HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS		QUICK SEARCH:   [advanced] Author: Keyword(s): Year:  Vol:  Page:

This Article Abstract Full Text (PDF) Alert me when this article is cited Alert me if a correction is posted Citation Map Services Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Citing Articles Citing Articles via Google Scholar Google Scholar Articles by Sinkora, M. Articles by Holtmeier, W. Search for Related Content PubMed PubMed Citation Articles by Sinkora, M. Articles by Holtmeier, W. Pubmed/NCBI databases Gene

Two Groups of Porcine TCR⁺ Thymocytes Behave and Diverge Differently¹

Marek inkora^2,*, Jana inkorová, Zdenk Cimburek and Wolfgang Holtmeier

^* Department of Immunology and Gnotobiology, Institute of Microbiology, Academy of Sciences of the Czech Republic, Nov Hrádek, Czech Republic; Art and Research Technologies, Nov Hrádek, Czech Republic; Department of Immunology and Gnotobiology, Institute of Microbiology, Academy of Sciences of the Czech Republic, Prague, Czech Republic; and Johann-Wolfgang Goethe University, Medical Clinic I, Department of Gastroenterology, Frankfurt, Germany

	Abstract

Top Abstract Introduction Materials and Methods Results Discussion Disclosures References

 
Developmental pathways of T cells are still unknown, largely because of the absence of recognized lineage-specific surface markers other than the TCR. We have shown that porcine thymocytes can be divided into 12 subsets of the following two major groups: 1) CD4^– thymocytes that can be further subdivided according to their CD2/CD8 phenotype, and 2) CD4⁺ thymocytes that are always CD1⁺CD2⁺CD8⁺ and have no counterpart in the periphery. In this study, we have analyzed thymocyte subsets with respect to behavior during cultivation, cell cycle status, and lymphocyte-specific transcripts. The group of CD4^– thymocytes gives rise to all T cells found in the periphery. Proliferating CD2⁺CD8^–CD1⁺CD45RC^– thymocytes are a common precursor of this group. These precursors differentiate into CD2⁺CD8⁺, CD2⁺CD8^–, and CD2^–CD8^– T cell subsets, which subsequently mature by loss of CD1 and by eventual gain of CD45RC expression. In contrast, the group of CD4⁺ thymocytes represents transient and independent subsets that are never exported from thymus as TCR⁺ T cells. In accordance with the following findings, we propose that CD4⁺CD8⁺ thymocytes extinguish their TCR expression and differentiate along the T cell lineage program: 1) CD4⁺ thymocytes are actively dividing; 2) CD4⁺ thymocytes do not die, although their numbers decreased with prolonged cultivation; 3) CD4⁺ thymocytes express transcripts for RAG-1, TdT, and TCR; and 4) CD4⁺ thymocytes are able to alter their phenotype to TCR⁺ thymocytes under appropriate culture conditions.

	Introduction

Top Abstract Introduction Materials and Methods Results Discussion Disclosures References

 
Early T cell progenitors, i.e., pro-T cells, are derived from stem cells in primary hemopoietic centers such as yolk sac, fetal liver, or bone marrow, and migrate to the thymus, where further differentiation takes place (1, 2). We and others showed that the embryonic thymus is colonized discontinuously by several successive waves of hemopoietic progenitors during embryogenesis, which correspond with the shift of hemopoiesis from one primary center to another (1, 2, 3). The early stages of T lymphoid development are well characterized in mice and partially in humans, but are relatively poorly understood in other species because of the following: 1) comparison with murine development has been hampered by different immunophenotypic selection criteria, and 2) the expression of typical differentiation markers such as CD25 does not appear to be required for successful generation of normal T cells in other species, including pigs (3, 4, 5). However, an increasing amount of data collected from different species indicates that the critical differentiation stages and developmental pathways of T cell development are consistent with generally accepted models of intrathymic T cell differentiation (5). Compared with the lineage, knowledge about early stages of T lymphoid development is less complete, because of the following: 1) the absence of recognized lineage-specific surface markers other than TCR, and 2) T cells mature much faster than T cells (3, 5). Ontogenetic studies in pigs have shown that thymocytes require 15 days to fully differentiate, whereas thymocytes do so in <3 days without any CD3^low or TCR^low transitional stage (3). T cells therefore appear in the thymus and populate the periphery before T cells (3, 6). The early appearance of TCR indicates that maturation and diversification of lymphocytes occur after a full expression of TCR, whereas the opposite is true for lineage (5, 6). Recently, we have described 12 subsets of porcine thymocytes based on their expression of CD1, CD2, CD4, CD8 isoforms, and CD45RC, and have shown that thymocytes can be divided into the two following major groups: 1) numerous CD4^– thymocytes that can be further subdivided according to their CD2/CD8 phenotype, and 2) scarce CD4⁺ thymocytes that are always CD1⁺, CD2⁺, and CD8⁺ (6). According to ontogenetic data, the maturation of thymocytes in the CD4^– group appears to begin with proliferation of a CD2⁺CD8^–CD1⁺CD45RC^– common precursor. This stage is followed by diversification into CD2⁺CD8⁺, CD2⁺CD8^–, and CD2^–CD8^– subsets that further mature by losing expression of CD1 and by increasing expression of CD45RC. Therefore, individual stages of development include CD1⁺CD45RC^–, CD1^–CD45RC^–, and CD1^–CD45RC⁺ cells. In contrast, thymocytes of CD4⁺ group exclusively bear CD8⁺, never mature into CD1^– cells, and have no counterpart in the periphery. Our observations also indicate that all peripheral CD8⁺ T cells express the CD8 homodimer and that two developmentally distinct subsets of these peripheral T cells that differ in MHC-II expression appear to exist as follows: one that is MHC-II^– directly develops in the thymus, whereas the MHC-II⁺ subset acquires CD8 in the periphery (6).

In this study, we show that the maturation of thymocytes occurs after successful expression of TCR. Proliferating CD2⁺CD8^–CD1⁺CD45RC^– thymocytes are precursors of CD4^– thymocytes that give rise to all TCR⁺ subsets found in the periphery. These precursors are particularly sensitive to apoptosis during culture. Each of the three CD4^– thymocyte subsets defined by CD2 and CD8 expression is shown to develop in the thymus through separate differentiation pathways originating from CD1⁺CD45RC^– to CD1^–CD45RC^– and progressing to CD1^–CD45RC⁺ cells. Our data also demonstrate that CD4⁺ thymocytes represent a transient and independent population that extinguishes their TCR expression and differentiates along the T cell lineage program. These data suggest that productive rearrangement and expression of TCR genes do not necessarily determine lineage commitment.

	Materials and Methods

Top Abstract Introduction Materials and Methods Results Discussion Disclosures References

 
Animals

Animals used in the study were as follows: 1) crossbred gilts bred in Nov Hrádek (3, 6, 7), and 2) large white/Landrace crossbred gilts. Fetuses were obtained by hysterectomy. Gestation age (day of gestation (DG))³ was calculated from the day of mating (gestation in swine is 114 days). All animal experiments were approved by the Ethical Committee of the Institute of Microbiology, Czech Academy of Science, according to guidelines in the Animal Protection Act.

Immunoreagents

The following mouse anti-pig mAbs, whose source and specificity were described earlier (3, 6, 7), were used as primary immunoreagents: anti-CD1 (76-7-4, IgG2a), anti-CD2 (MSA4, IgG2a or 1038H-5-37, IgM), anti-CD3 (PPT3, IgG1 or PPT6, IgG2b), anti-TCR (PPT17, IgG1 or PPT16, IgG2b), anti-CD4 (10.2H2, IgG2b), anti-CD8 (76-2-11, IgG2a), anti-CD8 (PPT23, IgG1), and anti-CD45RC (MIL5, IgG1). In some cases, mAbs were also labeled with biotin N-hydroxysuccinimide ester (3, 6).

Goat polyclonal Abs specific for mouse Ig subclasses (Southern Biotechnology Associates or Molecular Probes) labeled with FITC, PE, cyanine 5 (Cy5), or Alexa Fluor 350 were used as secondary immunoreagents. Biotinylated primary Abs were detected by a streptavidin-PE/Cy5 tandem conjugate (SpectralRed; Southern Biotechnology Associates).

Preparation of cell suspensions, cell cultures, staining, and flow cytometry (FCM)

Cell suspensions from thymus were prepared and cultured essentially as previously described (3, 6, 7, 8). Thymic stromal cultures were established using freshly isolated thymic suspensions. After 6 days of incubation, the nonadherent cells were removed and the monolayer was, after extensive washing, then used as feeder cells. Partial thymectomy was used for preparation of syngeneic thymic stromal cultures, followed by using thymic suspensions and thymocytes from the same donor, whereas allogeneic cultures were prepared using thymic stromal cells from different fetal donors. Allogeneic cultures were used in most of the studies because origin of thymic stromal cells did not affect the patterns of thymocyte behavior in cultures in examined parameters. Input stromal cultures were always tested for the absence of donor thymocytes to preclude cross-contamination. Surface staining of cells for FCM analysis and sorting was also performed, as described previously (3, 6, 7, 8), by indirect subisotype staining. The DNA content of multicolor-stained cells for cell cycle studies was determined using the DNA intercalating probe 7-aminoactinomycin D (7-AAD) (3, 6). Apoptotic cells were also detected using annexin-V and propidium iodide, according to the manufacturer protocol (Roche Diagnostics). In some experiments, propidium iodide was omitted because its spectral overlaps with other dyes. Samples were measured and/or sorted on standard FACSCalibur, FACS-LSR-II, or FACSVantage flow cytometers and analyzed by PC-LYSYS software (BD Immunocytometry Systems). Damaged and dead cells were excluded from analysis using propidium iodide fluorescence. A doublet discrimination module was used in DNA content analysis allowing single cell events measurement.

DNA fragmentation assay

Gel electrophoresis DNA fragmentation assay was used for detection of apoptosis and/or necrosis. The pellets containing the same number of examined cells were resuspended in 400 µl of lysis buffer (0.2% Triton X-100, 10 mM Tris-HCl (pH 8.0), and 1 mM EDTA). Cell lysates were immediately centrifuged, and the supernatants containing small DNA fragments were separated immediately from the pellets. DNA fragments from supernatants were precipitated by isopropanol, collected by centrifugation, washed with 70% ethanol, air dried, resuspended in 20 µl of Tris-EDTA buffer (10 mM Tris-HCl (pH 8.0) and 1 mM EDTA), and mixed with loading buffer. Electrophoresis was performed on 1% agarose gels.

PCR and RT-PCR

Preparation of RNA and cDNA was performed, as described previously (8, 9). First-strand cDNA synthesis was performed using 2 µg of total RNA or total RNA prepared from the same amount of sorted cells and primed with 200 ng of random hexamer primer. Genes or gene segments of interest were amplified using previously described primer pairs (8, 10, 11). The only exceptions were primers for pT (sense, ctgcagctgggtcctgcctc; antisense, agtctccgtggccgggtgca; designed to amplify all three spliced variants, pT1, pT2, and pT3), TCRV (sense, tggtayvkmcagtaycc; antisense, tctcagctggtacacagc; designed to amplify the most frequently encountered subgroups, V01, V02, V03, and V14 (12)), and TCRC (sense, gctgtgtaccagctgaga; antisense, agtttaggttcatatctg). As controls for determining relative transcript expression and efficiency of amplification, portion of -actin was amplified from cDNA (8). All PCR were performed using 200 ng of template DNA or 10% of cDNA preparations with components recommended by TaqDNA polymerase manufacturer (Invitrogen Life Technologies) in a T-Gradient thermal cycler (Biometra). Annealing temperatures were set to 55°C for amplification of TdT and RAG-1, 61°C for C_H, or 58°C for all other amplification. PCR products were analyzed in 1.5% agarose gels. In some cases, products were excised from the gel and purified using the Wizard Plus Miniprep DNA Purification System (WIZARD kit; Promega). Purified PCR products were used for cloning (see below) or for high-resolution CDR3-V length analysis, as described previously (10), using a similar principle as for CDR3 length analysis of V_H genes (8).

Cloning, hybridization, and sequencing

Cloning was performed, as described previously (8, 9), using reamplified gene segments by Pfu-polymerase (Stratagene) and cloning into EcoRV-digested pBluescript-II SK phagemids by blunt-end ligation. The ligation mixture was used to transform DH5-competent cells; the positive recombinant clones identified by blue/white selection were transferred to nylon membranes; and the membranes were hybridized with ³²P end-labeled gene-specific oligonucleotide probes. Positive clones were identified by autoradiography, and single-gene inserts were confirmed by XhoI/XbaI restriction digest after isolation of plasmid DNA. Positive recombinant clones of interest were sequenced (9).

	Results

Top Abstract Introduction Materials and Methods Results Discussion Disclosures References

 
Twelve subsets of thymocytes may be distinguished in thymus

Our earlier work (6), based on up to four-color FCM, demonstrated that thymocytes could be divided into 12 subsets according to their expression of CD1, CD2, CD4, CD8 isoforms, and CD45RC (Fig. 1). In this study, we have directly confirmed these findings using six-color staining through use of FITC, PE, Cy5, Alexa Fluor 350, PE/Cy5, and propidium iodide (Fig. 2). This analysis is based on the observation that CD8 expression is restricted to all CD2⁺CD4⁺ thymocytes (6), so the expression pattern for CD8⁺ is the same as for CD4⁺ thymocytes. Fig. 2 demonstrates that subsets 1–4 can be identified on dot plot B-IV, subsets 5 and 6 on dot plot A-V, subsets 7–9 on dot plot A-III, and subsets 10–12 on dot plot B-III. These findings were confirmed in all experiments regardless of age and breed of animals.

View larger version (27K): [in this window] [in a new window]   FIGURE 1. Separation of porcine thymocytes into 12 principal subsets (subsets 1–12) according to their surface phenotype (6 ).

View larger version (28K): [in this window] [in a new window]   FIGURE 2. FCM analysis of TCR, CD2, CD8, CD1, and CD45RC (row A), and TCR, CD8, CD8, CD4, CD1, and CD45RC (row B) expression by porcine thymocytes isolated from fetuses at DG110. Only live (propidium iodide-negative) TCR+ thymocytes were gated and analyzed for expression of CD2/CD8 (A-I) or CD8/CD8 (B-I). In additional analysis, the cells in dot plot A-I and B-I were further gated by three regions (R1–R3), and the thymocytes in these regions were analyzed for expression of CD1/CD45RC (column III for R1, column IV for R2, and column V for R3). CD1/CD45RC expression for all live thymocytes is also shown (column II). The analysis of CD4 expression is not shown because all CD8+ thymocytes are also CD4+, so the expression pattern for CD8+ (B-IV) is the same as for CD4+ thymocytes. These results were confirmed in all experiments regardless of age and breed of animals.

Only CD1⁺CD45RC^– thymocytes have capacity to divide

Cell cycle analysis of freshly isolated thymocytes based on simultaneous detection of surface phenotype and DNA staining using 7-AAD (Fig. 3) showed that dividing thymocytes all belonged to the CD1⁺CD45RC^– subset (Fig. 3A-II). There were no cycling thymocytes that expressed CD45RC (Fig. 3, A-III and A-V) and/or were negative for CD1 (Fig. 3, A-IV and A-V). Analysis for CD2 and CD8 demonstrated that CD2^–CD8^– thymocytes were not cycling (Fig. 3B-IV), whereas a substantial part of those that were CD2⁺CD8^– were actively dividing (Fig. 3B-II). Because all CD8⁺ T cells are also CD2⁺ (Fig. 3B-I, region R2), CD2⁺CD8⁺ thymocytes did not divide (Fig. 3C-II), similarly to CD2^–CD8^– thymocytes. Interestingly, almost all CD2⁺CD8⁺ thymocytes were actively proliferating (Fig. 3C-III). Combining these findings with previous results on phenotype of thymocyte subsets (Fig. 1), it appears that only three subsets of thymocytes were proliferating, as follows: subsets 1, 3, and 7 (Fig. 1). Because subsets 1 and 3 belong to the group of CD4⁺ thymocytes, subset 7 is the only one belonging to the group of CD4^– thymocytes.

View larger version (29K): [in this window] [in a new window]   FIGURE 3. FCM analysis of CD1/CD45RC (row A), CD2/CD8 (row B), and CD8/CD8 (row C) expression by fetal TCR+ thymocytes in relationship to cell proliferation. thymocytes isolated from 90-day-old fetuses were cell surface stained with combination of anti-TCR and appropriate mAbs (different rows) and fixed in 70% ethanol, and their DNA was visualized using 7-AAD. In each staining (row), single TCR+ thymocytes were gated (data not shown) and analyzed for cell surface expression by individual dot plots (column I) on the base of which thymocytes were further gated by three to four regions (R1–R4). The DNA content (7-AAD fluorescence) for single thymocytes in individual regions for corresponding staining (row) is shown in column II (for R1), column III (for R2), column IV (for R3), and column V (for R4). Positions of cells in G0-G1, S, and M-G2 cell cycle phase according to 7-AAD fluorescence are indicated on x-axis of each histogram. The similar results were obtained in all experiments independently of animal age and breed.

Apoptosis of the CD4⁺CD8⁺ thymocytes is negligible, whereas CD4^– thymocytes die substantially

Cell cycle analysis showed that a substantial portion of CD4⁺ thymocytes was dividing, whereas only one subset of CD4^– thymocytes proliferated (Fig. 3). In further analyses, we examined the extent of cell death among individual thymocyte population in short-term culture (Fig. 4). The cell death is almost nondetectable in freshly isolated cells (Fig. 4B, 0 h), but is pronounced with culture in a time-dependent manner (Fig. 4B, 4–20 h). The majority of all dying thymocytes were found to be within the T lineage cells, i.e., CD4⁺TCR^– (Fig. 4C) and CD4^–TCR^– (Fig. 4E), and also within the group of CD4^– thymocytes (Fig. 4F). In contrast, CD4⁺ thymocytes rarely died in culture (Fig. 4D), and the vast majority of them were in S + G₂-M phase of the cell cycle. After 20 h of culture, only 4% of CD4⁺ thymocytes were dying (Fig. 4D) in comparison with 45% of CD4^– thymocytes (Fig. 4F). Despite facts that CD4⁺ thymocytes were vigorously cycling and did not die, noteworthy was their stable frequency (Fig. 4A, region R2) and even decreasing mitotic activity during culture without accumulation in other cell cycle phases (Fig. 4D). DNA fragmentation analyses showed that thymocyte subsets succumb to apoptosis (gel strips in Fig. 4, B–F) because FCM-sorted and cultured subsets of thymocytes gave the DNA ladder profile characteristic of apoptosis. No DNA smear typical of necrosis was detected. Moreover, because we had used fresh, sorted cells, these data indicate that thymocytes die spontaneously and do not require coculture with other thymocyte subsets to produce this effect. The apoptotic nature of cell death for cultured thymocytes was also shown by staining with annexin-V and propidium iodide (data not shown).

View larger version (43K): [in this window] [in a new window]   FIGURE 4. Analysis of thymocytes and their subsets in relationship to cell death. Thymocytes isolated from 112-day-old fetuses were used fresh or were cultured for 4, 8, 12, or 20 h, stained with anti-TCR and anti-CD4, and fixed in 70% ethanol, and their DNA was visualized using 7-AAD. Individual thymocyte populations were gated by R1–R4 regions (dot plot A), and DNA content of these subsets during culture was further analyzed (histograms in C–F) with respect to sub-G (dying cells), G0-G1 (quiescent/growing cells), and S plus G2-M (cycling cells) phase of the cell cycle. Analysis of DNA content for all thymocytes is also shown (B). Contribution of individual thymocyte subset frequency to each quadrant (A) and phases of the cell cycle (B–F) during culture is indicated on the right of particular dot plot or histogram. Freshly isolated thymocytes in R1–R4 regions were also sorted by FCM as pure populations and thereafter cultured for 12 h, and the extent of apoptosis and/or necrosis after culture was determined by gel electrophoresis DNA fragmentation assay (gel strips are given on extreme right of each histogram). This analysis showed that, during short-term culture, all thymocytes die exclusively by apoptotic type of cell death. The apoptosis of individual thymocyte subsets was also examined by multicolor staining (surface labels PE, Cy5, and/or Alexa Fluor 350) that included the detection with annexin-V FITC and propidium iodide (G–I). Twelve principal thymocyte subsets (numbered 1–12 on the x-axis in the same manner as in the Fig. 1 and text) were analyzed. The proportion of individual thymocyte subsets among all thymocytes (G), the proportion of individual thymocyte subsets among apoptotic thymocytes (H), and the proportion of apoptotic cells among individual thymocyte subset (I) are shown. Individual bars represent average values obtained from six animals tested in three independent experiments.

CD2⁺CD8^–CD1⁺CD45RC^– thymocytes are common precursors of CD4^– thymocytes, giving rise to all TCR⁺ subsets that may be found in the periphery

Cell cycle analysis (Fig. 3) showed that only one subset of CD4^– thymocytes had the capacity to divide (subset 7, Fig. 1). This particular subset is also most sensitive to apoptosis (Fig. 4I). Among the group of CD4⁺ thymocytes, there are two proliferating subsets (subsets 1 and 3), but they have a very low frequency of apoptosis (Fig. 4I). We therefore investigated individual thymocyte subsets during prolonged culture (Fig. 5). In agreement with previous findings (Fig. 4G) and published results (6), the majority of freshly isolated thymocytes represented subset 7, whereas other CD4^– thymocytes (subsets 5–12) were minor at this time point (Fig. 5A, day 0). Regarding CD4⁺ thymocytes (subsets 1–4), the majority of them were located among subsets 1 and 3 (Fig. 5A, day 0). With prolonged culture, the frequency of subset 7 gradually decreased to zero at day 14 of culture as the proportion of subsets 5, 6, 11, and 12 increased (Fig. 5A). However, the frequency of subsets 5 and 11 reached their maximum at day 7 of culture, after which they gradually disappeared, whereas the proportions of subsets 6 and 12 persistently increased to become the only two subsets that remained after 20 days of culture (Fig. 5A). The same overall pattern of thymocyte behavior in culture shown in Fig. 5 was independently confirmed in other experiments in which animals of different age and breed were tested. The only exception was in postnatal pigs, in which the developmental pattern of the culture was the same, but the frequencies of the resultant cells observed at day 20 in fetal animals were reached earlier on approximately day 12. Because above mentioned data have indicated that subset 7 is a common precursor of CD4^– thymocytes, we have sorted this subset by FCM (Fig. 6A) and cultured it in complete RPMI 1640 medium for up to 7 days. The obtained results from culture (Fig. 6B) showed the similar pattern as shown in Fig. 5A, with majority of cells in subsets 5, 6, 8, 11, and 12 and no CD4⁺ thymocytes present.

View larger version (37K): [in this window] [in a new window]   FIGURE 5. The frequencies of thymocytes and their individual subsets during prolonged culture (0–20 days). The proportion of individual thymocyte subsets (numbered 1–12 on the x-axis in the same manner as in the Fig. 1 and text) among all thymocytes (A), the proportion of thymocytes among all thymocytes cultured either alone (in complete RPMI 1640 medium only) or on thymic stromal feeder monolayers (B), and the proportion of CD45RC+ cells among TCR+ (detected as CD3highTCR– cells) and TCR+ thymocytes (C) are shown. Individual bars represent average values obtained from 18 animals tested in four independent experiments.

View larger version (26K): [in this window] [in a new window]   FIGURE 6. The frequencies of individual subsets of thymocytes arisen from FCM-sorted thymocyte subset 7 (A) during culture in complete RPMI 1640 medium only (B) or in the presence of thymic stromal feeder monolayer cells (C). A, Shows sorting strategy when freshly isolated thymocytes (ORIGINAL) were stained for TCR, CD1, and CD8, and gated for thymocytes by region R1, and subset 7 (CD2+CD8–CD4–CD1lowCD45RC–) was sorted from gated cells according to CD1/CD8 phenotype as region R2. Sorted cells (SORTED) were reanalyzed in the same manner before culture to check the purity. The proportion of individual thymocyte subsets after culture (B and C) was determined by FCM analysis, as described in Fig. 2. Individual bars in graphs B and C represent the proportion of individual thymocyte subsets (numbered 1–12 on the x-axis in the same manner as in the Fig. 1 and text) among all thymocytes during different time points of culture (z-axis). Bars represent average values obtained in three independent experiments.

We have also investigated surface expression of CD45RC on and thymocytes during culture because CD45RC is a maturation marker of the T cell lineage. Fig. 5C shows the proportion of CD45RC⁺ cells among TCR⁺ and TCR⁺ thymocytes. Although a number of CD45RC⁺ thymocytes remained constant during culture (Fig. 5C, ), the proportion of CD45RC⁺ thymocytes increased from almost zero to 90–95% in 15 days of culture (Fig. 5C, ).

The CD4⁺CD8⁺ thymocytes contain transcript for productive rearranged TCR, whereas transcript for rearranged TCR declines in these cells

Phenotypic and cell cycle studies suggest that CD4⁺ thymocytes possess unique features in comparison with other thymocytes. They exclusively bear CD8, are always CD1⁺, may coexpress CD45RC, have no counterpart in the periphery, are actively dividing, and although their number decreases during prolonged culture, they do not die. In further analysis, we examine the occurrence of lymphocyte-specific transcripts in FCM-sorted subsets of thymocytes (Fig. 7). Suspensions of all thymocytes were stained, and individual thymocyte subsets were sorted by FCM according to their CD4 and TCR phenotype into four populations in the same manner as indicated in Fig. 4A (regions R1–R4). Following FCM sorting, RT-PCR was performed (Fig. 7). This analysis showed that CD4⁺TCR^– thymocytes (Fig. 7, first line) were composed of T lineage cells (TCR⁺ and TCR⁺), including less differentiated precursors (RAG-1⁺ and TdT⁺). We were unable to amplify transcripts typical for B cells (VDJ_H, constant portion of Ig L chain (LC)-, LC-) or T cells (TCR). In contrast, CD4^–TCR⁺ thymocytes (Fig. 7, third line) were composed of mature T cells (TCR⁺) with no TCR, TCR, RAG-1, TdT, VDJ_H, LC-, or LC- expression. CD4^–TCR^– thymocytes (Fig. 7, fourth line) were composed of mixtures of cells with B cell (VDJ_H⁺, LC-⁺, LC-⁺) and T cell (TCR⁺, TCR⁺) phenotype, including less differentiated precursors (RAG-1⁺, TdT⁺). Most interestingly, CD4⁺TCR⁺ thymocytes (Fig. 7, second line) contained transcripts typical for T cells (TCR⁺), but at the same time, they also contained transcripts for T cells (TCR⁺) and transcripts for rearrangement-associated genes (RAG-1⁺, TdT⁺). Because the same number of sorted cells was used for analysis of each subset and CD4⁺TCR⁺ thymocytes clearly bear TCR on their surface, intensity of TCR and TCR bands indicates that CD4⁺TCR⁺ thymocytes actively transcribe rearranged TCR locus, whereas TCR transcripts are in decline (Fig. 7, compare intensity of TCR and TCR bands for CD4⁺TCR⁺ thymocytes with other thymocytes, i.e., CD4^–TCR⁺ subset). Interestingly, although CD4⁺TCR⁺ thymocytes clearly bear a rearranged TCR, they were negative for TCR (Fig. 7, second line). However, when TCR transcript amplified from CD4⁺TCR⁺-sorted cells (Fig. 7, second line) was cloned, amplified in vivo, and sequenced, all of 20 obtained sequences were found to be productive (data not shown).

View larger version (32K): [in this window] [in a new window]   FIGURE 7. Detection of transcripts for pre-TCR (pT), TCR V region (TCR), TCR V region (TCR), TCR V region (TCR), RAG-1, TdT, Ig H chain V region (VDJH), and LC- and LC- in the sorted thymocyte subsets. -actin was used as a control for determining relative transcript transcription. Thymocytes isolated from 100-day-old fetuses were stained with anti-TCR and anti-CD4, and individual CD4/TCR thymocyte subsets were sorted by FCM in the same manner as indicated in Fig. 4A. All sorted cells were reanalyzed to ensure quality and efficiency of the sort (>99%). The results are representative of five experiments using DG80, DG95, DG100, and DG112 fetuses, and 14-day-old piglets.

The CD4⁺CD8⁺ thymocytes alter their phenotype and differentiate along the T cell lineage program

Because unique features of CD4⁺ thymocytes resemble T lineage cells in their development, we have sorted this subset by FCM (Fig. 8, A and B), cultured it for up to 14 days, and inspect resultant cells from culture for presence of T cells (Fig. 8C). Because of the current unavailability of a mAb specific for TCR, putative T cells in this work were detected as CD3^highTCR^– cells, as it was shown that this approach is sufficient for identification of T cells (6, 7). The obtained results showed that at least a part of CD4⁺ TCR⁺ thymocytes could revert their phenotype to become T cells (Fig. 8C; CD3⁺TCR^–). Reanalysis of resultant cells from culture that kept TCR⁺ phenotype (Fig. 8C; CD3⁺TCR⁺) showed that they remain CD4⁺ (data not shown). The culture in complete RPMI 1640 medium (Fig. 8) or in the presence of thymic stromal feeder monolayers (data not shown) did not result in any differences. To additionally prove the T cell nature of resultant cells from the culture (Fig. 8C), TCR and TCR transcripts were amplified (data not shown). VDJ rearrangements of amplified TCR transcripts were further analyzed by high-resolution CDR3 length analysis (Fig. 8D). Semipolyclonal patterns were a common feature of all samples, indicating a broad TCR repertoire resulting from different T cell clones.

View larger version (17K): [in this window] [in a new window]   FIGURE 8. Analysis of FCM-sorted CD4+ thymocytes after culture in complete RPMI 1640 medium. Note that cells cultured in the presence of thymic stromal feeder monolayers (data not shown) did not result in any differences. Freshly isolated thymocytes were stained for TCR and CD4, gated for CD4+ thymocytes, and sorted by FCM (A). Sorted cells were reanalyzed before culture to check the purity (B). The resultant cells after 14 days of culture in complete RPMI 1640 medium were stained for TCR and CD3 for analysis of the proportion of (CD3+TCR+) and (CD3+TCR–) T cells (C). The second part of resultant cells was used for RNA isolation, amplification of TCR transcripts, and analysis of VDJ rearrangements by high-resolution CDR3 length analysis (D). The results are representative of three independent experiments.

	Discussion

Top Abstract Introduction Materials and Methods Results Discussion Disclosures References

This is the first known study that directly demonstrates that CD4^– thymocytes mature and diverge into different CD2/CD8 subsets from CD2⁺CD8^–CD1⁺CD45RC^– precursors (subset 7) to give rise to all peripheral T cell subpopulations. Our earlier work (3) showed that T cells require 15–18 days to differentiate from pro-T cells to immature CD3^–CD4⁺CD8⁺ and then to CD3^lowCD4⁺CD8⁺ double-positive stage. These then mature into CD4⁺ or CD8⁺ single-positive lymphocytes expressing CD3 in high density. In contrast, thymocytes develop without any CD3^low transitional stage and require less than 3 days for expression of CD3 in high density (3, 6). This study complements such kinetic data by showing that thymocytes need 15 additional days to diverge into CD2/CD8 subpopulations and to reach terminal stages of development. This is based on in vitro culture studies (Fig. 6B) and changes in CD45RC expression (Fig. 5C) that directly correspond to the appearance of thymocytes in vivo during fetal ontogeny (6). Early expression of TCR is consistent with studies in other species (13, 14, 15) that indicate that T cells require a shorter time period for expression of TCR than T cells. Although T cells do not follow T cell progenitors through MHC-dependent positive and negative selection (5), our studies indicate that both and cell lineages require 15–18 days for full maturation.

Mitotically active thymocytes of either or lineage occupy the developmental stage before, during, or immediately after the rearrangement of the TCR genes (16, 17). Although this stage can be found only in CD3^–TCR^– thymocytes of lineage (3, 16, 17), significant proliferation of lineage occurs after the expression of TCR (Fig. 3F) (3, 6, 18). Only one subset of CD4^– thymocytes (subset 7) had capacity to divide and was also most sensitive to spontaneous apoptosis. This subset was shown to contain the common T cell precursors because they were sufficient for the generation of CD1^–CD45RC^– intermediate stages (subsets 5, 8, and 11), and subsequently the CD1^–CD45RC⁺ mature stages (subsets 6, 9, and 12). All belonged to CD4^– thymocyte group. In this respect, the following are significant: 1) intermediate stages were still somewhat sensitive to spontaneous apoptosis, whereas mature stages were resistive, and 2) immature and intermediate stages reached their maximum numbers during culture and thereafter disappeared, whereas only mature stages survived in long-lasting culture. The precursor subset 7 might correspond to cycling CD4^–CD8^–heat-stable Ag⁺CD90⁺ population recognized in mice (18), although there is no evidence that this population can generate other T cells. Studies in humans showed that CD1⁺ thymocytes express transcripts for rearrangement-associated genes (RAG-1), behave as functionally inert cells, and can differentiate into CD1^– thymocytes (19, 20). This follows the pattern of this study, although human studies did not provide the evidence that such population contains progenitors of all T cells. Moreover, a mixture of CD1⁺ T cells was used for transfer experiments in humans (20) with no additional information about CD2, CD8, CD4, or CD45Rx phenotype. This led authors to speculate that CD1⁺CD4⁺ T cells may be an obligatory early stage of development for all T cells. We show in this study that CD4⁺ thymocytes are not able to alter their CD4⁺ phenotype. Perhaps this phenomenon develops independently among T cells and at least a portion of them alters their phenotype to become T cells.

The maturation of CD4^– thymocytes from immature CD1⁺CD45RC^– to intermediate CD1^–CD45RC^– cells is restricted to the thymus, as no CD1⁺ T cells are exported to the periphery (6). In contrast, the maturation step from intermediate CD1^–CD45RC^– to mature CD1^–CD45RC⁺ cells is thymus independent, as our ontogenetic data show that CD1^–CD45RC^– can be exported from thymus into the periphery (6). The dependence on thymus environment of the former event is controversial because the maturation of thymocytes can also occur in vitro in complete medium only (Fig. 6B). Thus, thymic factors such as stromal cells or IL-7 shown to be necessary for TCR gene rearrangement and expression (21, 22, 23, 24) may not be needed for subsequent maturation of thymocytes expressing TCR. This is consistent with studies showing that thymic factors are not required for proliferation or survival TCR⁺ lymphocytes (24, 25). Surprisingly, the presence of stromal cells facilitated development of thymocytes along CD4⁺ thymocyte pathway (Fig. 6C), which may reflect a migration route for thymocytes in the thymus. Because vast majority of TCR⁺ thymocytes are located in medulla and corticomedullary border, an area poor in cortical stromal cells, these thymocytes may differentiate inside the CD4^– thymocyte group. Residing in the cortex may impair normal T cell development, and these cells may develop into CD4⁺ thymocytes. This is in agreement also with observation that CD4⁺ thymocytes always constitute only a minor fraction of all T cells in the thymus (Fig. 5) (6). Even the outcome of studies involving culture on stromal cells can be an artifact of the type and concentration of stromal cell used; the presence of stromal cells necessary for rearrangement and expression of TCR genes (21, 22, 23, 24) may also provide an environment necessary for reactivation of TCR rearrangement. This may lead to development of CD4⁺ thymocytes that may escape from their lineage program, alter their phenotype, and in turn adopt differentiation along the lineage. This maturation step is subsequently independent of stromal cells (Fig. 8). In this respect, it is significant that although CD4⁺ thymocytes bear productively rearranged TCR genes, no such T cells can be found in the periphery (6). This may be related to their permanent CD1 expression that prevents them from being exported from the thymus (26) while keeping them on the route to becoming T cells. Alternatively, once they adopt an aberrant lineage program and acquire CD4 and CD8 expression, their lack of complete TCR rearrangements keeps them in the CD1⁺ immature stage until they alter their TCR phenotype. This would agree with regulation of CD4 and CD8 expression during T cell development by c-myb/hairy and enhancer of split homologue/silencer-associated factor transcriptional factors (27).

Our current data are insufficient to determine the precise developmental progression of CD4⁺ thymocytes. In vitro studies (Fig. 8) suggest that at least some become CD3^highTCR^– T cells because they contain productively rearranged V and V. Because this population of CD4⁺ thymocytes is positive for pT and productively rearranged TCR, TCR, and TCR transcripts (Fig. 7), it probably contains a mixture of TCR⁺ cells expressing or missing TCR and/or pT genes in all different combinations. Perhaps those cells coexpressing TCR and pre-TCR go on to express the mature TCR, whereas others are blocked in development. However, the fact that CD4⁺ thymocytes are very rare suggests that their T cells’ progeny would not significantly influence the total T cell pool. In this respect, the absence of TCR transcript in CD4⁺TCR^– thymocytes, which are progenitors of peripheral T cells, indicates that the rearranged TCR genes in the T cells are very infrequent. This is probably caused by deletion of the TCR locus during TCR rearrangement (5) and/or by the low frequency of CD4⁺ thymocyte progeny. It is significant that a similar population of thymocytes extinguishing their TCR expression was predicted based on studies of wild-type, transgenic, and knockout mice (28, 29, 30, 31, 32, 33). However, there is no TCR⁺CD4⁺CD8⁺ transitional stage, but only a minor population of CD4⁺CD8⁺CD3^–CD25^low thymocytes being regarded as progenies. Also, when different species are compared, it is notable that CD1⁺CD4⁺CD8⁺ thymocytes occur in humans (20) and fetal mice (14, 34) and often display different properties than conventional CD4^– T cells. Furthermore, CD4⁺ T cells are also found in human fetal liver, which is regarded as an alternative site of T cell development for this species (35, 36). CD4⁺CD8⁺ thymocytes can also be found after transgene insertion and chromosomal deletion (32). Chromosome mapping of such animals indicated that the vs lineage decision can be independent of successful gene rearrangement. This would support our data that suggest, at least in the case of CD4⁺ thymocytes, productive rearrangement and expression of TCR genes do not necessarily determine lineage commitment. In any case, the results of our studies in a nontraditional species provide new insight into the development of T cells that may be valuable in understanding this process in many species.

	Acknowledgments

 
We gratefully acknowledge Dr. John E. Butler from University of Iowa (Iowa, IA) for critical reading of the manuscript. Our warm gratitude is also extended to Lucie Poulová and Marta Stojková for excellent technical assistance. We are grateful to the following researchers for the gifts of mAbs: H. Yang from Institute of Animal Health (Pirbright, U.K.), J. K. Lunney from Animal Parasitology Institute (Beltsville, MD), K. Haverson from University of Bristol (Bristol, U.K.), C. R. Stokes from University of Bristol, M. D. Pescovitz from Indiana University (Indianapolis, IN), and C. A. Huang from Massachusetts General Hospital (Charlestown, MA).