The Journal of Immunology, 1999, 162: 1448-1459.
Copyright © 1999 by The American Association of Immunologists
Distinct Effects of Jak3 Signaling on 
ß and 
Thymocyte Development1
Elizabeth E. Eynon*,
Ferenc Livák*,
Keisuke Kuida2,*,
David G. Schatz*,
and
Richard A. Flavell3,*,
*
Section of Immunobiology and
Howard Hughes Medical Institute, Yale University School of Medicine, New Haven, CT 06520
 |
Abstract
|
|---|
Janus kinase 3 (Jak3) plays a central role in the transduction of
signals mediated by the IL-2 family of cytokine receptors. Targeted
deletion of the murine Jak3 gene results in severe reduction of 
ß
and complete elimination of 
lineage thymocytes and NK cells. The
developmental blockade appears to be imposed on early thymocyte
differentiation and/or expansion. In this study, we show that
bcl-2 expression and in vivo survival of immature
thymocytes are greatly compromised in Jak3-/- mice. There
is no gross deficiency in rearrangements of the TCR and
certain loci in pre-T cells, and a functional TCR transgene
cannot rescue lineage differentiation in Jak3-/-
mice. In contrast, a TCRß transgene is partially able to restore
ß thymocyte development. These data suggest that the signals
mediated by Jak3 are critical for survival of all thymocyte precursors
particularly during TCRß-chain gene rearrangement, and are
continuously required in the lineage. The results also emphasize
the fundamentally different requirements for differentiation of the
ß and T cell lineages.
|
Introduction
|
|---|
Intrathymic
differentiation is characterized by the sequential emergence
of distinct cell lineages. In each lineage, alternating steps of
differentiation and proliferation ensure the development of appropriate
numbers of mature cells. A small number of bone marrow-derived cells
give rise to the NK, intrathymic dendritic cell, and ß and
T cell lineages (1). Phenotypically, most of these precursors are
characterized by the lack of CD4, CD8, and CD3 (triple-negative
(TN)4 cells) and the presence
of various combinations of CD25, CD44, and CD117 markers that define an
ordered developmental progression (stage I-IV) of these cells (2). At,
or shortly after, the onset of commitment to the T lineage, rapid
proliferation occurs (stage II or pro-T), followed by rearrangement of
the genes encoding the ß-, -, and -chains of the TCR (stage III
or pre-T) (3). Precursors with functional rearrangement of the and
genes diverge at this stage and give rise to lineage T cells
(4, 5). Cells with productive rearrangement of the ß-chain gene
progress into stage IV (postpre-T) (6) and develop into ß T cells
(7). The TCRß-chain, in association with the invariant preT-chain
(pre-TCR) (8), appears to induce differentiation and extensive
proliferation of TN precursors into the large number of CD4/CD8
double-positive (DP) thymocytes (9, 10).
Both cellular interactions between the stromal and lymphoid
components as well as humoral factors influence early thymocyte
development, although the relative contribution of each of them is
poorly understood. Of the cellular interactions, the TCR- or
pre-TCR-mediated effects have been best characterized (9, 10, 11, 12, 13), although
other molecules may also play important roles (14, 15). Of the humoral
contributors, several cytokines have been implicated, but
gene-targeting experiments or naturally occurring mutations have proved
the critical role of only IL-7 (16, 17) and the c-kit ligand
(18). IL-7 binds to its receptor, which is composed of a unique -
and shared common -chains (reviewed in 19 . This latter chain
also contributes to the receptors of IL-2, IL-4, IL-9, and IL-15
(reviewed in 19 . After ligand binding, the signal from a cytokine
receptor is transmitted to the nucleus via specific combinations of the
Jak kinases and STAT molecules. The signals from the IL-7R are
primarily relayed by the Jak1, Jak3, and STAT5 molecules (reviewed in
Refs. 19–21). Targeted deletion of genes encoding IL-7 (16), the
(17, 22)- or (23, 24, 25)-chains of its receptor, and Jak3 (26–29 and
this study) generally confirms the singular significance of this
pathway in thymocyte development. Importantly, IL-7 has recently been
shown to have a direct effect on cell survival by inducing expression
of the antiapoptotic gene, bcl-2 (30, 31, 32). Most of the
studies until recently, however, did not address the potential
molecular mechanisms that lead to the observed severe defects. Most
importantly, it has not been resolved whether the IL-7 signaling
cascade affects differentiation, proliferation, or both, and whether it
does so directly or indirectly by cooperating with other, most notably
the TCR or pre-TCR-associated, signaling pathways.
To resolve some of these questions, we have generated and studied mice
that are deficient in the Jak3 gene (29). We have been particularly
interested in the following questions: 1) To what extent does Jak3
influence proliferation and survival of thymocyte precursors? 2) Are
TCR gene rearrangements reduced in Jak3-/- thymocytes? 3)
Are and pre-TCR-dependent signals defective in the absence of
Jak3? Our results indicate that immature Jak3-/-
thymocytes, before initiation of TCR gene rearrangement, proliferate
extensively, but die rapidly at or just before the point of
TCRß-chain selection. This is accompanied by a significant reduction
of intracellular bcl-2 expression. Furthermore, we show that
TCRß- and -chain gene rearrangements are not severely reduced in
Jak3-/- thymocytes. Introduction of functionally
rearranged TCRß or TCR/ transgenes, however, has dramatically
different consequences on the restoration of development of the
corresponding thymocyte lineages in Jak3-/- mice. These
data suggest that Jak3 is critical primarily for survival of all
prothymocytes and for the survival/expansion of TCR-dependent T
cells, but not for differentiation or survival of the pre-TCR-dependent
thymocytes.
|
Materials and Methods
|
|---|
Mice
Jak3-deficient mice (Jak3-/-) were produced at
Yale University mouse facility, as previously described (29). All mice
analyzed were between 4–8 wk of age. The D10 TCRß-chain transgenic
mice were provided by D. Sant’Angelo (New Haven, CT) and
were maintained on a mixed background (33). The G8 TCR
transgenic mice were generated by S. Hedrick, obtained from
L. LeFrançois (Farmington, CT) (34), and maintained at Yale
University mouse facility on H-2 d/d, b/d, or b/b genetic background
after backcrossing to C57BL/10.D2 or C57BL/6 mice,
respectively.
Antibodies
Anti-CD4 (GK1.5), CD8 (53-6.7), and anti-MHC class II
(M5/114) (American Type Culture Collection, Manassas, VA) were grown in
Bruff’s media with 5% FCS and purified over Sepharose/protein G
(Pharmacia Biotech, Piscataway, NJ) column. Anti-CD44, anti-CD25,
anti-bcl-2, anti-NK1.1, anti-TCRß,
anti-TCR/, anti-TCR Vß8, anti-Mac-1, anti-CD90,
Fc block, and streptavidin were all purchased from PharMingen (San
Diego, CA) with biotin, FITC, Cy-chrome, APC, or PE conjugations.
Annexin V-FITC was purchased from BioWhittaker (Walkersville, MD). 7AAD
was purchased from Calbiochem (La Jolla, CA).
Cell depletions
Thymocytes were depleted of mature cells by addition of
biotinylated anti-CD4, anti-CD8, anti-NK1.1,
anti-TCRß, anti-TCR/, anti-MHC class II,
anti-FcR, and anti-Mac-1, followed by goat anti-rat and
goat anti-mouse IgG magnetic beads (PerSeptive Biosystems,
Framingham, MA). After incubation in ice, the suspensions were placed
next to a magnet at 4°C. After one round of depletion, cells were
labeled with streptavidin-APC and then gated for APC-negative
populations.
FACS staining
Surface staining for FACS analysis was performed by addition of
50 µl of diluted Ab to 1–3 x 106 cells in
microtiter plates. Cells were incubated for 20 min on ice, followed by
three washes with PBS plus 5% FCS plus 0.02% sodium azide (FACS
buffer). Streptavidin with various fluorochromes were added in the same
manner. For 7AAD labeling, the cells were fixed with 1%
paraformaldehyde (Sigma, St. Louis, IL) in PBS for 15 min on ice,
washed twice in FACS buffer, and resuspended in 5 µg/ml 7AAD in 0.1 M
sodium citrate, 0.1% Triton X-100 (Pierce, Rockford, IL), plus 50
µg/ml RNase (Sigma) in PBS. The cells were incubated for at least 30
min before analysis. For bcl-2 analysis, cells were first
stained for surface Ags, fixed with 2% paraformaldehyde for 20 min on
ice, and washed twice in FACS buffer. Cells were then permeabilized
using 0.3% saponin (Sigma) in FACS buffer. Anti-bcl-2 Abs
were added in saponin buffer for 20 min on ice, followed by two washes
in saponin buffer and one final wash in FACS buffer. As a control,
hamster anti-trinitrophenol (TNP) was used in the same manner.
Analysis was performed on a FACSort or FACScaliber (Becton Dickinson,
San Jose, CA) and data were analyzed using Cellquest software (Becton
Dickinson).
Immunocytochemistry
Thymi from Jak3-/-, heterozygous littermates,
/ TCR transgenic mice, and Jak3-/- x / TCR
transgenic mice were fixed in 1% PLP (0.1 M phosphate buffer, pH 7.4,
0.2 M L-lysine, 1% paraformaldehyde) overnight, dehydrated
in 60% sucrose in three stages, and frozen in OCT media (Tissue-Tek
Sakura Company, Torrence, CA) in a bath of 2-methylbutane (Sigma) over
frozen CO2. Frozen sections (7 µm) of thymus were cut on
a cryostat (Leica Instruments, Nussloch, Germany) and fixed onto
silane-coated slides and stored at -70°C until use. After thawing,
the slides were rehydrated in 0.1 M Tris, pH 7.5, with 0.01% Triton
X-100. Nonspecific reactivity was blocked using the above buffer with
3% BSA added. Anti-TCRß, anti-TCR/, anti-Thy-1, as
well as rat IgG and hamster IgG were added onto the sections for 2
h in a humidified chamber at RT. Sections were washed three times for 5
min each with 0.1 M Tris, pH 7.5. Biotinylated anti-Ig reagents
were added for 1.5 h at RT, followed by three washes.
Streptavidin-alkaline phosphatase was added for 30 min at RT, followed
by three washes. The color reagent Histomark Red (Kirkegaard & Perry
Laboratories, Gaithersburg, MD) was used according to the
manufacturer’s directions. The sections were counterstained with
Gill’s hematoxylin #1 (Sigma) and washed in NH4OH
containing water to produce a blue counterstain. Photographs are at x4
magnification.
Polymerase chain reaction
High m.w. DNA was prepared from total and electronically sorted
TN thymocytes of wild-type (wt) and Jak3-/- mice using
standard protocols (35). PCR primers used for analysis of V and V
gene rearrangements and the control RAG-2 gene were as published (5, 36, 37, 38). Serial dilutions of 150, 75, and 25 ng DNA were amplified for
30 cycles in a thermocycler (MJ Research, Watertown, MA). PCR products
were separated on a 1.5% composite sieving agarose gel stained with
SYBR Green dye (Molecular Probes, Eugene, OR) and quantitatively
analyzed on a fluorimager with ImageQuant 1.2 software (Molecular
Dynamics, Sunnyvale, CA).
PCR restriction fragment length polymorphism (RFLP)
PCR RFLP (39) was performed on DNA purified from total and
electronically sorted thymocyte populations, according to the published
modified protocol (36, 37), except that V4- and V5-J1
reactions were restriction digested with AluI and
Eco47III enzymes (New England Biolabs, Beverly, MA),
respectively. The control wt and TCRß-/- samples had
been analyzed and published with similar results on multiple occasions
(36, 37). Quantitative analysis was performed on a PhosphorImager with
ImageQuant 3.0 software (Molecular Dynamics).
|
Results
|
|---|
In vivo proliferation and survival of intrathymic precursors from
Jak3-/- mice
The initial characterization of Jak3-/- mice has
been published (26, 27, 28, 29). We have now extended the phenotypic analysis
to the earliest, TN stages of thymocytes, since TN thymocyte
differentiation appears to be particularly severely affected in mice
deficient in the IL-7 signaling pathway (16, 17). We observed two
phenotypes with compensated and severe defects in thymocyte development
in Jak3-/- mice. The compensated developmental phenotype
had cell numbers approaching 10% of wt, tightly organized forward/side
scatter FACS profiles similar to normal thymocytes; TN populations were
variable, but had no marked differences from wt, with 
70% of cells
in stages III and IV (compare in Fig. 1
,
D and E, and Table I) as well as normal CD4/CD8 ratios (not
shown). Thymi from mice with the severe phenotype exhibited less than
1% of normal thymus cellularity, a "spread out" pattern of forward
and side scatter (shown in Fig. 1, A–C and Table I), and
increased percentages of cells negative for CD4 and CD8 (Table I).
Another commonly seen phenotypic characteristic of mice with the severe
phenotype is a diminution of the stage III to IV transitional,
CD44low/CD25low, TN population that is normally
present in wt (or compensated Jak3-/-) mice. The severe
phenotype predominated in the Jak3-/- mice, with 71% of
the mice shown in Table I having the severe pattern of development. The
severe and compensated patterns of thymocyte development did not
correlate with any other phenotype such as sex, age, or the development
of enlarged spleen (27, 29). The severely altered forward and side
scatter profiles could be explained in part by either increases in
stromal cells as a percentage of the total cell population or increases
in dead or dying cells in the thymus. Although cells analyzed for
stages of T cell development were gated using a lymphocyte cell gate
(outlined in Fig. 1, A–C), we cannot exclude that some
portion of the thymocytes analyzed includes nonlymphoid cells.
View larger version (48K):
[in this window]
[in a new window]
|
FIGURE 1. FACS profiles of Jak3-/- TN thymocytes. FACS analysis of
thymocytes from Jak3+/- (A and
D) (6 wk), Jak3-/- compensated
(B and E) (4 wk), and
Jak3-/- severe (C and F) (4
wk) mice. Forward (x-axis) and side scatter profiles
(y-axis) (A, B, and
C) with lymphoid-gated populations are shown.
Two-dimensional dot plots of CD44-FITC (y-axis) versus
CD25-PE (x-axis) on the TN lymphoid-gated cells
(D, E, and F) from the
same mice as shown in A–C. Percentages of cells present
in the four quadrants stages I-IV are shown in the upper right
corner of each plot in the same pattern as the quadrants.
|
|
The above data indicated a dramatic reduction in the numbers of cells
at all stages of thymocyte differentiation in both phenotypes
described, which could be due to reduced proliferation or impaired
survival of intrathymic precursors. To distinguish between these two
possibilities, we investigated both the proliferative status as well as
viability of various stages of TN thymocytes from wt and
Jak3-/- mice. Thymocytes were depleted of CD4-, CD8-,
TCRß-, TCR-, NK1.1-, B220-, and Mac-1-positive cells by
Ab-mediated magnetic bead depletion and FACS gating. The cells (from
Fig. 1
) were further stained for CD44 and CD25 as well as 7AAD. 7AAD, a
fluorescent analogue of actinomycin D, intercalates into DNA in amounts
proportional to the amount of DNA in the nucleus. Thus, cells in the
S/G2 stage of cycle will have twice as much 7AAD as cells
in G0/G1, whereas apoptotic cells with
subdiploid DNA content will incorporate less 7AAD than normal
G0/G1 cells. In Jak3+/- mice, the
majority of stage I and II thymocytes were in
G0/G1, with up to 20% of the cells in
G2/S. Interestingly, in the Jak3-/- mice,
stage I and stage II populations were almost all in cycle (100 and
82%) (Fig. 2C).
Jak3+/- littermates had varying proportions of cycling
thymocytes in stages III and IV (8–18%), with less than 1% of the
cells with subdiploid DNA (Fig. 2A). This is in part due to
the fact that dying thymocytes are rapidly removed from the thymus by
stromal cells (40). In stages III and IV, Jak3-/- mice
had a very large increase in the numbers of cells with subdiploid DNA
(85% of the cells in the severe mouse shown in Fig. 2). There were no
increases in the percentage of cells with subdiploid DNA in either
stage I or II in Jak3-/- thymocytes. Therefore, in the
Jak3-/- mice, most cells died at or around the
TCRß-chain selection point of development, in the transition from
stage III to IV. In contrast, at earlier stages in development, most
Jak3-/- cells were in cycle.
Decreased bcl-2 expression in Jak3-/-
mice
Since IL-7 has been shown to be partially responsible for the high
levels of bcl-2 expression in early thymocytes, we wanted to
determine whether any of these TN populations were deficient in
bcl-2. Stage III thymocytes in Jak3+/- mice
were >90% bcl-2 positive (Table II). The level of bcl-2 in
each cell, as measured by the mean channel of fluorescence intensity,
was also high (177). In Jak3+/- stage IV thymocytes, the
percentage of cells positive for bcl-2 remained high (93%),
but the expression level decreased (115). Jak3-/- mice
have >90% of stage III thymocytes positive for bcl-2, but
the level of expression was markedly reduced compared with their +/-
littermates (66 versus 177). Stage IV thymocytes from
Jak3-/- mice had a decreased percentage of cells positive
for bcl-2 (68%) also with lower expression levels than in
+/- littermates (88 versus 115). These data indicated that in the
absence of Jak3, stage III thymocytes do express bcl-2, but
at very low levels, despite the lack of IL-7 signaling, which was not
maintained in stage IV. These findings were in accordance with the
striking increase in cell death of Jak3-/- cells in the
transition between stage III and IV, as measured by 7AAD staining.
Rearrangement of the TCR genes in Jak3-/- thymocytes
Rearrangements of the TCR, on the one hand, and the
ß-chain genes on the other, are required for proper maturation of the
immediate precursors of the (13) and ß (10) T cell
lineages, respectively. Since Jak3 deficiency had a profound effect on
the survival of cells undergoing TCR gene rearrangement, as shown
above, it was possible that Jak3 is required for initiation of
rearrangement of some or all TCR loci. The extent of rearrangement of
the most common TCR (V4, V5, and V8 to J1) and TCR (V4 and V7 to
J1 and V1 to J4) genes (41) was compared in electronically sorted TN
thymocytes from Jak3+/- and Jak3-/- mice.
Serial dilutions of the DNA template and a control PCR for the
nonrearranging gene, RAG-2, were used to achieve semiquantitative
results. Rearrangements of all analyzed TCR genes appeared to be
comparable in Jak3-/-TN thymocytes with that of wt
controls (Fig. 3). In contrast, only
V1-J4 rearrangements were observed at significant levels (about
50% of wt), whereas V4 and V7 to J1 recombination was reduced
at least 20–50-fold. These results demonstrated that TN thymocyte
precursors are selectively deficient in rearrangement of the
V4/7-J1-C1 locus. This conclusion was suggested from the
analysis of total thymocytes in common -chain-deficient (42) and
IL-7R-chain-deficient (22) mice. However, we have also demonstrated
the presence of complete V1-J4 and several V-J1
rearrangements with the potential for generation of TCR
complexes. Thus, the major deficiency in T cells in
Jak3-/- mice (27) cannot solely be explained by the
blockade of TCR and gene rearrangements.
Analysis of rearrangement of Vß8 to Jß1.1–6 gene segments did not
show dramatic reduction in Jak3-/- mice compared with wt
samples (data not shown). There was also no change in the pattern of
TCR or ß expression in mature T cells by FACS analysis of splenic
T cells (not shown).
Depletion of productive TCR and gene rearrangements in
Jak3-/-, ß thymocytes
Productive rearrangements of TCR and genes promote
development of T lineage cells, and therefore are selected in
these cells. In contrast, ß lineage T cells show depletion of
productive TCR and joints (4, 5, 43), either because the
TCR biases lineage decisions (4, 5) or interferes with V-DJß
recombination (36). The distribution of in frame (productive) versus
out of frame (nonproductive) rearrangements of TCR and genes
within a given thymocyte population can be revealed by PCR RFLP
analysis (4, 5, 37). To determine whether these rearrangements had any
effect on precursor thymocyte development, we performed PCR RFLP
analysis on total and electronically sorted DP thymocytes of
Jak3-/- mice and compared the results with those obtained
from total thymocytes of Jak3+/- littermates. As has been
shown previously (4, 5, 37), wt total (i.e., essentially entirely
ß lineage) thymocytes exhibited depletion of in frame V4,
V5, and V8 to J1 and V1 to J4 joints (Fig. 4). In contrast, thymocytes from
TCRß-/- mice, in which most thymocytes are selected by
the TCR (36, 44), showed characteristic enrichment of in frame
joints of the same rearrangements (Fig. 4). ß lineage (total or
DP) thymocytes from Jak3-/- mice exhibited two distinct
patterns with PCR RFLP analysis. In frame V4 and V5 to J1
rearrangements were distributed randomly (Fig. 4, A and
C), whereas V8 to J1 and V1 to J4 rearrangements
showed depletion of in frame joints (Fig. 4B) similar to
that seen in wt mice (Fig. 4C). Since extremely low numbers
of thymocytes are found in Jak3-/- mice, the
latter results suggest that at least certain TCR- and -chain
combinations prevent differentiation of intrathymic precursors into the
ß lineage pathway, despite their inability to promote T
cell maturation.
View larger version (31K):
[in this window]
[in a new window]
|
FIGURE 4. PCR RFLP analysis of TCR and gene rearrangements in wt
and Jak3-/- thymocytes. A and
B, DNA purified from wt total (T),
Jak3-/- total (Jk-T), Jak3-/-DP
(Jk-DP), and TCRß-/- total (ß-) thymocytes was amplified for the indicated TCR
and rearrangements, followed by restriction digestion with the
indicated enzymes. Note that only V1-J4 joints were analyzed, as
rearrangements of the other TCR locus are extremely low in
Jak3-/- mice. Dashes indicate the position of the in
frame (productive) rearrangements, as determined from their
distribution in TCRß-/- thymocytes (where all TCR-based
selection occurs through the TCR). C, Graphical
representation of the quantitative analysis of the distribution of in
frame versus out of frame joints. The vertical axis indicates the
proportion of in frame joints, as determined from A and
B. The horizontal line marks the 33% random
distribution level. Note that V4- and V5-J1 rearrangements do
not exhibit significant deviation from the random 33% in frame joint
distribution in Jak3-/- samples. In contrast, V8-J1
and V1-J4 rearrangements are shown, which show depletion of in
frame joints in Jak3-/- thymocytes similar to wt
samples.
|
|
The effect of a TCR transgene on Jak3-/-
thymocyte development
The presence of TCR and gene rearrangements in
Jak3-/-TN thymocytes and, at least for some of these
rearrangements, the depletion of productive joints in ß lineage
thymocytes suggested that maturation, survival, and/or expansion of
lineage cells may be impaired even in the presence of a
functional TCR in Jak3-/- mice. This could explain
the profound loss of T cells in Jak3-/- mice. To
test this hypothesis, we introduced the functionally rearranged
TCR transgenes derived from the G8 T cell clone (34) into
the Jak3-/- background. G8 transgenic mice have been
shown to develop increased proportions of transgenic thymocytes
and peripheral T cells on the MHC class I H-2d background,
but not on the MHC class I H-2b background, which is
thought to represent a negative selecting environment for this
transgene (34, 45, 46). We analyzed thymocyte development of wt and
Jak3-/-, G8 transgenic mice on both H-2d and
H-2b backgrounds. To our surprise, G8 transgenic,
Jak3-/- mice had extremely small thymi, with average
cellularity at or even below that of the nontransgenic, severe
Jak3-/- mice (Table I). Hematoxylin/eosin staining of
frozen tissue sections of these rudimental thymi showed no distinct
cortico-medullary organization (not shown). Stainings with Abs specific
for TCRß (Fig. 5, a,
c, e, and g) and TCR (Fig. 5, b, d, f, and h) were
performed to assess the extent of thymocyte development. No
TCRß-positive (Fig. 5g), very few Thy-1 bright (not
shown), and some TCR-positive (Fig. 5h) cells were
detected in G8 transgenic, Jak3-/- H-2d
sections. In contrast, most of the thymocytes from G8 transgenic,
Jak3+/- H-2d littermates were strongly
positive for Thy-1 and TCR (Fig. 5f) and negative for
TCRß (Fig. 5e). Nontransgenic, Jak3+/- mice
had normal thymic organization and staining of TCRß and (Fig. 5, a and b), whereas nontransgenic,
Jak3-/- mice had less organized thymi with strong TCRß
and scattered staining (Fig. 5, c and d).
While G8 transgenic, Jak3+/- littermates on the
H-2b background showed intense staining for TCRß (data
not shown), there was no difference in overall cellularity, structure,
and staining pattern of thymi from G8 transgenic, Jak3-/-
mice, regardless of the MHC background (data not shown).
Flow-cytometric analysis corroborated these observations showing
20–40% TCR-positive DN and few TCR-negative DP thymocytes
in G8 transgenic Jak3-/- mice (Table I and Fig. 6D). Thus, it appears that in
contrast to nontransgenic mice, enforced expression of a TCR
results in some TCR-positive thymocytes in the
Jak3-/- background (Fig. 6, D and
E). The total number of TCR-positive cells, however,
remains extremely low, about 100-fold less than is observed in the G8
transgenic, Jak3+/- littermates (Fig. 6E).
These observations indicate that the lack of reconstitution of
the thymic lineage in the Jak3-/- mice is not
simply due to the lack of expression of the transgenic TCR.
The effect of TCRß transgene on Jak3-/- thymocyte
development
In contrast to T cells, a significant extent of ß
lineage differentiation has been observed in Jak3-/-
mice, especially with the compensated phenotype (Fig. 1), suggesting
that some pre-TCR- or TCRß-mediated signals can function in these
mice. To confirm that pre-TCR-mediated expansion of ß T cell
precursors is Jak3 independent, we introduced the functionally
rearranged Vß8+, TCRß transgene derived from the D10
ß T cell clone (33) into the Jak3-/- background.
Forward and side scatter profiles showed that D10 TCRß
Jak3-/- transgenic mice have a phenotype corresponding to
the compensated Jak3-/- mice (compare Figs. 1B
and 7C). TCR Vß8 expression from the whole thymus (Fig. 7, B and D) in
Jak3+/- and Jak3-/- TCRß transgenic mice
showed equivalent proportions of TCR low and high populations.
Nontransgenic mice have 5–10% TCR Vß8+ cells in the
whole thymus (not shown), whereas all TCRß+ thymocytes in
D10 transgenic mice were Vß8+ (96 and 97%,
respectively). Absolute cell numbers for the samples shown in Fig. 7, A–D, as well as a control nontransgenic
Jak3-/- mouse show the partial rescue of TCR-positive
thymocytes in the Jak3-/-D10 TCRß transgenic mice (Fig. 7E).
View larger version (21K):
[in this window]
[in a new window]
|
FIGURE 7. Forward and side scatter profiles and Vß8 expression in
Jak3-/- x TCR D10 ß-chain mice. Forward and side
scatter profiles (A and C) with Vß8
(representing the transgenic ß-chain) staining (B and
D) of Jak3+/- D10 ß TCR transgenic
thymocytes (A and B) or
Jak3-/- D10 TCRß transgenic thymocytes
(C and D) are shown. Jak3+/-
D10 ß TCR thymocytes were 96% Vß8 positive (22% TCR hi), and
Jak3-/- D10 ß TCR thymocytes were 97% Vß8 positive
(20% TCR hi). Mice shown were 8 wk of age. The total number of
TCRß-positive thymocytes (assuming complete allelic exclusion in the
transgenic mice) is shown for the samples displayed in
A–D, and for a nontransgenic Jak3-/-
mouse (stained separately with a ubiquitous TCRß Ab)
(E). Note the logarithmic scale of the cell
numbers.
|
|
Wt, nontransgenic Jak3-/-, TCRß transgenic
Jak3+/-, and TCRß transgenic Jak3-/- mice
were compared for other aspects of thymocyte development. Total
cellularity increased 18-fold in TCRß transgenic
Jak3-/- mice, compared with nontransgenic (severe
phenotype) littermates, although the cell numbers were still
approximately 10-fold reduced compared with wt or TCRß transgenic
Jak3+/- mice and only 2-fold higher than compensated
phenotype Jak3-/- mice (Table I and Fig. 7E). This may
reflect the severe reduction in precursors reaching stage III and
TCRß-chain expression. In the TCRß transgenic Jak3-/-
mice, thymocyte populations in stages I-III were somewhat diminished
(Fig. 8), but staining with 7AAD (Fig. 8)
and Annexin V (data not shown) shows a pattern indistinguishable from
that of wt or TCRß transgenic Jak3+/- mice. Some
individual mice have higher numbers of cells in cycle in stages I and
II with a concomitant decrease in the percentages of cells in
G0/G1.
Stage III TN thymocytes from TCRß Jak3-/- mice were
>90% positive for bcl-2, but the level of expression was
not increased by the TCRß transgene (67 versus 66, see Table II). In
TCRß transgenic Jak3+/- mice, the percentages of
bcl-2+, stage IV TN thymocytes were slightly
decreased compared with stage III cells (89%), whereas TCRß
transgenic Jak3-/- mice were intermediate between wt and
nontransgenic Jak3-/- mice with 80% of the same stage
cells bcl-2+ (Table II). Levels of bcl-2 were also
not increased in stage IV in the D10 TCRß transgenic
Jak3-/- mice.
These data suggest that in contrast to the TCR transgene, a
rearranged TCRß transgene is able to increase the total number of
thymocytes and enhance expansion and differentiation of cells subject
to the selection through the pre-TCR. This effect is, in part, mediated
by increased survival of both stage III and stage IV thymocytes. The
increased survival is coincident with the increased expression of the
transgenic TCRß-chain, but it is apparently not mediated by increased
levels of bcl-2 expression. Although thymocyte numbers are
not substantially increased, there is a significant decrease in cells
dying at stage IV in the presence of a rearranged TCRß-chain. These
data indicate that the coordinated and extensive expression (insofar as
the transgene required no rearrangement and can be expressed as early
as signals allow) of a fully rearranged TCRß-chain is enough to
rescue these cells from the death caused by the severe decrease in
bcl-2 expression in the absence of a functional pre-TCR
complex.
|
Discussion
|
|---|
We have performed molecular and genetic analyses on
Jak3-/- mice to elucidate some of the mechanisms
responsible for the severe defects in thymocyte development observed in
these mice (26, 27, 28, 29). Our results suggest that Jak3 must play a critical
role in the survival of intrathymic precursors before TCR gene
rearrangement and in precursors of lineage T cells even after
expression of the TCR. In contrast, Jak3 does not appear to be
critical for initiation of TCRß, , and at least some gene
rearrangements, nor for the signals mediated by the pre-TCR that allow
expansion and differentiation of ß lineage precursors.
Rearrangement of the TCRß, , and loci (3, 46), and commitment
to the ß and T cell lineages (47, 48) take place in the
narrow window of stages II-IV, TN thymocyte differentiation. The
IL-7/IL-7R signaling cascade, including Jak3, has been shown to be most
critical to these stages of thymocyte development (16, 17, 23, 24, 25, 26, 27, 28, 29),
primarily affecting survival of thymocytes by inducing expression of
the antiapoptotic gene, bcl-2 (30, 31, 32). In addition, it has
been shown that in vitro administration of IL-7 induces expression of
some TCR (e.g., V4) loci in fetal liver cultures (49). IL-7
produced by stromal cells is required for initiation of TCR gene
rearrangements in fetal thymic reaggregate cultures (50), and TCR
gene rearrangements are selectively reduced in vivo in mice that lack
the common cytokine receptor -chain (22). These studies have
collectively implied that IL-7 is a principal factor for initiation of
various TCR gene rearrangements either by direct activation of the
V(D)J recombination mechanism or by increasing the viability of
thymocytes actively engaged in V(D)J recombination. They have also
provided a partial explanation for the profound defect of T cell
differentiation observed in IL-7R (22, 23)- and Jak3 (27)-deficient
mice.
In our studies, we have found a significant extent of recombination at
the TCR and some loci in Jak3-/-TN thymocytes.
Depletion of productive rearrangements of some, although not all, of
these loci, similar to that seen in wt mice, was also evident in the
ß lineage compartment of Jak3-/- mice. Finally,
introduction of a functionally rearranged TCR did not restore
T cell development on the Jak3-/- background.
These results are in accordance with a recent report (42) on mice
lacking the -chain of the common cytokine receptor. Thymocytes of
these mice have reduced but not completely absent rearrangements of the
TCR loci, and introduction of a rearranged TCR transgene did
not restore development of T cells (42).
The simplest interpretation of our results indicates that a failure of
rearrangement is not sufficient to explain the complete lack of
lineage T cells. Since Jak3 is not absolutely necessary for
V1-J4 rearrangements (see Fig. 3) and TCR locus recombination
in Jak3-/-TN thymocytes, it is entirely conceivable that
some TCR complexes are formed in these cells. Our transgenic
studies indicate that Jak3 is continuously required for development and
maintenance of lineage in the adult thymus. Thus,
precursors, even with a functional TCR, probably die rapidly in
Jak3-/- mice. This results in the depletion of productive
rearrangements of certain TCR and genes in the ß lineage
compartment (see Fig. 4). Our data suggest that these complexes could
contain V8 and V1, but not V4 or V5 proteins. The
implication of these results is that the two most typical V gene
products of the adult thymus (V4 and V5) cannot dimerize with the
V1-J4-C4 TCR-chain even in the absence of other TCR proteins.
Some reports do indeed suggest that in adult T cells, V4
predominantly pairs with V7, whereas V5 pairs with V4 (51, 52). The preferred V pairs of the V1 protein are currently
unknown, although V6 (53) and V8 (this study) are definite
candidates. Recently, it has been reported that IL-7 may affect
recombination of some, but not other, variable genes within the Ig
heavy chain locus (54), a finding reminiscent of the selective blockade
of rearrangement of certain, but not all TCR loci. It will be
interesting to determine what factors mediate the IL-7 dependence of
rearrangement of these loci during early lymphoid differentiation.
The Jak3 signaling pathway is also critical for ß lineage
development. IL-7 enhances recombination of the TCRß locus (55) and
is required for stromal cell-mediated initiation of TCRß gene
rearrangements in fetal thymic reaggregate cultures (50). Introduction
of a fully rearranged TCRß transgene on the IL-7R-chain (56),
the common cytokine receptor -chain (57), or the Jak3 (58, 59)-deficient backgrounds demonstrated significant restoration of
ß lineage development with increased numbers of, albeit
functionally defective, peripheral T cells. It has, however, recently
been shown that a transgenic TCRß does not completely mimic the
function of the pre-TCR (60); therefore, we introduced a single TCRß
transgene into the Jak3-/- background. In accordance with
the above reports, we have found significant restoration of ß
lineage differentiation. This increase in cell number is predominantly
caused by the increase in survival of stage IV thymocytes and not of
any increase in earlier precursors. We have also demonstrated, however,
that the increase in number and survival of stage IV, TN thymocytes is
not accompanied by increased levels of bcl-2 expression, a
well-documented target of IL-7. It appears that the TCRß transgene
does not substitute for the missing IL-7/IL-7R/Jak3 signaling pathway,
but rather accelerates expansion and transition of the few surviving
precursors into a stage in which survival no longer depends on Jak3,
but instead on TCR-mediated signals. We did not observe significant
differences in the level of V-DJß rearrangements in
Jak3-/- versus wt, TN thymocytes; nor did we observe any
skewing of the TCRß-chain usage in mature T cells in
Jak3-/- mice (data not shown). A current model suggests
that V to DJß recombination may occur only in a small fraction of
stage III thymocytes, at or near to the end of their natural life span
(61); therefore, small differences in the level and/or timing of V to
DJß rearrangements may not have been detected in our studies. It is
possible that the differences between the severe and compensated
phenotypes may reflect the ability of individual mice to accomplish
TCRß-chain rearrangement in a timely fashion. Future experiments,
combined with a more detailed definition of initiation of V to DJß
rearrangements in stage III, TN thymocytes, will be required to
determine the role of Jak3 signals in direct or indirect promotion of
TCRß locus recombination.
Although IL-7 has documented effects on proliferation of TN thymocytes,
there are other cytokines that use the common -chain and Jak3. It
has recently been reported, for example, that the IL-15/IL-2 signaling
pathway plays an important role in the maturation of the three closely
related lineages of NK, NK1.1 ß TCR+, and
intraepithelial lymphocytes (62). Similarly, the few surviving stage II
TN thymocytes of common -chain-deficient mice are apparently
dependent on c-kit receptor, since elimination of the
c-kit receptor-mediated signaling in these mice results in
complete blockade of thymocyte development in all lineages with
accumulation of stage I and II TN cells (18). In the transition from
stage II to III, down-regulation of c-kit (2, 63), followed
by loss of the IL-7 pathway (this study) may confer special sensitivity
of thymocytes to alternative survival signals, such as those provided
by the pre-TCR. This mechanism would ensure that only pre-T cells with
a functional pre-TCR/CD3 complex could expand and enter the next stage
of differentiation.
From recently accumulating studies and our present data, we propose the
following model to explain the interdependence of thymocyte
differentiation on both cytokine receptor and TCR-mediated signals
(Fig. 9). Pro-T cells express both IL-7R
and c-kit, but not TCR. These cells are absolutely dependent
on IL-7/Jak3 and c-kit signals (18) for proliferation and
survival (Fig. 9). Proliferation, in part, is mediated by the
ß-integrin/TCF signals (64), whereas survival, in part, is mediated
by the IL-7/Jak3-induced expression of bcl-2 (30, 31, 32).
Initiation of TCR gene recombination (first the TCR and loci;
Livák et al., submitted) allows the cells to express the
TCR and enter the lineage pathway. Although most TCR and some
loci are rearranged in Jak3-/- pre-T cells, this
process may, in part, be IL-7 dependent (22). However, maturation and
expansion of pre-T cells are absolutely dependent on IL-7/Jak3
signals. Neither c-kit nor a functional TCR can
rescue this lineage (Fig. 9). The dependence of T cells on
IL-7/Jak3 signals is also suggested by the finding that all G8
TCR transgenic thymocytes (immature and mature) express high levels of
bcl-2 throughout development (data not shown). After
productive rearrangement of the TCRß locus, pre-T cells (stage III)
express the pre-TCR, which allows the differentiation of ß lineage
T cells (7). Coincident with acquisition of the pre-TCR, these cells
switch from IL-7/Jak3 (and probably also c-kit)-dependent
signals to pre-TCR-dependent signals. The pre-TCR is critical for
proliferation (6) and survival (61) of postpre-T cells. The effect on
survival appears to be mediated by mechanisms distinct from
up-regulation of bcl-2 expression (Table II). Accordingly,
introduction of a functional TCRß transgene into the
Jak3-/- background is sufficient to increase the number
of ß lineage cells (providing uniform pre-TCR signals in all pre-T
(stage III) cells independent of TCRß locus recombination), but not
the production of pre-T cells (stage III) (which are IL-7/Jak3, but not
TCR dependent; see Fig. 9). The molecular mechanism of this switch from
cytokine-dependent to TCR-dependent differentiation remains unclear. It
is clear that neither nor ß TCR transgenes are able to affect
bcl-2 expression. This conclusion has important implications
for the divergence of the ß and T cell lineages. Since many
lineage T cells emerge from a common precursor at the
pro-T/early pre-T stage (Livák et al., submitted), most of the
T cells develop with absolute dependence on cytokine signals,
but less stringent requirements for the TCR-mediated signals.
However, the gain of a TCR is not a neutral event since it
appears to be selected for in the lineage and has severe
consequences in the Jak3-/- mice (Fig. 9). In contrast,
ß lineage differentiation, initiated after productive TCRß
rearrangement, proceeds with decreasing dependence from cytokine
signals. Ultimate maturation of ß lineage thymocytes eventually
becomes exquisitely sensitive to TCR-mediated signals to allow positive
selection of a highly specific TCRß repertoire.
|
Acknowledgments
|
|---|
We thank D. Sant’Angelo for the D10 ß T cell receptor transgenic
mice. We also thank A. Kruisbeek and A. Hayday for critical
reading of this manuscript.
|
Footnotes
|
|---|
1 This work was supported by a National Institutes of Health training grant fellowship to E.E.E. (5T32 AI07019), a National Institutes of Health grant to R.A.F. (1P01 AI30548), and a Presidential Faculty Fellows Award from the National Science Foundation to D.G.S. K.K. was an Associate, D.G.S. is an Associate Investigator, and R.A.F. is an Investigator of the Howard Hughes Medical Institute. 
2 Current address: Vertex Pharmaceuticals, Inc., 130 Waverly St., Cambridge, MA 02139.
3 Address correspondence and reprint requests to Dr. Richard A. Flavell, Section of Immunobiology, Yale University School of Medicine, 310 Cedar St., Box 208011, New Haven, CT 06520-8011. E-mail address:
4 Abbreviations used in this paper: TN, CD3/CD4/CD8 triple negative; 7AAD, 7-amino actinomycin D; DP, CD4/CD8 double positive; Jak, Janus kinase; PE, phycoerythrin; RFLP, restriction fragment length polymorphism; RT, room temperature; wt, wild-type.
Received for publication June 1, 1998.
Accepted for publication October 21, 1998.
|
References
|
|---|
-
Shortman, K., L. Wu. 1996. Early T lymphocyte progenitors. Annu. Rev. Immunol. 14:29.[Medline]
-
Godfrey, D. I., J. Kennedy, T. Suda, A. Zlotnik. 1993. A developmental pathway involving four phenotypically and functionally distinct subsets of CD3-CD4-CD8 triple negative adult mouse thymocytes defined by CD44 and CD25 expression. J. Immunol. 150:4244.[Abstract]
-
Godfrey, D. I., J. Kennedy, P. Mombarets, S. Tonegawa, A. Zlotnik. 1994. Onset of TCR-ß gene rearrangement and role of TCR-ß expression during CD3-CD4-CD8- thymocyte differentiation. J. Immunol. 152:4783.[Abstract]
-
Dudley, E. C., M. Girardi, M. J. Owen, A. C. Hayday. 1995. [Alpha]-ß and - T cells can share a late common precursor. Curr. Biol. 5:659.[Medline]
-
Livák, F., H. T. Petrie, I. N. Crispe, D. G. Schatz. 1995. In-frame TCR gene rearrangements play a critical role in the ß/ T cell lineage decision. Immunity 2:617.[Medline]
-
Hoffman, E. S., L. Passoni, T. Crompton, T. M. J. Leu, D. G. Schatz, A. Koff, M. J. Owen, A. C. Hayday. 1996. Productive T cell receptor ß chain gene rearrangement: coincident regulation of cell cycle and clonality during development in vivo. Genes Dev. 10:948.[Abstract/Free Full Text]
-
Dudley, E. C., H. T. Petrie, L. M. Shah, M. J. Owen, A. C. Hayday. 1994. T cell receptor ß chain gene rearrangement and selection during thymocyte development in adult mice. Immunity 1:83.[Medline]
-
Saint-Ruf, C., K. Ungewiss, M. Groettrup, L. Bruno, H. J. Fehling, H. von Boehmer. 1994. Analysis and expression of a cloned pre-T cell receptor gene. Science 266:1208.[Abstract/Free Full Text]
-
Fehling, H. J., A. Krotkova, C. Saintruf, H. von Boehmer. 1995. Crucial role of the pre-T-cell receptor gene in development of -ß but not - T cells. Nature 375:795.[Medline]
-
Mombaerts, P., A. R. Clarke, M. A. Rudnicki, J. Iacomini, S. Itohara, J. J. Lafaille, L. Wang, Y. Ichikawa, R. Jaenisch, M. L. Hooper, S. Tonegawa. 1992. Mutations in T-cell antigen receptor genes and ß block thymocyte development at different stages. Nature 360:225.[Medline]
-
Shinkai, Y., G. Rathbun, L. Kong-Peng, E. M. Oltz, V. Stewart, M. Mendelsohn, J. Charron, M. Datta, F. Young, A. M. Stall, F. W. Alt. 1992. RAG-2-deficient mice lack mature lymphocytes owing to inability to initiate V(D)J rearrangement. Cell 68:855.[Medline]
-
Mombaerts, P., J. Iacomini, R. S. Johnson, K. Herrup, S. Tonegawa, V. E. Papaioannou. 1992. RAG-1-deficient mice have no mature B and T lymphocytes. Cell 68:869.[Medline]
-
Itohara, S., P. Mombaerts, J. Lafaille, J. Iacomini, A. Nelson, A. R. Clarke, M. L. Hooper, A. Farr, S. Tonegawa. 1993. T cell receptor gene mutant mice: independent generation of ß T cells and programmed rearrangements of TCR genes. Cell 72:337.[Medline]
-
Boismenu, R., M. Rhein, W. H. Fischer, W. L. Havran. 1996. A role for CD81 in early T cell development. Science 271:198.[Abstract]
-
Gravestein, L. A., W. van Ewijk, F. Ossendorp, J. Borst. 1996. CD27 cooperates with the pre-T cell receptor in the regulation of murine T cell development. J. Exp. Med. 184:675.[Abstract/Free Full Text]
-
Freeden, -J, v. U. effry, P. Viera, L. A. Lucian, T. McNeil, S. E. G. Burdach, R. Murray. 1995. Lymphopenia in interleukin (IL)-7 gene-deleted mice identifies IL-7 as a nonredundant cytokine. J. Exp. Med. 181:1519.[Abstract/Free Full Text]
-
Peschon, J. J., P. J. Morrissey, K. H. Grabstein, F. J. Ramsdell, E. Marakovsky, B. C. Gliniak, L. S. Park, S. F. Ziegler, D. E. Williams, C. B. Ware, J. D. Meyer, B. L. Davison. 1994. Early lymphocyte expansion is severely impaired in interleukin 7 receptor-deficient mice. J. Exp. Med. 180:1955.[Abstract/Free Full Text]
-
Rodewald, H. R., M. Ogawa, C. Haller, C. Waskow, J. P. DiSanto. 1997. Pro-thymocyte expansion by c-kit and the common cytokine receptor chain is essential for repertoire formation. Immunity 6:265.[Medline]
-
O’Shea, J. J.. 1997. Jaks, STATs, cytokine signal transduction, and immunoregulation: are we there yet?. Immunity 7:1.[Medline]
-
Ihle, J. N., B. A. Witthuhn, F. W. Quelle, K. Yamamoto, O. Silvennoinen. 1995. Signaling through the hematopoietic cytokine receptors. Annu. Rev. Immunol. 13:369.[Medline]
-
Ivashkiv, L. B.. 1995. Cytokines and STATs: how can signals achieve specificity?. Immunity 3:1.[Medline]
-
Maki, K., S. Sunaga, Y. Komagata, Y. Kodaira, A. Mabuchi, H. Karasuyama, K. Yokomuro, Y. Miyazaki, K. Ikuta. 1996. Interleukin 7 receptor-deficient mice lack T cells. Proc. Natl. Acad. Sci. USA 93:7172.[Abstract/Free Full Text]
-
Cao, X., E. W. Shores, J. Hu-Li, M. R. Anver, B. L. Kelsall, S. M. Russell, J. Drago, M. Noguchi, A. Grinberg, E. T. Bloom, et al 1995. Defective lymphoid development in mice lacking expression of the common cytokine receptor chain. Immunity 2:223.[Medline]
-
DiSanto, J. P., W. Muller, D. Guy-Grand, A. Fischer, K. Rajewsky. 1995. Lymphoid development in mice with a targeted deletion of the interleukin 2 receptor chain. Proc. Nati. Acad. Sci. USA 92:377.
Top
Abstract
Introduction
