells and 1.5% among CD8+ cells.There was also a statistically significant skewness in the usage of the Vg6.7,Vp8 and Vg12 gene segments towards the CD4+ Tcell population, but not as dominating as for the Vg5.1. Lymphocytes from umbilical cord blood showed similar skewed reactivities for the Vg5.1, Vg6.7 and Vgl2-specific monoclonal antibodies. Aspects of positive and negative selection, as possible explanations for these findings, are discussed.

1 Introduction Tcells recognize foreign antigens as peptides in the context of self MHC, the CD4+ Tcells being MHC class I1 restricted and the CD8+ Tcells MHC class I restricted.To achieve this specific reactivity, thymocytes are educated in the thymus through only partly understood mechanisms. The TcR on matureT lymphocytes is expressed as a disulfide-linked a@ chain heterodimer, each chain having a V and a C domain. To create the variable part of the TcR, the a / p genes rearrange one of 50-100 V with one of 60-80 J segments, in the case of the a chain, and one of 75-100 V with one of 2 D and one of 13 J segments, for the p chain [l]. Cells reacting incorrectly with self-MHC are eliminated in the thymus, creating clonal deletions [2]. This elimination process of autoreactiveTcells involves mainly the immature CD4+CD8+thymocytes [2,3], containing the precursorsof mature CD4+ or CD8+ Tcells. Maturing thymocytes are also positively selected, as shown in mouse models, whereT cells using certain Vp gene segments could be selected this way [4, 51. The accessory molecules, CD4 and CD8, bind to nonpolymorphic portions of MHC class I1 and I molecules, respectively, and enhance the binding of the TcR to its ligand [6]. MHC class I and class I1 molecules are considered to direct the maturing thymocytes to either the CD4+ or the CD8+ Tcell subpopulation [7].Treatmentwith mAb to block CD4 or CD8 molecules in the thymus can prevent the development of CD4+ or CD8+ (singlepositive) peripheral T cells, respectively [3, 8, 91. [I 88621

*

This work was supported by the Cancer Society in Stockholm, the Swedish Cancer Society and the Soderberg Foundation.

Correspondence: Johan Grunewald, Department of Immunology, Karolinska Institute, Box 60400, S-10401, Stockholm, Sweden 0 VCH Verlagsgesellschaft mbH, D-6940 Weinheim, 1991

In this study, we analyzed the preference by which certain V, or Vg gene segments were expressed on the CD4+ and the CD8+ human peripheral blood T cell subpopulations. Such a skewed usage has been described previously in mice [4, 10, ll].We used seven recently available mAb, directed against different gene products of the V part of the human TcR, in a double-staining immunofluorescence test. Altogether the mAb detected an average of 21% of the CD4+ and 15% of the CD8+ peripheral Tcells.

2 Materials and methods 2.1 mAb

The following antibodies against the V part of the TcR a / p chain were used: LC4 (IgG1, reactive withVg5.1, [ 12]), 1C1 (IgG1, reactive withVg5.2 Vg5.3 [13, 14]),W112 (IgG,, reactive with Vg5.3 [15]), OT145 (IgG1, reactive with an allotypic epitope, Vg6.7 [16]), 16G8 (IgGzb, reactive with theVg8 family [15]) and S511 (IgGa,reactive with theVg12 family [17]), were all provided by TCell Sciences Inc. (Cambridge, MA). The mAb F1 (IgGZ,, reactive with a V,2.3 gene product) was produced in our laboratory [18]. The OKT3 (anti-CD3) hybridoma (AmericanType Culture Collection, Rockville, MD) was grown in our laboratory. The TcR 61 mAb (T Cell Sciences Inc.) was used as a marker for theTcR y/6. FITC-conjugatedF(ab')2 fragments of rabbit anti-mouse Ig were purchased from Dakopatts A/S (Glostrup, Denmark). PE-conjugated Leu-2a antiCD8) and Leu3a (anti-CD4) were obtained from Becton Dickinson (Mountain View, CA). Normal mouse serum, produced from BALBk mice, was used at a dilution of 1 :500 in PBS.

+

2.2 Isolation of cells

Heparinized peripheral blood was obtained from 24 blood donors and umbilical cord blood (n = 4) was taken at 0014-298019 110303-0819$3.SO + ,2510

820

Eur. J. Immunol. 1991. 21: 819-822

J. Grunewald, C. H. Janson and H. Wigzell

partus. PBMC were separated by Ficoll-Hypaque (Pharmacia, Uppsala, Sweden) gradient centrifugation, washed twice in RPMI 1640 medium (Gibco, Paisley, GB) and diluted in PBS.

2.3 Immunofluorescence test and FCM analysis Cells were kept at + 4 "Cduring the following procedures: cells were incubated for 30 min with unlabeled antibody, washed twice with PBS and incubated with FITC-conjugated F(ab')z fragments of rabbit anti-mouse Ig for 30 min. After washing the cells three times with PBS, normal mouse serum, diluted 1 :500, was added and incubated for 10 min, to block remaining rabbit anti-mouse Ig. PE-conjugated Leu-2a or Leu3a was added, cells incubated for 20 min and washed twice in PBS. A FACScan flow cytometer (Becton Dickinson) and a Hewlett Packard 300 computer (Palo Alto, CA) was used for the FCM analysis. Lymphocytes were gated out by forward and side scatter setting and loJ cells were analyzed. Normal mouse serum was used as negative control. Optimal compensation was set for green and orange fluorescence. The CD4+ and CD8+ subpopulations were defined as Leu-3a+, OKT3+, TcR 61- and Leu-2a+, OKT3+,TcR 61-, respectively.

The Vg6.7-specific antibody was reported to react with an allotype [16], and in line with this one of the investigated individuals had a reactivity close to 0%. The median reactivity for theVg6.7-specific mAb was 4.4% of the CD4+ and 1.2% of the CD8+ Tcells. The Vg8 reactivity was also biased towards the CD4+ subpopulation, with a median reactivity of 4.7%, compared to 3.1% for the CD8+ cells. The difference, although not numerically great, was statistically significant (p < 0.001).TheVgene segment products detected by the Va2.3-specific mAb F1 (Fig. la),Vg5.2 + 5.3- and Vg5.3-specific mAb did not show any statistically significant preferential usage within any of the two T cell subsets (Table 1).

In Fig. 1, the reactivities of V,2- and Vg5.1-specific mAb within CD4+ and CDW subpopulations from individual blood samples are compared. In all blood donors, the reactivities with the CD4+, compared to the CD8+ cells, were higher for theVg5.1-specific LC4 mAb (Fig. lc). Such a completely dominating skewness towards one subpopulation was not seen for the other mAb.Thus, the anti-mAb showed a biased usage in 21 of 24, the anti-Vg8 in 20 of 24 12

lja)

0

I(b) I

3 Results Analyzing the reactivity of the anti-TcR mAb in the CD4+ and CD8+ subgroups of 24 blood donors, we found that some mAb had a skewed reactivity towards one of the subsets, whereas others showed no preferential staining. In Table 1 the median values of the reactivities of the different mAb with the twoTcell populations and thep values for the differences between the respective reactivity are shown. Four mAb (specific for Vg5.1,Vg6.7,Vg8 and Vg12) reacted with a statistically significant higher portion of CD4+ than with CD8+ Tcells. The greatest difference could be seen with the Vg5.1specific mAb LC4, with a median value of 4.8% within the CD4+ and 1.5% within the CDW subgroup. All individuals investigated with LC4 showed a higher reactivity within the CD4+ compared to the CDW group (Fig. lc). The Vg12specific mAb reacted with 1.9% of the CD4+ cells and 1.2% of the CDW cells, with a homogeneous staining pattern, as reflected by the minimum and maximum values (Table 1).

3

0

A00

I

.I

P .I

*

c)

2

12

,JCl

Paired individual observations

Figure I. Percent reactivity of mAb detecting V,2.3 (a and b) and Vg5.1 (c and d) gene products in CD4+ (0) and CD8+ (0) lymphocytes, from adult blood donors 1-24 (a and c) and from umbilical cord blood 1-4 (b and d).

Table 1. Median. minimum and maximum values of TcR-specific mAb reactivity with CD4+ and CD8+ lymphocytes in 24 blood donors. p values for the difference between the CD4+ and CD8+ populations are calculated by using the Wilcoxon-Mann-Whitney two-tailed test Specifity of mAb

46 CD4+')

% CD8+

p value

CD4 > CDgb1

Vp5.1 Vp6.7 VP8 vp12 Vg5.2 + 5.3 Vp2.3 Vp5.3

4.8 (1.8-6.1) 4.4 (0.3-8.8) 4.7 (3.1-15.7) 1.9 (0.2-2.4) 2.2 (1.63.5) 3.5 (2.3-5.5) 0.8 (0.4-1.6)

1.5 (0.3-3.8) 1.2 (0.0-12.3) 3.1 (1.4-9.3) 1.2 (0.5-2.1) 2.6 (1.1-15.1) 3.1 (0.6-10.1) 0.8 (0.2-13.2)

< 0.001 < 0.001 < 0.001 < 0.001

24 21 20 20 11 12 12

0.13 0.31 0.76

a) The bold numbers represent median values. Minimum and maximum values are shown within brackets . b) Figures represent number of individuals (total = 24) with higher reactivity of anti-TcR mAb within the CD4+ compared to the CD8+ subpopulation.

Biased T cell receptor V gene usage in T lymphocytes

Eur. J. Immunol. 1991. 21: 819-822

V l (V,21 4.2

Vp 5.1

1

5.0

821

thymus, or the result of a postthymic event, leading to the expansion of CD4+ Vg5.1+ cells. Considering the first explanation, positive selection of thymocytes expressing certain Vg gene segments has been shown in murine systems [4, 5 , 121. It was for example demonstrated that Vg6-expressing T cells were positively selected by MHC class I1 determinants, resulting in increased numbers of CD4+ Vg6+ matureTcells [5]. Could our results be explained by the opposite argument, namely a negative selection against CD8+V$. 1+ thymocytes? We believe not, since in several cases a negative selection of autoreactive cells has been shown to occur at the CD4+CD8+ (double-positive) immature thymocyte stage, resulting in the deletion of both CD4+ and CD8+ subsets. This was for instance shown for theVpl7a gene segment in a mouse model [2, 31.

Considering a post-thymic selection process, one might notice the superantigens, such as staphylococcal enterotoxins, that are known to selectively stimulate Tcells expressFITC-fluorescence (log) ing certain V gene families, irrespective of MHC class I1 Figure 2. FCM profiles from a representative experiment of haplotype [13]. A superantigen could thus be seen as double staining PBL. Fluorescence intensity of V,2.3- and Vg5.1- driving a VgS.l+ subpopulation linked to CD4+ in a specific mAb are shown on the abscissa and the PE fluorescence selective manner.Toinvestigate this possibility,we analyzed due to CD8-and CDCspecific mAb on the ordinata. Figures in umbilical cord blood lymphocytes, presumably never in right comers show percent reactivity within each subpopulation. contact with exogenous antigens, reflecting a "virgin" state of reactivity. In the umbilical cord blood, the same imbalance,when not accentuated, as in blood samples from and the Vg6Vg12 in 20 of 24 individuals tested, while the adults was seen (Fig. lb, d). Thus, it seems likely that the other mAb (Va2.3, Fig. la,Vg5.2 5.3 and Vg5.3 showed preferential usage of the Vg5.1 gene segment by CD4+ T no preferential usage (Table 1). cells is the result of positive selection events within the thymus, during maturation of T cells. A representative double-staining experiment is shown in Fig. 2. For this individual, the Va2.3-specific mAb F1 The MHC products are important for the selection proreacted with 3.8% and 4.2%, and the anti-Vg5.l with 2.6% cesses in the thymus. The fact that every individual had and 5.0% ,of the CD8+ and CD4+ subpopulations, respec- Vg5.1 being present in a higher percentage of CD4+ T vs. tively. CD8+ T cells, excludes restricted, polymorphic MHC determinants in the selective process. Accordingly, this The added reactivity of the TcR mAb (F1 LC4 1C1 + positive selection mechanism in the thymus is possibly OT145 16G8 S511) was calculated to median values mediated through the influence of non-polymorphic MHC (plO-p90) of 20.9% (17.6-25.8) of the CD4+ and 15.4% determinants. For the other Vgenes showing a preferential, (11.8-22.8) of the CD8+ cells.There was a greater variation although less completely skewed expression, polymorphic in the reactivities within the CD8+ cell populations, while determinants may play more decisive roles [ll]. the CD4+ cells had a more homogeneous reactivity (range). Lymphocytes from umbilical cord blood (taken at partus) Alternatively, Mls-like gene products [ 191 could selectively had similar reactivities as peripheral blood from adults, stimulate (or inhibit) certain Vg gene segments. These with a bias towards the CD4+ cells for theVg5.1 (Fig. Id), proteins might cross-link non-polymorphic parts of the Vg6.7 and Vgl2-specific mAb. MHC with a certainVg gene product, hereby driving thisVp gene expressing thymocyte towards MHC class I or class 11. The result would be a matureTcel1,expressing a specificVg gene product preferentially on CD4+ or CD8+ cells. 4 Discussion

+

+

+

+

+

Using mAb directed against different gene products of the variable part of the TcR alp, we have calculated the expression of thesevgene segments within the two major T cell subpopulations, the MHC class I-restricted CD8+ and the MHC class 11-restricted CD4+ cells. For some of the antibodies, there was a statistically significant skewed usage between the two subpopulationsinvestigated, but for others, no difference was detected.To explain this phenomenon (in the following concentrating on the finding of the Vg5.1-specific mAb reactivity), two major possibilities should be considered: the skewness could either be the result of a negative or positive (or both) selection within the

Also self-peptides are critically involved in selection of the T cell repertoire [20]. There are different sources of antigens, as well as cellular antigen-processing machineries, related to the MHC class I and class I1 molecules. Accordingly, it could be expected that certain self-peptides bind differentially to MHCclass I and class 11, resulting in a corresponding selection of a specific V gene expression towards either the CD4+ or the CD8+ subset. A possible linkage of theVg5.1 gene to a favored Dg or Jg gene segment during TcR rearrangement, or a preferential pairing with a V, gene segment product, must also be taken in consideration.

J. Grunewald, C. H . Janson and H. Wigzell

822

In murine models, the findings of other groups support our results. Thus, specific Vp gene segments predominantly used by CD8+ Tcells [lo] as well as by CD4+ Tcells [4, 111 have been described. In conclusion, we tested the reactivity of seven mAb, directed against different gene products of the variable part of the human TcR, within the CD4+ and the CD8+ Tcell subsets of human PBMC and umbilical cord blood. We found a biased reactivity of some of the mAb to the CD4+ subset, being most pronounced with theVg5.1 gene product where all individuals investigated showed a higher reactivity in the CD4+ subset compared to the CD8+. A positive selection mechanism in the thymus, possibly through the influence of non-polymorphic MHC determinants, is the most likely explanation for this phenomenon. We are presently exploring this possibility. Received August 25, 1990; in revised form October 8, 1990.

Eur. J. Immunol. 1991. 21: 819-822 5 MacDonald, H. R., Lees, R., Schneider, R., Zinkernagel, R. and Hengartner, H . , Nature 1988. 336: 471. 6 Gabert, J., Langlet, C., Zamoyska, R., Parnes, J., SchmittVerhulst, A.-M. and Malissen, B., Cell 1987. 50: 545. 7 Teh, H. S., Kisielow, F!, Scott, B., Kisihi, H., Uematsu, Y., Bliithmann, H. and Von Boehmer, H., Nature 1988. 335: 229. 8 Zuniga-Pfliicker, J. C., Jones, L., Longo, D. and Kruisbeek, A . , J. Exp. Med. 1990. 171: 427. 9 Zuniga-Pflucker, J., McCarthy, S., Weston, M., Longo, D., Singer, A . and Kruisbeek, A . , J. Exp. Med. 1989. 169: 2085. 10 Liao, N.-S., Maltzman, J. and Raulet, D., J. Immunol. 1990. 144: 844. 11 Tomonari, K., Lowering, E. and Spencer, S., Immunogenetics 1990. 31: 333. 12 Maecker, H. and Levy, R., J. Immunol. 1989. 142: 1395. 13 Kappler, J., Kotzin, B., Herron, L., Gelfand, E., Bigler, R., Boylston, A . , Carrel, S., Posnett, D., Choi,Y. and Marrack, F!, Science 1989. 244: 811. 14 Boylston, A . , Borst, J.,Yssel, H., Blanchard, D., Spits, H. and DeVries, J., J. lmmunol. 1986. 137: 741. 15 Tian, W.-T., Skibbens, R., Kubinec, J., Henry, L.,KO, J.-L.,

Yeh, G. and Ip, S., Monoclonal antibodies specific to humanT cell antigen receptor V-beta gene products. FASEB J. 3:

5 References Kimura, N . , Toyonaga, B.,Yoshikai, Y., Du, R. and Mak, T., Eur. J. Immunol. 1987. 3: 375. Kappler, J., Roehm, N. and Marrack, F!, Cell 1987. 49: 273. Fowlkes, B. J., Schwartz, R. H. and Pardoll, D. M., Nature 1988. 334: 620.

Blackman, M., Marrack, I? and Kappler, J., Science 1989.244: 214.

A486. 16 Posnett, D.,Wang, C. and Friedman, S., Proc. Natl. Acad. Sci., USA 1986. 83: 7888. 17 Bigler, R., Fisher, D.,Wang, C., Rinnooy Kan, E. and Kunkel, H., J. Exp. Med. 1983. 158: 1OOO. 18 Janson, C. H . , Tehrani, M., Mellstedt, H. and Wigzell, H., Cancer lmmunol. Immunother. 1989. 28: 225. 19 Janeway, C.,Yagi, J., Conrad, P.,Katz, M., Joes, B.,Vroegop, S. and Buxner, S., lmmunol. Rev. 1989. 107: 61. 20 NikoliC-ZugiC, J. and Bevan, M., Nature 1990. 344: 65.

Biased expression of individual T cell receptor V gene segments in CD4+ and CD8+ human peripheral blood T lymphocytes.

The expression of seven different alpha and beta gene segments of the T cell receptor on normal human CD4+ and CD8+ T lymphocytes, respectively, was e...
379KB Sizes 0 Downloads 0 Views