|
|
||||||||
Laboratory of Immunology, National Institute of Allergy and Infectious Diseases, National Institutes of Health, Bethesda, MD 20892
| Abstract |
|---|
|
|
|---|
| Introduction |
|---|
|
|
|---|
-,
CD44high, CD3intermediate
(CD3int),3 CD45RBlow, and
CD62Llow (4, 5, 6). Moreover, NK1.1+ T cells
preferentially use V
14 (7, 8, 9), associated with Vß8, Vß7, or
Vß2, in their TCRs (10, 11). These cells were initially found and
studied in the thymus and spleen (2, 3, 4), but the liver is now known to
be rich in NK1.1+ T cells (4, 5, 12, 13). The positively
selecting ligand for NK1.1+ T cells appears to be the
conserved MHC class I-like molecule CD1 (14, 15, 16), which normally
associates with ß2-microglobulin
(ß2m) on the cell surface
(17). Indeed, surface expression of CD1 is impaired, and the number of
NK1.1+ T cells is undetectably low in the spleen and thymus
of young mice homozygous for a disrupted ß2m gene
(ß2m-KO mice) (18, 19) as well as in the spleen and
thymus of CD1-KO mice (20, 21, 22). Prompt IL-4 production after primary
stimulation of anti-CD3 injection does not occur in either
ß2m-KO mice (23) or CD1-KO mice (20, 21, 22), further
supporting the importance of NK1.1+ T cells in early IL-4
production. A major deficiency of NK1.1+ T cells has also been reported in SJL mice, which display little or no prompt production of IL-4 after anti-CD3 injection (24). Similar to ß2m-KO mice, SJL mice display a marked diminution in IgE production after anti-IgD injection, suggesting that IL-4 production by NK1.1+ T cells plays a major role, directly or indirectly, in producing the IL-4 which is important in switching to this isotype (25). It should be noted, however, that CD1-KO mice do display IgE production in response to anti-IgD, although they have a striking deficit in thymic NK1.1+ T cells (20, 21, 22). The lack of NK1.1+ T cells in SJL mice may be due in part to a large deletion in the Vß genetic region of their TCRs that includes the genes for Vß8 (26). However, SWR mice with a similar Vß deletion do display NK1.1+ T cells and produce IL-4 in response to injection of anti-CD3 (24, 27). Recently, SJL mice have been reported to express cells that lack both NK1.1 and the capacity to produce IL-4 but express other properties of NK1.1+ T cells. The lack of expression of NK1.1 and the lack of IL-4 production have been reported to be independently controlled by distinct recessive genes (27).
Here we report that the defects in expression of IL-4-producing NK1.1+ T cells observed in young adult ß2m-KO and SJL mice are partially repaired in the livers of these mice as they age. We describe the features of the NK1.1+ T cells that appear in these older mice, and show that these cells are capable of producing IL-4 after anti-CD3 injection.
| Material and Methods |
|---|
|
|
|---|
ß2m-KO mice on a C57BL/6 background were bred and maintained in the animal facilities of National Institute of Allergy and Infectious Diseases, National Institutes of Health. C57BL/6 and SJL mice were obtained from the Division of Cancer Treatment, National Cancer Institute (Frederick, MD).
Cell preparation
To obtain hepatic mononuclear cells (MNC), mice were anesthetized with ether and sacrificed. Livers were removed, pressed through a 40-µm nylon cell strainer (Falcon, Lincoln Park, NJ), and suspended in PBS (0.1 M, pH 7.4). After one washing with PBS, MNC were isolated from hepatocytes and hepatic nuclei by Ficoll-Isopaque density gradient centrifugation (density of 1.090). To avoid selective cell loss during the gradient centrifugation, it was important to dilute the liver cell suspensions (i.e., 30 ml per liver) before overlaying the sample on the gradient cushion. MNC collected from the interface were then suspended in MEM. Spleen cells were also collected by the Ficoll-Isopaque method, while thymocytes were obtained by forcing the thymus through a nylon cell strainer.
Fluorescence analysis
For the detection of NK1.1+ T cells, MNC were
stained by FITC-anti-CD3, -anti-CD4, or -anti-CD8 and by
phycoerythrin (PE)-anti-NK1.1 after blocking FcR binding with the
mAb 2.4G2. For further characterization of NK1.1+ T cells,
three-color staining was done. Cy-Chrome-labeled anti-CD3 and
PE-labeled anti-NK1.1 were used in combination with FITC-labeled
anti-IL-2R
, anti-IL-2Rß, anti-Vß8, anti-CD62L,
or anti-CD44. Fluorescence staining of 105 cells was
performed at 4°C in 100 µl of PBS supplemented with 3% FCS and
0.5% NaN3. Fluorescence analysis was conducted using a
FACScan Flow Cytometer (Becton Dickinson, Mountain View, CA).
Conjugated mAbs were purchased from PharMingen (San Diego,
CA).
Cell proliferation assay
NK1.1+CD3+ T cells and NK1.1-CD3+ T cells were freshly sorted from the livers of older ß2m-KO, SJL, and C57BL/6 mice. Cells (104) were mixed with 5 x 104 CD1-expressing L cells or control parental L cells that had been treated with 100 µg/ml mitomycin C for 1 h at 37°C to prevent them from overgrowing the culture. These mixtures were cultured in the presence of 20 U/ml IL-2 in a 96-well U-plate for 5 days. [3H]thymidine uptake during the last 48 h was measured.
Analysis of expression of IL-4
Using the guanidinium method as described (28), total RNA was prepared from the spleen and liver T cells of young and older C57BL/6, ß2m-KO, or SJL mice that had been injected i.v. with 1.33 µg of the anti-CD3 mAb 2C11 or with the same amount of PBS as a control. T cells were enriched from MNC using murine T cell enrichment columns (R&D Systems, Minneapolis, MN). For analysis of expression of IL-4 mRNA in spleen and liver T cells, mRNAs were amplified by a modified standard RT-PCR amplification procedure as described previously (3). Primers specific for murine IL-4 and ß-actin were as follows: IL-4: 5' primer (GAATGTACCAGCAGC) and 3' primer (CTCAGTACTACGAGTAATCCA); ß-actin: 5' primer (GATGACGATATCGCTGCGCTG) and 3' primer (GTACGACCAGAGGCATACAGG). After reverse transcription, samples were amplified for 30 cycles of 1 min at 94°C, 1 min at 55°C, and 1 min at 72°C. After amplification, PCR products were separated by electrophoresis in 8% acrylamide gels and visualized by UV light illumination after ethidium bromide staining.
| Results |
|---|
|
|
|---|
To examine the effect of aging on the number of
NK1.1+ T cells in ß2m-KO mice, we
examined young (4-wk-old) and fully mature (8-mo-old) mice. The number
of NK1.1+CD3+ T cells in the spleens of
8-mo-old ß2m-KO mice was <1%, similar to their
frequency in 4-wk-old mice (Fig. 1
A). However, a
significant number of liver NK1.1+CD3+ T cells
appeared in the 8-mo-old ß2m-KO mice. Indeed, the ratio
of these NK1.1+ T cells to the total number of liver
lymphocytes increased with age, reaching 11% at 8 mo of age (Fig. 1
B). There was a comparable increase in the absolute
number of NK1.1+ T cells; 1-mo-old ß2m-KO
mice had 1.0 ± 0.4 x 104 NK1.1+ T
cells, while 8-mo-old ß2m-KO mice had 4.2 ±
0.8 x 104 NK1.1+ T cells.
|
Surface phenotype of liver NK1.1+ T cells in older ß2m-KO mice
We compared the expression of cell surface markers of liver
NK1.1+ T cells from older ß2m-KO mice
with that of NK1.1+ T cells from C57BL/6 mice. Liver
NK1.1+ T cells obtained from older ß2m-KO
mice were CD3low, CD44high,
IL-2Rß+, and IL-2R
-, which was similar to
that expressed in NK1.1+ T cells from young C57BL/6 mice
(Fig. 2
). Liver NK1.1+ T
cells from ß2m-KO mice preferentially used Vß8 to a
degree that was similar to that displayed in the liver
NK1.1+ T cells of normal mice (Fig. 2
).
|
We also examined the age-dependent appearance of
NK1.1+ T cells in SJL mice. As in ß2m-KO
mice, NK1.1+ T cells that were essentially undetectable in
the spleen and liver of 1-mo-old mice accumulated in the livers of
older SJL mice. The great majority of liver NK1.1+ T cells
of older SJL mice were CD4-CD8-.
NK1.1+ T cells were rare in the spleens (Fig. 3
A) and thymi (data not
shown) of older SJL mice. The phenotype of liver NK1.1+ T
cells in old SJL mice was similar to that observed in other
NK1.1+ T cells. These cells were CD3int,
IL-2R
-, IL-2Rß+, and
CD44bright (Fig. 3
B). Vß8+
T cells were not detected in the liver NK1.1+ T cells of
older SJL mice, which is to be expected, because these mice have a
large deletion in the Vß region of the TCR gene complex that includes
Vß8 (26).
|
To determine whether liver NK1.1+ T cells from
older ß2m-KO and SJL mice respond to CD1, we conducted
cell proliferation assays of sorted NK1.1+ T cells using
both L cells expressing CD1 (L-CD1) and parental L cells as
stimulators. Liver NK1.1+ T cells from older
ß2m-KO and SJL mice responded to L-CD1, but only about
one-third as well as did liver NK1.1+ T cells from C57BL/6
mice (Fig. 4
).
|
An important question is whether the liver
NK1.1+ T cells that appear in older
ß2m-KO mice and SJL mice are functionally similar to the
NK1.1+ T cells of normal mice. One characteristic function
of "normal" NK1.1+ T cells is that they produce IL-4
promptly in response to the injection of anti-CD3 (3). We examined
the induction of IL-4 mRNA in spleen and liver T cells from young and
older ß2m-KO and SJL mice 90 min after anti-CD3
injection and compared it with the response of spleen and liver T cells
from young C57BL/6 mice. Spleen and liver T cells from 5-wk-old normal
mice that had been injected with anti-CD3 promptly expressed IL-4
mRNA, as analyzed by RT-PCR (Fig. 5
). Previous studies have
demonstrated that virtually all the IL-4 mRNA induced by anti-CD3
is due to NK1.1+ T cells (3, 23). IL-4 mRNA was not
detected in spleens or livers of young ß2m-KO or SJL mice
in response to injection of anti-CD3 (Fig. 4
). By contrast, IL-4
mRNA was induced in the livers but not in the spleens of older
ß2m-KO and SJL mice in response to anti-CD3.
|
| Discussion |
|---|
|
|
|---|
It has been previously demonstrated that both ß2m-KO and
SJL mice have a defect in the number of NK1.1+ T cells in
their spleens and thymi (12, 14, 18, 23). This defect of
NK1.1+ T cells also exists in the livers of
ß2m-KO mice (12, 13) and SJL mice (Fig. 4
; 27 . The
marked diminution in the number of NK1.1+ T cells in
ß2m-KO mice can be explained by the absence, or striking
reduction, in the expression of CD1, which is believed to be critical
for positive selection of these cells; the lack of expression of
NK1.1+ by T cells in SJL is genetically linked to the
NKR-P1 gene complex (27). The failure of SJL T cells to promptly
produce IL-4 in response to anti-CD3 challenge is reportedly
controlled by an independent genetic locus (27).
In this study, we found that liver NK1.1+ T cells
accumulate in older ß2m-KO and SJL mice. By 7 to 8 mo of
age, they have reached a frequency of
10% of liver MNC. These cells
are phenotypically similar to the NK1.1+ T cells of normal
mice. They are CD3int, IL-2Rß+,
IL-2R
-, CD44high, and
CD62Llow. The ß2m-KO NK1.1+ T
cells also preferentially use Vß8 in their TCRs. Furthermore, liver T
cells produce IL-4 promptly after injection of anti-CD3, strongly
suggesting that liver NK1.1+ T cells from both
ß2m-KO and SJL mice resemble NK1.1+ T cells
in their cytokine-producing capacity.
The observation that 48% of the NK1.1+ T cells appearing in the liver of old ß2m-KO mice express Vß8, a percent comparable with the percent of Vß8+ cells among C57BL/6 NK1.1+ T cells, is enigmatic, since it had been assumed that the skewed TCR expression on these cells reflected selection by CD1 (4, 5). Similarly, NK1.1+ T cells from older ß2m-KO mice showed reactivity in vitro to L-CD1 but not L cells, implying that they have a degree of specificity for CD1. These results may be explained by the recent report that CD1 has a non-ß2m-associated form as well as ß2m-associated form (17); the former has been reported to be largely expressed in peritoneal B cells (31). If similar expression exists in a compartment in which NK1.1+ T cells undergo selection, it could explain both the reactivity of liver NK1.1+ T cells of ß2m-KO mice with CD1 and their skewed TCR expression.
Indeed, it is interesting to speculate on the possibility that the liver NK1.1+ T cells from ß2m-KO mice may represent a "primitive" cell type, in which positive selection or stimulated expansion may occur not in the thymus but does possibly occur in the peritoneum or liver under the stimulatory influence of B cells expressing non-ß2m-associated CD1.
The liver NK1.1+ T cells found in older ß2m-KO and SJL mice are almost all CD4-CD8-, while NK1.1+ T cells from other mouse strains are composed of both CD4+ and CD4-CD8- T cells. Although no clear functional differences between CD4+ and CD4-CD8-NK1.1+ T cells have been reported, it is possible that selection of CD4+NK1.1+ T cells may have a more stringent dependence upon CD1 expression than that of double-negative NK1.1+ T cells.
| Acknowledgments |
|---|
| Footnotes |
|---|
2 Address correspondence and reprint requests to Dr. William. E. Paul, Laboratory of Immunology, National Institute of Allergy and Infectious Diseases, National Institutes of Health, 10 Center Drive-MSC 1892, Bldg 10, Rm 11N311, Bethesda, MD 20892-1892. ![]()
3 Abbreviations used in this paper: CD3int, CD3intermediate; ß2m, ß2-microglobulin; KO mice, knockout mice; MNC, mononuclear cells; PE, phycoerythrin. ![]()
Received for publication July 3, 1997. Accepted for publication November 20, 1997.
| References |
|---|
|
|
|---|
ß- T cells receptor + CD4-CD8- T cells produce IL-4. J. Immunol. 149:1211.[Abstract]
/ß+ cells: new clues to their origin, specificity, and function. J. Exp. Med. 182:63.
chain is used by a unique subset of major histocompatibility complex class I-specific CD4+ and CD4-CD8- T cells in mice and humans. J. Exp. Med. 180:1097.
14+ T-cell antigen receptor
chain in mice. Proc. Natl. Acad. Sci. USA 88:7518.
14+TCR
chain in NK1.1+ T cell populations. Int. Immunol. 7:1157.
ß-bearing thymocytes which predominantly express a single Vß gene family. Nature 329:251.[Medline]
/ß+ cells in the liver of mice. J. Exp. Med. 166:577.
ßint liver lymphocytes: influence of thymus, ß2-microglobulin and NK1.1 expression. Int. Immunol. 7:1729.This article has been cited by other articles:
![]() |
H. Y. Kim, M. Pichavant, P. Matangkasombut, Y. I. Koh, P. B. Savage, R. H. DeKruyff, and D. T. Umetsu The Development of Airway Hyperreactivity in T-bet-Deficient Mice Requires CD1d-Restricted NKT Cells J. Immunol., March 1, 2009; 182(5): 3252 - 3261. [Abstract] [Full Text] [PDF] |
||||
![]() |
Y. I. Koh, H. Y. Kim, E. H. Meyer, M. Pichavant, O. Akbari, T. Yasumi, P. B. Savage, R. H. DeKruyff, and D. T. Umetsu Activation of Nonclassical CD1d-Restricted NK T Cells Induces Airway Hyperreactivity in {beta}2-Microglobulin-Deficient Mice J. Immunol., October 1, 2008; 181(7): 4560 - 4569. [Abstract] [Full Text] [PDF] |
||||
![]() |
I. Tsunoda, L.-Q. Kuang, and R. S. Fujinami Induction of Autoreactive CD8+ Cytotoxic T Cells during Theiler's Murine Encephalomyelitis Virus Infection: Implications for Autoimmunity J. Virol., November 13, 2002; 76(24): 12834 - 12844. [Abstract] [Full Text] [PDF] |
||||
![]() |
T. Dao, M. Exley, W. Z. Mehal, S. M. A. Tahir, S. Snapper, M. Taniguchi, S. P. Balk, and I. N. Crispe Involvement of CD1 in Peripheral Deletion of T Lymphocytes Is Independent of NK T Cells J. Immunol., March 1, 2001; 166(5): 3090 - 3097. [Abstract] [Full Text] [PDF] |
||||
![]() |
H. S. Kim, J. Garcia, M. Exley, K. W. Johnson, S. P. Balk, and R. S. Blumberg Biochemical Characterization of CD1d Expression in the Absence of beta 2-Microglobulin J. Biol. Chem., April 2, 1999; 274(14): 9289 - 9295. [Abstract] [Full Text] [PDF] |
||||
![]() |
C. Angenieux, J. Salamero, D. Fricker, J.-P. Cazenave, B. Goud, D. Hanau, and H. de la Salle Characterization of CD1e, a Third Type of CD1 Molecule Expressed in Dendritic Cells J. Biol. Chem., November 22, 2000; 275(48): 37757 - 37764. [Abstract] [Full Text] [PDF] |
||||
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |