Reprints Available Directly from the Publisher Photocopying Permitted by License Only Tcr Vcz-and V//-gene Segment Use in T-cell Subcultures Derived from a Type-iii Bare Lymphocyte Syndrome Patient Deficient in Mhc Class-ii Expression

Previously, we and others have shown that MHC class-II deficient humans have greatly reduced numbers of CD4+CD8– peripheral T cells. These type-III Bare Lymphocyte Syndrome patients lack MHC class-II and have an impaired MHC class-I antigen expression. In this study, we analyzed the impact of the MHC class-II deficient environment on the TCR V-gene segment usage in this reduced CD4+CD8– T-cell subset. For these studies, we employed TcR V-region-specific monoclonal antibodies (mAbs) and a semiquantitative PCR technique with V α and V ß amplimers, specific for each of the most known V α- and V ß;-gene region families. The results of our studies demonstrate that some of the V α-gene segments are used less frequent in the CD4+CD8– T-cell subset of the patient, whereas the majority of the TCR V α- and V ß-gene segments investigated were used with similar frequencies in both subsets in the type-III Bare Lymphocyte Syndrome patient compared to healthy control family members. Interestingly, the frequency of TcR V α12 transcripts was greatly diminished in the patient, both in the CD4+CD8– as well as in the CD4–CD8+ compartment, whereas this gene segment could easily be detected in the healthy family controls. On the basis of the results obtained in this study, it is concluded that within the reduced CD4+CD8– T-cell subset of this patient, most of the TCR V-gene segments tested for are employed. However, a skewing in the usage frequency of some of the V α-gene segments toward the CD4–CD8+ T-cell subset was noticeable in the MHC class-II deficient patient that differed from those observed in the healthy family controls.


INTRODUCTION
During T-cell development, positive and negative selection events occur within the thymic microenvironment that influence in part the composition of the mature peripheral T-cell compartment (Fink and Bevan, 1978;Zinkernagel et al., 1978;Sprent et al., 1988;Schwartz, 1989;Nikolic-Zugic and Bevan, 1990;Teh et al., 1991). Interactions between T-cell receptor (TcR) and MHC class-I or class-II antigens play a pivotal role in this process. As a result, only T cells bearing TcRs with moderate affinity for "self" MHC class I or class II can be found in the'periphery . *Corresponding author. The avidity of a given TcR, expressed on the double-positive thymocyte (CD4+CD8+), for a MHC class-I or class-II antigen dictates the resulting single-positive phenotype (CD4+CD8or CD4-CD8+) of the mature T cell (Kaye et al., 1989;von Boehmer et al., 1989;Teh et al., 1990). However, the specificity of TcRs for MHC class-I or-II antigens can be influenced in part by the coexpressed accessory molecules (CD4 or CD8, respectively;Robey et al., 1991). Interference in the interactions involved in this selection process has a dramatic effect on the composition of the peripheral T-cell compartment. (1) Disturbance of the TcR/MHC class-II antigen interaction in neonates due to masking of I-A antigens with anticlass-II monoclonal antibodies results in diminished numbers of peripheral CD4+CD8-T cells (Kruisbeek et al., 1983(Kruisbeek et al., , 1985. (2)  lack class-I heavy-chain expression due to a disrupted 2-microglobulin (2m) gene have greatly reduced numbers of CD4-CD8+ peripheral T cells (Koller et al., 1990;Zijlstra et al., 1990).
The diversity of TcRs is generated in part through the use of various different Vowand V]/gene segments This V-gene segment use however is not random, but is influenced by the MHC haplotype, as has been shown in mice (Benoist and Mathis, 1989;Bill and Palmer, 1989). As a result, use of certain TcR V-gene segment families can be strongly biased to either the CD4+CD8-or CD4-CD8+ population or com-plet61y deleted in both subsets (Liao et al., 1990;Singer et al., 1990). Analysis of the TcR V-gene segment use of clones sharing reactivity to a certain peptide showed that in mice these clones often use a limited Vo and Vfl repertoire (Fink et al., 1986;Engel and Hendrik, 1988). In man, also a restricted TcR Vowand Vfl-gene segment use in clones reactive to a defined peptide has been reported (Wucherpfennig et al., 1990;Ben-Nun et al., 1991).
The impact of MHC class-II antigen expression on the selection of the peripheral T-cell repertoire in man can be studied in patients with the Bare Lymphocyte Syndrome (BLS) (also referred to as MHC class-II deficiency syndrome). This syndrome is a lethal combined immunodeficiency often associated with opportunistic infections and malabsorptiofi. The main characteristic of this syndrome is defective MHC class-II antigen expression (type-II BLS), sometimes in combination with a reduced MHC class-I expression (type-III BLS), on cells that normally express these antigens, including those involved in thymic differentiation (Touraine et al., 1978;Schuurman et al., 1979;Schuurman et al., 1985;Rijkers et al., 1987). The lack of MHC class-II in conjunction with aberrant class-I expression was also noticeable in established EBV transformed B-cell lines and T-cell lines (Bull et al., 1990;Lambert et al., 1990;Lambert et al., 1991). Previously, we and others (Zegers et al., 1984;Rijkers et al., 1987;Lambert et al., 1991) have shown that in a human environment devoid of MHC class-II antigen expression, the numbers of peripheral CD4+CD8-T cells are greatly reduced. In addition, these CD4+CD.8-cells have an in vitro proliferative disadvantage compared .to CD4+CD8-cells derived from healthy controls (Lambert et al., 1991).
In this study, we have analyzed the impact of the MHC class-II deficient environment on the relative frequencies of TCR Vowand V]/-gene segment use in the CD4+CD8peripheral T-cell subset of a type-III BLS patient.
Aliquots were thawed and stained with anti-CD4 and anti-CD8 monoclonal antibodies. CD4+CD8-and CD4-CD8+ T cells were isolated by FACS sorting. CD4+CD8-clones were established by limiting dilution of stained and sorted The profile of the PCR cycles was as follows: denaturation at 95 C for 60 s, annealing at 55 C for 60 s and elongation at 72 C for 60 s in a Bio-Med Thermocycler 60. Amplifications were performed for a maximum of 40 cycles to determine trace amounts of specific V-gene segment usage. After 25 (Vc) or 20 (V/J) cycles of amplification, aliquots were drawn and used for Southern blot analysis. The PCR products were size fractionated on a 0.8% agarose gel and transferred to Gene Screen Plus (NEN). TcR-specific DNA sequences were detected by hybridization using an internal 32p labeled Co or C/J probe, respectively (Yoshikai et al., 1984;Yanagi et al., 1985).
Transfer and hybridization were performed according to the manufacturer's instructions.
MHC class II transcription analysis (DR, DP, and DQ A and B chains) was performed as described (Lambert et al., 1991). Transfer and hybridizations were performed as described before. MHC class-II specific sequences were vis-ualized using probes from the Xth International HLA workshop.

Flow Microfluorometric Analysis
Cells were analyzed by indirect immunofluorescence on a flow cytometer (FACScan, Becton-Dickinson). Cells were washed with PBS containing 1% BSA and 0.1% sodium azide. Aliquots of 1 105 cells were incubated at 4 C with monoclonal antibodies (mAb) for 30 min, either directly analyzed when conjugated mAbs were used, or washed and reincubated with fluorescein-isothiocyanate-conjugated rabbit antimouse immunoglobulin F(ab) fragments (Dakopatts).
After another 30 min, cells were washed three times and analyzed. The used monoclonal antibodies were anti-CD4 Ft and anti-CD8 eE (Becton-Dickinson) and antibodies that are specific for family members belonging to the TcRVflS-, V/6-, Vfl8-, Vf112-, and VcC2-gene segment family (Diversi-T, T cell Science) and Vc12 (DerSimonian et al., 1991).

Immunofluorescence Analysis
In previous studies, we have shown that despite the apparent lack of MHC class-II antigen expression in the typeSIII Bare Lymphocyte Syndrome (BLS) patient MBI, CD4+CD8-T cells can be found among PBMC, albeit their numbers are reduced (ratio CD4+/CD8+=0.75; Lambert et al., 1991). Of interest was the observation that within the CD4+CD8-subset, 48% of the CD4+CD8-T cells coexpressed the CD45RO marker, which is characteristic for activated T cells. As shown in Fig. 1, the T-cell lines derived from patient MBI (type-III BLS) lacked detectable levels of MHC class-II transcripts as determined by polymerase chain reaction (PCR) as well as staining with anti-HLA class-II specific monoclonal antibodies (Lambert et al., 1991).
To address the question whether the lack of MHC class-II antigen expression has an impact on the composition of this impaired peripheral CD4+ T-cell subset with respect to employment of the various TcR V-gene segments, we studied the frequency of TcR Vowand V]J-gene segment use in these cells. PBMC were stained with mon-  show that all tested TcR V-region families were expressed in the patient-derived PBMC. The observed frequencies in the patient-derived material were within normal range. Variation in the frequency of use of the different V-gene families was noticeable when the different individuals were compared. In general, the percentage of T cells carrying a particular V-gene segment was always less than 10%, except for 8 in the patient-derived PBMC (see Table 1). The relative high frequency of V]J8 positive cells seen in the patient and healthy controls was also observed in PBMC derived from other, unrelated individuals using the same antibody (data not shown).
To determine the TcR V]/and V o use in the individual CD4+CD8-and CD4-CD8+ T-cell populations, in vitro expanded T cells were isolated by FACS sorting and analyzed (Table 2). In the patient-derived material, all tested Vowand   (Fig. 2). In general, TcR VI/gene segment families with the most members showed the strongest amplification. An exception was V119, which has only one member and was strongly amplified in 011 individuals, both in the CD4+CD8-and CD4-CD8+ T-cell population. Some TcR Vl/-gene segment families were used at almost equal frequencies in both CD4+ and CD8+ populations in all individuals, whereas other gene segment families showed an apparent skewing to one subpopulation in some but not all individuals. For instance, V]/1 showed a skewing to the CD4-CD8+ population in KBI, FBI, and MBI, whereas Vf114 was preferentially used in the CD4--CD8+ population in KBI, FBI, and ZBI but not MBI. Likewise, Vf118 was skewed to the CD4+CD8population in FBI, MBI, and ZBI (Fig.   2).
To study the TcR Vcz-gene segment use, 22 different Vcz-gene segment families were analyzed. The results of these studies'are depicted in Fig. 3. Differences in the frequency of use of the various TcR Vcz families are more pronounced in the CD4+CD8-and CD4-CD8+ populations within one individual, compared to the TcR V-gene segment use (Fig. 3). For instance, in the sibling FBI, Vo15 and Vcz16 seem to be more frequently used in the CD4-CD8+ subset, whereas in the father ZBI, amplification of Vcz15 and 16 was comparable in both subsets (Fig. 3). In the CD4+CD8-subculture of the type-III BLS patient, a clear diminished use of the Vzl, 2, 3, 7 families could be observed, which was not seen in the healthy family controls in addition to biased expression in the CD4-CD8+ subset of various other TcR Vcz-gene segments not seen in healthy controls. The most striking difference, however, was the apparent low frequency of use of the TcR Vcz12-gene segment in the patient both in the CD4+CD8-and CD4-CD8+ subsets.
To confirm the random TcR Vcz and V]/use in the CD4+CD8-T-cell subset in the type-III BLS patient, the V-gene segment use in TcR expressed on several CD4+CD8-T-cell clones was analyzed. The clonality of these clones was tested by gene rearrangement analysis of the TcRfl chain using the restriction enzymes Eco RI and Hind III and a C]/probe (data not shown). As in the bulk culture, no preferential or restricted use of a given TcR Vczand Vl/-gene family could be detected in these clones (Table 3). Of note is that in three clones, the Vf119-gene segment could be amplified specifically, which might account for the preferential use observed in the analysis of the bulk culture. As in the CD4+CD8-and CD4-CD8+ bulk cultures, none of the clones analyzed used the TcR Vcd2-gene segment.

DISCUSSION
The type-III Bare Lymphocyte Syndrome provides a human model to study the influence of the lack of MHC class-II antigen expression on the composition of the peripheral T-cell compartment. In this study, we have analyzed the impact of the MHC class-II deficient environment on the TcR V-gene segment usage in the CD4+CD8-and CD4-CD8+ T-cell subsets. Our analysis showed that for most of the V regions tested, no restricted or u.nusual skewing in the frequency of employment of the various TcR Vetand Vfl-gene segments could be detected. Both within the CD4+CD8as well as in the CD4-CD8+ T-cell subset of the patient; the observed frequency of TCR V-gene use for the majority of the TCR Vgene segments tested was comparable to the healthy family controls. However, in the patient, most of the Vc-gene segments were used at higher frequencies in the CD4-CD8+ T-cell subset, in particular, the Vow1, 2, 3, and 7 genes. Of note is that the Vct12-gene segment was used at much lower frequencies both in the CD4+CD8and CD4-CD8+ subsets, in comparison to healthy controls.
Our studies showed that in the frequency of TcR Vo-gene segment use, more variation was noticeable compared to V]/-gene segment use between the CD4+CD8-T cells and CD4-CD8+ T cells within one individual. This could be related because most of the Vo-gene segment families contain only one member, whereas the majority of the VJ-gene segment families have more members. As a result, differences in usage frequency of the different Vfl subfamilies can be masked by the amplification of other members of the same family. The results from this study indicate that in man, the impact of MHC class-II antigen expression on T-cell development is more pronounced on the TcR Vowrather than on the V]/gene segment use. This in spite of the fact that both the TcR Vet and V]/ chain seem to be involved in binding of the peptide/MHC complex (Chotia et al., 1988;Davis and Bjorkman, 1988).

Although our analyses show variety in TcR
Vowand Vfl-gene segment use between the single CD4+CD8-and CD4-CD8+ T-cell population, the skewing of a particular V-gene segment to a subset is less strong as observed in some mouse strains. Several factors could contribute to this difference between man and mouse. First, the composition of the peripheral T-cell population is not only influenced by the individual haplotype, but also by the exposure to environmental antigens and immunological activation at a given moment. It should be noted that the analyzed mice of various strains have been kept in a germfree state. Second, exposure to "self-superantigens" results in a skewing of certain TcR gene segment positive T-cells to either the CD4,CD8+ or CD4+CD8-subset (Benoist and Mathis, 1989;Bill and Palmer, 1989;Blackman et al., 1989;Liao et al., 1989) or to a complete deletion of T cells bearing a TcR with certain gene segments in both the CD4+CD8-and CD4-CD8+ population Kisielow et al., 1988;MacDonald et al., 1988aMacDonald et al., , 1988bPircher et al., 1989). In mice several "self superantigens" are identified as endogenous retroviral sequences (Dyson et al., 1991;Frankel et al., 1991;Marrack et al., 1991;Woodland et al., 1991), but no human analogue has been found so far. Third, in mice, lack of use of a particular V-gene segment also resulted from deletion of part of the Vgene segments on the genomic level as determined by Southern blot analysis of the TcR loci both in inbred and in wild mice (Behlke et al., 1986;Haqqi et al., 1989aHaqqi et al., , 1989bPullen et al., 1990). In man, no major deletion of V-gene segments could be detected in most individuals (Concannon et al., 1987;Baccala et al., 1991).
Fourth, most of the analyzed mice are homozygous, whereas our analyzed individuals are heterozygous. This heterozygocity could mask the influence of the haplotype on the peripheral T-cell compartment. The presence of single CD4+CD8-T cells in the periphery of this MHC class-II deficient patient is in seemingly contradiction with studies in the mouse in which interference in the TcR/MHC class-II interaction leads to an almost complete lack of CD4+CD8-T cells in the periphery (Kruisbeek et al., 1983(Kruisbeek et al., , 1985. Recently, MHC class-II deficient mice have been generated by homologous recombination. Here, the lack of MHC class-II antigen expression in these mice resulted in a strong reduction in the number of single CD4+CD8-T cells in the periphery (Cosgrove et al., 1991;Grusby et al., 1991). Of note is the observation that despite the lack of expression of the to-date known mouse MHC class-II structures that are responsible for the T-CELL SUBCULTURES DERIVED FROM A TYPE-III 235 generation of the CD4+CD8-T-cell subset, the periphery of these mice is not devoid of CD4+CD8-T cells. In these mice, small numbers of CD4+CD8-T cells can be found that exhibit random use of the TCR V-gene segments tested for (Cosgrove et al., 1991;Grusby et al., 1991).
These observations are reminiscent of the results from this study in which we have demonstrated that in a human model with no detectable peripheral MHC class-II expression, CD4+CD8-T cells can be found in the periphery, albeit their numbers are greatly reduced. Furthermore, these CD4+CD8-T cells are.not restricted in their TcR V-gene segment use, indicating that these cells are not derived from an oligoclonal outgrowth of a few CD4+CD8-T cells that have escaped from the thymic selection process. It is at present not clear whether the observed skewing in the use frequency of some of the TcR V-gene segments in the T-cell subsets is a direct consequence of aberrant thymic selection processes or that it is resulting from a lack of MHC class-II mediated postthymic modification of the peripheral T-cell compartment.