IL-4-Producing γδ T Cells That Express a Very Restricted TCR Repertoire Are Preferentially Localized in Liver and Spleen

Mice

C57BL/6 (B6) mice Tg for a rearranged Vγ1Jγ4Cγ4 chain (Vγ1Tg) and TCRδ knockout mice were maintained in our animal facilities. DBA/2 and (B6 × DBA/2)F₁ (B6D2F₁) mice were obtained from Iffa Credo (L’Abresle, France). (DBA/2 × B6)F₁ (D2B6F₁) Vγ1Tg were produced by mating DBA/2 females with B6 Vγ1Tg males. F₁ mice were typed for the presence of the transgene by PCR analysis on tail DNA using primers specific for the Vγ1 gene and for the Vγ1Jγ4 junction (see below). Mice were used between 7–15 wk of age.

Antibodies

Anti-CD4 (RL.174), anti-CD8 (HO-2.2), anti-heat-stable Ag (anti-HSA; J11d), anti-Cβ (H57-597), anti-Cδ (3A10), anti-Vγ1 (2.11), and anti-Vδ6.4/Vδ6.3 (9D3) were prepared and used as previously described (33). PE-labeled anti-Cδ (GL3), FITC- or PE-labeled anti-HSA, anti-Thy-1, anti-CD62 ligand (anti-CD62L), anti-CD44, anti-CD69, anti-CD3ε, anti-CD45RB, anti-NK1.1, anti-CD4, and anti-CD8 were purchased from PharMingen (San Diego, CA). Goat anti-mouse IgM was purchased from Sigma (St. Louis, MO). For the cytokine-specific ELISA we used the following mAb: anti-IFN-γ (clones R46A2, and AN18) and anti-IL-4 (clones BVD4 and BVD6; a gift from Dr. P. Minoprio, Unité d’Immunoparasitologie, Institut Pasteur, Paris, France).

Immunofluorescence staining and FACS

Cells (10⁵-10⁶) were incubated in staining buffer (PBS, 3% FCS, and 0.1%NaN₃) with the indicated labeled mAbs for 30 min on ice and washed twice. When biotin-conjugated mAbs were used, the cells were further incubated with either PE-labeled streptavidin (Southern Biotechnology Associates, Birmingham, AL) or streptavidin Tricolor (Caltag, South San Francisco, CA) for 15 min on ice. After another wash, cells were resuspended in staining buffer containing 1 μg/ml propidium iodide and analyzed using a FACScan flow cytometer (Becton Dickinson, Mountain View, CA). Dead cells were gated out either by their staining with PI or by their forward and angle light scatter profile. Data were analyzed using the CellQuest program.

For FACS sorting, cell suspensions were prepared as indicated above and incubated with the appropriate mAb at a concentration of 30 × 10⁶ cells/ml for 30 min on ice. After two washes, cells were resuspended in PBS-10% FCS and sorted in a FACStar^Plus cell sorter (Becton Dickinson). The purity of the sorted populations was >95%.

Cell purification, cell culture, and cytokine-specific ELISA

Single-cell suspensions were prepared from the thymus, spleen, lymph nodes, and liver according to standard procedures. CD4⁻CD8⁻ (DN) and CD8⁻HSA⁻ thymocytes were prepared by complement-mediated killing as previously described (8). Liver mononuclear cells were isolated by centrifugation in a discontinuous gradient of Percoll (Pharmacia, Uppsala, Sweden). Briefly, total liver cells were resuspended in 100% isotonic Percoll solution and overlaid with 70 and 40% isotonic Percoll solutions. After centrifugation for 30 min at 500 × g, cells at the 40–70% interface were collected, washed twice, and further purified from contaminating RBC by density gradient centrifugation using Lympholyte M (Cedarlane, Hornby, Canada). The cells were washed twice in medium containing 5% FCS. Splenic RBC were lysed by incubating spleen cells for 2 min in 5 ml of NH₄Cl solution. Lymph node cells and RBC-depleted spleen cells were resuspended in medium containing 10% FCS at a concentration of 2.5–10 × 10⁶ cells/ml and incubated for 90 min in petri dishes (Optilux, Falcon 1005, Oxnard, CA) previously coated with anti-Cβ and anti-IgM Ab (each at 5 μg/ml in PBS), with sporadic agitation. Nonadherent cells were recovered and washed twice before use.

FACS-sorted cells (1.5 × 10⁵ cells/ml) were cultured in flat-bottom microtiter plates previously coated with 10 μg/ml anti-Cδ mAb (3A10) in complete medium, i.e., DMEM with Glutamax-I medium (Life Technologies, Gaithersburg, MD) supplemented with sodium pyruvate, 5 × 10⁻⁵ M 2-ME, nonessential amino acids, and antibiotics (all from Life Technologies) and 10% FCS (Boehringer Mannheim, Meylan, Germany). Mouse rIL-2 was added at a final concentration of 100 U/ml. Supernatants from 3-day cultures were tested for the presence of IL-4 and IFN-γ by ELISA as previously described (34).

Production of the 9D3 mAb

TCRδ KO mice were immunized i.p. twice at 2-wk intervals with 5 × 10⁶ DTN40 (Vγ1/Vδ6.4) hybridoma cells and once i.v. with 10⁶ DTN40 hybridoma cells resuspended in PBS. The DTN40 hybridoma was obtained by fusing Thy-1^dull γδ thymocytes of DBA/2 origin with BW5147 thymoma cells (8). Three days after the last injection, spleen cells were fused with SP2/0 cells as described previously (33). The cells were then distributed in 96-well flat-bottom plates in complete medium supplemented with hypoxanthine-aminopterin-thymidine. Culture supernatants from growth-positive wells were tested for their ability to bind to the immunizing hybridoma cell but not to a TCR-negative variant of the same hybridoma. Binding of the Ab to the hybridoma cells was detected with FITC-labeled goat anti-mouse Ig (Caltag) and analyzed with a FACScan. The fine specificities of the selected Ab were determined by their ability to stain γδ T cell hybridomas that express different TCRγ and TCRδ chains.

Oligonucleotide primers and PCR conditions

The following oligonucleotide primers were used: Vδ6, TCTGTAGTCTTCCAGAAATCA; Vδ6.4, GTTTTCCTTATTCGACAAACA; Cδ, CGAATTCCACAATCTTCTTG; JS1, GTTCCTTGTCCAAAGACGAG; Vγ1, CCGGCAAAAAGCAAAAAAGTT; and Vγ1Jγ4Tg junction, CCCATGATGTGCCTGACCAG. PCR was performed using a GeneAmp PCR system 9600 (Perkin-Elmer/Cetus, Norwalk, CT). Each cycle consisted of incubation at 92°C for 20 s, followed by 55°C for 30 s and 72°C for 30 s. Before the first cycle, a 2-min 94°C denaturation step was included, and after the 35th cycle the extension at 72°C was prolonged for 4 min.

Nucleic acid preparation and population analysis of TCRδ rearrangements

Total cellular RNA from sorted cells was extracted with RNA-B (Bioprobe Systems, Montreuil, France). Before RNA extraction, sorted cells were mixed with 10⁶ SP2/0 cells as a carrier. cDNA was synthesized with oligo(dT) (Pharmacia) using superscript reverse transcriptase (Life Technologies) according to the manufacturer’s instructions. The population analysis of TCRδ rearrangements has been previously described in detail (8, 20).

Vγ1Jγ4Cγ4 Tg animals of appropriate genetic background contain a population of Thy-1^dull. γδ thymocytes

Given the paucity of γδ cells in the peripheral lymphoid organs in the mouse, it was difficult to study the peripheral representation of the Thy-1^dull γδ population in normal mice. To overcome this problem we took advantage of the availability of a mouse Tg for a rearranged Vγ1Jγ4Cγ4 chain that we produced previously (35). The DNA inserted to produce this Tg line consisted of a 45-kb cosmid containing the rearranged Vγ1Jγ4Cγ4 gene flanked by 10 kb of sequence upstream of the Vγ1 segment and 26 kb of sequence downstream of the last Cγ4 exon (Fig. 1⇓A). Thus, this construct should contain all the regulatory elements required to ensure normal expression of the transgene. The cosmid clone was isolated from a T cell hybridoma of B6 origin (T3.13.1) (33), and the Vγ1Jγ4 junctional sequence is identical with one of the two more common Vγ1Jγ4 junctional sequences found among the Thy-1^dull γδ thymocytes in DBA/2 mice (8). αβ T cell development in these mice appears quite normal, as indicated by the similar number of total thymocytes (not shown) and of the major thymocyte populations defined by double staining with CD4 and CD8 mAbs in Tg animals and littermate controls (Fig. 1⇓B). In contrast, a 3- to 10-fold increase in the number of γδ thymocytes is present in Tg mice compared with littermates, and most of these cells express the transgenic chain (Fig. 1⇓C). This increase in the absolute numbers of γδ T cells in the Tg mice is also evident in the periphery, where Tg animals contain 3- to 10-fold more γδ T cells than control littermates; here again, most of the γδ T cells in Tg mice express the Vγ1 chain (Fig. 1⇓C).

• Download figure
• Open in new tab
• Download powerpoint

FIGURE 1.

A, Restriction map of the γ2F construct used to generate Vγ1Tg mice. Variable and constant regions are indicated beneath the map. The sizes of the variable and constant regions are approximate. Not all the restriction sites are necessarily shown. A, ApaI; F, FspI; K, KpnI; N, NotI; R, EcoRI; S, SnaBI. B, Expression of CD4 and CD8 in the thymus of Vγ1Tg mice and non-Tg littermates. Thymocytes from the indicated mice were stained with FITC-labeled anti-CD4 and PE-labeled anti-CD8 mAb and analyzed with a FACScan. Numbers indicate the percentage of positive cells in each quadrant. C, Expression of the γδ TCR and the Vγ1 chain in different organs of Vγ1Tg mice and non-Tg littermates. Cells isolated from the indicated organs were stained with FITC-labeled anti-CD3 and biotin-labeled anti-δ mAb (left panels) or with FITC-labeled anti-δ and biotin-labeled anti-Vγ1 Ab (right panels) followed by streptavidin-PE. Dot plots show the log₁₀ of fluorescence intensity and were produced with the CellQuest program. Numbers indicate the percentages of CD3⁺γδ⁺ and CD3⁺γδ⁻ (left panels) and of γδ⁺Vγ1⁺ and γδ⁺Vγ1⁻ among total cells present in the lymphocyte gate. Some 20,000 to 100,000 cells were analyzed in each panel.

Before performing phenotypic, functional, and TCR repertoire analysis in the periphery of the Tg mice, we first established that the introduction of the transgene did not modify the representation of the Thy-1^dull γδ thymocytes. As shown in Fig. 2⇓, around 50 and 8% of the γδ thymocytes expressed low levels of Thy-1 in B6D2F₁ and B6 Tg animals, respectively. These numbers correlate well with those previously found in normal B6D2F₁ and B6 mice (8). The somewhat higher percentages of Thy-1^dull γδ thymocytes found in Tg mice are probably due to the fact that virtually all γδ T cells in these mice express the Vγ1 gene product. Furthermore, most of the Thy-1^dull γδ thymocytes isolated from B6D2F₁ Tg mice express a HSA⁻ CD62L⁻ CD44⁺ phenotype (Fig. 2⇓B), characteristic of the Thy-1^dull γδ thymocytes in normal animals (8).

• Download figure
• Open in new tab
• Download powerpoint

FIGURE 2.

Vγ1Tg mice of D2B6F₁ genetic background contain a major population of Thy-1^dull thymocytes. DN thymocytes from the indicated Vγ1Tg mice were stained with PE-labeled anti-Thy-1, biotin-labeled anti-δ, and FITC-labeled anti-HSA, anti-CD62L, or anti CD44 mAb, followed by streptavidin-Tricolor and analyzed in a FACScan. A, Dot plots of γδ vs Thy-1 staining in B6 and D2B6F₁ Vγ1Tg mice. Numbers indicate the percentage of Thy-1^dull γδ thymocytes among total γδ⁺ cells (mean ± 1 SD of 10 animals analyzed individually). B, Profiles of expression of the indicated marker on γδ⁺Thy-1^bright (upper panels) and γδ⁺Thy-1^dull (lower panels) thymocytes from D2B6F₁ Vγ1Tg mice.

When we analyzed the periphery of B6D2F₁ Tg mice for the presence of Thy-1^dull γδ T cells, we found that only around 10% of the γδ T cells present in the spleen, lymph nodes, liver, and bone marrow expressed low levels of Thy-1 (not shown). These data could imply that cells expressing similar phenotypic and functional properties as the Thy-1^dull γδ thymocytes rarely seed the periphery. Alternatively, this could also indicate that Thy-1^dull γδ thymocytes modulate their levels of Thy-1 in the periphery. To distinguish between these two possibilities we sought an independent marker that could specifically identify the Thy-1^dull γδ thymocyte population and putative peripheral descendents. Because the most characteristic feature of this population is the expression of a very restricted TCR repertoire, with almost exclusive utilization of the Vγ1 chain together with certain members of the Vδ6 subfamily (8), we decided to produce an Ab able to recognize these Vδ6 subfamily members.

The 9D3 mAb recognizes Vδ6.3 and Vδ6.4 chains

To produce such an mAb we immunized γδ-deficient mice with a Thy-1^dull γδ T cell hybridoma of DBA/2 origin. Spleen cells from immunized mice were fused with SP2/0 cells, and the resulting hybridomas were tested for their ability to bind to the immunizing cells but not to a TCR-negative variant of the same hybridoma. The fine specificities of the selected mAbs were analyzed by staining a large panel of T cell hybridomas of B6 and DBA/2 origin previously characterized for their Vγ and Vδ usage (P. Pereira, unpublished observations). One hybridoma, termed 9D3, specifically stained some γδ T cell hybridomas expressing a Vδ6 chain but did not stain γδ T cell hybridomas expressing other Vδ chains, including Vδ2, Vδ4, Vδ5, and Vδ7 (not shown). Further analyses with γδ T cell hybridomas expressing different members of the Vδ6 subfamily showed that 9D3 stained all γδ T cell hybridomas of B6 origin expressing the Vδ6.3 chain (36) and all γδ T cell hybridomas of DBA/2 origin expressing the Vδ6.4 (8) chain but not hybridomas expressing any other member of the Vδ6 subfamily, including the related B6 Vδ6.1 (37) chain and the DBA/2 Vδ6.5 and Vδ6.6 chains (8) (Table I⇓). Furthermore, 9D3 stained a sizable proportion of γδ thymocytes, splenocytes, and intestinal intraepithelial lymphocytes in DBA/2 and B6 strains as well as in B10 and B10.D2 mice that share the same TCRα/δ haplotype as B6 mice (39, 40, 41), but did not stain any γδ T cells in BALB/c, C3H, CBA, or 129/Sv strains of mice, which share a different TCRα/δ haplotype (39, 40, 41), suggesting that the 9D3 mAb does not recognize the Vδ6.2 chain (37) (not shown).

Three-color staining of DN thymocytes from D2B6F₁ Tg mice with anti-γδ, anti-Thy-1, and 9D3 mAb showed that 80–90% of Thy-1^dull γδ thymocytes were 9D3⁺, consistent with the predicted specificity of the 9D3 mAb. Conversely, around 90% of the γδ⁺9D3⁺ thymocytes expressed low levels of Thy-1 (Fig. 3⇓). These data indicated that it would be possible to enrich for Thy-1^dull γδ T cells by separating the cells on the basis of 9D3 expression.

• Download figure
• Open in new tab
• Download powerpoint

FIGURE 3.

Correlation between 9D3 expression and low levels of Thy-1. DN thymocytes from D2B6F₁ Vγ1Tg mice were stained with FITC-labeled anti-δ, PE-labeled anti-Thy-1, and biotin-labeled 9D3 mAb followed by streptavidin-Tricolor and analyzed in a FACScan. A dot plot of γδ vs 9D3 staining in total DN cells (left) and Thy-1 expression by γδ⁺9D3⁺ (R2) and γδ⁺9D3⁻ (R3) populations is shown. R2 and R3 denote the electronic gates used for the analysis.

Frequency of γδ⁺9D3⁺ cells in different organs

We next analyzed the frequency of 9D3⁺ cells among γδ lymphocytes in different organs in D2B6F₁ Tg mice. As shown in Table II⇓, 9D3⁺ cells represent roughly 50, 30, and 20% of the γδ T cells in liver, spleen, and peripheral lymph nodes, respectively. In other sites, such as bone marrow and peritoneal cavity, the frequency of γδ T cells was too low to allow accurate quantification even in Tg mice, although γδ⁺9D3⁺ cells were present at those sites.

γδ⁺9D3⁺ cells in different organs express a restricted TCRδ repertoire

One of the characteristics of the Thy-1^dull γδ thymocytes is their very restricted TCR repertoire. Besides their almost exclusive use of the Vγ1 chain and one or two members of the Vδ6 chain subfamily, their VδDδJδ junctional sequences show limited diversity, and the great majority of them show identical length (8, 20). To investigate the putative relationship between the Thy-1^dull γδ thymocytes and the γδ⁺9D3⁺ cells in the periphery we examined the junctional length of the Vδ6 transcripts present in sorted γδ⁺9D3⁺ and γδ⁺9D3⁻ cells isolated from thymus, spleen, liver, and peripheral lymph nodes as well as from a Thy-1^dull γδ hybridoma (DTN40). RT-PCR reactions with Vδ6- and Cδ-specific primers on cDNA isolated from all sorted populations were performed. An aliquot of each amplification product was submitted to a run-off reaction in the presence of a 6-carboxy-fluorescein (FAM)-labeled Jδ1 primer, and the fluorescent products were resolved on a denaturing acrylamide gel cast on an automated sequencer.

Typical of polyclonal VδDδJδ junctions, the profiles obtained from sorted γδ⁺9D3⁻ populations isolated from different organs showed 11–15 defined peaks that form a Gaussian-type curve (Fig. 4⇓). All adjacent peaks differ in size by three nucleotides, and thus, the profiles show mainly in-frame junctions. In contrast, the profiles obtained for the sorted γδ⁺9D3⁺ populations showed a prominent peak at a CDR3 length identical with that displayed by the DTN40 hybridoma cell line, although the relative representation of this peak varied depending on the organ from which the cells were isolated. Calculation of the area of this peak relative to the area of all peaks in each organ indicated that 70, 35, 50, and 25% of the Vδ6(D)Jδ1 rearrangements present in γδ⁺9D3⁺ populations isolated from thymus, spleen, liver, and lymph nodes, respectively, showed this particular CDR3 size.

• Download figure
• Open in new tab
• Download powerpoint

FIGURE 4.

Population analysis of Vδ6(D)Jδ1 rearrangements expressed by γδ⁺9D3⁺ and γδ⁺9D3⁻ populations isolated from different anatomical sites. Total RNA samples isolated from sorted γδ⁺9D3⁺ and γδ⁺9D3⁻ populations isolated from the indicated anatomical sites and from the DTN40 hybridoma cells were converted to cDNA and amplified with Vδ6- and Cδ-specific primers. An aliquot of the amplified products was submitted to a run-off reaction with a FAM-labeled Jδ1 primer, and the final products were resolved with an automated sequencer. Plots show the profile of fluorescence intensity vs the length of the fragments. LN, lymph node cells

γδ⁺9D3⁺ cells in the periphery produce IL-4 and IFN-γ upon activation with anti-γδ mAbs and IL-2

We have previously shown that Thy-1^dull γδ thymocytes secrete high levels of IL-4 and IFN-γ upon activation in vitro with coated anti-γδ mAbs and IL-2 (8). To investigate whether γδ⁺9D3⁺ cells in the periphery have the same functional abilities as the Thy-1^dull γδ thymocyte population, sorted γδ⁺9D3⁺ and γδ⁺9D3⁻ cells from thymus, spleen, lymph nodes, and liver of D2B6F₁ Tg mice were cultured in anti-γδ mAb-coated plates in the presence of IL-2, and the levels of IL-4 and IFN-γ present in the culture supernatants were quantitated (Table III⇓). Regardless of the organ from which they were isolated, γδ T cells secreted substantial amounts of IFN-γ, although 2- to 4-fold higher levels were detected in γδ⁺9D3⁻ cells than in γδ⁺9D3⁺ cells. In contrast, larger differences were observed in the levels of IL-4 produced by these two γδ T cell subsets in the different organs. Thus, in thymus, spleen, and lymph nodes, γδ⁺9D3⁺ cells secreted 10- to 20-fold higher levels of IL-4 than γδ⁺9D3⁻ cells, while the difference was only 2-fold in the liver. It is likely that in the thymus, the low levels of IL-4 detected in the γδ⁺9D3⁻ cultures arise from the roughly 10% of Thy-1^dull γδ thymocytes in D2B6F₁ animals, which express the Vδ6.6 gene product and are not recognized by the 9D3 mAb. Similarly, the low levels of IL-4 detected in the γδ⁺9D3⁻ cultures from spleen and lymph nodes could arise from Vδ6.6-bearing cells, although we do not have direct evidence for this interpretation. Taken together, these experiments suggest that the production of IL-4 by peripheral γδ T cells is primarily accomplished by cells expressing the Vδ6 chain and a restricted TCR repertoire.

Phenotypic analysis of γδ⁺9D3⁺ cells at different anatomical sites^.

We then investigated the surface phenotype of the γδ⁺9D3⁺ and γδ⁺9D3⁻ cells present at different anatomical sites, and their FACS profiles in thymus, spleen, and liver are illustrated in Fig. 5⇓. Invariably, most γδ⁺9D3⁺ cells showed an HSA^low CD44^bright CD62L^low CD45RB^int CD69⁺ phenotype, similar to that of NK T cells and activated T cells. However, none of these markers could unambiguously define the γδ⁺9D3⁺ population. In fact, the same activated phenotype is expressed by the vast majority of γδ T cells in the liver, and CD44^bright cells represent an important fraction of the γδ⁺9D3⁻ cells in thymus and spleen.

• Download figure
• Open in new tab
• Download powerpoint

FIGURE 5.

Phenotypic analysis of γδ⁺9D3⁺ and γδ⁺9D3⁻ populations. Cells were prepared as described in Materials and Methods and three color stained as described in Fig. 2⇑. Shown are the profiles of the indicated Ab in γδ⁺9D3⁺ and γδ⁺9D3⁻ populations isolated from the indicated organs. Staining results of lymph node cells were similar to those presented for spleen cells.

γδ⁺9D3⁺ cells are heterogeneous with regard to the expression of other cell surface markers (Fig. 6⇓). Thus, approximately half these cells express the NK1.1 marker in every organ tested. However, a similar proportion of the γδ⁺9D3⁻ cells present in the liver are also NK1.1⁺. We have previously shown that, unlike other γδ thymocytes, around half the Thy-1^dull γδ thymocytes express the CD4 coreceptor (8). Similarly, a fraction of the γδ⁺9D3⁺ cells present in the spleen and liver expresses the CD4 molecule, although the fraction of CD4⁺γδ⁺ cells appears to be lower in the liver than in the spleen. Finally, most γδ⁺9D3⁺ cells are negative for the expression of CD8, even though CD8⁺γδ⁺ cells can be readily detected among the γδ⁺9D3⁻ cells in liver and spleen, confirming previous results (42).

• Download figure
• Open in new tab
• Download powerpoint

FIGURE 6.

CD4, CD8, and NK1.1 expression by γδ⁺9D3⁺ and γδ⁺9D3⁻ populations. A, D2B6F₁ Tg thymocytes treated with anti-HSA and anti-CD8 mAbs and complement were stained with FITC-labeled anti-CD4, PE-labeled anti-δ, and biotin-labeled 9D3 followed by streptavidin-Tricolor. The histogram shows CD4 expression in γδ⁺9D3⁺ cells. B, DN thymocytes from D2B6F₁ Tg mice were stained with FITC-labeled anti-NK1.1, PE-labeled anti-δ, and biotin-labeled 9D3 mAb followed by streptavidin-Tricolor. Histograms show NK1.1 expression in γδ⁺9D3⁺ and γδ⁺9D3⁻ populations. C, Cells were prepared as described in Materials and Methods and three color stained as described above. Shown are the profiles of the indicated Ab in γδ⁺9D3⁺ and γδ⁺9D3⁻ populations isolated from the indicated organs. All cells were analyzed with a FACScan. Staining results of lymph node cells were similar to those shown for spleen cells.

γδ⁺9D3⁺ cells expressing a restricted TCRδ repertoire exist in the periphery of normal mice

Two types of experiments were performed to investigate the presence of this type of γδ T cells in the periphery of normal mice and to estimate their numbers. First we performed RT-PCR reactions on total RNA isolated from different organs of DBA/2 mice with primers specific for the Vδ6.4 and Cδ gene segments. Because most αβ T cells have deleted the TCRδ locus in both chromosomes, this PCR is expected to mainly, if not exclusively, amplify transcripts expressed by γδ T cells. The PCR amplifications were submitted to run-off reactions with a FAM-conjugated Jδ1 primer, and the labeled fragments were separated on a sequencing gel. The profiles obtained from one such experiment are shown in Fig. 7⇓. As can be seen, a major peak protruding from the Gaussian-type curve and corresponding to a CDR3 length identical with that of the DTN40 hybridoma cell line is evident in thymus, spleen, and liver. The same peak is also evident, although to a lesser extent, in the lymph nodes and in the peritoneal cavity. These experiments suggested that cells expressing a restricted TCRδ repertoire are present in these organs in normal mice.

• Download figure
• Open in new tab
• Download powerpoint

FIGURE 7.

Cells with a restricted TCRδ repertoire exist in the periphery of normal mice. RNA isolated from the indicated organs of DBA/2 mice was amplified by RT-PCR with Vδ6.4- and Cδ-specific primers. An aliquot of the amplified products was submitted to a run-off reaction with a FAM-labeled Jδ1 primer, and the final products were resolved with an automated sequencer. Plots show the profile of fluorescence intensity vs the length of the fragments. LN, lymph node cells; PC, peritoneal cavity cells.

The second set of experiments concerned the quantification of γδ⁺9D3⁺ cells by FACS analysis in different anatomical sites, and the results are presented in Table IV⇓. The proportion of 9D3⁺ cells among γδ T cells in different organs in normal mice compared well with those found in Tg mice, keeping in mind that all γδ T cells in the Tg mice express the Vγ1 gene product. From this and the previous set of experiments we estimate that in normal DBA2 mice, roughly 2–4 × 10⁵ thymocytes, and between 0.5 and 2 × 10⁵ cells in the spleen and liver share the characteristics previously described for the Thy-1^dull γδ thymocytes.

In conclusion, our studies have shown that γδ T cells able to simultaneously produce IL-4 and IFN-γ upon activation in vitro exist in the periphery of mice Tg for a rearranged Vγ1 chain. These cells, which are found predominantly in spleen and liver, express a restricted repertoire of TCR composed of the Vγ1 gene product mainly associated with the Vδ6.4 chain and exhibit limited junctional sequence diversity. γδ T cells with a similarly restricted TCR repertoire are present in the same organs in normal DBA/2 mice. These cells probably represent the descendants of the previously described Thy-1^dull γδ thymocytes (8), although they seem to up-regulate Thy-1 expression upon seeding the periphery. Our experiments, however, did not directly address this issue, and such a relation between Thy-1^dull γδ thymocytes and the cells studied here is, therefore, not formally established. Although this γδ T cell subset expresses an activated/memory phenotype, none of a variety of markers tested could unambiguously define this T cell population.

The phenotype, TCR diversity level, and functional properties found in this γδ T cell subset are also characteristic of the T NK population, and both T cell populations seem to home preferentially to liver and spleen. These common properties may reflect a similar differentiation program and/or a related function. However, the likely difference in their specificities implies separate aspects of function. Additional experiments will aim at characterizing the specific ligands recognized by this γδ T cell population.