Research Article
807
The BTB-MATH protein BATH-42 interacts with RIC-3
to regulate maturation of nicotinic acetylcholine
receptors
Anna Shteingauz, Emiliano Cohen, Yoav Biala and Millet Treinin*
Department of Physiology, Hebrew University, Hadassah Medical School, Jerusalem, 91120, Israel
*Author for correspondence (e-mail: [email protected])
Journal of Cell Science
Accepted 5 November 2008
Journal of Cell Science 122, 807-812 Published by The Company of Biologists 2009
doi:10.1242/jcs.036343
Summary
RIC-3 is a member of a conserved family of proteins that affect
nicotinic acetylcholine receptor maturation. In yeast and in
vitro, BATH-42, a BTB- and MATH-domain-containing protein,
interacts with RIC-3. BATH-42 is also known to interact with
the CUL-3 ubiquitin ligase complex. Loss of BATH-42 function
leads to increased RIC-3 expression and decreased activity of
nicotinic acetylcholine receptors in Caenorhabditis elegans vulva
muscles. Increased expression of RIC-3 is deleterious for activity
and distribution of nicotinic acetylcholine receptors, and thus
the effects of BATH-42 loss of function on RIC-3 expression
explain the associated reduction in receptor activity.
Introduction
Nicotinic acetylcholine receptors (nAChRs) are a diverse family of
ligand-gated ion channels mediating excitatory or modulatory roles.
Biogenesis of functional nAChRs is a complex, time-consuming
and inefficient process (Merlie and Lindstrom, 1983). This process
is assisted by a number of ER-resident chaperones, including
nonspecific chaperones such as BIP and calnexin, and the nAChRspecific chaperone RIC-3 (Blount and Merlie, 1991; Chang et al.,
1997; Halevi et al., 2002). RIC-3 homologs increase the efficiency
of the maturation process, increasing both steady-state levels of
receptor subunits and surface expression of mature receptors
(Halevi et al., 2003; Cheng et al., 2005; Castillo et al., 2005; Cohen
Ben-Ami et al., 2005). Effects of RIC-3 homologs are not restricted
to quantitative effects, because the human RIC-3 homolog enables
maturation of α7 nicotinic receptors in cells that will otherwise
express no functional α7 receptors (Williams et al., 2005), and RIC3 was also shown to affect properties of the C. elegans DEG-3/DES2 receptor (Cohen Ben-Ami et al., 2005). Thus RIC-3 family
members are key players in the maturation of nAChRs in both
vertebrates and invertebrates.
Using a two-hybrid screen in yeast, we identified BATH-42 as
a protein that interacts with the RIC-3 C-terminus. BATH-42, which
contains BTB/POZ (broad-complex, Tramtrack and bric-a-brac/Pox
virus and zinc finger) and MATH (meprin-associated Traf
homology) domains, was previously identified to interact with CUL3 (Xu et al., 2003). CUL-3 functions as a scaffold within a cullinring ligase (CRL) complex, whose function is to recruit E2ubiquitin-conjugating enzymes to specific substrates. Substrate
specificity of CRLs requires a substrate receptor (Petroski and
Deshaies, 2005). Specifically, for CUL-3, the substrate receptors
are BTB-domain proteins. These proteins interact with CUL-3 via
Overexpression of BATH-42 is also detrimental to nicotinic
acetylcholine receptor function, leading to decreased pharyngeal
pumping. This effect depends on the C-terminus of RIC-3 and
on CUL-3. Thus, our work suggests that BATH-42 targets RIC3 to degradation via CUL-3-mediated ubiquitylation. This
demonstrates the importance of regulation of RIC-3 levels, and
identifies a mechanism that protects cells from the deleterious
effects of excess RIC-3.
Key words: C. elegans, Proteasome, CUL-3
their BTB domain and recruit the substrate via another proteininteraction domain (Xu et al., 2003; Petroski and Deshaies, 2005).
Similarly to BATH-42, MEL-26 contains BTB and MATH domains,
and targets the microtubule-severing protein MEI-1/katanin for
CUL-3-dependent degradation, thus enabling meiosis-to-mitosis
transition (Pintard et al., 2003).
BATH-42 might therefore also target specific proteins, such as
RIC-3, for degradation by the ubiquitin proteasome system. To
evaluate this hypothesis, we examined the in vivo implications of
interactions between RIC-3 and BATH-42 by analyzing the effects
of BATH-42 loss of function or overexpression. This analysis
supports our hypothesis, and suggests that interactions between
BATH-42 and RIC-3 enable regulation of the levels of RIC-3.
Moreover, we show that this mechanism is active under
physiological conditions. In addition, we show that excess RIC-3
is deleterious for nAChR function and distribution, motivating the
need for such regulation.
Work by others has shown regulation of receptor subunits by
the proteasome (reviewed by Yi and Ehlers, 2005). Our work
highlights the need to regulate not only the receptors themselves,
but also the chaperones required for their assembly, and
specifically identifies a mechanism that protects cells from excess
RIC-3. Moreover, this suggests that the proteasome indirectly
regulates synaptic transmission mediated by nAChRs via
regulation of RIC-3.
Results
RIC-3 and BATH-42 interact in vitro
The C-terminus of all RIC-3 homologs contains one or two coiledcoil domains (Halevi et al., 2003), which are known to function
in protein-protein interactions (Stefancsik et al., 1998). To
808
Journal of Cell Science 122 (6)
Journal of Cell Science
Fig. 1. X-gal assays and GST pull-down assays show an interaction between
BATH-42 and RIC-3 (C-terminus). (A) X-Gal assay on yeast cells expressing
the bath-42 prey alone, both the ric-3 bait and the bath-42 prey, or the ric-3
bait alone. (B) In vitro transcribed, translated and labeled bath-42 was loaded
in lane 1. Labeled BATH-42 bound to GST:RIC-3, GST:ODZ or GST beads
was loaded in lanes 2, 3 or 4, respectively.
identify proteins that interact with the RIC-3 C-terminus, we used
it as bait in a yeast two-hybrid screen. This screen identified
BATH-42 (C50C3.8) as a RIC-3 C-terminus interactor (Fig. 1A).
To verify this interaction, we used GST pull-down of in vitro
transcribed and translated BATH-42. This analysis validates the
two-hybrid results, showing a strong interaction between BATH42 and the RIC-3 C-terminus when fused to GST (Fig. 1B).
Interaction between GST:RIC-3 and BATH-42 is specific, because
a much weaker interaction was seen using GST alone or GST
fused to the unrelated ODZ protein (Levine et al., 1994) (Fig.
1B). The conserved coiled-coil sequences are likely to be
responsible for the observed interactions, because GST fusions
and two-hybrid baits containing only the first coiled-coil and short
flanking sequences are sufficient for this interaction (results not
shown).
Expression of BATH-42 and RIC-3 overlap
In yeast cells and in vitro, BATH-42 interacts with the RIC-3 coiledcoil domain. For this interaction to occur in C. elegans, the two
proteins should be expressed within the same cells. To examine
whether BATH-42 expression overlaps with RIC-3 we fused the
intergenic sequences upstream of BATH-42 to GFP. Analysis of
this bath-42p:GFP reporter shows restricted expression in specific
muscles, namely the pharyngeal and the vulval muscles (Fig. 2A,C).
Occasionally expression was also seen in a few neurons in the head
and body regions. In addition, bath-42p:GFP expression was seen
in non-excitable cells, namely the posterior part of the intestine and
the seam cells (Fig. 2B). Vulval muscles, pharyngeal muscles, and
most neurons are known to express RIC-3 and to require RIC-3 for
maturation of nAChRs (Halevi et al., 2002). Specifically, within
the pharyngeal muscle, RIC-3 is required for maturation of the EAT2 nAChR and therefore for pharyngeal pumping, and within the
vulva muscles, RIC-3 is required for maturation of the levamisolesensitive muscle nAChR. Thus an interaction between BATH-42
and RIC-3 within these cells has the potential to affect nAChRmediated signaling.
Fig. 2. bath-42p:GFP expression in L4 stage or young adult C. elegans.
(A) Pharyngeal muscles. Scale bar: 10 μm. (B) Seam cells (fluorescence on top
right of the picture represents gut auto-fluorescence). Scale bar: 5 μm.
(C) Vulval muscles. Scale bar: 10 μm.
bath-42 loss of function enhances RIC-3 levels and reduces
nAChR function in vulval muscles
To examine the functional implications of BATH-42–RIC-3
interaction we studied a bath-42 loss of function [bath-42(lf)] allele
made available by the National Bioresource Project, Japan
(www.shigen.nig.ac.jp/c.elegans/index.jsp). This mutant was
outwardly wild type, with a normal growth rate, the normal number
of eggs (Fig. 3B) and a normal rate of pharyngeal contractions
[2.9±2.6% of bath-42(lf) were pumping defective relative to
3.3±2.6% of N2 (n=35, N=5 each)]. Thus BATH-42 is not essential
for function of the pharyngeal or vulval muscles under normal
growth conditions.
However, two assays revealed significant differences relative
to wild-type nematodes. First, RIC-3 levels in the vulva muscles,
assessed by immunohistochemistry, were increased slightly but
significantly when compared with body wall muscles and ventral
cord neurons (Fig. 3A). Second, the egg-laying response to
levamisole, an agonist of the UNC-29-containing nAChRs known
to function in the vulva muscles (Waggoner et al., 2000), is
decreased (Fig. 3B). This effect is due to a difference in the
response to levamisole, because no significant difference in egg
laying is seen on food or in M9 buffer without levamisole (Fig.
3B). To show that this second effect is not a result of background
mutations in the bath-42(lf) strain, we also examined the effects
of bath-42 dsRNA feeding on the egg-laying response to
levamisole. Bath-42 knockdown in wild-type N2 animals resulted
in a reduction of the egg-laying response to levamisole 60±10%
relative to N2 fed with vector dsRNA (n=33-36, N=3, P<0.01).
This reduction is similar to the phenotype of bath-42(lf) mutants.
Thus, we conclude that reduced BATH-42 expression leads to
reduced sensitivity of the egg-laying response to levamisole.
Levamisole sensitivity serves as an assay for function and
expression levels of the levamisole-sensitive receptors (Waggoner
et al., 2000), suggesting that reduced bath-42 expression, as seen
in the bath-42(lf) mutant or following treatment with bath-42
dsRNA, leads to a reduced level or function of the vulva muscle
receptor.
Our results show that bath-42(lf) leads to an increase in RIC-3
levels while reducing expression or function of the levamisolesensitive nAChR in vulva muscles. The first result is consistent
BATH-42 regulates RIC-3 expression
809
Journal of Cell Science
Fig. 4. Increased RIC-3 expression leads to reduced nAChR activity in vulva
muscles. (A) Number of eggs laid per animal by N2 (wild-type), ric3(md1181) overexpressing ric-3 following 5 ng/μl injection, or ric-3(md181)
overexpressing ric-3 following 20 ng/μl injection. Differences between the
three strains are significant P<0.01 (following Bonferroni correction) using the
Student’s t-test for two samples (n=51-54 each, N=5). (B) Representative
images of RIC-3:GFP distribution in vulval muscles of transgenics injected
with 5 ng or 20 ng RIC-3:GFP. Arrows indicate representative puncta. Scale
bar: 10 μm.
Fig. 3. bath-42(lf) leads to an increase in RIC-3 levels and to reduced
levamisole sensitivity in vulva muscles. (A) Normalized staining intensity
of RIC-3 in vulval muscles of wild-type (N2, black bars), or bath42(tm2360), a bath-42(lf) mutant (grey bars). For each stained animal
average staining intensity in vulval muscles was divided by the average
staining intensity in the adjacent body-wall muscles and ventral cord
neurons. Difference is significant P<0.01 using the Student’s t-test for two
samples (n=38-40 each, N=3). Representative images of RIC-3 staining in
vulval muscles are shown on the right. Scale bar: 10 μm. (B) Number of
eggs laid per animal by N2 or bath-42(lf) on food, M9 or M9 supplemented
with 50 μM levamisole. Also shown is egg-laying in M9 supplemented with
50 μM levamisole for bath-42(lf) homozygotes that are heterozygous for
ric-3(md1181), a ric-3(lf) mutant (white). Significant differences were
found between number of eggs laid in the presence of levamisole for N2
compared with bath-42(lf), P<0.01 (n=72 each, N=6), and for
bath-42;ric-3(md1181)/+ compared with bath-42(lf), P<0.05 (n=67-92,
N=7-8) using the Student’s t-test for two samples.
with a role for BATH-42 in targeting RIC-3 to degradation, and
shows that in the absence of BATH-42, steady-state levels of RIC3 increase. However, the second result appears inconsistent with
this hypothesis, unless increased RIC-3 expression is detrimental
for its function. To examine this possibility, we looked at the effects
of overexpressing RIC-3 on the egg-laying response to levamisole.
This analysis showed that increased RIC-3 expression, achieved by
expressing RIC-3 from different copy numbers of a RIC-3:GFP
transgene, was associated with a decrease in the number of eggs
laid in response to levamisole (Fig. 4). Thus, decreased degradation
of RIC-3 in bath-42(lf) mutants leading to an increase in RIC-3
levels can explain the decreased egg-laying response to levamisole.
If indeed bath-42(lf) egg-laying defects are a result of RIC-3
overexpression, we expect that reducing RIC-3 expression will
suppress this defect. To examine this prediction, we preformed
levamisole-dependent egg-laying assays on bath-42(lf) homozygous
animals that are heterozygous for a ric-3(lf) mutation. Results of
this assay show that levamisole-dependent egg-laying is increased
in these mutants relative to bath-42(lf) mutants carrying two wildtype copies of ric-3 (Fig. 3B). These results support our hypothesis
that the effects of bath-42(lf) on levamisole sensitivity are a result
of its effects on RIC-3 expression.
Overexpression of RIC-3 influences nAChR distribution and
function
Next, we examined the distribution of overexpressed RIC-3. In vulva
muscles, overexpressed RIC-3 accumulated in puncta, which are
likely to be aggregates of RIC-3 (Fig. 4B). These RIC-3 aggregates
might sequester nAChR subunits away from the assembly process.
To examine this suggestion, we looked at effects of RIC-3
overexpression on localization of the DEG-3 nAChR subunit in
PVD cell bodies (Yassin et al., 2001). This analysis shows that
overexpression of RIC-3 affects the distribution of DEG-3, leading
to a less-homogeneous distribution (Fig. 5A,B). Moreover, DEG3 accumulations overlap with RIC-3 accumulations, supporting the
idea that excess RIC-3 sequesters nAChR subunits away from the
assembly process. Overexpression of RIC-3 reduced the number of
DEG-3(u662)-dependent degenerations, as assayed by counting the
number of swollen (degenerating) neurons, i.e. degenerations that
are caused by constitutive activity of this mutant nAChR (Fig. 5C)
(Treinin and Chalfie, 1995). This effect is consistent with our
suggestion that RIC-3 overexpression interferes with RIC-3 function
and thus with nAChR distribution and function.
BATH-42 overexpression reduces pharyngeal pumping in a
RIC-3- and CUL-3-dependent manner
Analysis of bath-42(lf) mutants is consistent with a role for BATH42 in targeting RIC-3 for degradation. To further examine this
hypothesis, we examined the effects of BATH-42 overexpression
using the hsp-16-2 promoter (Jones et al., 1986). Heat-shock
induction of BATH-42 expression in adults led to decreased growth
(fewer progeny and reduced rate of development). This effect was
associated with a large decrease in pharyngeal pumping (Fig. 6), a
phenotype similar to that seen in ric-3(lf) mutants (Halevi et al.,
2002). Heat-shock of wild-type animals not expressing hsp-162p:BATH-42 did not significantly reduce growth rate or pharyngeal
pumping (30±4.6% were defective in pumping following heat shock,
compared with to 19±2.6 % without heat shock; n=43-46, N=6),
showing that this effect required BATH-42 overexpression.
Effects of overexpressing BATH-42 on pharyngeal pumping are
consistent with the hypothesis that BATH-42 overexpression leads
810
Journal of Cell Science 122 (6)
Journal of Cell Science
Fig. 6. Overexpression of BATH-42 leads to reduced pharyngeal pumping in a
ric-3- and cul-3-dependent manner. Effects of hsp-16-2p:BATH-42 transgene
on pharyngeal pumping in wild-type (N2), ric-3(md158) mutants [ric-3(lf)],
ric-3(lf) mutants also expressing a truncated RIC-3 TM protein, animals
expressing the RIC-3 ΔN protein, N2 treated with cul-3 dsRNA or N2 treated
with vector dsRNA. Grey, control animals without heat shock; black, heatshocked (HS) animals. Effects of heat-shock on pumping in N2 animals
expressing hsp-16-2p:BATH-42 are significant P<0.01 using the Wilcoxon
two-sample test (n=45-90 each, N=3-5). Effects of heat-shock on pumping in
animals expressing RIC-3 ΔN are also significant P<0.05 (n=75-78 each, N=5)
Fig. 5. Increased RIC-3 expression leads to defects in DEG-3 distribution and
function. (A) Representative images of N2 or RIC-3:GFP-overexpessing
transgenics. DEG-3 staining in green, GFP staining (for RIC-3:GFP) in red,
merge in yellow. Scale bar: 2 μm. (B) Coefficient of variation of DEG-3
staining in N2 or in transgenics injected with 20 ng/μl RIC-3:GFP. Difference
is significant P<0.01 using the Student’s t-test for two samples (n=19, N=2).
(C) Number of degenerating cells in deg-3(u662) animals expressing high
copy number RIC-3:GFP (20ng/μl) compared with deg-3(u662) animals
lacking this transgene. Difference is significant P<0.01 using the Student’s ttest for two samples (n=41-42, N=2).
to RIC-3 degradation. Unfortunately, our RIC-3 antibodies did not
provide clear and reproducible staining of the pharynx, and
expression of functional RIC-3:GFP transgenics showed a large
variability between cells and between animals. Thus, we used
genetic tools to validate our assumption that BATH-42 mediates its
affects on pharyngeal pumping via its interaction with RIC-3. First,
we examined whether effects of BATH-42 overexpression were
additive to the effects of ric-3(lf). This analysis showed no additive
effects, and is thus consistent with our hypothesis (Fig. 6). Second,
we examined whether the effects of BATH-42 overexpression
required the RIC-3 C-terminus. For this purpose, we examined the
effects of BATH-42 overexpression in animals expressing a
truncated version of RIC-3 (RIC-3 TM), lacking the C-terminal
domain. Here, we note that the conserved coiled-coil, the domain
that is likely to be responsible for the interaction of RIC-3 with
BATH-42 (see above), also had a receptor-specific role in enhancing
the interaction of RIC-3 with nAChRs. Specifically, RIC-3 TM did
not promote maturation of the levamisole-sensitive receptors found
in the vulva muscles, but did promote maturation of the pharyngeal
muscle nAChR (Biala et al., 2008). Thus we can only examine
whether the effects of overexpressing BATH-42 require the
conserved coiled-coil domain in pharyngeal muscles but not in vulva
muscles. Analysis presented in Fig. 6 shows that animals expressing
the RIC-3 TM transgene are resistant to the effects of BATH-42
overexpression. To show that this resistance was not due to the
presence of RIC-3 on a multi-copy extrachromosomal array, we
also examined the effects of BATH-42 overexpression on another
RIC-3 derivative, RIC-3 ΔN, differing from RIC-3 TM in having
an additional 65 amino acids encoding the conserved coiled-coil
domain. This RIC-3 derivative rescues the ric-3(lf)-dependent
defect as well as wild-type RIC-3 (Biala et al., 2008). Animals
expressing this RIC-3 ΔN derivative were sensitive to the effects
of BATH-42 overexpression (Fig. 6). Thus we conclude that
interaction of BATH-42 with the RIC-3 C-terminus, probably with
the conserved coiled-coil domain, is required for effects of BATH42 overexpression on pharyngeal pumping.
BATH-42 was previously shown to interact with CUL-3, and is
thus likely to function as a substrate specific adaptor, targeting
specific proteins for degradation in a CUL-3-dependent manner (Xu
et al., 2003). Effects of overexpression of BATH-42 on pharyngeal
pumping are similar to the effects of ric-3 loss of function. Thus
we suggest that BATH-42 targets RIC-3 for degradation in a CUL3-dependent manner. To examine this, we investigated whether the
effects of BATH-42 overexpression are suppressed by depletion of
CUL-3. For this purpose, we examined the effects of BATH-42
overexpression in animals fed with cul-3 dsRNA. CUL-3 was
required for the effects of BATH-42 overexpression on pharyngeal
pumping, because cul-3 dsRNA treatment overnight before heat
shock suppressed the effects of BATH-42 overexpression (Fig. 6).
Control experiments using dsRNA treatment of vector sequences
showed no suppression of the effects of BATH-42 overexpression.
Thus, the effects of BATH-42 overexpression require CUL-3. These
results, together with our previous work, suggest that the effects of
BATH-42 overexpression are likely to be a result of CUL-3dependent ubiquitylation and degradation of RIC-3.
Discussion
BATH-42 is a BTB- and MATH-domain-containing protein,
previously shown to interact with CUL-3 (Xu et al., 2003). BTB
proteins are postulated to function as substrate-specific receptors,
enabling recognition of specific substrates by CUL-3, thereby
directing specific proteins to CUL-3-dependent ubiquitylation
(Petroski and Deshaies, 2005). Here we show that BATH-42
interacts with RIC-3, an effector of nAChR maturation. Effects of
BATH-42 loss of function or overexpression are consistent with a
role of BATH-42 in targeting RIC-3 for degradation in a CUL-3dependent manner. We suggest that under physiological conditions,
BATH-42 activity maintains optimal levels of RIC-3 by targeting
Journal of Cell Science
BATH-42 regulates RIC-3 expression
excess RIC-3 for degradation. Excess RIC-3 interferes with function
of nAChRs, and thus, loss of BATH-42 function also interferes with
nAChR function.
Members of the RIC-3 family have a key role in nAChR
maturation in vertebrates and invertebrates. RIC-3 and its homologs
were shown to affect maturation of many different nAChRs. Effects
of RIC-3 family members on nAChRs include changes in surface
expression, steady-state levels, receptor properties and the formation
of functional receptors (Halevi et al., 2003; Williams et al., 2005;
Cheng et al., 2005; Castillo et al., 2005; Cohen Ben-Ami et al.,
2005). Thus, the regulation of RIC-3 levels is likely to have major
effects on nAChR-mediated synaptic signaling.
Moreover, results presented here suggest that overexpression of
RIC-3 leads to formation of RIC-3 aggregates that colocalize with
nAChR subunits, and are therefore likely to reduce the amount of
receptor subunits available for formation of functional receptors. Thus,
tight regulation of RIC-3 levels is important. We note that RIC-3 is
not the only chaperone whose quantity is regulated by the proteasome.
The quantity of UNC-45, a chaperone for muscle myosin, is also
regulated by the proteasome (Hoppe et al., 2004). Overexpression of
UNC-45, similarly to overexpression of RIC-3, is detrimental for its
function. Specifically, overexpression of UNC-45 interferes with
myosin assembly (Hoppe et al., 2004). The detrimental effects of
overexpression of RIC-3 or UNC-45 suggest that excess chaperone
levels sequester interacting proteins away from the assembly process.
Indeed, overexpression of functional RIC-3:GFP, in addition to its
effects on nAChR function in vulval muscle, and the PVD neurons,
also affects nAChR function in body-wall muscles (Y.B., unpublished
results). Thus, the maintenance of optimal RIC-3 levels by the
proteasome is important for normal cholinergic signaling.
Regulation of synaptic protein turnover by ubiquitylation enables
regulation of synaptic plasticity. In post-synaptic membranes
ubiquitylation was shown to regulate turnover of glutamate receptors
(reviewed by Yi and Ehlers, 2005). Specifically, BTB- and Kelchdomain-containing proteins were shown to affect the levels of
glutamate receptors in C. elegans neurons and in mammalian cells
(Schaefer and Rongo, 2006; Salinas et al., 2006). In addition to
effects of the ubiquitin proteasome pathway on mature receptors
within the synapse, this pathway was also shown to affect maturation
of receptors. Specifically, proteasome inhibition increased the
quantity of nAChRs, suggesting that a balance between degradation
and assembly regulates the level of mature nAChRs (Christianson
and Green, 2004). RIC-3 family members also affect the level of
mature nAChRs (Cheng et al., 2005; Castillo et al., 2005; Cohen
Ben-Ami et al., 2005), suggesting that an interaction with RIC-3
protects nAChR subunits from degradation. Therefore, it is possible
to regulate the quantity of mature nAChRs both directly, through
ubiquitylation and degradation of nAChR subunits and indirectly,
through ubiquitylation and degradation of RIC-3. Our results
showing increased RIC-3 levels in BATH-42 loss-of-function
mutants suggest that RIC-3 levels in vivo are controlled by the
proteasome. Such regulation, in addition to protecting cells from
the detrimental effects of excess RIC-3, provides the means for
regulating the synaptic activity mediated by many different nAChRs.
Materials and Methods
Two-hybrid screens
For the yeast two-hybrid assay, the C-terminus of RIC-3, 112 amino acids containing
the two coiled-coil domains [PCR amplified from YK266d12 (Halevi et al., 2002)],
was cloned into pAS-1 and served as bait following transformation into strain y1024
(MATa, gla4Δ, trp1-901, ura3-52, ade2-101, URA3::GAL1-lacZ, LYS2::GAL1-HIS3,
his3-200 leu-2-3,112, cyhr (kindly provided by Yona Kassir, Technion, Haifa, Israel).
811
The bait-containing cells were then transformed with a C. elegans cDNA λACT-RB2
library (kindly provided by Robert Barstead, OMRF, University of Oklahoma,
Oklahoma City, OK) and plated on –His, –Leu, –Trp with 20 mM 3-amino-1,2,4triazole. Colonies growing on these plates contain a bait plasmid and a prey plasmid
enabling HIS3 expression. To further examine these colonies they were transferred
to nitrocellulose for LacZ assays as described below. Colonies producing blue color
in this assay, an indication of LacZ (β-galactosidase) expression, were further studied.
Only plasmids that on their own did not enable LacZ expression but did enable LacZ
expression upon isolation and retransformation to a bait-containing strain were further
analyzed. For LacZ assays shown in Fig. 1, strains were grown on YEPD plates
overnight, overlaid with nitrocellulose filters, and then crushed by placing the filter
in liquid nitrogen; nitrocellulose filters were then placed on 3MM filter paper saturated
with Z-buffer with X-gal, and allowed to develop overnight.
Nematode growth and analysis
Wild-type (N2 Bristol) and all other strains were grown on NGM plates seeded with
OP50 at 20°C unless stated otherwise (Wood, 1988). Induction of dsRNA expression
was done in LB + ampicillin using 4 mM IPTG for 3 hours. The bacteria (HT115DE3)
expressing cul-3, bath-42, or vector dsRNA were then spun down, resuspended using
M9 buffer (0.02 M KH2PO4, 0.042 M Na2HPO4, 0.085 M NaCl and 1 mM MgSO4)
+ 5 mM IPTG, and seeded on NGM plates. For dsRNA feeding experiments, L4
animals were transferred to plates seeded with dsRNA expressing bacteria and grown
overnight at 20°C before egg-laying assays or heat-shock and pumping assays. Heat
shock was done for 0.5 hours at 32°C followed by 2 hours recovery at 20°C. Pumping
rate was observed under high magnification and animals whose pumping rate was
lower than 70 pumps per 30 seconds were considered pumping defective (normal
pumping rate is ~110 pumps per 30 seconds). Egg-laying experiments were conducted
on adults picked the night before to fresh plates and then assayed on fresh NGM
plates or in multiwell plates in the presence of M9 or in M9 + 50 μM levamisole, 1
animal per plate or well for 2 hours (Waggoner et al., 2000). bath-42(lf) homozygous
that are heterozygous for ric-3(md1181) were obtained by crossing bath-42(lf) males
with bath-42(lf) ric-3(md1181) double mutant hermaphrodites, progeny of this cross
not showing the ric-3(md1181) phenotypes (coiling movement and other ric-3 lossof-function phenotypes are recessive) were picked for levamisole assays as described
above. For neuronal degeneration assays, the high copy number (20 ng/μl) RIC-3:GFP
functional transgene was crossed into a deg-3(u662) background. Degenerations were
observed in early L1 larva, 1-2 hours after hatching, using Nomarski optics (Treinin
and Chalfie, 1995). As controls, we used larva from the plates that did not carry the
transgene, as seen by the absence of the GFP marker. In all experiments n is number
of animals examined, and N is the number of independent experiments. Results are
given as average ± s.e.m.
Molecular biology
To generate RIC-3:GST fusions, a BamHI site was inserted upstream of the first,
conserved, coiled-coil domain following PCR amplification from yk719-F3 (Halevi
et al., 2002). Either a BamHI-EcoRI fragment (entire C-terminus) or a BamHI-SalI
fragment (first, conserved, coiled-coil domain) was cloned into pGEX-3X. A 1.3 kb
EcoRI-BglII fragment of BATH-42 from the two-hybrid isolated clone was cloned
into pET-17b for in vitro transcription and translation. Purification of the GST fusions,
in vitro expression of labeled proteins, and GST pull-down experiments were done
as described (Jimenez et al., 1997). Ponceau staining was performed on gels to confirm
that all GST fusions were expressed similarly. BATH-42p:GFP was generated
following PCR amplification of a 1.1 kb fragment from cosmid T06F1, which was
then cloned into the BamHI and SalI sites of pPD95.75 (Fire et al., 1990). This
fragment covers the intergenic region between BATH-42 and C50C3.6 lying upstream
to BATH-42. For dsRNA studies, the 1.3kb EcoRI-BglII fragment of BATH-42 from
the two-hybrid isolated clone was cloned into the pPD129.36 plasmid, enabling
dsRNA expression in bacteria; controls contained vector alone (Fire et al., 1998). For
overexpression studies, the coding region of BATH-42 was amplified (and inserted
into plasmid pPD49.78 downstream of the hsp16-2 promoter (Jones et al., 1986; Fire
et al., 1990). RIC-3 overexpression was achieved by injecting 5 ng/μl or 20 ng/μl of
a functional RIC-3:GFP transgene into ric-3(md1181) animals [GFP was inserted inframe into the SalI site in a ric-3 rescuing clone, this clone rescues ric-3(lf) function
(Halevi et al., 2002)]. RIC-3 TM and RIC-3 ΔN transgenics contain the RIC-3 TM
or the RIC-3 minimal sequence as previously described (Cohen Ben-Ami et al., 2005)
upstream and in-frame of the GFP-coding sequence and between the noncoding
regulatory regions of the ric-3 rescuing clone (Halevi et al., 2002). These RIC-3
transgenes were crossed into a strain also containing the hsp-16-2p:BATH-42
transgene and the ric-3(md1181) mutation for RIC-3 TM.
Immunohistochemistry
The C-terminus of RIC-3 fused to GST (see above) was purified as described above,
and used to immunize rabbits. Immunohistochemistry was done after picric acid
fixation as previously described (Yassin et al., 2001), using 1:1000 dilution of the
anti-RIC-3 antibody and 1:250 dilution of Alexa Fluor 488 anti-rabbit antibodies
(Invitrogen 48619A). For double staining, we used 1:150 dilution of anti-DEG-3
(Yassin et al., 2001) and anti-GFP (Roche Applied Sciences; 1:50,000). Before
staining, RIC-3 antibodies were incubated overnight with ric-3(md1181) animals to
812
Journal of Cell Science 122 (6)
remove non-specific antibodies. Images were recorded using a CCD camera
(Hamamatsu ORCA ER) and a SimplePCI image-acquisition and analysis program
(Compix). Intensity of staining is average intensity per pixel for the vulval muscles
normalized to the average intensity per pixel in an adjacent region within the same
animal containing body wall muscles and ventral cord neurons. Coefficient of variation
is the s.d. divided by the mean intensity of staining within the PVD.
We thank Lionel Pintard for the cul-3 dsRNA feeding vector
(pLP36), Shohei Mitani, Japan National Bioresource Project, for the
bath-42(lf) mutant, Robert J. Barstead for the two-hybrid library, Andy
Fire for reporter and overexpression and dsRNA expression vectors,
Yona Kassir for yeast strains, and Zeev Paroush for clones used in and
advice on GST pull-downs. This research was supported by the Israel
Science Foundation (grant No. 379/06).
Journal of Cell Science
References
Biala, Y., Liewald, J. F., Cohen Ben-Ami, H., Gottschalk, A. and Treinin, M. (2008).
The conserved RIC-3 coiled-coil domain mediates receptor-specific interactions
with nicotinic acetylcholine receptors. Mol. Biol. Cell [Epub ahead of print]
doi:10.1091/mbc.E08-08-0851.
Blount, P. and Merlie, J. P. (1991). BIP associates with newly synthesized subunits of
the mouse muscle nicotinic receptor. J. Cell Biol. 113, 1125-1132.
Castillo, M., Mulet, J., Gutierrez, L. M., Ortiz, J. A., Castelan, F., Gerber, S., Sala,
S., Sala, F. and Criado, M. (2005). Dual role of the RIC-3 protein in trafficking of
serotonin and nicotinic acetylcholine receptors. J. Biol. Chem. 280, 27062-27068.
Chang, W., Gelman, M. S. and Prives, J. M. (1997). Calnexin-dependent enhancement
of nicotinic acetylcholine receptor assembly and surface expression. J. Biol. Chem. 272,
28925-28932.
Cheng, A., McDonald, N. A. and Connolly, C. N. (2005). Cell surface expression of 5hydroxytryptamine type 3 receptors is promoted by RIC-3. J. Biol. Chem. 280, 2250222507.
Christianson, J. C. and Green, W. N. (2004). Regulation of Nicotinic receptor expression
by the ubiquitin-proteasome system. EMBO J. 23, 4156-4165.
Cohen Ben-Ami, H., Yassin, L., Farah, H., Michaeli, A., Eshel, M. and Treinin, M.
(2005). RIC-3 affects properties and quantity of nicotinic acetylcholine receptors via a
mechanism that does not require the coiled-coil domains. J. Biol. Chem. 80, 28053-28060.
Fire, A., Harrison, S. W. and Dixon, D. (1990). A modular set of lacZ fusion vectors for
studying gene expression in Caenorhabditis elegans. Gene 93, 189-198.
Fire, A., Xu, S., Montgomery, M. K., Kostas, S. A., Driver, S. E. and Mello, C. C.
(1998). Potent and specific genetic interference by double-stranded RNA in
Caenorhabditis elegans. Nature 391, 806-811.
Halevi, S., McKay, J., Palfreyman, M., Yassin, L., Eshel, M., Jorgensen, E. and Treinin,
M. (2002). The C. elegans ric-3 gene is required for maturation of nicotinic acetylcholine
receptors. EMBO J. 21, 1012-1020.
Halevi, S., Yassin, L., Eshel, M., Sala, F., Sala, S., Criado, M. and Treinin, M. (2003).
Conservation within the RIC-3 gene family: effectors of nAChR expression. J. Biol.
Chem. 36, 34411-34417.
Hoppe, T., Cassata, G., Barral, J. M., Springer, W., Hutagalung, A. H., Epstein, H. F.
and Baumeister, R. (2004). Regulation of the myosin-directed chaperone UNC-45 by
a novel E3/E4-multiubiquitylation complex in C. elegans. Cell 118, 337-349.
Jimenez, G., Paroush, Z. and Isch-Horowicz, D. (1997). Groucho acts as a corepressor
for a subset of negative regulators, including Hairy and Engrailed. Genes Dev. 11, 30723082.
Jones, D., Russnak, R. H., Kay, R. J. and Candido, E. P. (1986). Structure, expression,
and evolution of a heat shock gene locus in Caenorhabditis elegans that is flanked by
repetitive elements. J. Biol. Chem. 261, 12006-12015.
Levine, A., Bashan-Ahrend, A., Budai-Hadrian, O., Gartenberg, D., Menasherow,
S. and Wides, R. (1994). Odd Oz: a novel Drosophila pair rule gene. Cell 77, 587598.
Merlie, J. P. and Lindstrom, J. (1983). Assembly in vivo of mouse muscle acetylcholine
receptor: identification of a subunit species that may be an assembly intermediate. Cell
34, 747-757.
Petroski, M. D. and Deshaies, R. J. (2005). Function and regulation of cullin-ring ubiquitin
ligases. Nat. Rev. Mol. Cell. Biol. 6, 9-20.
Pintard, L., Willis, J. H., Willems, A., Johnson, J. L., Srayko, M., Kurz, T., Glaser,
S., Mains, P. E., Tyers, M., Bowerman, M. and Peter, M. (2003). The BTB protein
MEL-26 is a substrate-specific adaptor of the CUL-3 ubiquitin-ligase. Nature 425, 311316.
Salinas, G. D., Blair, L. A. C., Needleman, L. A., Gonzales, J. D., Chen, Y., Li, M.,
Singer, J. D. and Marshall, J. (2006). Actinfilin is a Cul3 substrate adaptor, linking
GluR6 kainate receptor subunits to the ubiquitin-proteasome pathway. J. Biol. Chem.
281, 40164-40173.
Schaefer, H. and Rongo, C. (2006). KEL-8 is a substrate receptor for CUL3-dependent
ubiquitin ligase that regulates synaptic glutamate receptor turnover. Mol. Biol. Cell 17,
1250-1260.
Stefancsik, R., Jha, P. K. and Sarkar, S. (1998). Identification and mutagenesis of a
highly conserved domain in troponin T responsible for troponin I binding: Potential role
for coiled coil interaction. Proc. Natl. Acad. Sci. USA 95, 957-962.
Treinin, M. and Chalfie, M. (1995). A mutated acetylcholine receptor subunit causes
neuronal degeneration in C. elegans. Neuron 14, 871-877.
Waggoner, L. E., Dickinson, K. A., Poole, D. S., Tabuse, Y., Miwa, J. and Schafer, W.
R. (2000). Long-term nicotine adaptation in Caenorhabditis elegans involves PKCdependent changes in nicotinic receptor abundance. J. Neurosci. 20, 8802-8811.
Williams, M. E., Burton, B., Urrutia, A., Shcherbatko, A., Chavez-Noriega, L. E.,
Cohen, C. J. and Aiyar, J. (2005). Ric-3 promotes functional expression of the nicotinic
acetylcholine receptor alpha7 subunit in mammalian cells. J. Biol. Chem. 280, 12571263.
Wood, W. B. (1988). The Nematode Canorhabditis Elegans. Cold Spring Harbor, NY:
Cold Spring Harbor Laboratory Press.
Xu, L., Wei, Y., Reboul, J., Vaglio, P., Shin, T. H., Vidal, M., Elledge, S. J. and Harper,
J. W. (2003). BTB proteins are substrate-specific adaptors in an SCF-like modular
ubiquitin ligase containing CUL-3. Nature 425, 316-321.
Yassin, L., Gillo, B., Kahan, T., Halevi, S., Eshel, M. and Treinin, M. (2001).
Characterization of the DEG-3/DES-2 receptor: a nicotinic acetylcholine receptor that
mutates to cause neuronal degeneration. Mol. Cell. Neurosci. 17, 589-599.
Yi, J. J. and Ehlers, M. D. (2005). Ubiquitin and protein turnover in synapse function.
Neuron 47, 629-632.