| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
*
Howard Hughes Medical Institute and Departments of Medicine and Microbiology/Immunology, University of California, San Francisco, CA 94143; and
Genome Science Department, Lawrence Berkeley National Laboratory, Berkeley, CA 94720
 |
Abstract |
|---|
 Top Abstract IntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
|
Introduction |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Although the alleles of a given cytokine may be independently
expressed, Th2 cells and lines frequently coexpress IL-4, IL-13, and
IL-5 cytokines distinguished by their genomic proximity
(8). These three cytokines are clustered within an
120-kb interval on mouse chromosome 11 and the syntenic region in
humans on chromosome 5q31 (9). Together, the expression
pattern and genomic organization suggest a comparative study of the
sequences within these regions that are conserved across species might
elucidate the mechanisms that coordinately regulate these genes. To
test this hypothesis, we used mice containing human 5q31 yeast
artificial chromosome
(YAC)3
(3) transgenes that included the cytokine cluster to
examine the expression of these human genes in mouse T cells
(9). Remarkably, these human ILs were expressed faithfully
in CD4+ T cells in vitro and in vivo. These data
support the existence of conserved regulatory elements near the
cytokine cluster itself that enables the activation and/or stable
expression of the type 2 cytokine genes in a cell- and lineage-specific
manner.
|
Materials and Methods |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Human YAC containing 350 kb (854G6-F1) and 450 kb (A94G6) of chromosome 5q31 were used to create transgenic mice by microinjection of fertile FVB/N pronuclei using standard techniques (9). Both human YAC transgenes contain the human IL-4 and IL-13 genes; the A94G6 transgene also contains human IL-5. Founder animals were bred and screened using described primers (9). Transgenic mice were backcrossed five times to the BALB/c background and maintained at the University of California, San Francisco, specific pathogen-free barrier facility before use.
Lymphocyte purification and analysis
Thymus, spleen, and lymph nodes were collected from designated
mice and dispersed through a 70-µm nylon mesh to produce single-cell
suspensions. After hypotonic lysis of RBC, the remaining cells were
labeled with conjugated mAbs FITC-anti-CD4 (CT-CD4; Caltag, South
San Francisco, CA) and TriColor-anti-CD8 (CT-CD8
; Caltag), and,
for the spleen and lymph node cells, additionally with
FITC-anti-CD19 (1D3; PharMingen, San Diego, CA) and PE-NK1.1
(PK136; PharMingen). Cells were analyzed by forward scatter, side
scatter, and fluorescence phenotype using flow cytometry (FACScalibur;
Becton Dickinson, Mountain View, CA).
Highly purified, naive CD4+ T cells were obtained
by sorting pooled spleen and lymph node cells for
L-selectinhigh (mAb MEL-14; PharMingen) and
CD4+ cells that had a small, resting profile on
forward-side scatter parameters as described (Mo-Flo MultiLaser flow
cytometer; Cytomation, Fort Collins, CO) (10). Sorted
cells were >98% CD4+,
L-selectinhigh, TCRß+
small cells. Where designated, purified naive cells were preincubated
in PBS with 5 µM carboxyfluorescein diacetate succinimidyl ester
(CFSE; Molecular Probes, Eugene, OR) for 8 min at room temperature and
washed extensively in large volumes of culture medium (RPMI 1640
supplemented with 10% heat-inactivated FCS, 2 mM
L-glutamine, 0.1 mM sodium pyruvate, 50 µM 2-ME, and
antibiotics) before use (11).
Generation of Th1 and Th2 populations in vitro
Spleens from designated mice were dispersed into single-cell populations. After lysis of RBC, samples were depleted of CD8+ and NK1.1+ cells by incubation with anti-CD8 (3.155; American Type Culture Collection, Manassas, VA) and anti-NK1.1 mAb (PharMingen), followed by low-toxicity rabbit and guinea pig complement (Cedarlane Laboratories, Hornby, Ontario, Canada). After purification over Ficoll, the resulting enriched populations contained 45–50% CD4+ T cells with the remaining cells mostly B cells.
Polarized CD4+ T cell lines were generated by incubating 1.5 x 106 CD4-enriched cells in 1 ml culture medium containing 50 U/ml recombinant human IL-2, anti-TCRß (H57.597; 1 µg/ml), and anti-CD28 (37N51.1; 5 µg/ml) mAb in 24-well plates. For Th1 conditions, cells were also incubated with recombinant murine IL-12 (7 ng/ml) and anti-murine IL-4 mAb (11B11; 100 µg/ml), or, for Th2 conditions, with recombinant murine IL-4 (10 ng/ml). After 5 days, cells were washed and purified over Ficoll to remove dead cells and resuspended in fresh culture medium. The resulting Th1 and Th2 cell lines were >93% CD4+ T cells.
Where designated, highly purified naive CD4+ T cells were stimulated using Th2 conditions, as above, but with the addition of irradiated T cell-depleted spleen cells as APC (10), and with or without either 1 µg/ml each neutralizing Abs against human IL-4 and IL-13 (R&D Systems, Minneapolis, MN) or, 100 ng/ml each recombinant human IL-4 and IL-13 (R&D Systems).
Activation of NK T cells in vivo
NK T cells in designated mice were activated using i.v. anti-CD3 mAb as described elsewhere (12). In brief, after determining optimal dose and timing, 1 µg anti-CD3 mAb (2C11) was injected i.v. into groups of mice. The spleens were removed after 90 min and, after lysis of RBC, the numbers of cells that produced murine and human IL-4 were determined using enzyme-linked immunospot (ELISPOT) assays.
Activation of Th2 cells in vivo
Designated mice were inoculated s.c. with 500 third-stage larvae of Nippostrongylus brasiliensis in 0.2 ml PBS (13). After 12 days, the mice were killed and the numbers of adult worms in the intestines were enumerated using inverted microscopy. The lungs and mesenteric lymph nodes were excised and dispersed into single-cell suspensions. After lysis of RBC, cytokine production was quantified as designated below.
Intracellular cytokine determination
Intracellular cytokine analysis was performed as described
previously(14). In brief, Th1 and Th2 cell lines or cell
suspensions prepared from N. brasiliensis-infected mice were
stimulated with PMA (40 ng/ml; Sigma, St. Louis, MO) and ionomycin (2
µg/ml; Sigma). After 2 h, brefeldin A (10 µg/ml; Epicentre
Technologies, Madison, WI) was added to promote the intracellular
accumulation of secreted cytokines. After an additional 2 h, cells
were washed twice in PBS/1% FCS, incubated on ice with
TriColor-anti-CD4 mAb, washed, fixed in 4% formaldehyde for 15 min
and washed twice in PBS/1% FCS. Cells were lightly permeabilized by
incubation at room temperature for 10 min in PBS/0.5% saponin and then
incubated for 30 min with designated anti-cytokine Abs diluted in
PBS/0.5% saponin. Anti-cytokine mAb included FITC- or PE-conjugated
anti-mouse IL-4 (11B11; PharMingen), FITC-conjugated anti-mouse
IFN-
(XMG1.2; PharMingen), PE-conjugated anti-human IL-4
(MP4–25D2; PharMingen), and PE-conjugated anti-human IL-13
(JES10–5A2; PharMingen), or FITC-conjugated and PE-conjugated isotype
control mAb. Cells were washed and suspended in PBS/1% FCS for flow
cytometric analysis.
Cytokine ELISA
Enriched CD4+ T cells were incubated in duplicate cultures of 106 cells on plates precoated with anti-TCRß and anti-CD28 mAb. Supernatants were collected after 48–72 h and assayed by ELISA for murine IL-4 using mAb 11B11 and BVD6-24G2 and for murine IL-13 using mAb 413 (R&D Systems) and polyclonal IgG BAF413 (R&D Systems) as described elsewhere (10). Assays were normalized to recombinant cytokine controls. The limits of detection were 50 pg/ml for IL-4 and IL-13. ELISA for human IL-5 was performed according to the manufacturer’s instructions (R&D Systems).
Cytokine ELISPOT assays
Spleen single-cell suspensions from designated mice were used for cytokine ELISPOT assays (10). Abs for detecting murine IL-4 were as described previously (10), mAb 413 and polyclonal IgG BAF413 (R&D Systems) were used for murine IL-13, and mAb 604 and BAF204 (R&D Systems) were used for human IL-4. After development, individual spots were counted from duplicate determinations using inverted microscopy.
Statistics
When appropriate, Student’s paired t test was used to determine the statistical significance of the differences between groups.
|
Results |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Examination of the transgenic mouse strains revealed no
differences in size or morphology of the thymus, spleen, or lymph
nodes. Analysis of the spleens showed normal numbers of the major
lymphoid populations, including CD19+ B cells,
NK1.1+ NK cells, and CD4+
and CD8+ T lymphocytes (Fig. 1
A). Flow cytometric analysis
of thymi showed normal percentages of CD4+,
CD8+, and CD4+,
CD8+ double-positive cells in the transgenic mice
(Fig. 1B). Serum IgM, IgG, and IgE were not significantly
different among the transgenic mice and negative littermate control
mice (data not shown). Thus, inclusion of the human 5q31 transgene does
not overtly influence development of the major mouse lymphocyte
populations.
|
CD4+ T cells were stimulated in vitro under
conditions that favored the development of IL-4- and IL-13-secreting
Th2 cells or IFN--secreting Th1 cells. Analysis of Th2 cell lines
from nontransgenic littermates and both human 5q31 transgenic mice in
various experiments revealed that 60–90% of such cells expressed
intracellular murine IL-4 after restimulation; no significant
differences were noted among the cell lines derived from the various
strains of mice. Gating on CD4+ T cells revealed
that expression of both human IL-4 and IL-13 occurred only in
CD4+ T cells and almost completely within
CD4+ T cells that expressed mouse IL-4 (Fig. 2A). Although reagents are not
available to discriminate reliably between mouse and human IL-5 using
intracellular detection, human IL-5 could be readily determined by
ELISA of supernatants collected 48 h after stimulation of cells
prepared under Th2 conditions from the A94G6 mice (121 pg/ml human
IL-5), but not the 854G6 transgenic mice which lack the human IL-5
gene. Thus, the human 5q cytokines were faithfully expressed in murine
Th2 cells.
|
-secreting cells (Fig. 2
B); <5% of cells expressed
mouse IL-4 under these conditions (data not shown). Similarly, <5% of
CD4+ T cells expressed human IL-4 (Fig. 2B) or IL-13 (data not shown), and expression was
concentrated within cells that did not express mouse IFN-.
Furthermore, human, like murine, IL-5 was not detected in supernatants
taken from stimulated Th1 transgenic T cells. Thus, the human
cytokines, like their murine counterparts, were expressed under
Th2-specific conditions. Human type 2 cytokines do not modulate murine Th development
IL-4, and, to a lesser extent, IL-13, themselves influence the
differentiation of Th2 cells (8, 15, 16), raising the
possibility that the human cytokines might be contributing to effects
on expression of the mouse cytokines in the transgenic cells. To be
sure that these cytokines do not impede interactions of the murine
cytokines with their endogenous receptors, highly purified naive
CD4+ T cells were stimulated under Th2 conditions
in the absence or presence of human IL-4 and IL-13 or neutralizing Abs
to human IL-4 and IL-13. Using cells labeled with the vital dye CFSE to
mark discrete cell divisions, cytokine expression could be analyzed
kinetically (Fig. 3A). Neither
the addition nor neutralization of the human cytokines had any
consistent effect on the timing or numbers of control or transgenic T
cells that began to express murine IL-4 after stimulation.
|
A). When examined
kinetically at 2 and 3 days after stimulation, cells that expressed
human IL-4 consistently appeared later and in lesser frequency as
compared with cells that expressed murine IL-4 (Fig. 3B);
similar findings occurred with IL-13 (data not shown). With time in
culture, comparable numbers of murine IL-4-expressing cells could be
generated, however (Fig. 2A). These data are consistent with
competition between the murine and human type 2 cytokine cluster for
expression, particularly early after stimulation of naive T cells. When
analyzing cells stained with CFSE, however, dual staining for murine
and human IL-4 was not possible due to bleaching by the bright CFSE
fluorochrome. Expression of human IL-4 in murine NK T cells
Among lymphocyte populations, NK T cells are unique in their
capacity to produce IL-4 rapidly following initial cross-linking of the
TCR (12, 17). This effect is best revealed with systemic
administration of anti-CD3 Ab or superantigens that activate large
numbers of T cells. Despite such widespread activation, essentially all
of the IL-4 produced in the first 3 h is derived from NK T cells
(12). To assess whether the human cytokines were
faithfully expressed in NK T cells, 5q31 transgenic and
littermate-negative mice were treated with optimal doses of
anti-CD3 and the spleen cells were analyzed after 90 min for
expression of murine and human IL-4 using ELISPOT assays (Table I). Under these conditions, human IL-4
was readily detected, although expression was not as robust as murine
IL-4. Thus, the mechanisms regulating IL-4 expression in mouse NK T
cells were sufficient to activate the human gene.
|
Worm infections powerfully drive the differentiation and
activation of Th2 cells in vivo; such cells are important for the
elaboration of IL-4 and IL-13 that are required for expelling
intestinal helminths (18). Human 5q31 transgenic and
littermate-negative control animals were infected with N.
brasiliensis and examined 12 days later for their ability to
control worm infection. Adult worms were expelled from both groups of
mice, and Th2 responses were evident among CD4+ T
cells using intracellular cytokine detection (Fig. 4). In the transgenic
CD4+ T cells, expression of the human cytokines
was readily detected in cells prepared from both the lung and
mesenteric lymph nodes. In each of 11 individually analyzed mice in
three separate experiments, we consistently found fewer cells from the
5q31 transgenic mice expressed murine IL-4 than those cells from
control nontransgenic mice. Supernatants collected from cells
stimulated in vitro revealed human IL-5 in the supernatants taken from
the A94G6 transgenic cells which contained the human gene (lung cells,
31.0 ± 4.0 pg/ml; mesenteric lymph node cells, 15.7 ± 2.1
pg/ml). Human IL-5 was undetectable in supernatants of stimulated cells
from uninfected mice (data not shown). Thus, the human 5q31 cytokines
IL-4, IL-13, and IL-5 were each produced in response to a Th2 stimulus
capable of activating the endogenous murine cytokines in vivo.
|
|
Discussion |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
The coexpression of IL-4, IL-13, and IL-5 is common in Th2 cells, but mechanisms that allow such coordinate regulation remain unknown. DNA-binding factors that are shared among several of these cytokine gene promoters, such as GATA-3 (19, 20) or Stat6 (21, 22, 23), might contribute to their coregulation, although it is clear that promoters and enhancers are not reliably shared by these three genes. For example, c-Maf, critically implicated in IL-4 expression in T lymphocytes, was not essential for regulation of IL-13 which lies only 12.5 kb downstream (24). Although none of these cytokines are expressed by naive T cells, T cell activation is known to result in substantial nuclear reorganization whereby groups of genes become coordinately repositioned into areas accessible to the general transcription machinery (25, 26, 27). It is possible that the 5q31 cytokine gene cluster might be repositioned during cell activation such that expression can occur in the presence of the appropriately available factors, such as c-Maf, GATA-3, NF-ATc, or Stat6.
The experiments presented here suggest that conserved elements within
the human 5q31 YAC were sufficient to confer activation- and
lineage-specific expression of the IL-4, IL-13,
and IL-5 genes in murine T lymphocytes. The two YAC
transgenes share in common the sequences of five complete genes and
overlap in a region from 85 kb centromeric of the IL-4
gene to 100 kb downstream of IL-13 (9). None of
the other three genes in this region, Septin 2, a
cyclin-like gene and KIF3A, are known to be involved in
regulating expression of the cytokine genes, suggesting that activation
of the human cytokines is exerted solely by trans-activating
mouse factors. Within murine lymphocytes, expression of the cytokines
in this cluster is essentially limited to Th2 cells and NK T cells. The
latter are distinguished by their capacity to produce IL-4 rapidly
after TCR engagement, which contrasts with the delay in IL-4 expression
after stimulation of typical naive ß T cells (7, 12, 17). As shown here, elements of the IL-4 locus that
enable it to be expressed rapidly in NK T cells were also functional in
the human locus, suggesting again that genomic sequences conserved
between the species were sufficient to confer lineage-specific
regulation. Despite this, the ratio of NK T cells that expressed human,
as compared with the mouse, IL-4, was substantially less than the ratio
of human- to mouse-expressing cells among conventional ß T cells,
whether assessed in vitro or in vivo. Whether this reflects differences
in the experimental procedures used to measure cytokine production in
the two cell populations or true differences in regulation of the locus
itself will require further study.
Two intriguing aspects of these studies are additionally noteworthy.
First, previous studies of the endogenous IL-4 gene have
demonstrated independent expression among alleles, leading to the
development of both mono- and biallelically expressed clones (1, 2). If the human YAC transgenic mice were adding a third such
allele, we would have expected to find cells that expressed the human
IL-4 allele but neither of the endogenous mouse alleles.
However, such cells were not found (Fig. 2A), suggesting
that sufficient genetic drift may have occurred between the species
such that activation of the human genes occurs only in those cells that
optimally activated the mouse genes, perhaps reflecting the presence of
a limiting factor(s) that binds with greater affinity to murine DNA
recognition sequence(s). Second, early after activation in vitro (Fig. 3) and in vivo (Fig. 4), T cells that contained human transgenes less
efficiently activated the mouse genes. Evidence for competition after
activation of the human locus was also suggested by in vitro analysis
of murine and human IL-4 and IL-13 mRNA levels and in vivo using a
murine asthma model (28). It is possible that some
rate-limiting factor is required to optimize stable gene expression of
these cytokines which must compete between conserved murine and human
DNA recognition sites early after T cell activation.
The critical role of the cytokines expressed from the 5q31 locus, in particular IL-4 and IL-13, in mediating protection against intestinal helminths has been well defined (18). Additionally, these same cytokines, as well as IL-5, have been implicated in a number of pathologic processes related to asthma and allergic diseases (28, 29, 30). In murine models of asthma, either IL-4 or IL-13 could provoke most of the cardinal manifestations of asthma pathology, including recruitment of airway eosinophils, hyperplasia of mucus-secreting goblet cells, and induction of airway hyperresponsiveness (31, 32). These cytokines were overexpressed in airway lymphocytes and inflammatory cells of patients with asthma. Genetic studies have established linkage to human 5q31 among various populations with asthma and atopic disease (33, 34), and analysis of these same transgenic mice strongly supported the contributions of the regions near the cytokines themselves in establishing the asthma phenotype (28).
The finding that the human 5q31 transgenic mice faithfully express
these type 2 cytokines in vitro and in vivo suggests that these animals
are important reagents for examining regulation of the locus in a
defined experimental setting. Whether IL-4,
IL-13, and IL-5 are coordinately regulated
through use of common promoter elements or through mechanisms that more
globally regulate access to the entire gene cluster, such as locus
control regions (35), remains an important area of study.
Both mechanisms may apply to regulation of these genes. The two in vivo
assays reported here demonstrated consistently lower expression of the
endogenous murine cytokines in cells that contained the human
transgenes, suggesting the possibility of competition for shared
factors required for stable gene expression. Under optimal conditions,
as shown by the in vitro studies, the development of cells expressing
the murine cytokines was comparable between transgenic and
nontransgenic CD4+ T cells (Fig. 2).
Identification of common sequences responsible for the regulation of
the type 2 cytokine locus will be important in defining mechanisms that
underlie the stereotyped expression of these cytokines during allergic
inflammatory disease.
|
Acknowledgments |
|---|
|
Footnotes |
|---|
2 Address correspondence and reprint requests to Dr. Richard M. Locksley, University of California, San Francisco, Box 0654, C-443, 521 Parnassus Avenue, San Francisco, CA 94143-0654.
3 Abbreviations used in this paper: YAC, yeast artificial chromosome; CFSE, carboxyfluorescein diacetate succinimidyl ester; ELISPOT, enzyme-linked immunosorbent spot; GATA, transcription factor family recognizing the consensus sequence (A/T)GATA(A/G); c-Maf, cellular homologue of the avian viral oncogene v-maf; KIF3A, kinesin superfamily proteins.
Received for publication December 3, 1999. Accepted for publication February 23, 2000.
|
References |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
derived from CD4+ T cells is sufficient to mediate T helper cell type 1 development. J. Exp. Med. 188:1651., and Stat6 are required for the expulsion of the gastrointestinal nematode parasite Nippostrongylus brasiliensis. Immunity 8:255.[Medline]
This article has been cited by other articles:
|
|
 |
|
  M. Mas, P. Cavailles, C. Colacios, J.-F. Subra, D. Lagrange, M. Calise, M.-O. Christen, P. Druet, L. Pelletier, D. Gauguier, et al. Studies of Congenic Lines in the Brown Norway Rat Model of Th2-Mediated Immunopathological Disorders Show That the Aurothiopropanol Sulfonate-Induced Immunological Disorder (Aiid3) Locus on Chromosome 9 Plays a Major Role Compared to Aiid2 on Chromosome 10 J. Immunol., May 15, 2004; 172(10): 6354 - 6361. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 D. F. Rogan, D. J. Cousins, S. Santangelo, P. A. Ioannou, M. Antoniou, T. H. Lee, and D. Z. Staynov Analysis of intergenic transcription in the human IL-4/IL-13 gene cluster PNAS, February 24, 2004; 101(8): 2446 - 2451. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 M. A. Nobrega and L. A. Pennacchio Comparative genomic analysis as a tool for biological discovery J. Physiol., January 1, 2004; 554(1): 31 - 39. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 C. P. Leo, S. Y. Hsu, and A. J. W. Hsueh Hormonal Genomics Endocr. Rev., June 1, 2002; 23(3): 369 - 381. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 W. O. C. Cookson Asthma Genetics Chest, March 1, 2002; 121(2007): 7S - 13S. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 W. O.C. Cookson and M. F. Moffatt Genetics of asthma and allergic disease Hum. Mol. Genet., October 1, 2000; 9(16): 2359 - 2364. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 G. G. Loots, I. Ovcharenko, L. Pachter, I. Dubchak, and E. M. Rubin rVista for Comparative Sequence-Based Discovery of Functional Transcription Factor Binding Sites Genome Res., May 1, 2002; 12(5): 832 - 839. [Abstract] [Full Text] [PDF]
|
|
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
