| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Department of Microbiology and Immunology, University of California, San Francisco, CA 94143
 |
Abstract |
|---|
 Top Abstract IntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
|
Introduction |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Chemokine receptors are coupled to heterotrimeric G proteins, which
consist of 
, ß, and 
subunits. Four families of G subunits
can be distinguished based on their function and amino acid sequence
homology, and are termed Gs, Gi, Gq, and G12. In vitro
studies suggest that Gi family members are essential for mediating
chemoattractant responses (10). For G subunits to
function, they must switch between an inactive GDP-bound and an active
GTP-bound conformation. Essential elements in this GDP/GTP cycling are
positive regulators that accelerate the nucleotide exchange on G and
negative regulators that increase the intrinsic GTPase activity of
G. While activated G protein-coupled receptors (GPCRs) themselves
act as nucleotide exchange factors (GEFs), GTPase-activating proteins
(GAPs) for heterotrimeric G proteins have only recently been recognized
and were named regulators of G protein signaling (RGS) molecules
(11, 12, 13, 14). RGS proteins contain a highly homologous region
of about 120 aa, referred to as the RGS domain, that confers the GAP
activity of the protein by stabilizing the GTP-to-GDP transition state
of G subunits. Most RGS proteins are relatively small, consisting of
200 aa, but an increasing number of larger RGS proteins have been
described that contain additional functional domains such as a PDZ
domain in RGS12, a Dbl homology domain in p115RhoGEF, and a
B-raf homology domain in RGS14 (15). Studies on
the specificity of RGS molecules for different G proteins to date have
indicated that human RGS1, RGS3, RGS4, and RGS16 predominantly interact
with Gi subunits, RGS2 with Gq, RGSZ1 with Gz, and p115RhoGEF
with G12/13. These relationships may not be rigid, however, as some
studies have indicated that RGS1 and RGS3 can inhibit signaling via
Gq-coupled receptors (16, 17, 18), while RGS2 may
antagonize some Gi-coupled receptors (13). Evidence is
accumulating that the specificity of RGS proteins can also be regulated
through interactions with the GPCRs themselves (16, 19, 20). Further work is therefore needed to understand how
individual RGS proteins regulate individual GPCRs.
Lymphoid tissues and cells have been found to express multiple RGS molecules (21). Human RGS1 (BL34/1R20) was first identified through its expression in activated B cells (22, 23), and RGS3 was isolated by screening a B cell cDNA library with an RGS domain probe (13). RGS2 (GOS8) was first isolated as a gene induced by treatment of human blood mononuclear cells with the T cell mitogen, Con A (24). Further work has shown that expression of human RGS1 is induced in B cells by treatment with anti-IgM Abs as well as by IL-4, cAMP, or platelet-activating factor (13, 22, 23). RGS1 is strongly inducible by phorbol esters, whereas RGS2 is induced more strongly by calcium ionophore (25). RGS14 is also highly expressed in spleen as well as in brain and lung (26). Recently, it has also been reported that RGS16 is expressed in some lymphoid cell types (27).
To explore whether RGS molecules might function downstream of B cell surface receptors in regulating B cell responsiveness to lymphoid chemokines, we have characterized the expression pattern and activity of several mouse RGS proteins. We report the sequence of mouse RGS1 and describe a novel short isoform of RGS3, RGS3s, present in both mice and humans. Mouse RGS1 and RGS3 are expressed at high levels in spleen and lymph nodes. The expression patterns of mouse RGS2 and RGS14 are also examined, and RGS1, RGS2, RGS3, and RGS14 transcript levels are shown to be rapidly modulated in B lymphocytes in response to Ag challenge in vivo. Using cell lines transfected with RGS1-, RGS2-, and RGS3-GFP fusion proteins, we demonstrate that RGS1 and RGS3 can attenuate migration to the lymphoid chemokines BLC, ELC, and SDF-1. These findings suggest an important role for RGS molecules in helping regulate cell positioning in lymphoid organs during the immune response.
|
Materials and Methods |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Mouse rELC (macrophage-inflammatory protein-3ß) was purchased
from R&D Systems (Minneapolis, MN). Mouse rBLC was isolated as a
HIS-tagged protein, as described (28). Human SDF-1
(N33A) was synthesized by chemical ligation and was a gift from M.
Siani (Gryphon Sciences, South San Francisco, CA). Note human and mouse
SDF-1 only differ in 1 aa, and SDF-1 (N33A) has identical activity
to native human and mouse SDF-1. The anti-CD40 mAb (clone FGK)
(29) was a gift from Antonius Rolink (Basel Institute for
Immunology, Basel, Switzerland). Hen egg lysozyme (HEL) was purchased
from Sigma (St. Louis, MO).
Mice and animal challenge
C57BL/6 MD4 Ig transgenic (IgHEL) mice carry transgenes encoding IgMa and IgDa heavy and light chains specific for HEL (30). C57BL/6 HEL transgenic mice were of the ML5 line, which carries a transgene encoding HEL under the metallothionine promotor and contains HEL at 10–30 ng/ml in serum (30). C57BL/6 MD4 and ML5 transgenic mice were mated to obtain double-transgenic (IgHEL/HEL) mice. Transgenic mice were screened by tail bleeding, followed by flow cytometry analysis to assess the Ig or double-transgenic nature of B cells, as previously described (31). C57BL/6 mice were obtained from Jackson ImmunoResearch (West Grove, PA). For activation experiments, 6–10-wk-old IgHEL or nontransgenic mice were injected i.p. with 4.5 mg HEL, 200 µg anti-CD40 mAb, or 200 µg control rat mAb in RPMI 1640 medium.
Clone identification and sequence analysis
Advanced BLAST searches of the NCBI mouse expressed sequence tag (EST) database using TBLASTN (32) with the respective template, retrieved mouse ESTs for RGS1, RGS2, RGS3s, RGS3p, and RGS14, as indicated in the result section. The indicated mouse EST clones were obtained from Genome Systems (St. Louis, MO) as EcoRI-NotI inserts in the pT7T3-Pac vector, and sequenced. Similarity scores were calculated using Clustalw with the Blosum matrix. To obtain a probe for detecting RGS14 expression, the database was searched with the mouse cDNA sequence that had previously been deposited in GenBank (Accession U85055). One EST clone was identified (AA981480) containing the 3'-untranslated region and sequence encoding the C-terminal 5 aa of mouse RGS14.
Expression of murine RGS proteins
GFP-RGS fusion proteins were generated by cloning cDNA fragments encoding the open reading frame of RGS1, RGS2, or RGS3 into the pEGFP-C1 plasmid (Clontech, Palo Alto, CA). To adapt the respective cDNA in frame to the sequence encoding EGFP, compatible restriction sites were introduced by PCR at the 5' and 3' end of the coding region in the RGS cDNAs. In addition, a truncated version of RGS1 was designed that retains the N-terminal 60 aa, but lacks the RGS domain. Primers used were as follows (written 5'-3'): RGS1or RGS1T60 sense, CCGGGATCCAGATCTCCAGGAATGTTCTTTTCTGCTAGCCCAAAGG; RGS1 antisense, CGCGGATCCACTAGTCTGTCAGCCAGGAATCACTCACTTTAAAGT; RGS1T60 antisense, GTTCTGCAGTCAAAGTATGTCTTTGGACTTGGCCGATTTC; RGS2 sense, CCGTCTCGAGCTCAAAGTGCCATGTTCCTGGCTGTCC, and antisense, CGCGGATCCGGGGGACTCCTGGTCTCATGTAGC; and RGS3s sense, CCGTCTCGAGCTCTCCGAGGCATGTACCTCACACGCA, and antisense, CGCGGATCCGGATACTGGTGGTCCCTAGAGCGG. Cloning sites used were for RGS1 BglII-BamHI, for RGS1T60 BglII-PstI, and for RGS2 and RGS3s XhoI-BamHI. Obtained expression constructs were verified by sequence analysis.
Cell culture and transient transfection
The murine B cell lines were maintained in RPMI 1640 supplemented with 10% FBS, 2 mM L-glutamine, 50 µg/ml streptomycin, 50 U/ml penicillin, 10 mM HEPES, and 50 µM 2-ME. 2PK3 and 300-19 cells were transfected by electroporation with the following amounts of plasmid DNA: 10 µg pEGFP-C1, 30 µg pEGFP-C1-mRGS1, 40 µg pEGFP-C1-mRGS1T60, 40 µg pEGFP-C1-mRGS2, and 30 µg pEGFP-C1-mRGS3s. Cells (1–1.5 x 107 in 0.4 ml) were pulsed at 250 V and 960 µF. Assays were started 10–14 h after transfection.
Cell purification and flow cytometry
Mouse lymphocyte cell suspensions were obtained from spleen or
lymph node by passing teased tissues through a 70-µM cell strainer
(Falcon; Becton Dickinson, Franklin Lakes, NJ). B cells were purified
from spleen cell suspensions by staining non-B cells with anti-CD43
biotin and anti-CD11c biotin (PharMingen, San Diego, CA) and
depletion using streptavidin-MACS beads and a MACS (Miltenyi Biotec,
Auburn, CA). T cells were isolated from spleen or lymph nodes by
staining non-T cells with biotinylated Abs to B220, Mac-1 (Caltag
Laboratories, Burlingame, CA), CD11c, and Gr-1 (PharMingen), followed
by depletion by MACS. Flow cytometry analysis on a FACScan (Becton
Dickinson, San Jose, CA) confirmed that purity of the isolated B cells
was >92% B220+ cells, and of isolated T cells
>93% CD4+ or CD8+ cells.
When mice were challenged with HEL or anti-CD40 mAb before
isolation of B cells, purified B cells were also tested for the
induction of CD69 using anti-CD69 FITC (PharMingen). For
integrin and CXCR5 expression analysis, the following Abs were used:
anti-CD11a (LFA-1 L chain) biotin (PharMingen), followed by
streptavidin-PE (Caltag); anti-CD29 (ß1
integrin chain), followed by anti-hamster IgG PE (PharMingen); and
anti-CXCR5 (Burkitt’s lymphoma receptor 1) (33),
followed by anti-rabbit Ig biotin (PharMingen) and
streptavidin PE.
Northern blot analysis
A total of 10–15 µg of RNA from mouse tissues or purified
cells was subjected to gel electrophoresis and transferred to a
Hybond-N+ membrane (Amersham Pharmacia Biotech,
Piscataway, NJ). Blots were probed successively with random-primed
32P-labeled mouse DNA probes (>2 x
106 cpm/ml) at 65°C for 1–4 h in ExpressHyb
solution (Clontech). Membranes were washed according to the
manufacturer’s instructions, exposed to a Phosphor-Screen (Molecular
Dynamics, Sunnyvale, CA), and quantified by PhosphorImager analysis.
The cDNA fragments used as probes were from the mouse EST clones in the
pT7T3-Pac vector, as follows: RGS1 XbaI-PvuII
(809 bp); RGS2 BstXI-BanII (605 bp); RGS3
XhoI-HindIII (359 bp; the XhoI site is
in the vector polylinker; the fragment encodes aa 222–316 of the
predicted full-length mouse RGS3 protein); RGS3s AvrII-XmaI
(197 bp; this probe includes 145 bp specific to RGS3s); and RGS14
EcoRI-NdeI (480 bp; the EcoRI site is
in the vector polylinker). To control for loading and RNA integrity,
membranes were reprobed with a mouse elongation factor-1 probe.
Chemotaxis assay
Chemotaxis assays were performed in 5-µm Transwell plates (Corning Costar, Cambridge, MA). Transfected cells were collected and resuspended in RPMI 1640 containing 0.5% BSA and 10 mM HEPES (migration media) at a concentration of 7.5 x 106/ml (live cells). Chemokines were resuspended in migration media to the indicated concentration, and 600 µl was aliquoted into 24-well plates forming the bottom chamber during the assay. The 5-µm-pore polycarbonate Transwell inserts were transferred to the wells containing media alone or plus chemokine, 100 µl of cells were added into the Transwell insert (top chamber), and cells were allowed to migrate through the porous bottom for 3 h at 37°C. Cells that migrated to the bottom chamber were enumerated, and GFP-positive cells were identified by collecting events for a fixed time (60 s) on a FACScan. By counting a 1/5 dilution of input cells in the same manner, the absolute number of cells that migrated to the bottom chamber could be determined. The percentage of migrated cells was calculated by dividing the number of transmigrated GFP-negative, GFP-dull, and GFP-bright cells by the number of input cells expressing the same GFP levels. Each data point is the mean of duplicate wells. To simplify the comparison between the different transfected cell populations, the GFP-negative fractions of each GFP-RGS-transfected cell population were normalized to the GFP-negative fraction of the control GFP-transfected cells. This was achieved by first determining the percentage of input GFP-negative cells that migrated in the control GFP-transfected population and then dividing this number by the percentage of input GFP-negative cells that migrated in each GFP-RGS-transfected population. The percentage of cells in each of the GFP-negative, dull, and bright gates that migrated was then multiplied by this number. The data presented for the different GFP-RGS-transfected population in the GFP-dull and GFP-bright compartment are the mean from duplicate wells normalized to the migration of GFP-negative cells in the same wells. Dead cells identified by their decreased forward scatter were not included in the analysis.
|
Results |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
To study the expression and function of murine RGS1 and
RGS3, we searched the mouse NCBI EST database by using human RGS1
(22, 23), or human RGS3 (13) as a template.
Twelve mouse ESTs homologous to human RGS1 were identified
(AA154742, AA110076, AA120409,
AA118893,
AA199350,
AA289401,
AA200913,
AA547074, AA608105, AA561820,
AA915687, and AA684124) that could be aligned into a contig, and
sequence analysis of clone AA154742 revealed a cDNA of 1274 bp encoding
a predicted protein of 196 aa (Fig. 1
A). The high nucleotide
identity (76%) and amino acid identity (87%) of this sequence with
human RGS1 (Fig. 1A) support its designation as murine RGS1.
The search for mouse RGS3 led to identification of EST clones W49391,
AA880930, and AA789696 that contained a 3'-untranslated region and
nucleotides encoding for 297 aa homologous to the C terminus of human
RGS3. This partial mouse RGS3 cDNA showed 90% amino acid identity to
human RGS3 (aa 222–519 of human RGS3) and 79% nucleotide identity (nt
591-2638 of human RGS3). We also identified three ESTs (AA278130,
AA717912, and AI529982) encoding a novel variant of RGS3 that lacks the
N-terminal 326 aa encoded by exon 3 in human RGS3 (34) and
contains an alternative 5' region encoding 21 aa, giving rise to a
predicted protein of 192 aa (Fig. 1B). A search of the human
EST database revealed two ESTs (H08458 and N87218) encoding an
identical NH2-terminal region to that identified
in the short form of mouse RGS3 and also containing highly similar
5'-untranslated sequence (data not shown). The existence of this
variant RGS3 sequence in both mouse and human indicates a conserved
role and leads us to name this form of RGS3 as RGS3s. Finally, we
isolated and sequenced mouse RGS2 from the EST clone AA221794. The
sequence was found to differ from the previously reported mouse RGS2
cDNA (35) by 4 nt in the coding region. The EST cDNA
contained the nucleotides GG instead of CC at position 137/138 and GC
instead of CG at position 254/255, and these changes alter the amino
acid sequence at four positions (Fig. 1A). The new sequence
at these positions is identical to the sequence in human RGS2,
indicating that human and mouse RGS2 may differ by as little as 8 aa
(96% amino acid identity). When the mouse RGS protein sequences were
aligned together with their human counterparts (Fig. 1), it was
apparent that the RGS domains were highly conserved, with 11 aa changes
between mouse RGS1 and human RGS1, 7 between mouse RGS3 and human RGS3,
and only 1 difference between mouse RGS2 and human RGS2.
|
Human or rat molecules of RGS1, RGS2, RGS3, and RGS14 have been
reported to be expressed in lymphoid tissues (22, 23, 24, 26).
To examine the expression pattern of these RGS members in murine
lymphoid tissues, Northern blot analyses were performed (Fig. 2). RGS1 was most abundant in spleen,
lymph node, and intestine, with moderate expression in Peyer’s patches
and skeletal muscle. The high expression in
RAG1-/- spleen indicates that a significant
amount of the constitutive RGS1 expression in spleen is in non-B or T
lymphocytes, and consistent with this, purified T cells expressed only
intermediate amounts of RGS1, and expression in purified B cells was
low to undetectable (Figs. 2A and 3B). Low
expression of RGS1 was also detected in thymus, lung, peritoneal cells,
white blood cells, and bone marrow cells (Fig. 2A).
Comparison of two murine B cell lines, the 300-19 pre B cell line
(36) and the IgG+ mature B cell
line, 2PK3 (37), revealed a striking difference in RGS1
levels, with strong expression in the 300-19 cells, but no detectable
expression in 2PK3 cells (Fig. 2A). RGS2 was found to be
strongly expressed in brain, heart, lung, spleen, white blood cells,
and bone marrow cells (Fig. 2A) in accordance with previous
studies (35). Again strong expression was detected in
RAG1-/- spleen and expression was low in
purified lymphocytes. However, in contrast to RGS1, basal RGS2
expression was detected in freshly isolated B cells, and this level was
typically higher than in T cells (Fig. 2A). 300-19 pre-B
cells expressed detectable levels of RGS2, whereas the 2PK3 cell line
differed from mature B cells in lacking measurable RGS2 (Fig. 2A). Analysis of full-length (3.5-kb) mouse RGS3
transcripts revealed their presence in most tissues, with lymphoid
tissues typically expressing high levels (Fig. 2, A and
B). B and T lymphocytes also appeared to make only a small
contribution to total splenic RGS3 based on expression levels in
RAG1-/- spleen, although RGS3 expression was
evident in both subsets of lymphocytes (Figs. 2, A and
B, and 3B). The 300-19 and 2PK3 cell lines
expressed little mouse RGS3 (Fig. 2, A and B).
Mouse RGS3s (1.8 kb) was also expressed in most lymphoid tissues,
but typically at lower levels than full-length RGS3 (Fig. 2B). High expression of mouse RGS3s was detected in heart,
brain, and lung (Fig. 2B). Finally, RGS14 expression was
examined because of a report indicating it was highly expressed in rat
spleen (26). Mouse RGS14 was constitutively present at
high levels in most lymphoid tissues and cells, including thymus,
spleen, lymph node, peritoneal cells, white blood cells, and bone
marrow cells. High basal expression was detected in freshly purified B
and T lymphocytes, and consistent with this, the level of RGS14 was
reduced in total RNA from RAG1-/- spleen
compared with wild-type spleen (Fig. 2A). Low expression of
RGS14 was detected in Peyer’s patches, lung, heart, skeletal muscle,
intestine, and the 2PK3 and 300-19 cell lines (Fig. 2A).
|
To analyze the expression pattern of RGS1, RGS2, RGS3, and RGS14
during Ag stimulation of B lymphocytes, we used
IgHEL transgenic mice carrying Ig heavy and light
chain transgenes encoding IgMa and
IgDa specific for the protein Ag HEL
(30). Consistent with previous studies reporting
spontaneous changes in RGS1 and RGS2 in human PBMC following in vitro
culture (25), we found that expression of RGS1, 2, and 14
underwent rapid changes in murine B cells upon in vitro culture (data
not shown). To circumvent this problem, B cells were activated in vivo
by i.p. challenge of IgHEL transgenic mice with
HEL, and splenic B cells were then purified at various time points and
used directly for RNA preparation. Intraperitoneal injection of HEL Ag
was effective in activating splenic B cells in an Ag-specific manner,
as established by the increase in expression of the activation marker
CD69 on B cells in IgHEL transgenic mice, but not
in nontransgenic mice (Fig. 3A, left panels).
Northern blot analysis of total RNA from the isolated B cell
populations revealed marked differences in the pattern of RGS molecule
expression (Fig. 3, B and C). RGS1 transcripts
were expressed at very low levels in B cells from unchallenged mice,
but were rapidly induced to high levels after in vivo exposure of
IgHEL transgenic B cells to HEL Ag (Fig. 3, B and C) and persisting at elevated levels for at
least 6 h (Fig. 3, B and C). The time course
of RGS2 induction in HEL-stimulated IgHEL B cells
was similar, except RGS2 mRNA levels became elevated only 2-fold
over quiescent levels compared with about a 40-fold increase in RGS1
transcript levels (Fig. 3C). In marked contrast, RGS3 and
RGS14 transcripts were present in naive B cells at substantial levels
and HEL stimulation suppressed expression (Fig. 3, B and
C). Although RGS3 mRNA was only moderately down-regulated in
response to HEL activation, to about half basal levels, RGS14
transcripts showed a more substantial 6-fold decrease. Thus, the RGS1
and RGS2 genes vs the RGS3 and RGS14 genes exhibit reciprocal patterns
of expression in naive and Ag-stimulated B cells in vivo.
|
A, right
panels). In striking contrast to the effects of activation by Ag,
challenge of mice with anti-CD40 mAb for various times did not
change RGS1 expression in B cells and caused a modest decrease in RGS2
mRNA levels (Fig. 3D). RGS3 and RGS14 transcripts appeared
to be transiently reduced in response to CD40 stimulation, as observed
for BCR-activated B cells.
Chronic exposure of B cells to self Ag leads to a state of anergy
and causes changes in cell migration and survival (2). To
examine the effect of chronic self Ag exposure on the expression levels
of RGS mRNA in B cells, B lymphocytes were isolated from
IgHEL/HEL double-transgenic mice. These mice
carry in addition to the IgHEL transgenes a
second transgene encoding soluble HEL as a circulating self Ag in a
form and amount sufficient to trigger anergy, but not deletion
(30). In chronically stimulated B cells isolated from
IgHEL/HEL double-transgenic mice, levels of RGS1
and RGS2 transcripts were elevated by about 2- or 3-fold, respectively
(Fig. 3, E and F). Interestingly, the extent of
up-regulation was reciprocal to that observed following acute
activation: RGS2 transcript levels were relatively higher in
chronically vs acutely activated B cells, whereas RGS1 levels were
substantially lower in the chronically activated B cells. Both RGS3 and
RGS14 appeared to be expressed at slightly reduced levels in
chronically stimulated compared with naive B cells (Fig. 3, E and F), although the reductions appeared less
marked than seen in acutely activated B cells.
RGS1 and RGS3 inhibit chemotaxis of transfected B cell lines to lymphoid chemokines
To test whether RGS1, RGS2, or RGS3 could regulate B cell
chemotactic responses to lymphoid chemokines, we first sought to
identify a transfectable B cell line that was responsive to BLC, ELC,
and SDF-1. We compared the 300-19 pre-B cell line previously used in
chemotaxis studies (41), with the
IgM+ WEHI231 B cell line (42), the
IgG+ mature B cell lines A20 (43)
and 2PK3 (37), the surface Ig-negative M12 B cell line
(43), and the J558L plasmacytoma cell line
(44). Using 5-µM-pore Transwell chemotaxis chambers, we
found that the 2PK3 cell line responded well to all three lymphoid
chemokines, whereas 300-19 pre-B cells responded only weakly to SDF-1
and ELC and failed to respond to BLC. Among the other cell lines, A20
and J558L cells showed little ability to migrate to any of the
chemokines, M12 cells exhibited high background migration but were able
to respond to SDF-1 and BLC, and WEHI231 cells responded to SDF-1 and
weakly to BLC and ELC (data not shown). 2PK3 cells were therefore
chosen for most of our experiments, although, because of their previous
characterization, 300-19 cells were also tested in some cases. To be
able to measure quantitatively levels of transfected RGS protein, we
tagged the N terminus of mouse RGS1, mouse RGS2, and mouse RGS3s with
green fluorescent protein (GFP). 2PK3 cells were transiently
transfected with the chimeric GFP-RGS constructs or as controls with
GFP or GFPmRGS1T60, a RGS1 chimera that lacks the RGS domain and hence
the GAP activity of the molecule (Fig. 4A). The chemotactic response
of transfected cells to BLC, ELC, or SDF-1 was measured in Transwell
migration assays, and the migratory cells were divided into those
expressing intermediate (GFP-dull, second quadrant) or high
(GFP-bright, third quadrant) amounts of RGS-GFP fusion protein. During
our analysis, we found that each batch of transfected cells showed
differences in the overall migration levels, assessed by the response
of the GFP-negative cells in each population. To compensate for this,
the migration of each transfected population was normalized so that the
response of the GFP-negative cells was matched (see Materials and
Methods). We found that mouse RGS1 was a strong inhibitor of the
migratory response to both ELC and BLC when expressed at sufficiently
high levels (GFP-bright) (Fig. 4B), inhibiting migration by
about 90% at all concentrations of BLC or ELC compared with migration
of control cells expressing GFP alone (Fig. 4B). The
inhibition was dose dependent, as cells expressing lower amounts of
RGS1 molecules (GFP-dull) were less compromised in their response to
ELC or BLC, demonstrating an inhibition of about 60% (Fig. 4B). The inhibitory effect observed was strongly dependent
on the RGS domain, as the truncated GFPmRGS1T60 molecule had little
effect (Fig. 4B). Similar to RGS1, expression of mouse RGS3s
also inhibited migration to ELC and BLC, although the inhibition was
less pronounced, being 20–23% at low RGS3s expression (GFP-dull) and
about 70% at higher RGS3s expression levels (GFP-bright). Both RGS1
and RGS3s limited 2PK3 cell chemotaxis to SDF-1, but to a lesser extent
than in response to ELC or BLC (Fig. 4B). In agreement with
RGS1 and RGS3s inhibiting chemotaxis through effects on chemokine
responsiveness rather than by causing changes in other molecules needed
for the chemotactic response, flow-cytometric analysis showed that
levels of LFA-1, ß1 integrin, and CXCR5 in the
transfected cells were unchanged (Fig. 5). In contrast with the effect of mouse
RGS1 and mouse RGS3s, mouse RGS2 had weak effects on 2PK3 cell
chemotaxis. Compared with the RGS domain-deficient GFPmRGS1T60
construct, GFPmRGS2 had no effect on the response to SDF-1 and only a
small effect on the response to BLC and ELC (Fig. 4B). These
findings for RGS2 are consistent with other studies suggesting RGS2 is
relatively specific for Gq family members and has little activity
against Gi proteins (16, 45). To confirm that the
GFPmRGS2 protein was active as a GAP, we tested its ability to inhibit
signaling by the Gq-coupled M1 muscarinic receptor. Jurkat cells
stably expressing the M1 receptor (46) were transiently
transfected with GFPmRGS2 or GFP alone, and the GFP-bright cells were
tested for ability to flux calcium in response to the M1 receptor
agonist, carbachol. A substantial reduction in the calcium flux was
observed in the RGS2-transfected cells compared with cells containing
GFP alone (data not shown), providing strong evidence that the GFPmRGS2
protein was functional.
|
|
C). Notably, although expression of GFPmRGS1
had an inhibitory effect on 300-19 cell migration (Fig. 4C),
the effect was not as strong as in 2PK3 cells, with about 30%
inhibition for GFP-dull cells and 50% inhibition for GFP-bright cells.
In this regard, it is important to note that 300-19 cells, in contrast
to 2PK3 cells, express substantial levels of endogenous RGS1 mRNA (Fig. 2A). High endogenous RGS1 protein expression might cause
this inhibitory pathway to be near saturation even without the
introduction of further RGS1 by transfection. In agreement with this
possibility was the very low chemotactic response to SDF-1 of
untransfected 300-19 cells compared with 2PK3 cells. At optimal SDF-1
concentrations, only 1% of total 300-19 cells migrated (Fig. 4C) compared with 35% of total 2PK3 cells (Fig. 4B). Interestingly, transfection of 300-19 cells with the
GFPmRGS1T60 mutant caused an increase in the migration to SDF-1
compared with cells transfected with GFP alone (Fig. 4C).
This effect was not seen in the 2PK3 cells that do not express
endogenous RGS1 (Fig. 4B). As RGS1T60 lacks the GAP activity
of RGS1, but retains the unique NH2-terminal
region, it may compete with endogenous RGS1 for receptor/Gi
association, while being unable to negatively regulate Gi responses.
Taken together, these findings provide evidence that the endogenous
RGS1 gene can be expressed at sufficient levels to have effects on cell
migration. |
Discussion |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
i-dependent chemotactic response to lymphoid chemokines BLC, ELC,
and SDF-1. These observations suggest that Ag receptor-induced changes
in RGS molecule expression may function to regulate B cell
responsiveness to lymphoid chemokines and thereby help direct
cell-positioning events in lymphoid organs.
To study the activity of RGS1 and RGS3 molecules in regulating
chemokine responsiveness in murine lymphocytes, it was first necessary
to isolate their mouse homologues. A single form of mouse RGS1 was
identified in the EST database that was similar over its entire length
to its human homologue. In addition to a conserved RGS domain, this
included an 60-aa amino-terminal domain that is unique to RGS1.
Searches for mouse homologues of RGS3 led to identification of two
forms of RGS3, a partial clone that appeared to correspond to
full-length human RGS3, and several clones encoding a short form of
mouse RGS3. Human ESTs encoding an analogous short form of human RGS3
were also identified, and we have designated this new RGS3 variant as
RGS3s. In this variant, the amino-terminal 348 aa encoded by exons 3
and 4 of human RGS3 (34) are replaced by a predicted 21-aa
domain (Fig. 1B). The 5' untranslated sequence of mouse and
human RGS3s is also highly conserved and is distinct from sequence
previously reported in human RGS3 exon 1, 2, or 3 (data not shown). The
mechanisms leading to the expression of this new form of RGS3 are
presently unclear, although they presumably involve splicing of one (or
more) novel exon to the splice acceptor of what has previously been
defined in human RGS3 as exon 5 (34). Future genomic
sequence analysis should establish how many exons make up the unique
region of RGS3s as well as determining their location with respect to
the previously defined exons. Interestingly, a truncated version of
RGS3 (termed RGS3T) has previously been identified by 5' race analysis
(47). The untranslated sequence of RGS3T was found to
begin within exon 3, and translation was suggested to start at either
of three ATG codons within the 3' end of exon 3. The RGS3s sequence
defined in this study is distinct from the predicted RGS3T because
RGS3s contains sequence (and putative exons) not present in the
original RGS3, whereas the sequence encoded within RGS3T, although
truncated, is identical to the original RGS3. Because RGS3s and RGS3T
transcripts appear to be of similar size, they may both contribute to
the prominent lower m.w. RGS3 hybridization bands seen on Northern
blots of human heart, brain, lung, kidney, and liver RNA (13, 38). As discussed further below, the different amino-terminal
regions may have significant effects on the function of RGS3. Thus, in
future studies, it will be important to determine the relative amounts
of the different RGS3 isoforms in B cells and other cell types.
By expressing GFP fusion proteins of RGS1, RGS2, and RGS3s in the 2PK3
B lymphoma cell line, we have been able to establish that RGS1 and
RGS3s antagonize chemotactic responses mediated by SDF-1, BLC, and ELC,
chemokines that signal through three different receptors (CXCR4, CXCR5,
and CCR7, respectively). By contrast, RGS2 had only limited effects,
causing slight reductions in the response to BLC and ELC and little or
no reduction in SDF-1 responses. These findings are in close agreement
with a recent report from Bowman et al., who found that human RGS1, 3,
and 4, but not human RGS2, diminished the chemotactic response of
transfected L1.2 pre-B cells to fMLP, IL-8, and monocyte
chemoattractant protein-1 (48). In another recent study,
RGS3 was shown to antagonize the migratory response of intermedullary
collecting duct kidney cells to lysophosphatidic acid
(49). Interestingly, in the study of Bowman et al., RGS3
was more effective at inhibiting chemokine responses than RGS1
(48), whereas in our experiments, the RGS1 construct was
more effective than the RGS3 construct (Fig. 4). These differences may
reflect specific properties of the transfected cell lines (L1.2 vs 2PK3
and 300-19) or of the different chemokine receptors under study.
However, it should be noted that the construct used by Bowman et al.
contained the original large RGS3 isoform, whereas our construct was
made with the shorter RGS3s variant. The
NH2-terminal domain of RGS3 has recently been
shown to translocate to cell membranes in response to increases in
intracellular calcium (17), suggesting it may enhance the
ability of RGS3 to inactivate membrane-associated G proteins. It
therefore seems possible that the stronger inhibitory effect of RGS3
than RGS3s reflects intrinsic differences in the Gi inhibitory
properties of the two proteins. In addition to effects on chemotactic
responses, RGS1, 2, and 3 have previously been shown to inhibit
IL-8-induced activation of extracellular signal-regulated kinase (ERK)
in transfected 293T cells (13). In biochemical studies,
RGS1 and 3 functioned as GAPs for Gi (50, 51), whereas
RGS2 was more effective as a GAP for Gq (16, 45).
However, some studies have indicated that RGS1 and RGS3 can also have
functional effects on Gq family members (16, 17, 18, 47),
and RGS2 may be able to antagonize some Gi family members
(13). Because the chemotactic response requires signaling
by pertussis toxin-sensitive Gi-coupled receptors, our findings
support the conclusion that RGS1 and 3 can function as GAPs for Gi
molecules, whereas RGS2 is an inefficient GAP for this family of
proteins.
The physiological relationship between BCR-induced changes in RGS
molecule expression and changes in B cell responsiveness to
chemokines is presently not understood. BCR triggering in B cells leads
to a decrease in the magnitude of the response to the CXCR4 ligand,
SDF-1, and an increase in the response to the CCR7 ligands, ELC and SLC
(7, 52, 53). It seems possible that the rapid increase in
RGS1 induced by BCR signaling works together with other processes, such
as phosphorylation-mediated down-regulation of CXCR4 receptors
(53), to decrease the chemotactic response to SDF-1. In
this respect, it is noteworthy that germinal center B cells have high
levels of RGS1 (22) and CXCR4, but fail to migrate to
SDF-1 (52). Consistent with this is the low chemotactic
response to SDF-1 of 300-19 pre-B cells, which express high
constitutive levels of endogenous RGS1 (Figs. 2A and
4C). Likewise, the ability of the truncated form of RGS1
(RGS1T60) to augment the response of 300-19 pre-B cells to SDF-1 (Fig. 4C) indicates a possible physiological role for RGS1 in
regulating CXCR4 responsiveness, because this deletion mutant might be
expected to antagonize interactions needed for full-length RGS1 to
function. 2PK3 cells did not have detectable endogenous RGS1 expression
and, as predicted by this model, overexpression of RGS1T60 in these
cells failed to augment the response to SDF-1. However, it remains to
be established in which cell populations RGS1 regulates CXCR4
responsiveness in vivo.
Despite the increase in RGS1 message following BCR engagement and the
ability of RGS1 to antagonize the response of transfected 2PK3 cells to
ELC, chemotaxis of mature B cells to ELC is increased after BCR
stimulation. This suggests that the action of RGS1 in inhibiting
chemokine receptor-stimulated migration may be more selective in mature
B cells than in the transformed 2PK3 cell line. RGS selectivity for
responses mediated by particular chemokine receptors could arise at
several levels, such as the G protein-coupling propensity of the
chemokine receptor (54, 55), or through direct interaction
of RGS molecules with chemokine receptors, as has been suggested in
other GPCR systems (16, 19, 20). Alternatively, the levels
of RGS1 induced in mature B cells by Ag may not be sufficient to
inhibit the ELC response. Reciprocally, perhaps the small decrease in
RGS3 expression that occurs in B cells activated in vivo by Ag
contributes to the enhanced chemotactic response to ELC. The decreased
expression of RGS14 might also be important in this enhancement,
although it has yet to be shown whether RGS14 is an antagonist of
chemokine-mediated chemotaxis. Alignment of the RGS14 GAP domain with
other RGS molecules (data not shown) indicates is has highest
similarity with RGS12, which can regulate Gi (19),
consistent with a possible role for RGS14 in regulating chemotaxis.
Further to the changes in RGS expression in B cells activated acutely
by Ag, small differences in RGS molecule expression were evident in
chronically stimulated, anergic, B cells (Fig. 3).
IgHEL/HEL double-transgenic mice contain fewer B
cells than IgHEL transgenic mice, and the purity
of the B cell preparations obtained from the double-transgenic animals
is typically a few percentages lower than from
IgHEL transgenic mice, making it possible that
the differences seen in RGS molecule expression are due to differences
in B cell purity. We think this unlikely, however, because if the
elevation in RGS1 and 2 in anergic cells was attributable to
contaminating cells, then it might be expected that RGS3 and 14 should
also be elevated, as all four RGS molecules are expressed at high
levels in total spleen cells (Fig. 2). Yet RGS3 and RGS14 levels were
not elevated in the anergic cells and may even be reduced (Fig. 3).
Previous studies in human PBMC have shown that RGS1 is strongly induced
by phorbol esters, activators of the ERK pathway, whereas RGS2 is
strongly up-regulated by treatment with ionomycin (25).
Anergic B cells have a small constitutive elevation in ERK activity and
in intracellular calcium (56), and these signals may
therefore contribute to the elevated basal RGS1 and RGS2 expression. It
seems possible that the differences in RGS1 molecule expression could
contribute to the reduced competitiveness of anergic B cells for
migrating into lymphoid follicles (57). The significance
of the elevated RGS2 expression induced by acute and chronic Ag
exposure remains an intriguing question because the role of signaling
by Gq family molecules in B cells is poorly defined. One group has
suggested that Gq molecules may regulate signaling via Bruton’s
tyrosine kinase (Btk) (58, 59), raising the possibility
that RGS2 functions in activated and anergic B cells to limit the
extent of Btk activity.
In summary, our studies together with others suggest that regulated expression of RGS1 and RGS3, and possibly RGS14, plays a role in changing B cell responsiveness to chemokines during the response of B cells to foreign or self Ags. Alterations in RGS2 expression during acute and chronic B cell activation appear more likely to regulate other, presently unknown, GPCR signals in B cells. In future studies, it will be important to characterize whether the additional RGS family members that have recently been identified in lymphoid tissues, including RGS12 (19, 26) and RGS16 (27), are also modulated during B lymphocyte activation. A more complete understanding of RGS function in regulating B cell migration, differentiation, and tolerance will also require greater knowledge of the interplay between these proteins and other molecules, such as GPCR kinases, protein kinase Cs, protein kinase As, and arrestins, that also regulate GPCR signaling.
|
Acknowledgments |
|---|
|
Footnotes |
|---|
2 The sequences presented in this paper have been submitted to GenBank/EMBL/DDBJ under accession numbers AF215667 (mRGS1), AF215668 (mRGS2), AF215669 (mRGS3s), and AF215670 (mRGS3).
3 Address correspondence and reprint requests to Dr. Jason G. Cyster, Department of Microbiology and Immunology, 513 Parnassus Avenue, Box 0414, Room HSE301, University of California, San Francisco, CA 94143.
4 Abbreviations used in this paper: BLC, B lymphocyte chemoattractant; BCR, B cell receptor; ELC, EBV-induced molecule 1 ligand chemokine; ERK, extracellular signal-regulated kinase; EST, expressed sequence tag; GAP, GTPase-activating protein; GEF, nucleotide exchange factor; GFP, green fluorescent protein; GPCR, G protein-coupled receptor; HEL, hen egg lysozyme; RAG, recombination-activating gene; RGS, regulator of G protein signaling; SDF-1, stromal cell-derived factor-1; SLC, secondary lymphoid tissue chemokine.
Received for publication September 8, 1999. Accepted for publication February 18, 2000.
|
References |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
i3, is a member of a protein family with a highly conserved core domain. Proc. Natl. Acad. Sci. USA 92:11916.1 (the G-protein subunit). Mol. Cell. Biol. 16:5194.[Abstract]
11. Mol. Cell. Biol. 19:714. proteins and inhibits pheromone signaling in yeast. J. Biol. Chem. 272:8679. dimers released by activation of Gi-coupled receptors. Proc. Natl. Acad. Sci. USA 94:14495.
+ lymphomas: sensitivity to cortisol and phytohemagglutinin. J. Immunol. 110:1470. function. Proc. Natl. Acad. Sci. USA 94:14389.-subunits. Nature 383:172.[Medline]
chemotaxis and promotes protein kinase C (PKC)-induced internalization of CXCR4. J. Exp. Med. 189:1461.
), mouse
RGS1 (
), mouse RGS2 (
), mouse RGS3s (•), or mouse RGS1T60,
which does not contain the RGS domain (
). Transfected cell
populations were subjected to chemotaxis through 5-µm-pore Transwell
filters. The number of input and migrated cells in each GFP gate, as
outlined in A, was determined by flow cytometry. Results
are expressed as the percentage of input cells in each GFP gate that
migrated to the bottom chamber. The data were normalized to the
GFP-negative populations, as described in Materials and
Methods. C, Effects of RGS1 and RGS1T60 on
300-19 cell chemotaxis to SDF-1. 300-19 cells were transiently
transfected with GFP, GFPmRGS1, or GFPmRGS1T60, and chemotaxis was
measured as described for 2PK3 cells in B. The results
in B are representative of three or four independent
experiment, and data points represent the mean of assay points
performed in duplicate; and results in C are the mean of
seven experiments with data points performed in duplicate.