| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Laboratory of Immunology, National Institute of Allergy and Infectious Diseases, National Institutes of Health, Bethesda, MD 20892
 |
Abstract |
|---|
 Top Abstract IntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
557. Thus,
this tyrosine-based sequence can function as a mobile Stat6 activation
cassette. However, mutants bearing this sequence induced CD23
expression much less efficiently than did wild-type IL-4R, requiring
150-fold more IL-4 to reach maximal CD23 expression. Our results
indicate the importance of juxtatyrosine residues in IL-4R signaling
and argue for an essential role of extended domain structure in the
recognition and function of juxtatyrosine
sequences. |
Introduction |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
-chain, which binds IL-4 with high
affinity (2, 3, 4), and the 
c chain (5, 6). The latter is shared by
receptors for IL-2, IL-7, IL-9, and IL-15. IL-4 binding to the IL-4R
-chain is believed to cause the association of the c chain with
the IL-4/IL-4R -chain complex and to be followed by the activation
of the receptor-associated kinases, Janus kinase-1
(JAK-1)4 and JAK-3. These
kinases are thought to then induce the phosphorylation of critical
tyrosines in the IL-4R -chain and the transcriptional activator
Stat6, with resulting IL-4-induced gene expression (reviewed in 7 . Indeed, mice in which genes for the IL-4R -chain (8), the c
chain (5), or JAK-3 (9, 10, 11) have been disrupted show major impairment
of IL-4 function as do cell lines in which JAK-1 has been mutated (12, 13).
We previously reported that distinct regions of the cytosolic portion
of the IL-4R -chain were principally responsible for IL-4-induced
growth and differentiation (Fig. 1
A) (14, 15, 16). We found that
the region of the IL-4R -chain between amino acids (aa) 437 and 557
was critical for growth control and that a sequence within this region,
centering on Y497 (Y1), played a central role, acting as a docking site
for phosphotyrosine-binding (PTB) domain-containing proteins (15). Y1
is surrounded by an extended sequence that is homologous to sequences
in the insulin and IGF-1 receptors; this sequence has been designated
the insulin/IL-4 receptor (I4R) motif (15). Mutations at Y1 eliminate
the capacity of the receptor to cause the phosphorylation of a critical
substrate, IRS-1. Such receptors generally fail to signal cell growth.
A chimeric receptor consisting of a truncated IL-2R ß-chain and aa
439 to 555 of the IL-4R -chain conveyed to transfected 32D cells the
capacity to take up [3H]thymidine in response to IL-2,
establishing the role of this region in cytokine-induced growth (16).
|
-chain between aa 557 and 657 was shown by
deletion mutations to play an important role in IL-4-induced activation
of Stat6 DNA-binding activity and the expression of a series of genes,
including CD23, IgC
(I), and I-Ad (14). A truncation
mutation terminating at aa 657 (657) had full gene activation
competence, while a 557 truncation mutant had only very weak (but
detectable) capacity to induce CD23, a key IL-4-induced gene (17).
Furthermore, a chimeric receptor consisting of a truncated IL-2R
ß-chain and aa 558–657 of the IL-4R -chain signaled induction of
CD23 in response to IL-2, but did not mediate IL-2-dependent
growth (16). Thus, aa 558–657 of the human IL-4R contains what we have
termed a "gene regulation domain."
The gene regulation domain contains three conserved Y residues (Y2, Y3,
and Y4) (Fig. 1
A). By introducing Y-to-F mutations in the
full length IL-4R, we found that any two of these Y residues could be
mutated without affecting gene expression, as long as these receptors
were expressed at high density (14). However, the loss of all three Y
residues (Y2,3,4F mutation) resulted in a receptor that
failed to signal detectable gene expression or activation of Stat6
DNA-binding activity (Ref. 14; and data not shown). Thus, a receptor in
which any one of these three Y residues is unmutated is capable
of signaling gene activation.
The correlation of decreased Stat6 DNA-binding activity with diminished gene expression in the truncation and point mutants tested implied that Stat6 function was required for expression of the genes in question, which has since been shown to be the case. IL-4-induced gene expression functions are lost in cells from mice in which the Stat6 gene has been disrupted (18, 19, 20). Lymphocytes from these mice can still display IL-4-mediated growth, although to an attenuated degree.
Having confirmed the role of the gene expression domain Y residues in
Stat6 activation and gene expression, we were struck by another feature
of this domain. Each of the three Ys was equidistantly spaced, 27 aa,
from one another. Furthermore, while there was little evolutionary
sequence conservation in the domain as a whole, the sequences
immediately surrounding each Y residue (the "juxtatyrosine"
sequence) were homologous and were evolutionarily conserved (Fig. 1B). A consensus sequence of GYK/QXF was found surrounding
Y2, Y3, and Y4. Interestingly, particularly in view of the weak gene
activation activity of the 557 truncation mutant, Y1 is embedded in
a related sequence, AYRXF. The observed sequence similarities may be
critical, as a mutation in the Y2 juxtatyrosine sequence (polymorphic
form of GYREF) has recently been linked to atopic disease
(21).
Here, we have examined the capacity of the consensus sequence
from the gene regulation domain to induce Stat6-dependent gene
expression, by preparing 557 mutants in which the sequence
surrounding Y1 has been replaced with test sequences. We show that the
extended sequence derived from Y3 (EAGYKAF) induces a substantial
increase in the capacity of the truncated receptor to induce gene
activation. Thus, this sequence can be regarded as a mobile Stat6
activating module. However, the resultant receptor, although displaying
such activity, is much less efficient than the wild-type receptor,
arguing for an important role of the extended structure of the gene
regulation domain in mediating its functions.
|
Materials and Methods |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
The murine B cell line M12.4.1 was the kind gift of Dr. Richard
Asofsky (National Institute of Allergy and Infectious Diseases,
National Institutes of Health, Bethesda, MD) and was maintained in
cRPMI (RPMI 1640 medium (Life Technologies, Grand Island, NY)
supplemented with 10% FBS, 2 mM L-glutamine, 100 u/ml
penicillin, 100 µg/ml streptomycin, and 1 mM sodium pyruvate, all
from Biofluids (Rockville, MD)). Mouse anti-human IL-4R was
purchased from Genzyme (Cambridge, MA). Horse anti-mouse Ig was
purchased from Vector Laboratories (Burlingame, CA). FITC-conjugated
streptavidin was purchased form R&D Systems (Minneapolis, MN).
FITC-conjugated rat anti-mouse CD23 (clone B3B4) and
FITC-conjugated rat anti-mouse I-Ad (clone AMS-32.1)
were purchased from PharMingen (San Diego, CA). Mouse anti-p85
subunit of phosphoinositol-3' kinase (clone UB93-3) Ab and rabbit
anti-mouse IRS-2 antisera were purchased from Upstate
Biotechnologies (Lake Placid, NY). Rabbit anti-mouse Stat6 was the
kind gift of Dr. James Ihle (St. Jude’s Research Hospital, Memphis,
TN). Mouse IL-4 expressed in the baculovirus system was affinity
purified as described previously (15) such that 1 U = 0.5 pg, as
determined by proliferation of IL-4-dependent CT.4S cells. Human IL-4
was the kind gift of Dr. Melanie Spriggs (Immunex, Seattle, WA).
FACS analysis
For IL-4R analysis, cells were washed once in FACS buffer
(PBS/3% FCS/0.1% sodium azide), and resuspended in 10 µl of FACS
buffer in 96-well V-bottom plates. Cells were incubated with 10 µl of
normal horse serum diluted 1:5 for 10 min at room temperature, followed
by mouse anti-human IL-4R at a final concentration of 10 µg/ml
for 30 min at 0°C. Cells were washed in FACS buffer, resuspended in
10 µl buffer, and incubated 30 min at 0°C with horse anti-mouse
Ig at a concentration of 10 µg/ml. Cells were washed in FACS buffer,
resuspended in 10 µl buffer, and incubated 30 min at 0°C with
FITC-streptavidin at a concentration of 10 µg/ml. Cells were washed,
resuspended in 200 µl FACS buffer, and analyzed in the presence of
propidium iodide with a FACScan (Becton Dickinson, San Jose, CA). For
assessment of CD23 or I-Ad expression, cells were first
incubated with 3 µl of 2.4G2 rat anti-mouse FcRII/III ascites
for 10 min at 4°C, followed by 10 µg/ml FITC-conjugated rat
anti-mouse CD23 or rat anti-mouse I-Ad for 30 min
at 4°C in FACS buffer. Cells were then washed twice and analyzed in
the presence of propidium iodide.
To determine the percentage of mouse IL-4 responsiveness in CD23 assays, FACS histograms were analyzed by setting a marker with a left-side boundary to include no more than 6% CD23-positive cells in the unstimulated control population. The fraction of mouse IL-4 responsiveness was then calculated by using the marker to determine the percentage of positive cells in each sample: % human IL-4-stimulated CD23+ cells - % unstimulated CD23+ cells/% mouse IL-4-stimulated CD23+ cells - % unstimulated CD23+ cells.
cDNAs/expression vectors
Wild-type and mutated human IL-4RA cDNAs were cloned into the vector pREP9 (Invitrogen, San Diego, CA).
Site-directed mutagenesis
Mutagenesis of specific residues in the human IL-4R was
conducted using the Altered Sites II in vitro mutagenesis system
(Promega, Madison, WI), according to the manufacturer’s
specifications. Oligonucleotides used for mutagenesis are indicated,
with mutated nucleotides in italics: 557.I4R-HIR, 5'-GTC ATC GCA GGC
AAC CCT GAA TAC CTC AGC GCC AGC AAC
TCC CTG AGC CAG; 557.Y1F, 5'-GGC AAC CCT GCT TTC CGC AGC
TTC AGC; 557.EA, 5'-CTC GTC ATC GCA GGC GAG
GCT GCT TAC CGC AGC T; 557.GYKAF, 5'-ATC GCA GGC AAC CCT
GGT TAC AAA GCC TTC AGC AAC TCC CTG;
557.EAGYKAF, 5'-CCC CTC GTC ATC GCA GGC GAG
GCT GGT TAC AAA GCC TTC AGC AAC TCC CTG AGC;
579, 5'-GGC TAT CAG GAG TTT GTA TAA CGC GTG
GAG CAG; Y2F (Y575 to F), 5'- CCC ACC AGT GGC TTT CAG GAG
TTT GTA; Y3F (Y603 to F), 5'- GGA GAG GCT GGT TTC AAG GCC
TTC TCA; Y4F (Y631 to F), 5'- GGG GAA GAG GGG TTT AAG CCT
TTC CAA.
Creation of stable transfectants
The mutants used in this study are shown in Figure 2. M12.4.1 cells (5 x
106) were transfected with 20 µg of uncut plasmid DNA
with a Genepulser electroporation device (Bio-Rad, Melville, NY) using
960 µFD and 200V. Cells were grown in 10 ml cRPMI overnight, then
selected for resistance to neomycin (G418; Life Technologies) at 800
µg/ml for 12 to 21 days. Colonies were screened by Ab staining and/or
125I-labeled IL-4 (125I-IL-4) binding.
|
To detect surface expression of human IL-4R, cells (1 x
107/ml) were incubated in HBSS/20 mM HEPES/2% FBS with 20
ng/ml 125I-IL-4 (kindly provided by Dr. Jacalyn Pierce,
National Cancer Institute, Bethesda, MD) for 60 min at 0°C.
Cells (5 x 106) were then layered on 250 µl of
phthalate oil solution (1x dioctyl phthalate and 1.5x dibutyl
phthalate, from Aldrich Chemical, Milwaukee, WI, and Sigma Chemical,
St. Louis. MO, respectively) in 0.4 ml microcentrifuge tubes (Fisher
Scientific, Pittsburgh, PA), and centrifuged 10 min at 15,000 rpm. Tube
bottoms were cut, and radioactivity in pellets was measured using a
gamma counter (Beckman Instruments, Fullerton, CA). Specifically bound
125I-IL-4 and receptor numbers per cell were determined by
subtracting binding to mock transfected cells.
Immunoprecipitation studies
For immunoprecipitation of the p85 subunit of phosphoinositol-3'-kinase (PI-3K), 2 x 108 cells were stimulated for 10 min at room temperature with 10,000 u/ml mouse IL-4 or 20 ng/ml human IL-4. Cells were subjected to immunoprecipitation with 5 µl mouse anti-p85 subunit of PI-3K followed by electrophoresis and Western blotting with 1 µg/ml rabbit anti-mouse IRS-2 essentially as described (15) and developed using enhanced chemiluminescence (Rockford, IL). The blot was then stripped by incubation in a solution of 0.1 M 2-ME, 2% SDS, and .072 M Tris, pH 6.8, and reprobed with a 1:1000 dilution of mouse anti-p85 subunit of PI-3K.
Induction of CD23 and I-Ad
Cells (1 x 105/ml) were incubated for 48 h in the presence of 500 U/ml mouse IL-4 or 1 ng/ml human IL-4, followed by Ab staining and FACS analysis, as described above.
Induction of I mRNA
Cells (1 x 105/ml) were cultured for 48 h
in the presence of LPS (10 µg/ml) (Difco, Detroit, MI) 1000 µ/ml
mouse IL-4 or 10 ng/ml human IL-4, as indicated. RNA was harvested
using RNAzol (Tel-Test, Friendswood, TX) according to the
manufacturer’s specifications. Total RNA (10 µg) was electrophoresed
on a 1% formaldehyde agarose gel and transferred to supported
nitrocellulose (Schleicher and Schuell, Keene, NH). Blots were
probed with a 32P-labeled cDNA containing the
CH1 and CH2 elements.
Electrophoretic mobility shift assay (EMSA)
Cells (5–10 x 106) were stimulated with
IL-4 (10,000 U/ml mouse IL-4 or 20 ng/ml human IL-4) for 10 min at room
temperature and washed with 10 ml PBS. Cell pellets were resuspended in
an equal volume of lysis buffer (0.5% Nonidet P-40, 50 mM Tris, pH
8.0, 10% glycerol, 100 µM EDTA, pH 8.0, 50 mM NaF, 150 mM NaCl, 100
µM Na3VO4, 1 mM DTT, 400 µM PMSF, and 1
µg/ml pepstatin A, leupeptin, and aprotinin), and incubated on ice
for 60 min. Lysates were cleared by centrifugation at 15,000 rpm
for 15 min at 4°C, and supernatants were harvested, resuspended to a
final concentration of 5 µg/µl, and stored at -70°C. For assays,
5 to 25 µg of cell lysate was incubated with 100 ng
32P-labeled oligonucleotide in reaction buffer (40 mM KCl,
1 mM MgCl2, 0.1 mM EDTA, 0.5 mM DTT, 20 mM HEPES, pH 7.9,
6% glycerol, 1 mg/ml BSA, 0.1 mg/ml poly(dI:dC)) for 15 min room
temperature. Reactants were loaded (without loading dye) onto a 6%
polyacrylamide, 0.22x TBE gel, which had been prerun at 200 V for 60
min and electrophoresed at 300 V for 2 h. Gels were dried and
exposed directly to film. For these studies, we used a double-stranded
oligonucleotide corresponding to a -activated site (GAS)-like
element found in the mouse IL-4 promoter at position -162 to -152:
gatcAAGACC-162TTCACAGGAA-152CTTTAATCT
gatc, which has been shown to bind Stat6 specifically (22), or a
GAS-like element found in the human CD23 promoter: 5'
gatcAAGACC-227TTTCTAAGAA-217CTTTAATCTgatc,
which binds multiple STAT proteins. For studies using the CD23
oligonucleotide, lysates were confirmed to contain Stat6 by supershift
analysis with anti-Stat6 Abs. Oligonucleotides were synthesized
with a 5'-ACTG overhang on each end (denoted by lowercase letters) and
labeled using Klenow DNA polymerase and [32P]dCTP by
standard techniques.
Yeast two hybrid system
The yeast strain EGY48 (trp1, ura3, his3, leu2; provided by the laboratory of Roger Brent, Harvard Medical School) and yeast two-hybrid system methodology has been previously described (23, 24, 25). The insulin receptor ß bait containing the kinase domain (pLexA.IR.cyto) and the IRS-1 (pAcid.IRS-1.45-516) yeast expression vectors (provided by Thomas Gustafson, University of Maryland at Baltimore) have also been described (26). A human IL-4R/insulin receptor kinase composite bait containing wild-type or mutant forms of the human IL-4R I4R motif was constructed by inserting a double-stranded oligonucleotide into the EcoRI and internal XhoI site of the pLexA.IR.cyto plasmid (coding strand oligo: 5'-AATTGAGGCAGCCAGATGGGCCGCTCGAGCCGCTTGTCATCGCAGGCAACCCTGCTT ATCGCAGTTTCTCGAATTCGTTTCCATGCTCTGTGTACGTGCCGGACGAGTGGGAGGTGAGTCGA), replacing the insulin receptor I4R motif and destroying the 5' EcoRI and 3' XhoI sites. Underlined bases were those changed to create internal XhoI and EcoRI sites with minimal changes in amino acid sequence outside of the I4R motif. These internal EcoRI and XhoI sites were then used to insert 54-bp double-stranded mutant oligonucleotides, which encoded amino acid changes in the I4R motif. Point mutations were verified by sequence analysis. The effect of amino acid changes in the I4R motif on IRS-1 interaction in the yeast two-hybrid system was determined by assaying for ß-galactosidase expression as described (26).
|
Results |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Stable transfectants of M12 cells expressing 557 weakly induce
CD23 and I in response to IL-4. Although the single Y in this
truncated receptor (Y1) is part of the I4R motif also found in the
insulin and IGF-1 receptors, M12.4.1 transfectants expressing insulin
or IGF-1 receptors fail to induce either CD23 or I in response to
insulin or IGF-1 (14). This suggested that the precise sequence
surrounding Y1 in the IL-4R -chain had weak gene activation
activity, while the comparable sequence in the insulin or IGF-1
receptors did not. To test this, we prepared a site-directed 557
mutant (557.I4R-HIR) (see Fig. 2 for a list of all receptor mutants
used in this study) in which the sequence EYLSA from the human insulin
receptor I4R motif was substituted for the AYRSF of IL-4R I4R motif
(Figs. 1B and 3).
A stable transfectant of M12 expressing wild-type 557 (557.WT)
was weakly active, inducing, at 10 ng/ml of human IL-4, CD23 expression
on 25% as many cells as did mouse IL-4 acting on the endogenous IL-4R.
Not only did human IL-4 cause fewer cells to become CD23-positive, but
these cells, on the average, expressed less CD23. 557.I4R-HIR
receptors were significantly less active in this assay. The importance
of Y1 was most clearly demonstrated by mutating Y1 to F, creating a
receptor (557.Y1F) that failed to induce CD23.
These data indicate that the minimal gene expression activities
exerted by the IL-4R truncation mutant 557 are completely
dependent upon Y1 and that residues surrounding Y1 contribute to gene
expression. We reasoned that this site would be valuable for testing
the gene expression activity of sequences, derived from other parts of
the receptor, that had been inferred to have potent gene activation
capacity.
The IL-4R -chain contains a mobile Stat6 activation sequence
Mutation studies have shown that Y2, Y3, and Y4 are
important in gene activation (14). To determine whether the core
homologous sequence in which they are embedded (GYK/QXF) was a
functional Stat6 activation motif, we inserted the version of this
sequence from the Y3 site at Y1 in 557. Hou et al. demonstrated that
15-mer peptides derived from Y3 and Y4 but not Y2 were active in their
assay of the inhibition of Stat6 dimerization (27). Since human Y3 and
Y4 share an E at position -3 (where Y = 0), we tested its
importance in Stat6 activation and gene induction. In addition,
substituting the sequence EA (-3, -2) from Y3 also results in
the replacement of NP at the comparable location in Y1, possibly
important in that it should destroy the PTB domain-binding activity of
this sequence (28).
Three mutant receptors were created: 557, bearing an NP-to-EA
mutation (557.EA); 557, bearing an AYRSF-to-GYKAF mutation
(557.GYKAF); and 557, bearing an NPAYRSF-to-EAGYKAF mutation
(557.EAGYKAF) (Fig. 2). Each of these receptors was stably
transfected into M12 cells and assessed for its ability to activate
Stat6 DNA-binding activity in EMSA experiments. In transfectants
expressing 557.WT, human IL-4 elicited Stat6 DNA-binding activity
that was either undetectable (Fig.
4A) or very weak.
Transfectants expressing 557.EA and 557.GYKAF were no more active
than was 557.WT. By contrast, the 557.EAGYKAF mutant was clearly
active in eliciting a Stat6 response, although much less so than WT4R.
These data indicate that the sequence EAGYKAF can direct Stat6
activation when moved to a new location, arguing that the sequence
surrounding Y3 is a mobile Stat6 activation module. Whether EAGYKAF
constitutes the essential Y3 Stat6 activation sequence or its
superiority over the core GYKXF is accounted for because it replaces
the NP in the wild-type receptor remains to be determined. However, the
lack of sequence conservation at positions -3 and -2 (Fig. 1B) supports the latter possibility.
EAGYKAF can enhance IL-4-induced gene expression
To assess the ability of the Y3 juxtatyrosine residues to enhance
IL-4-induced gene expression events, we tested each 557 mutant for
activity in the CD23 expression assay (Fig. 4B), using 10
ng/ml of human IL-4. 557.EA and 557.GYKAF were no more active in
this assay than 557.WT. 557.EAGYKAF caused a substantially
greater degree of CD23 expression than did 557.WT. This result
indicated that the juxtaY3 sequence could not only direct Stat6
DNA-binding activity, but could also induce expression of an
IL-4-responsive gene.
|
557.EAGYKAF
transfectants for their IL-4-induced expression of I-Ad and
obtained similar findings to those observed for CD23 induction. While
557.WT induces little or no increase in expression of
I-Ad on M12.4.1 transfectants, 557.EAGYKAF receptors
increased I-Ad levels in response to human IL-4 to an
extent similar to WT4R (Fig. 4C). However, because of the
relatively narrow "window" of induction, we did not test the other
mutants for I-Ad up-regulation since it would be unlikely
that partial activation could be detected.
We also assessed induction of Ig class switching in response to IL-4
using the 557 mutants. M12 cells induce the expression of I
transcripts in response to stimulation with LPS and IL-4, a process
that has been demonstrated to be Stat6 dependent (18, 19, 20).
557.EAGYKAF mutants increased I mRNA expression over that
observed with 557.WT, but only modestly (Fig. 4D). Since
switching to IgE expression requires substantially higher
concentrations of IL-4 than does either CD23 or class II induction
(29), the relatively modest enhancement in I mRNA expression by
transfectants expressing 557.EAGYKAF was not unexpected.
Concentration dependence of IL-4 induction of CD23
The CD23 induction assay in M12 cells offers a sensitive measure of Stat6-dependent gene expression. We conducted a concentration-response analysis to more precisely characterize the signaling properties of the various mutant receptors. This analysis revealed a more complex picture of gene expression than was shown by the single high dose assay.
We tested concentrations of human IL-4 between 0.003 and 10 ng/ml
(Fig. 5). In selected experiments,
concentrations as high as 250 ng/ml were used. WT4R induced detectable
CD23 expression at 0.016 ng/ml of human IL-4. A peak response,
comparable to that induced by 0.5 ng/ml of mouse IL-4 (hereafter
referred to as "control response"), was stimulated by 
2 ng/ml of
human IL-4. The concentration required for 50% maximal induction was
0.06 ng/ml. Transfectants expressing 557.EAGYKAF receptors failed to
induce maximal responses at any concentration of human IL-4 tested, up
to 250 ng/ml (data not shown). They required 10 ng/ml of human IL-4 to
induce 50% of control CD23 expression levels, an amount 150 times
that required by transfectants expressing WT4R.
Transfectants expressing 557.WT, 557.EA, and 557.GYKAF failed
to induce responses that reached 50% of control levels (Fig. 5). No
further increases in induction of CD23 expression were observed with
any of the transfectants tested at 250 ng/ml of human IL-4 (data not
shown). Their concentration-response curves were essentially identical,
indicating that no advantage is conferred by substituting either the
core Stat6 sequence alone or by simply replacing the NP from the
juxtaY1 sequence.
|
). The concentrations of human
IL-4 required for 50% induction ranged from 0.15 to 0.45 ng/ml,
indicating that these receptors were considerably more active than
557.EAGYKAF but less active than the WT4R. Truncation of the gene expression domain inhibits Stat6 activation and CD23 expression
The lack of full CD23 or I induction by the 557.EAGYKAF
mutant suggested that the possession of a single functional
Stat6-activation sequence was insufficient for maximal IL-4-induced
gene expression. The much greater activity expressed by Y2,3F, Y3,4F,
and Y2,4F transfectants might be because each of these possessed a
"strong" Stat6 site and a "weak" Stat6 site (i.e., the juxtaY1
sequence). Alternatively, these mutants might retain an overall gene
regulation domain structure that was lost when the module was moved to
a new site. To examine this possibility, we created a mutant of
the IL-4R -chain truncated C terminally to the conserved portion of
the juxtaY2 sequence (579). This mutant should possess a strong and
a weak Stat6 site but has a disrupted gene regulation domain.
Transfectants expressing 579 exhibited diminished IL-4-induced Stat6
DNA-binding activity as compared with WT4R, Y3,4F, or 657 (Fig.
6A). As noted above, 657
possesses each of the three strong Stat6 sites and has full gene
activation capacity. 579 transfectants induced only partial CD23
expression upon stimulation with 10 ng/ml of human IL-4 (Fig. 6B), similar to that induced in 557.EAGYKAF mutants and
substantially less than M12 lines expressing either WT4R, Y3,4F, or
657 (Fig. 6C). We conclude that the presence of a strong
(Y2) and a weak (Y1) Stat6 site in the IL-4R -chain is, by itself,
insufficient for full Stat6 activation.
|
The relatively low level of gene expression activity of
557.EAGYKAF appears to reflect the need to retain the structure of
the gene regulation domain, or possibly the need for a more extended
sequence to construct a fully competent Stat6 site. However, our
studies did not directly assess the effect of replacing the NP at
-3,-2 of the juxtaY1 sequence on the binding of possible
competitive substrates. The related proteins IRS-1 and IRS-2 have been
shown to interact with the insulin receptor I4R sequence NPXY (30, 31).
To determine whether the NPXY of the I4R motif of the IL-4R was
important in binding PTB domain proteins, we studied the requirements
for these interactions using a yeast two-hybrid system and stable
557.EAGYKAF IL-4R transfectants.
Yeast strains expressing a "bait" construct composed of the IL-4R
I4R motif linked to the insulin receptor ß-chain kinase domain and a
"prey" construct containing the full length IRS-1 molecule were
established. Importantly, this construct allows tyrosine
phosphorylation of the IL-4R I4R motif by the insulin receptor kinase
(data not shown). Analysis of reporter gene expression using the yeast
two-hybrid system described by Brent and colleagues (23) revealed a
strong, direct interaction between IRS-1 and the
tyrosine-phosphorylated I4R motif of the IL-4R (Table I). Mutation of the central tyrosine of
the I4R motif (Y497) to F or E eliminated the interaction with IRS-1.
Similarly, mutating the N at the -3 position (N494) to Q or E blocked
the interaction of IRS-1 with the I4R motif. In contrast, mutation of
the P at -2 (P495) to E or A or the S at +2 (S499) to T or R reduced
IRS-1 binding only modestly as judged by reporter gene expression.
Thus, the binding of the PTB domain protein IRS-1 is critically
dependent on the N at -3 as well as on the central Y of the I4R motif
(Y1).
|
557.EAGYKAF, which lacks the N at position -3, to activate
IRS-2, the IRS-1 homologue expressed in hemopoietic cells. IRS-2
undergoes tyrosine phosphorylation upon IL-4 stimulation and
subsequently associates with multiple downstream signaling proteins,
including the regulatory subunit of PI-3K (p85) (32). Abs specific for
the PI-3K p85 will coprecipitate IRS-2 from IL-4-stimulated, but not
unstimulated, cells. As shown in Figure
7, human 557.WT receptors interacting
with human IL-4 induced IRS-2 association with PI-3K to the same extent
as did mouse IL-4 activation of the endogenous mouse IL-4R. However,
human 557.EAGYKAF receptors were substantially less active in this
assay. Therefore, substituting the "full" Y3 sequence for Y1
diminishes association of the competing protein IRS-2. Interestingly,
IRS-2 association with PI-3K was not fully inhibited, which may be
consistent with recent studies (31) indicating that IRS-2 (although not
IRS-1) possesses an alternative means of interacting with the insulin
receptor. Thus, the replacement of NP with EA in 557.EAGYKAF (and
probably in 557.EA) results in a diminished but not absent
activation of IRS-2 and thus may not fully ablate potential substrate
competition at the Y1 site. |
Discussion |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
that
are largely responsible for the control of IL-4-mediated growth and
gene expression, respectively (14). The growth domain, located between
aa 437–557 includes the I4R motif with its PTB domain-binding
sequence, NPXY. This I4R motif is conserved in the receptors for IL-4,
insulin, and IGF-1 (15), and all three of these receptors activate
IRS-1 and IRS-2 (14, 33). Interestingly, the IL-4R truncation mutant
557, lacking the gene expression domain, was able to weakly activate
IL-4-mediated gene expression in stable transfectants, while similar
transfectants expressing receptors for insulin or IGF-1 did not (14).
This led us to examine differences between these three receptors.
We noted that the juxtaY1 sequence, while being conserved at the
residues that define the I4R motif, was different at positions -1, +1,
and +3 (where Y1 is residue 0) in the IL-4R compared with homologous
residues in the insulin and IGF-1 receptors (Fig. 1). Given that STAT
activation has been shown to be dependent upon Y residues and key
surrounding amino acids (34), we anticipated that the juxtaY1
sequence would determine the ability of a truncated IL-4R to
activate Stat6 and the genes that it regulates.
By transferring sequences from the human insulin receptor or
introducing a Y-to-F mutation at Y1 in the 557 mutant of IL-4R,
we show that Y1 and its juxtatyrosine sequence are required for the
CD23 expression elicited by this receptor. While this result indicated
that juxtatyrosine sequences control Stat6 activation, Y1 is a weak
Stat6 activator. To locate strong Stat6 activation sequences, we looked
to the gene expression domain.
Located between aa 557 and 657, the gene regulation domain possesses three conserved Ys (Y2, Y3, Y4) whose juxtatyrosine residues are evolutionarily conserved and homologous to one another. These residues are equidistantly spaced 27 aa from one another; this spacing is conserved among the human, mouse, and rat IL-4R, while other Y spacing is not (35). All three Ys are embedded in the sequence GYK/QXF. Interestingly, the weak Stat6 activator Y1 is surrounded by the similar sequence AYRSF. This preservation of the Y+3 residue in all STAT-activating sequences may be particularly important, as this location has been shown to play an important role in SH2 domain binding (36), the mechanism for STAT association with receptors.
Wang et al. (16) demonstrated that the entire gene regulation domain could be moved to a truncated IL-2Rß. In response to IL-2 stimulation, transfectants expressing this receptor activated CD23, but did not proliferate. This important finding both confirmed the function of the domain and showed it was a mobile structure. We wanted to further these results by determining the minimal Stat6 activation cassette in the gene regulation domain.
Hou at al. (27) showed that peptides expressing extended sequences
surrounding Y3 and Y4 inhibited the induced DNA-binding activity of
Stat6 and we demonstrated that all three Ys function in Stat6
activation and gene expression. Thus, we suspected that the difference
in juxtatyrosine sequences between the IL-4R growth and gene regulation
domains would explain the failure of the growth domain to activate
Stat6 and the genes it regulates. Surprisingly, transferring the
conserved juxtaY3 sequence GYKAF to Y1 in 557 had no effect on the
ability of this mutant to activate Stat6.
A possible explanation of thus unexpected result could be that the NPXY
sequence within the I4R motif would allow the binding of PTB
domain-containing proteins, including IRS-1 and IRS-2. If these large
proteins were also attempting to "dock" to phosphorylated Y1, this
might inhibit Stat6 binding. Thus, we substituted EA (found at -3 and
-2 in the juxtaY3 sequence) for NP in the juxtaY1 sequence,
effectively destroying the PTB domain recognition site. Indeed, data
from a yeast two-hybrid analysis indicated that mutations at the N
(-3) and P (-2) residues greatly decrease IRS-1 binding to the
IL-4R I4R motif, giving hope that this mutation would block IRS-2
binding to IL-4R in M12 cells. However, 557.EA mutants also
showed no increase in gene expression abilities.
Since neither mutation alone enhanced gene expression, we transferred
the entire sequence to IL-4R, making a 557.EAGYKAF mutant that
possessed the conserved juxtatyrosine sequence, and should lack
IRS-2 activation capacity. In fact, we found that this mutant could
elicit some association of IRS-2 with PI-3K p85, although at reduced
levels compared with the wild-type receptor. This result is in keeping
with recent findings of a second receptor association domain present in
IRS-2 but not IRS-1 (31). How this domain binds the IL-4R remains
unknown, and the more complex question of IRS-2 association and its
effects on Stat6 activation remain to be answered. Nonetheless,
557.EAGYKAF did show enhanced Stat6 DNA binding activity and could
activate all of the genes tested. Thus, the sequence EAGYKAF can serve
as a Stat6 activation sequence and is mobile, at least within the
IL-4R.
Although introducing the extended Y3 sequence allows enhanced
gene expression, 557 receptors expressing this sequence do not
maximally induce such activation. Indeed, the 557.EAGYKAF mutant
only weakly activates I and fails to induce maximal CD23 expression
even at 19 nM IL-4, the highest concentration tested. Transfectants
expressing this receptor require greater than 150-fold more IL-4 to
induce CD23 expression on 50% of test cells than is required by
transfectants expressing wild-type receptor. By contrast, transfectants
expressing full length receptors in which 2 of the 3 Ys within Stat6
sites have been mutated to F achieve close to full activation at <1 nM
IL-4 and require only 3- to 10-fold more IL-4 for induction of 50%
activity than transfectants expressing the wild-type receptor. The
superiority of these point mutants might represent the need to maintain
more extended sequences, either because the Stat6 docking site is
actually larger than that inferred from sequence homology or because
such sequences are necessary to retain three dimensional structure.
Alternatively, the point mutants might be superior to
557.EAGYKAF because they actually express one strong Stat6-binding
site (Y2, Y3, or Y4) and one weak Stat6-binding site (the Y1 site),
whereas 557.EAGYKAF expresses only one site. This does not appear to
be the case. A truncated receptor (579) in which the truncation
occurs just after the end of the conserved sequence in the Y2
Stat6-binding site functions, similar to the 557.EAGYKAF rather than
to the double point mutants, is consistent with the need for more
extended conserved sequences rather than with the need for a strong and
a weak Stat6 site.
In this regard, it is important to point out that other studies of the mobility of receptor elements have generally not examined the relative efficiency of receptors expressing the transplanted elements or the ability of these receptors to activate STAT-specific gene expression. Thus, Gerhartz et al. (37), Stahl et al. (34), and May et al. (38) have each shown that short receptor-derived consensus elements can direct activation of Stat1, Stat3, and Stat5, respectively, but only to the extent of assessing STAT tyrosine phosphorylation or DNA binding. While these are two crucial events in the use of the JAK-STAT pathway, the end effect of gene expression has remained unknown.
We show that an IL-4R-derived consensus motif can confer Stat6
activation to a receptor that normally only weakly activates Stat6, and
more importantly, that this activation results in the expression of
three Stat6-specific genes. Further, our study offers evidence that
Stat6 activation and the resulting gene expression can occur as a
graded event, resulting in partial gene activation that requires a
higher concentration of IL-4 than does the wild-type IL-4R. The
genes affected showed a range of expression, with the more sensitive
CD23 and I-Ad expression elicited to a higher degree than
I. Thus, our study demonstrates that a short consensus element
cannot only direct specific activation of Stat6, but that the degree of
Stat6 activation determines the extent of Stat6-dependent gene
expression.
These studies point to the capacity of individual units of the
receptor to behave as modules, but also indicate that their optimal
function depends upon retention of certain elements of their native
structure. Indeed, it is striking that the three evolutionary variants
of the IL-4R that have been studied show a high degree of
conservation in both the number and the placement of the Stat6-binding
sites within the gene regulation domain. This strongly suggests that
the structural organization of the receptor is key to its efficient
function. Indeed, understanding the significance of the equal spacing
of the Stat6-binding sites in the absence of significant sequence
conservation of the Y2-Y3 and Y3-Y4 interval sequences to one another
or among species may be critical to understanding the normal
functioning and the evolution of the receptor.
It has recently been reported that a polymorphism exists in the juxta-Y2 sequence in humans and that individuals with an R at the +1 position (GYREF) are at much higher risk of atopic disease than individuals with a Q at that position (21). The authors suggest that this reflects a diminished capacity of the R polymorphic form to bind the phosphatase SHP-1 (SRC homology phosphatase-1). An alternative possibility is that the Q polymorphic form is a weaker Stat6 site. An examination of the detailed dose response properties of the Y2F mutant (14) could be very revealing. If the principal function of the Q polymorphic form is as a SHP-1 docking site, then the Y2F mutant might be expected to be more effective as an IL-4R then the wild-type form, since the SHP site would have been eliminated.
|
|
|
Acknowledgments |
|---|
|
Footnotes |
|---|
2 J.J.R. is a Cancer Research Institute/Miriam and Benedict Wolf Fellow.
3 H.H. is a Leukemia Society of America Fellow.
4 Abbreviations used in this paper: JAK, Janus kinase; aa, amino acid; PTB, phosphotyrosine-binding (proteins); IGF, insulin growth factor; I4R, insulin/IL-4 receptor; IRS, insulin receptor substrate; 125I-IL-4, 125I-labeled IL-4; PI-3K, phosphoinositol-3'-kinase; EMSA, electrophoretic mobility shift assay; I, germline C transcript; WT4R, wild-type IL-4R.
Received for publication January 8, 1998. Accepted for publication April 13, 1998.
|
References |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
subunit of the interleukin-4 receptor. New Engl. J. Med. 337:1720.This article has been cited by other articles:
|
|
 |
|
  J. Zamorano, M. D. Rivas, F. Setien, and M. Perez-G Proteolytic Regulation of Activated STAT6 by Calpains J. Immunol., March 1, 2005; 174(5): 2843 - 2848. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
J. Zamorano, M. D. Rivas, A. Garcia-Trinidad, C.-K. Qu, and A. D. Keegan Phosphatidylcholine-Specific Phospholipase C Activity Is Necessary for the Activation of STAT6 J. Immunol., October 15, 2003; 171(8): 4203 - 4209. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
A. L. Mora, L. M. Stephenson, B. Enerson, J. Youn, A. D. Keegan, and M. Boothby New Programming of IL-4 Receptor Signal Transduction in Activated T Cells: Stat6 Induction and Th2 Differentiation Mediated by IL-4R{alpha} Lacking Cytoplasmic Tyrosines J. Immunol., August 15, 2003; 171(4): 1891 - 1900. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
J. Zhu, L. Guo, C. J. Watson, J. Hu-Li, and W. E. Paul Stat6 Is Necessary and Sufficient for IL-4's Role in Th2 Differentiation and Cell Expansion J. Immunol., June 15, 2001; 166(12): 7276 - 7281. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 J. Zhu, H. Huang, L. Guo, T. Stonehouse, C. J. Watson, J. Hu-Li, and W. E. Paul Transient Inhibition of Interleukin 4 Signaling by T Cell Receptor Ligation J. Exp. Med., October 9, 2000; 192(8): 1125 - 1134. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 M. Pesu, K. Takaluoma, S. Aittomaki, A. Lagerstedt, K. Saksela, P. E. Kovanen, and O. Silvennoinen Interleukin-4-induced transcriptional activation by Stat6 involves multiple serine/threonine kinase pathways and serine phosphorylation of Stat6 Blood, January 15, 2000; 95(2): 494 - 502. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
H. Y. Wang, C. P. Shelburne, J. Zamorano, A. E. Kelly, J. J. Ryan, and A. D. Keegan Cutting Edge: Effects of an Allergy-Associated Mutation in the Human IL-4R{alpha} (Q576R) on Human IL-4-Induced Signal Transduction J. Immunol., April 15, 1999; 162(8): 4385 - 4389. [Abstract] [Full Text] [PDF]
|
|
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
