Cell Excitability Necessary for Male Mating Behavior
in Caenorhabditis elegans Is Coordinated
by Interactions Between Big Current
and Ether-A-Go-Go Family K+Channels
Brigitte LeBoeuf and L. Rene Garcia1
Howard Hughes Medical Institute, Department of Biology, Texas A&M University, College Station, Texas 77843
ABSTRACT Variations in K+channel composition allow for differences in cell excitability and, at an organismal level, provide flexibility
to behavioral regulation. When the function of a K+channel is disrupted, the remaining K+channels might incompletely compensate,
manifesting as abnormal organismal behavior. In this study, we explored how different K+channels interact to regulate the neuro-
muscular circuitry used by Caenorhabditis elegans males to protract their copulatory spicules from their tail and insert them into the
hermaphrodite’s vulva during mating. We determined that the big current K+channel (BK)/SLO-1 genetically interacts with ether-a-go-
go (EAG)/EGL-2 and EAG-related gene/UNC-103 K+channels to control spicule protraction. Through rescue experiments, we show
that specific slo-1 isoforms affect spicule protraction. Gene expression studies show that slo-1 and egl-2 expression can be upregulated
in a calcium/calmodulin-dependent protein kinase II-dependent manner to compensate for the loss of unc-103 and conversely, unc-
103 can partially compensate for the loss of SLO-1 function. In conclusion, an interaction between BK and EAG family K+channels
produces the muscle excitability levels that regulate the timing of spicule protraction and the success of male mating behavior.
through cross talk by the various molecular signaling compo-
nents. This signaling is essential for fine-tuning cell excitabil-
ity levels to generate appropriate responses and is partially
accomplished by a variety of K+channels. Organisms have
a large number of K+channels that have multiple isoforms
and ancillary subunits (Salkoff et al. 2006; Torres et al. 2007;
Fodor and Aldrich 2009). In excitable cells, a few K+channel
types are responsible for the maintenance of depolarization;
the rest are proposed to be modifiers, fine-tuning the partic-
ular cell or cell type for its specific function (Bargmann 1998;
Santi et al. 2003).
The importance of these modifiers is highlighted by the
multitude of human disorders that exist as a result of
defective K+channels. These disorders include epilepsy,
which can be caused by dysfunction of a variety of channels
HE neuronal and muscular networks that execute be-
havioral responses to stimuli are ultimately maintained
(Schmitt et al. 2000; Du et al. 2005), long Q-T syndrome,
one form of which can be caused by mutations in the HERG
gene and leads to cardiac arrhythmias and sudden death
(Sanguinetti et al. 1995), hypertension caused by kidney
malfunction (Grimm and Sansom 2010), and erectile dys-
function (Werner et al. 2005).
The K+channels, which fine-tune cell excitability, need to
work together. When there is a defective or absent channel,
the remaining K+channels might compensate in an incom-
plete manner. However, the structure and function of the
compensating channels likely differ from the conventional
one, resulting in imperfect regulation of cell excitability. De-
termining how defective K+channels change the dynamics
of the remaining K+channels will provide insights into how
the integration of their functions maintains appropriate ex-
To address how the regulation of cell excitability and
behavior is coordinated by fine-tuning K+channels, we used
Caenorhabditis elegans male mating as a model system. The
stereotyped steps of this behavior include contact response,
backward locomotion and vulval scanning, turning, vulva
sensing, repetitive spicule insertion attempts, complete spicule
Copyright © 2012 by the Genetics Society of America
Manuscript received September 7, 2011; accepted for publication December 3, 2011
1Corresponding author: Department of Biology, 3258 Texas A&M University, College
Station, TX 77843. E-mail: email@example.com
Genetics, Vol. 190, 1025–1041March 2012
penetration, and sperm transfer. Males utilize a pair of copu-
latory spicules to breach the hermaphrodite vulva. The spi-
cules must be maintained inside the male tail prior to
mating and during mating until vulva penetration (Liu and
Sternberg 1995). We have previously identified two K+chan-
nels, ether-a-go-go (EAG) and EAG-related gene (ERG), that
inhibit protraction until the vulva has been breached (Garcia
and Sternberg 2003; LeBoeuf et al. 2007). However, these
two K+channels do not account for all the regulation neces-
sary to coordinate spicule protraction. In this article we iden-
tify the big current (BK) SLO1 K+channel as an additional
regulator of male sex muscle excitability. Since these three K+
channels function in the mating circuit to regulate behavior,
we explored their effects on each other and elucidated how
their interplay maintains the cell excitability that allows
a male to successfully sire progeny.
Materials and Methods
Strains and culture methods
The following strains were used in this study: unc-103(n1213)
(Park and Horvitz 1986), unc-103(sy557) (Garcia and Stern-
berg 2003), and pha-1(e2123) (Schnabel and Schnabel 1990)
on LGIII; unc-43(sy574) (LeBoeuf et al. 2007) on LGIV; him-5
(e1490) (Hodgkin et al. 1979), slo-1(js379) (Wang et al.
2001), slo-1(rg432) (this work), egl-2(rg4) (LeBoeuf et al.
2007), and egl-2(n693) (Reiner et al. 1995) on LGV. C. elegans
is a naturally hermaphroditic species with a low incidence of
males; to generate a larger percentage of males, strains in this
study contained the him-5(e1490) allele. Animals were main-
tained on NGM plates seeded with Escherichia coli strain OP50
at 20? (Brenner 1974).
Protraction constitutive assay
A total of 20–30 virgin L4 males were isolated on NGM plates
with OP50. The males were allowed to develop into adults
overnight and scored as positive for spicule protraction if at
least one spicule partially extended from the cloaca. P values
were determined using GraphPad Prism (version 4.03).
Identification of the slo-1(rg432) mutation
We sequenced all of slo-1, including the promoter region,
from the genomic DNA of an unc-103(n1213); egl-2(rg4)
strain with a lower-than-normal instance of the protraction
constitutive (Prc) phenotype. We discovered that rg432
changes the sequence TGACATTTATATTATCATTTT to TGA-
CATTTATTTTATCATTTT in the intronic region between exons
11 and 12 (exons numbered according to Wang et al. 2001).
pBK1 contains the slo-1 promoter and isoform slo-1(A2;B0;
C1;D0) tagged with GFP and was provided by Michael No-
net (Washington University, St. Louis). To express slo-1(A2;
B0;C1;D0):GFP in specific tissues, we removed the slo-1 pro-
moter from pBK1 and added the XbaI restriction site in front
of the cDNA via single-site mutagenesis with primers
FpBK1XbaI and pBK1r to create plasmid pBL161. pBL161
was then cut with XbaI, blunt-ended, and Gateway vector
conversion reading frame cassette (RfC) A (Invitrogen,
Carlsbad, CA) was ligated into the site to create plasmid
pBL176. The tissue-specific constructs were created using
LR clonase (Invitrogen) to recombine the pBL176 destina-
tion vector with the vectors containing the tissue-specific
promoters. pBL176 was recombined with pLR35 (Paex-3,
pan-neuronal) (LeBoeuf et al. 2007), pLR21 (Punc-103E,
sex muscles) (Reiner et al. 2006), and pLR92 (Pacr-8,
body-wall muscles) (LeBoeuf et al. 2007) to create plasmids
pBL179, pBL180, and pBL181, respectively.
We cloned slo-1(A2;B0;C0;D0) into plasmid pSX322YFP
to generate pBL185cc. slo-1(A2;B0;C0;D0) was cloned from
PCR-amplified cDNA using primers Fslo1cDNASphI and
Rslo1cDNAXbaI. The cDNA was generated from poly(A)
RNA using Qiagen’s (Valencia, CA) Oligotex mRNA Mini
kit and Invitrogen’s SuperScript First-Strand Synthesis sys-
tem for RT–PCR. Both the plasmid and PCR product were
cut with SphI and XbaI and ligated together to generate
pBL185cc. In this manner, we cloned nine different slo-1
full-length cDNA isoforms, and we discovered a new exon
located between exons 2 and 3. We cut pBL185cc with NheI
and ligated Gateway RfC A (Invitrogen) in front of slo-1(A2;
B0;C0;D0) to create plasmid pBL192. pBL192 contained
point mutations in the slo-1 cDNA that changed amino acids.
To correct the slo-1 sequence, we cut pBL192 and a plasmid
containing slo-1(A2;B0;C0;D0) provided by Brandon Johnson
and Miriam Goodman (Stanford University, Palo Alto, CA)
with BseRI. The short fragment from digested slo-1(A2;B0;
C0;D0) was ligated to the long fragment from digested
pBL192 to create pBL224. We recombined a PCR fragment
of the slo-1 promoter generated with primers Fslo1p and
Slo1pr from N2 genomic DNA with pDG15 using BP clonase
(Invitrogen) to generate plasmid pBL153 (the same promoter
region reported in Wang et al. 2001). We performed an LR
reaction between pBL153 and pBL224 to generate plasmid
A plasmid containing a Gateway RfC, slo-1(A1;B0;C0;
D0), and GFP was created by cutting pBL176 and a plasmid
containing slo-1(A1;B0;C0;D0) provided by Miriam Good-
man with StuI and HpaI. The long restriction enzyme prod-
uct from cutting pBL176 was ligated to the short restriction
enzyme product from slo-1(A1;B0;C0;D0) to generate
pBL198. pBL198 was recombined with pLR21 and pBL153
via LR clonase to create plasmids pBL204 and pBL198,
Plasmids containing mutated unc-103 genomic DNAwere
generated using single-site mutagenesis on plasmid pLR67,
which contains unc-103 genomic DNA plus Gateway RfC C.1
(Invitrogen) (Reiner et al. 2006). The primers psy557A and
phosrevsy557A were used to create pLR74 containing the
sy557A mutation, and the primers psy557B and phosrev-
sy557B were used to create pLR75 containing the sy557B
B. LeBoeuf and L. R. Garcia
mutation. pLR74 was recombined with pLR21 and pLR28
(Punc-103F, sex neurons) (Reiner et al. 2006) using LR clo-
nase to create plasmids pBL35 and pBL36, respectively.
pLR75 was recombined with pLR21 and pLR28 using LR
clonase to create plasmids pBL34 and pBL37, respectively.
pTG44 containing the unc-103E promoter driving egl-2
cDNA was constructed as previously described (LeBoeuf
et al. 2007). We added the n693gf mutation by performing
single-site mutagenesis on pTG44 with primers fegl2n698gf
and regl2698gf to create plasmid pBL111 (LeBoeuf et al.
pBL160 was created via an LR reaction between pBL153
(Pslo-1) and pLR186 (Invitrogen Gateway RfC C.1:DsRed1-
E5) (LeBoeuf et al. 2011).
Plasmids containing the construct of interest were injected
into 1-day-old adult hermaphrodites of the appropriate
strain following standard procedures (Mello et al. 1991).
Either 50 ng/ml pBL66 or 100 ng/ml pBX1 were used as
a transgenic marker (Granato et al. 1994; LeBoeuf et al.
2007). pUC18 was used as carrier DNA to complete each
injection mixture to a final concentration of 200 ng/ml. For
each plasmid injected, concentrations were as follows: 20
ng/ml was used for pBL34, pBL35, pBL36, pBL37, pBK1,
pBL180, pBL179, pBL181, and pBL199 and 50 ng/ml was
used for pBL111, pBL160, pBL226, and pBL204. Once stable
transgenic lines were obtained, males from two or more in-
dependent transgenic lines were scored for the Prc pheno-
type following the procedure previously described.
To obtain a stable expression line to measure fluorescent
levels in males carrying Pslo-1:DsRed1-E5, the transgene
was integrated using trimethylpsoralen following standard
procedures (Anderson 1995) to create transgenic line rgIs2.
Image acquisition and quantification
Males carrying the rgIs2 transgene were isolated at mid-L4.
Images were either taken immediately, or the males were
allowed to develop into adults on NGM plates containing E.
coli OP50 for 1 or 2 days. Males were immobilized on 10%
Noble agar (dissolved in water) pads with Polybead poly-
styrene 0.1-mm microspheres (Polysciences, Warrington,
PA) and covered with a cover slip. Images were taken using
an Olympus BX51 microscope with a ·40 objective. The
image signal was split into two channels using the Dual View
Simultaneous Imaging system by Photometrics (Tucson, AR).
Pictures were taken by the Hamamatsu EM-CCD digital cam-
era ImagEM using HC Image (version 18.104.22.168) (Bridgewater,
NJ). All images were taken with the gain, sensitivity, and
exposure time held constant. The one exception is at the
2-day time point, when the red channel was overexposed at
the standard settings. One image was taken at the standard
settings and a second image was taken when the gain was
lowered so the red channel was no longer overexposed. Data
from both images were used to obtain the mean gray level for
the red channel (see below).
Images were analyzed using Hamamatsu Simple PCI
(version 22.214.171.124). A region of interest (ROI) was placed over
the male tail in both the red and green channels and in the
background of both the red and green channels to get the
background fluorescent level. The ROI was held to a constant
size using the “ROI clone” tool. The mean gray level was
recorded for each ROI, and the background fluorescence
was subtracted. For 2 day-old adult males, the mean gray
level was determined for the green channel under standard
settings as well as both the green and red channel with the
gain lowered. Since the green channel was not overexposed
in either image, a ratio between the high gain and low gain
image was obtained. This ratio was then combined with the
low gain red mean gray level to approximate the standard
gain red mean gray level. The analyzed images were not
modified in any way. Statistical analysis was performed using
GraphPad Prism (version 4.03).
Quantitative real-time PCR
Each strain of worms was cultivated on six 100-mm NGM
plates containing E. coli OP50. Worms were harvested be-
fore they starved by washing the plates with M9 buffer.
Worms were collected at 1000 rpm and the M9 and E. coli
were removed. The worms were washed with an additional
1 ml of M9 to remove additional E. coli. For experiments
where hermaphrodites were segregated from males, worms
were first placed on a 20-mM Nitex nylon filter to remove
larva, and then on a 35-mM Nitex nylon filter to separate
males and hermaphrodites (Sefar Filtration, Depew, NY).
Worms were divided into four 1.5-ml microcentrifuge tubes
and 250 ml Tri Reagent (Sigma-Aldrich, St. Louis) and 0.5
mm zirconium oxide beads (Next Advance, Cambridge, MA)
were added. The Bullet Blender (Next Advance) set for
3 min at speed 8 was used to break open the worms. The
debris, Tri Reagent, and beads were spun down for 1 min at
top speed in a microcentrifuge. The Tri Reagent and RNA
were moved to a fresh RNase-free 0.5-microcentrifuge tube.
Ten microliters of glycogen and 50 ml of chloroform were
added, and the mixture was shaken by hand and centrifuged
for 10 min at 12,000 rpm to separate the layers. The aque-
ous layer was transferred to a fresh 0.5-RNase-free micro-
centrifuge tube. Two ethanol precipitations were performed
to remove any salts that might interfere with the ability to
separate poly(A) RNA from total RNA. The quality of the
total RNA was determined using a 2% agarose gel and the
quantity was measured using a spectrophotometer. Poly(A)
RNA was obtained using Ambion’s Poly(A) Purist kit (Ap-
plied Biosystems, Austin, TX). cDNA was made using the
SuperScript First-Strand Synthesis system for RT–PCR (Invi-
trogen) with 200 ng of starting poly(A) RNA from each
sample and replicated at least one time. Reference genes
were determined as done in Vandesompele et al. (2002)
and Hoogewijs et al. (2008). Thirteen reference genes
were considered: act-1, gpd-3, F23B2.13, rrn-1.1, cdc-42,
Overlapping K+Channels in Mating
Y45F10D4F.3, ama-1, csq-1, eif-3.C, mdh-1, gpd-2, pmp-3,
and tba-1. geNorm was used to identify the most stable ref-
erence genes (http://medgen.ugent.be/~jvdesomp/genorm/).
The two most stable reference genes for comparison across
K+channel mutants were act-1 and mdh-1. Amplification
was performed using Bio-Rad’s ssoFast EvaGreen on a Bio-
Rad CFX96 Real-Time Detection system (Hercules, CA).
act-1 and egl-2 were amplified using the primers reported
in LeBoeuf et al. (2011). mdh-1, cdc-42, and pmp-3 were
amplified using the primers reported in Hoogewijs et al.
(2008). unc-103 was amplified using primers 2qPCRunc-
103F and 2qPCRunc-103R. slo-1 was amplified using pri-
mers F1-qPCRslo1ex11 and qPCRslo1ex12-1R. The primers
for slo-1 A1 were Fslo1a1ex9 and Slo1a1ex9r. The primers
for slo-1 A2 were Fslo1a2ex10 and Slo1a2ex10r. The pri-
mers for slo-1 B0 were Fslo1b0 and Slo1b0ex14r. The pri-
mers for slo-1 B1 were Fslo1b1 and Slo1b1r. The primers for
slo-1 C0 were Fslo1c0 and Slo1c0r. The primers for slo-1 C1
were Fslo1c1 and Slo1c0r. The primers for slo-1 D0 were
Fslo1d0 and Slo1d0r. The primers for slo-1 D1 were Fslo1d0
and Slo1d1r2. The primers for acr-18 were qPCRacr-18F
and qPCRacr-18R. The primers for gar-3 were 2qPCRgar-
3F and 2qPCRgar-3R. The primers for unc-29 were
qPCRunc-29F and qPCRunc-29R. The primers for unc-43
were qPCRunc-43F and qPCRunc-43R. The primers for
cat-2 were qPCRcat-2F and qPCRcat-2R. The primers for
let-363 were qPCRlet-363F and qPCRlet-363R. The primers
for sir-2.1 were 2qPCRsir-2.1F and 2qPCRsir-2.1R. Samples
were diluted 1:10 for all K+channel sequences, 1:1000 for
mdh-1, and 1:10,000 for act-1. All C(t) values were between
20 and 30. Each sample was done in triplicate, and each
experiment was repeated at least once from newly made
cDNA. Expression levels of each gene of interest were nor-
malized to act-1 and mdh-1 and then reported as relative
gene expression levels using CFX Manager Software Data
Analysis (Bio-Rad version 1.6) according to the CTmethod
(Livak and Schmittgen 2001). Statistical analysis was per-
formed using GraphPad InStat (version 3.06). The results
reported in Figure 3 are from one representative experiment
for each gene tested.
The primer designations are as follows:
Fslo1cDNASphI: 59- CGCGGCATGCGCTAGCATGGGCGAGA
Rslo1cDNAXbaI: 59- CGCGTCTAGAAAAGTGTCGTTTGCCC
A BK channel/slo-1 mutation results in increased male
mating circuit excitability
C. elegans males possess copulatory spicules that must be
held inside their tail until the hermaphrodite vulva is
breached. Protractor and retractor muscles are attached to
the base of the spicules and control their proper position.
Once the appropriate mating cues have been received and
integrated by both neurons and muscles, the protractors
contract, forcing the spicules out of the tail and into the
vulva (Sulston et al. 1980; Liu and Sternberg 1995; Liu
et al. 2011). Permanent spicule protraction in the absence
of mating cues interferes with successful sperm transfer and
thus reproduction. This abnormal spicule phenotype is re-
ferred to as Protraction constitutive (Prc) and arises when
mechanisms that tightly regulate the spicule protraction cir-
cuit begin to fail. The percentage of Prc males in a population
B. LeBoeuf and L. R. Garcia
is determined by isolating virgin larval males from hermaph-
rodites. The males are allowed to mature to adulthood, and
the phenotype is scored 18 to 24 hr later. A male that dis-
plays at least partial protraction of one spicule is designated
as displaying the Prc phenotype. We previously identified
two members of the ether-a-go-go (EAG) K+channel family,
EAG and ether-a-go-go–related gene (ERG), that regulate
spicule protraction (Garcia and Sternberg 2003; LeBoeuf
et al. 2007). In C. elegans, EAG is encoded by egl-2 and in
the male spicule protraction circuit is expressed in sex
muscles (Weinshenker et al. 1999); males with a null mu-
tation in the K+channel display no obvious mating defects (0%
of the males display the Prc phenotype) (Table 1) (LeBoeuf
et al. 2007). ERG is encoded by unc-103 and is expressed in
nearly all muscles and neurons (Reiner et al. 2006). In contrast
to egl-2 mutants, males with a null mutation in unc-103 display
spicule protraction defects (41% of the males display the Prc
phenotype) (Table 1) (Garcia and Sternberg 2003). Mutating
both unc-103 and egl-2 voltage-dependent K+channel genes
induce the Prc phenotype in 83% of males (Table 1) (LeBoeuf
et al. 2011). While the penetrance of the Prc phenotype is much
higher in the double mutant males than either single mutant,
not all males in a population lacking both K+channels dis-
played the Prc phenotype. This led us to ask whether other
K+channels are involved in maintaining cell excitability.
One likely candidate is the voltage- and calcium-activated
BK channel slo-1, which is broadly expressed in neurons and
muscles (Wang et al. 2001; Kim et al. 2009). slo-1 loss-of-
function mutants exhibit many phenotypes, including im-
paired movement caused by defective neurotransmitter re-
lease, alcohol resistance, and muscle degeneration (Wang
et al. 2001; Davies et al. 2003; Carre-Pierrat et al. 2006; Liu
et al. 2007a). We found that 70% of males with an early
nonsense mutation in slo-1 [allele js379, hereafter referred
to as slo-1(lf); Wang et al. 2001] spontaneously protract their
spicules in the absence of mating cues (Table 1), demonstrat-
ing that like EAG and ERG-like K+channels, the BK channel
also controls the excitability of male sex circuit components.
slo-1 has been shown in C. elegans and other systems to
interact with calcium/calmodulin-dependent protein kinase
II (CaMKII) (Hawasli et al. 2004; Liu et al. 2006, 2007a;
LeBoeuf et al. 2007). unc-43 encodes CaMKII in C. elegans
and regulates movement, defecation, and egg laying in ad-
dition to its role in male mating (Reiner et al. 1999). We
previously reported that loss-of-function mutations in CaM-
KII cause C. elegans males to display the Prc phenotype. The
sy574 allele in unc-43 changes a glycine to a glutamate in
the putative substrate binding site of CaMKII. A total of 38%
of unc-43(sy574) males displayed the Prc phenotype but
exhibit no other obvious defects (Table 1) (LeBoeuf et al.
2007). To determine whether the frequency of the unc-43
(sy574)–mediated Prc phenotype was related to reduced
SLO-1 function, we created a double mutant containing slo-
1(lf) and unc-43(sy574) to ask whether the K+channel ge-
netically acts downstream of CaMKII. In contrast to the single
mutant males, we found that the incidence of the Prc pheno-
type in double mutant males was 96% [P value ,0.0001 to
slo-1(lf)] (Table 1). This synergistic effect suggests that
while slo-1 might be a target of unc-43 in other behavioral
contexts, it is not the only CaMKII target in the male mat-
Specific isoforms of sex muscle-expressed BK
channel/slo-1 can reduce abnormal protraction
Since slo-1 is involved in regulating the timing of male sex
muscle contractions, we asked where it functions in the
mating circuit. We obtained a construct (Pslo-1:slo-1(A2;
B0;C1;D0)::GFP) containing the slo-1 isoform previously
used to restore function in hermaphrodites with movement
defects (Wang et al. 2001). slo-1 has four identified splice
includes the exons at A2 and C1 (Figure 1) (Johnson et al.
2011). Splice sites A–C were previously identified (Wang
et al. 2001), whereas we identified D during cloning of
full-length slo-1 cDNAs (Figure 1C) (Materials and Methods).
We introduced Pslo-1:slo-1(A2;B0;C1;D0)::GFP into slo-1(lf)
worms and discovered that spicule protraction was reduced
from 61 to 38% (P ¼ 0.02) (Table 2). We then generated
tissue-specific constructs to express slo-1(A2;B0;C1;D0) in
neurons (Paex-3), body-wall muscles (Pacr-8), and sex muscles
(Punc-103E) (Reiner et al. 2006; LeBoeuf et al. 2007). slo-1
(A2;B0;C1;D0) expression in neurons and body-wall muscles
did not reduce the Prc phenotype, whereas expression in the
sex muscles reduced the incidence of spontaneous protraction
Table 1 Spicule protraction induced by mutations in K+channels
Genotype % spicule protracted (n)P value, Fisher’s exact test
unc-103(0); egl-2(0) slo-1(lf)
0.0005 to unc-103(sy557)
,0.0001 to unc-103(0)
,0.0001 to slo-1(lf)
,0.05 to slo-1(lf)
,0.0001 to slo-1(lf)
0.7 to slo-1(lf)
Overlapping K+Channels in Mating
to 23% [P ¼ 0.0005 to slo-1(lf)] (Table 2). Thus, slo-1 is
partially regulating spicule protraction in the sex muscles.
Though slo-1(A2;B0;C1;D0) significantly reduces the in-
stance of spontaneous protraction, a large percentage of the
males (38%) expressing this isoform still displayed the mu-
tant phenotype (Table 2). slo-1(A2;B0;C1;D0) is one isoform
of at least 14 that exist in C. elegans (Wang et al. 2001;
Johnson et al. 2011). We asked whether another isoform
is capable of complete rescue of the mutant phenotype.
We tested two additional isoforms of slo-1, slo-1(A2;B0;C0;
Figure 1 slo-1 gene structure and splice
variants used in this study. (A) Boxes in-
dicate relative exon sizes and location. A
letter is used to designate an alternatively
spliced region; a number designates an
alternatively spliced site. A letter followed
by 0 indicates where splicing occurs when
the alternative exon is not included. For
splice region A, either the exon at A1 or
A2 is included in all isoforms. For splice
regions B–D, the exons are included in
some isoforms but not in others. The
shaded box indicates the exon newly de-
scribed in this work; when present in an
isoform, it is spliced between previously
described exons 2 and 3. The open boxes
indicate the specific exons differentially
spliced in the three isoforms analyzed in
this study. (B) Comparison of amino acid
sequences for exons 9 (splice site A1) and
10 (splice site A2). The open boxes indi-
cate amino acids that differ between the
two exons. (C) The amino acid sequence
for the newly described exon at splice site
D1. (D) The sequence change for the mu-
tant slo-1(rg432) located in an intron. The
top letter is the wild-type sequence, while
the bottom letter is the mutant sequence.
Table 2 Transgenic rescue of mutant BK/slo-1 and ERG/unc-103–induced spicule protraction
expression % spicule protracted (n)
Fisher’s exact test
Most neurons and muscles
Most neurons and muscles
Most neurons and muscles
0.02 to slo-1(lf)
0.0005 to slo-1(lf)
,0.0001 to slo-1(lf)
,0.0001 to slo-1(lf)
,0.0001 to slo-1(lf)
,0.0001 to unc-103(0)
Sex muscle and sex neurons
Sex muscle and sex neurons 62 (52) 0.008 to unc-103(0)
Sex muscle and sex neurons
,0.0001 to slo-1(lf) egl-2(0)
,0.0001 to slo-1(lf) egl-2(0)
aunc-103(0) animals contain pha-1(e2123).
B. LeBoeuf and L. R. Garcia
D0) and slo-1(A1;B0;C0;D0). slo-1(A2;B0;C0;D0) does not
have the exon at splice site C1 and was first reported in
Wang et al. (2001). slo-1(A1;B0;C0;D0) exchanges one con-
served hydrophobic region 38 amino acids in length in the C
terminus with another 38-amino-acid sequence of high ho-
mology (Figure 1, A and B). When expressed from the slo-1
promoter, both slo-1(A2;B0;C0;D0) and slo-1(A1;B0;C0;D0)
reduced spicule protraction (from 61% to 5 and 0%, respec-
tively) (Table 2). Additionally, tissue-specific expression of
slo-1(A1;B0;C0;D0) in the sex muscles via the unc-103E pro-
moter resulted in a 9% instance of the Prc phenotype [P
value ,0.0001, compared to slo-1(lf)] (Table 2). In conclu-
sion, the exon located at C1 reduces slo-1’s ability to inhibit
premature spicule protraction and demonstrates that iso-
form specificity does play a role in restoring male mating
BK channel/slo-1(lf)–induced sex muscle spasms can be
modified by EAG family K+channels
Since mutations in BK/slo-1, ERG-like/unc-103, and EAG/
egl-2 differentially regulate the excitability of the sex
muscles, we asked whether simultaneous loss of multiple
K+channels would have an additive or even a synergistic
effect on male mating circuit excitability. We made double
mutants between slo-1(lf) and strains that carry large dele-
tions of unc-103 and egl-2 [hereafter referred to as unc-103
(0) and egl-2(0)] (Park and Horvitz 1986; Garcia and Stern-
berg 2003; LeBoeuf et al. 2007). Contrary to our initial
expectations, both unc-103(0); slo-1(lf) and egl-2(0) slo-1
(lf) double mutants displayed a decrease in the instance of
the Prc phenotype [unc-103(0); slo-1(lf) and egl-2(0) slo-1
(lf) dropped the frequency of the Prc phenotype to 51 and
30%, respectively, lower than the slo-1(lf)–induced 70% pro-
traction] (Table 1). Thus, removing an ERG-like or an EAG
K+channel along with a BK channel from the male mating
circuit lowers rather than increases sex muscle excitability.
Males lacking all three K+channels, an unc-103(0); egl-2(0)
slo-1(lf) triple mutant, protracted their spicules 79% of the
time, a frequency similar to the slo-1(lf) single mutant and
the unc-103(0); egl-2(0) double mutant (Table 1). This sug-
gests that when either egl-2 or unc-103 is removed from the
slo-1 background, some aspect (transcriptional, translational,
or post-translational) of the remaining K+channels might be
upregulated to compensate for the loss.
The nonadditive effect of combining the slo-1(lf) mutation
with either unc-103(0) or egl-2(0) suggests the existence of
mechanisms that sense the level of cell excitability and
adjusts extant K+channels to partially compensate for miss-
ing or defective K+channels. To address this hypothesis, we
utilized an allele described in an earlier report, a neomorphic
loss-of-function allele (sy557) in the unc-103 ERG-like K+
channel gene. unc-103(sy557) contains two point mutations:
sy557A, a H165N change in the linker region between trans-
membrane domains two and three, and sy557B, a W244R
change in transmembrane domain five (Garcia and Sternberg
2003). unc-103(sy557) males displayed a Prc frequency of
75%, significantly higher than males completely lacking unc-
103 (P ¼ 0.0005, compared to unc-103(0) at 41%) (Table 1),
suggesting that the mutant K+channel gene adversely affects
compensating molecules. Since the frequency of unc-103
(sy557) males displaying the Prc phenotype was similar to
unc-103(0); egl-2(0) double and slo-1(lf) single mutant males,
we hypothesized that the sy557-encoded mutation(s) inter-
feres with the compensating properties of EGL-2 and/or SLO-1.
To address which sy557 mutation confers the neomor-
phic dominant negative properties to UNC-103, we mutated
a construct containing unc-103 genomic DNA to encode ei-
ther the sy557A H165N or sy557B W224R change (Garcia
and Sternberg 2003). We then expressed the mutated unc-
103 sequences in unc-103(0) male sex muscles using pro-
moter unc-103E [Punc-103E:unc-103(sy557A or B)] or sex
(sy557A or B)] to determine whether transgenic mutated
versions of unc-103 can interfere with compensating mech-
anisms and increase the Prc frequency (Reiner et al. 2006).
Neither UNC-103(H165N) nor UNC-103(W224R) expressed
individually in the sex muscles or neurons increased the rate
of spontaneous spicule protraction (Table 2). However,
expressing UNC-103(W224R), but not UNC-103(H165N),
in both sex muscles and neurons significantly increased pro-
traction from 26 to 62% (P ¼ 0.008, Fisher’s exact test)
(Table 2). Thus, disrupting unc-103 via a W244R change
in transmembrane domain five can interfere with additional
neural and muscle-compensating functions to consequently
increase the instance of unregulated spicule protraction.
To determine whether the compensating functions that
UNC-103(W224R) interferes with are related to EGL-2 and/
or SLO-1 activity, we transgenically introduced the neo-
morphic dominant negative UNC-103(W224R) into egl-2(0)
slo-1(lf) double mutant males. We then asked whether the
frequency of the Prc phenotype is similar to unc-103(0); egl-
2(0) slo-1(lf) triple mutants or higher. We hypothesized that
if there is a genetic interaction between wild-type slo-1, egl-2
and unc-103, then UNC-103(W224R) would interfere with
wild-type slo-1 and/or egl-2 activity and the males would
display a Prc frequency of ?70–80%. A testable prediction
of this hypothesis is that if UNC-103(W224R) is transgeni-
cally introduced into egl-2(0) slo-1(lf) double mutants, then
UNC-103(W224R) should only interfere with wild-type unc-
103, since the other two K+channels are already mutant;
consequently the transgenic males should also display a Prc
frequency of ?70–80%. If the hypothesis is incorrect, and
UNC-103(W224R) interferes with something other than egl-
2 and slo-1 function, the transgenic males should display
a Prc frequency greater than the unc-103(0); egl-2(0) slo-1
(lf) triple mutant. As previously stated, egl-2(0) slo-1(lf)
males were 30% Prc, whereas slo-1(lf) males were 70%
Prc (Table 1); this decrease in the Prc frequency in the
double mutant could be due to compensation by unc-103.
We expressed UNC-103(W224R) in male sex muscles and
neurons, individually and together, in egl-2(0) slo-1(lf)
males. UNC-103(W224R) expressed in neurons of egl-2(0)
Overlapping K+Channels in Mating
slo-1(lf) males had no effect on spicule protraction (Table 2).
However, UNC-103(W224R) expressed in the sex muscles
was sufficient to return spontaneous spicule protraction to
levels seen in slo-1(lf) mutants (72% of the males displayed
the Prc phenotype) (Table 2). Expressing UNC-103(W224R)
in both sex muscles and neurons was not significantly differ-
ent from muscles alone (81% of the males displayed the Prc
phenotype) (Table 2). In conclusion, UNC-103(W224R)
appears to interfere with only unc-103 in an egl-2(0) slo-1
(lf) background. Since the instance of spontaneous spicule
protraction was similar to that seen in unc-103(sy557) and
unc-103(0); egl-2(0) slo-1(lf) mutant males, unc-103(sy557)
likely also interferes with egl-2 and slo-1. These data are con-
sistent with the idea that there is a genetic interaction be-
tween slo-1, egl-2, and unc-103.
BK/slo-1 expression is increased in the absence
of ERG/unc-103 but not EAG/egl-2
The genetic experiments discussed in the preceding sections
suggested that BK channel/slo-1 function or expression is
possibly upregulated in the absence of EAG family K+chan-
nels unc-103 and egl-2. To ask whether slo-1 expression in
the male sex muscles might be changed in mutant back-
grounds, we transgenically expressed dsRed1-E5 from the
slo-1 promoter as a proxy of promoter activity. dsRed1-E5
shifts its emission spectra from green to red over time, and
we previously used it to differentiate fluorescent proteins
that were newly expressed from older accumulated proteins
(Terskikh et al. 2000; LeBoeuf et al. 2011). We expressed
integrated Pslo-1:dsRed1-E5 in wild type, unc-103(0), unc-
103(sy557), and egl-2(0) males. We then asked whether
removing the K+channels affected marker gene expression
in the tails over the course of L4 development and 2 days of
adulthood. In L4 wild-type males, we measured the marker
gene expression in the male anal depressor muscle, an ac-
cessory to the spicule protractor muscles. Since the anal de-
pressor muscle differentiates early in the embryo (Sulston
et al. 1983), dsRed1-E5 expression was higher in the red
channel than the green (Figure 2, A and D, P value ,0.05
for red vs. green for all genotypes, Mann–Whitney t-test).
However, the fluorescence intensity difference in the L4 anal
depressor between the green and red channels changed
when unc-103(0) was introduced into the genetic back-
ground (Figure 2A). Pslo-1:dsRed1-E5 expression in unc-
103(0) male tails was increased in both channels relative
to the wild type (Figure 2A). This suggests that in the ab-
sence of unc-103, the transgenic and possibly the endoge-
nous slo-1 promoter is more active and indicates that
more slo-1 gene product is made in an attempt to com-
pensate. In contrast, Pslo-1:dsRed1-E5 expression was un-
changed compared to wild type in egl-2(0) male tails
(Figure 2A). Pslo-1:dsRed1-E5 expression was also un-
changed in unc-103(sy557) males (Figure 2A), suggesting
that unlike the unc-103(0) allele, the presence of this non-
functioning UNC-103 protein does not promote an increase
in slo-1 expression.
In day 1 and day 2 wild-type adult males, we measured
marker gene expression in the fully developed male
protractor muscles and the anal depressor accessory muscle.
Pslo-1:dsRed1-E5 expression in the green channel was higher
than in the red channel, due to the differentiation of the sex
muscles (Figure 2, B and E, P value ,0.05, Mann–Whitney
t-test). There was no difference between the green and red
channels in males lacking EGL-2 or UNC-103 K+channels,
though expression in the red channel was significantly higher
than the green channel in unc-103(sy557) males (Figure 2B,
P value ,0.05, Mann–Whitney t-test). However relative to
wild-type animals, there was a significant increase in marker
gene expression when unc-103 was deleted (Figure 2B). In
contrast, marker gene expression in egl-2(0) males was not
significantly different from the wild type (Figure 2B). There-
fore, similar to L4 males, the deletion of unc-103 caused an
increase in transgenic and possibly endogenous slo-1 pro-
moter activity but the loss of egl-2 had no effect. Pslo-1:
dsRed1-E5 expression in unc-103(sy557) males was signifi-
cantly higher in the red but not the green channel, likely due
to the large variability in expression seen among these males.
Individual males could have differing responses to the loss of
excitability regulation induced by the unc-103(sy557) allele.
In day 2 wild-type adult males, expression in the red
channel was much higher than in the green channel, due to
the accumulation of dsRed1-E5 for 2 days (Figure 2, C and F).
Similar to L4 and 1-day-old adult males, Pslo-1:dsRed1-E5
expression was higher in unc-103(0) mutants but unaffected
in egl-2(0) mutants (Figure 2C). This is further evidence that
a higher rate of slo-1 expression is maintained in unc-103(0)
males as they age. In contrast, Pslo-1:dsRed1-E5 expression is
lower in 2-day-old unc-103(sy557) males (Figure 2C), a re-
versal of the slight increases in expression in the younger
adults. These data suggest that while slo-1 levels might be
similar to wild type in young unc-103(sy557) adult animals,
this rate is not maintained as they age.
Food deprivation can attenuate BK
channel/slo-1(lf)–induced spicule protraction
through EAG K+channel/egl-2
In previous work, we demonstrated that starvation can
suppress the excitability of the mating circuit in ERG-like K+
channel/unc-103 mutants through enhanced compensation
by EAG/egl-2 function (Gruninger et al. 2006; LeBoeuf et al.
2007, 2011). Since we established that there is a genetic
interaction between slo-1, egl-2, and unc-103, we asked
whether food-deprived BK channel/slo-1 mutant males, like
unc-103(0) animals, had a reduced frequency of the Prc
phenotype via enhanced egl-2 compensation. C. elegans
matures through four larval stages (L1–L4) to reach adult-
hood; the male sex muscles differentiate in the last stages of
L4. We starved males by picking them at the late L4 stage
and placed them on NGM plates lacking their food source,
E. coli OP50. Males develop normally under these starvation
conditions (Gruninger et al. 2006). After 18–22 hr, we scored
whether the starved males displayed the spontaneous spicule
B. LeBoeuf and L. R. Garcia
protraction phenotype. We found that 31% of food-deprived
slo-1(lf) males displayed the Prc phenotype, significantly
lower than slo-1(lf) males on food (70% displayed the Prc
phenotype, P value ,0.0001) (Table 3). Thus, like unc-103
mutants, starvation is able to partially suppress slo-1(lf)–
We then asked whether starvation was able to inhibit
protraction in males lacking both slo-1 and egl-2. We found
that depriving egl-2(0) slo-1(lf) males of food had no effect on
the instance of Prc (31 vs. 30%, P ¼ 0.7) (Table 3), consistent
with the idea that egl-2 mediates the effects of starvation on
the excitability of the mating circuit. To support the idea that
in the absence of food, enhanced egl-2 compensation can re-
duce slo-1(lf)–induced excitability defects, we expressed an
egl-2 gain-of-function allele in the male sex muscles and
asked whether the frequency of slo-1(lf)–induced spicule pro-
traction can be further suppressed by starvation.
The egl-2(gf) allele induces egg retention in hermaphro-
dites as well as defecation and chemotaxis defects (Reiner
et al. 1995; Weinshenker et al. 1995; Weinshenker et al.
Figure 2 slo-1 promoter expression increases
in an unc-103(0) mutant background. (A–C)
Mean gray level of DsRed-E5 fluorescent pro-
tein expression in the male tail. DsRed-E5 ini-
tially expresses in the green channel and then
shifts its emission spectra to the red channel. x-
axis indicates the genotype of the male and
fluorescent channel measured, while the y-axis
indicates the amount of fluorescence (mean
gray level) measured in the male tail (A.U., ar-
bitrary units). Error bars represent standard de-
viation. *P value ,0.05, **P value ,0.005,
***P value ,0.0001; Mann–Whitney t-test.
(A) L4 stage males. Wild type n ¼ 29, unc-
103(0) n ¼ 26, egl-2(0) n ¼ 38, and unc-103
(sy557) n ¼ 22. (B) One-day-old males. Wild
type n ¼ 34, unc-103(0) n ¼ 29, egl-2(0) n ¼
21, and unc-103(sy557) n ¼ 23. (C) Two-day-
old males. Wild type n ¼ 33, unc-103(0) n ¼
23, egl-2(0) n ¼ 21, and unc-103(sy557) n ¼
20. (D–F) Fluorescent images of wild-type male
tails. (Left) Green channel. (Right) Red channel.
Bars, 20 mm. Open boxes indicate area used to
determine mean gray level. The male is oriented
so that the top of the image is anterior and the
left is ventral. (D) Mid-L4 male tail (the stage in
L4 when the male tail spike completes its re-
traction). (E) One-day-old adult male tail. (F)
Two-day-old adult male tail.
Overlapping K+Channels in Mating
1999). In the male mating circuit, egl-2(gf) suppresses muscle
excitability when activated by starvation (LeBoeuf et al. 2011).
We made a double mutant containing both slo-1(lf) and egl-2
(gf) and found that the double mutant males were extremely
constipated and in 91% of the males, the proctodeum, includ-
ing the spicules, hemorrhaged from the cloacal opening.
slo-1 has a broad expression pattern in neurons and
muscles outside the male mating circuit, whereas egl-2 is
expressed in a few head neurons and intestinal muscles
(Weinshenker et al. 1999; Wang et al. 2001). The slo-1(lf)
and egl-2(gf) alleles appear to cause a synthetic effect result-
ing in chronic constipation and hemorrhaging. To circum-
vent this problem, we expressed egl-2(gf) cDNA only in the
sex muscles of egl-2(0) slo-1(lf) males using the unc-103E
promoter. A total of 66% of egl-2(0) slo-1(lf); rgEx253
[Punc-103E:egl-2(gf)] males display abnormal spicule pro-
traction, significantly lower than egl-2(gf) slo-1(lf) males
and similar to slo-1(lf) males (91 and 70%, P values
,0.005 and =0.7, respectively) (Table 3). Starvation re-
duced spicule protraction in egl-2(0) slo-1(lf); rgEx253
[Punc-103E:egl-2(gf)] males to 7% (P value ,0.05 to fed)
(Table 3). Thus, egl-2 compensation can be stimulated by
starvation to reduce sex muscle excitability, and this increase
in egl-2 activity can reduce slo-1(lf)–induced muscle spasms.
slo-1(rg432) encodes an intronic point mutation
that reduces sex muscle excitability
As we have previously stated, unc-103(0); egl-2(0) double
mutant males display spontaneously protracted spicules at a
frequency of 83%. In the course of working with the double
mutant, we realized that a spontaneous genetic modifier
(rg432) unknowingly got crossed into the strain. The mod-
ifier reduced the frequency of the unc-103(0); egl-2(0)
double mutant Prc phenotype to 46% [P value ,0.005,
compared to unc-103(0); egl-2(0)] (Table 3). In addition,
we discovered that starvation strongly suppressed the in-
stance of the Prc phenotype for unc-103(0); egl-2(0) animals
that contained the rg432 modifier [P value ,0.005, com-
pared to unc-103(0); egl-2(0)] (Table 3). We roughly map-
ped the genetic modifier (through conventional crosses,
data not shown), and not surprisingly, it mapped close to
slo-1. We sequenced slo-1 from the unc-103(0); egl-2(0)
rg432 line and found an A-to-T change in an intron between
exons 11 and 12 (Figure 1D). This mutation was not present
in the normal unc-103(0); egl-2(0) line. When unlinked to
egl-2(0) and unc-103(0), animals harboring the slo-1(rg432)
allele appeared behaviorally wild type under fed or starved
conditions (Table 3). In double mutant combinations, slo-1
(rg432) had no effect on the frequency of the Prc phenotype
of unc-103(0) or egl-2(0) males (Table 3). Since unc-103(0);
egl-2(0) slo-1(rg432) triple mutant males displayed a similar
Prc phenotype to unc-103(0) single mutant males (46 vs.
41%, P ¼ 0.7, Fisher’s exact test) (Table 3), it was possible
that the slo-1(rg432) allele facilitated increased SLO-1 com-
pensation in the absence of egl-2.
It has recently been reported that a point mutation in
a slo-1 intron can affect alternative splicing (Glauser et al.
2011). Since the slo-1(rg432) allele is located in an intron,
we asked whether alternative splicing was affected. The slo-1
genetic locus contains four alternatively spliced regions, la-
beled A–D, and the rg432 mutation does not map to any of
them (Figure 1D). We designed primers to distinguish be-
tween the differentially spliced exons and performed quanti-
tative RT–PCR on cDNA obtained from wild-type and slo-1
(rg432) nonstarved adult hermaphrodites or males. Total
slo-1 expression was greatly increased from males to her-
maphrodites, likely due to the difference in the number of
excitable cells (sex-specific muscles and neurons) between
the two sexes, and this was true for all isoforms (Figure 3,
A and B). Similar to wild type, slo-1(rg432) males showed
increased expression when compared to mutant hermaphro-
dites, and this effect was not isoform specific (Figure 3, A
and B). slo-1 transcript levels were decreased in slo-1
(rg432) hermaphrodites, with some splice sites being signif-
icantly affected while others were not (Figure 3, A and B).
Importantly, slo-1 mRNA levels were significantly lower in
Table 3 Effects of starvation on K+channel mutation-induced spicule protraction
% spicule protracted males (n)P value
Genotype Fed StarvedFisher’s exact testa
unc-103(0); egl-2(0) slo-1(rg432)
aComparing fed vs. starved of the same genotype.
B. LeBoeuf and L. R. Garcia
slo-1(rg432) males, resulting in a much more pronounced
effect on slo-1 expression levels by the mutant allele in males
vs. hermaphrodites (Figure 3, A and B). Thus, the slo-1
(rg432) allele results in a decrease in slo-1 transcript levels,
which is most profound in males and is not isoform specific in
Stable K+channel transcription is dependent on BK
channel/slo-1 and EAG K+channel/egl-2
Our experiments discussed so far suggest that K+channels
in the male sex muscles can be upregulated to compensate
for one another to control muscle excitability. The dsRed1-
E5 marker gene analysis indicated that increases in gene
expression via mRNA synthesis or stability could be one
method of upregulation. To determine whether K+channel
compensation via mRNA synthesis or stability is more gen-
eral and not just limited to the male spicule muscles, we
measured global relative changes in K+channel mRNA
abundance in wild type and unc-103, egl-2, and slo-1
mutants. We isolated poly(A) mRNA from well-fed mixed
gender and mixed staged whole worm extracts and per-
formed quantitative RT–PCR experiments.
First, we measured global unc-103 and slo-1 transcript
levels in egl-2(0) worms and found that K+channel tran-
script levels were significantly decreased (Figure 4, A, C, and
D). Thus, the relative normalcy of egl-2(0) worms cannot be
explained by an increase in unc-103 and slo-1 transcript
levels. Likewise, unc-103(sy557) worms displayed unaf-
fected (egl-2) or decreased (slo-1 and unc-103) levels of
transcript (Figure 4). However, egl-2 and slo-1 transcript
levels were increased in unc-103(0) worms (Figure 4, B–
D), indicating that gene expression is a possible mechanism
of attempted compensation in these worms. Additionally,
unc-103(sy557) males displayed a higher instance of the
Prc phenotype than unc-103(0) males, a difference that
could be due in part to the lack of K+channel upregulation
in unc-103(sy557) mutants.
We next measured K+channel transcript levels in slo-1
(lf) and slo-1(rg432) worms. While slo-1 transcript levels
were unaffected in slo-1(rg432) mutants, egl-2 and unc-
103 levels were decreased (Figure 4). The slo-1(rg432) al-
lele promotes regulation of male sex muscle excitability, but
it is not due to an increase in transcription or mRNA stability.
slo-1(lf) mutants display a high level of the Prc phenotype,
so it was unsurprising to see that unc-103 and egl-2 levels
were significantly decreased (Figure 4). Also as expected,
the levels were significantly lower than wild type in the
slo-1(lf) mutant containing an early nonsense mutation (Fig-
ure 4, C and D), likely as a result of nonsense-mediated
mRNA decay (Hodgkin et al. 1989; Maquat 2004; Longman
et al. 2007). The primers used to detect slo-1 transcripts start
at amino acid F641, considerably downstream from the
Q251stop mutation in the loss-of-function mutant. There is
little to suggest extensive read-through of the stop mutation,
Figure 3 Variations in male and
hermaphrodite gene transcript
levels. qRT–PCR was performed
on populations of young adult
hermaphrodites and young adult
males in wild type and slo-1
(rg432) backgrounds. Error bars
represent standard deviation. (A
and B) Transcript levels of slo-1
splice variants. (C and D) Tran-
script levels of various genes. (A)
x-axis indicates the slo-1 splice
site tested, while the y-axis indi-
cates the normalized fold expres-
sion. (B) x-axis indicates the gene
tested, while the y-axis indicates
the normalized fold expression.
(A and C) Dark blue bars, wild-
type hermaphrodites; light blue
bars, wild-type males; red bars,
pink bars, slo-1(rg432) males. (A
and C) Data here are reformatted
in table form in B and D for ease
of comparison. (B and D) Let-
ters next to numbers indicate a
P value of ,0.05, unpaired t-test
with Welch’s correction. Foot-
note symbols are as follows:
“a” compares wild-type males
“b” compares wild-type hermaphrodites to slo-1(rg432) hermaphrodites; “c” compares wild-type hermaphrodites to slo-1(rg432) males; “d” compares
wild-type males to slo-1(rg432) males, and “e” compares slo-1(rg432) hermaphrodites to slo-1(rg432) males. Herm, hermaphrodite.
Overlapping K+Channels in Mating
due to the severe defects shown by slo-1(lf) animals, indi-
cating functional SLO-1 K+channels are greatly reduced if
not abolished (Wang et al. 2001; Carre-Pierrat et al. 2006).
Since the frequency of the Prc phenotype was signifi-
cantly decreased in unc-103(0); egl-2(0) slo-1(rg432) as
compared to unc-103(0);egl-2(0) males (Table 3), we asked
how global slo-1 mRNA levels were affected in the triple and
double mutants relative to wild type and the single mutants.
In unc-103(0); egl-2(0) animals, the global slo-1 transcript
abundance were similar to the wild type. This could be an
additive consequence of the unc-103(0) mutation increasing
slo-1 transcript levels, balanced with the effect of egl-2(0)
decreasing them. Unexpectedly, slo-1 mRNA levels were
drastically lower in unc-103(0); egl-2(0) slo-1(rg432)
mutants relative to unc-103(0); egl-2(0) and slo-1(rg432)
animals (Figure 4, C and D). This indicates that the lack
of EGL-2 and UNC-103 activity has a synthetic interaction
with the rg432 allele to globally reduce slo-1 mRNA abun-
dance even more so than the rg432 allele alone. However,
the behavior of unc-103(0); egl-2(0) slo-1(rg432) triple
mutants was not worse than unc-103(0); egl-2(0) double
mutants, but rather, resembles unc-103(0) single mutants.
This suggests that the amount of slo-1 mRNA cannot be
solely rate limiting in controlling behavior. Thus a complex
interplay between the consequences of the three mutant
alleles might have a pleotropic effect on other genes that
control general neural and muscle excitability.
Finally, we asked whether a molecule that is known to
pleotropically affect neural and muscle cell excitability also
influences slo-1, egl-2, and unc-103 mRNA abundance. CaM-
KII is a calcium-activated protein and is known to influence
gene expression (Kapiloff et al. 1991; Ramirez et al. 1997;
Hughes et al. 2001; Zhang et al. 2004; Ronkainen et al.
2011). In C. elegans, loss-of-function mutations in CaMKII/
unc-43 have profound deleterious effects on virtually every
behavior of the worm (Reiner et al. 1999). The unc-43
(sy574) allele we previously isolated causes males to display
the Prc phenotype, but does not have any gross effect on
other behaviors in males or hermaphrodites (LeBoeuf et al.
2007, 2011). We therefore asked whether the sy574 allele,
which outwardly appears to affect one behavior, influences
global slo-1, egl-2, and unc-103 mRNA abundance. We dis-
covered that the transcript amount of all three K+channels
was decreased in animals that contain the unc-43(sy574) al-
lele (Figure 4). This result indicates that despite the pheno-
typic mildness of the unc-43(sy574) allele, it has broad effects
of mRNA levels, and that slo-1, egl-2, and unc-103 expression
is directly or indirectly connected with CaMKII function.
slo-1(rg432) preferentially affects genes involved
in regulating cell excitability in males
The above experiments do not explain how the slo-1(rg432)
allele, which results in decreased slo-1 transcript levels in
hermaphrodite and male populations and decreased egl-2
Figure 4 slo-1, egl-2, and unc-
103 K+channel transcript levels.
qRT–PCR performed on wild-
type and K+channel mutants.
(A–C) *P , 0.05, **P , 0.005,
***P , 0.0005, unpaired t-test,
mutant compared to wild type.
Error bars represent standard de-
viation. x-axis indicates the geno-
type tested, y-axis is the level
of normalized gene expression.
(A) ERG/unc-103 expression. (B)
EAG/egl-2 expression. (C) BK/
slo-1 expression. (D) The data in
A–C are reformatted as a table
for ease of comparison. (*P value
,0.05, statistically significant dif-
ference from wild type; unpaired
B. LeBoeuf and L. R. Garcia
and unc-103 mRNA levels in mixed-staged populations, is
able to reduce spicule protraction in unc-103(0); egl-2(0)
males. Decreasing the pool of available transcripts should
have a negative impact on the amount of protein produced,
and consequently a cell’s ability to maintain its polarized
state. One possibility is that, as a response to decreased
slo-1 transcript levels, cells generally lower the amount of
other components that promote cell depolarization, thus
resulting in a more stable system. To test this hypothesis,
we measured the transcript levels of genes involved in reg-
ulating the excitability of cells used in male mating behavior
and also of genes involved in more general cell processes.
(nAChR) a-subunit], gar-3 (ACh receptor coupled to gaq),
unc-29 (nAChR non–a-subunit), unc-43 (CaMKII), and cat-2
(tyrosine hydroxylase) as candidate genes that are enriched
in excitable cells (Sulston et al. 1975; Fleming et al. 1997;
Hwang et al. 1999; Lints and Emmons 1999; Reiner et al.
1999; Sze et al. 2000; Garcia et al. 2001; Mongan et al.
2002; LeBoeuf et al. 2007; Liu et al. 2007b, 2011). Neither
acr-18 nor cat-2 transcript levels were increased in males as
compared to hermaphrodites (Figure 3, C and D), although
both were reduced in slo-1(rg432) males. While C. elegans
males have many more excitable cells than hermaphrodites,
only a few of these express cat-2, which could account for
the lack of difference between sexes. acr-18 is expressed in
many male-specific neurons and muscles (Liu et al. 2011);
however, the acr-18 mRNA was slightly, but not significantly
increased in males as compared to hermaphrodites. The acr-
18 levels could be reduced in non–sex-specific cells in males,
resulting in transcript levels that are not significantly differ-
ent from hermaphrodites.
In contrast to acr-18 and cat-2 and similar to slo-1, the
genes gar-3, unc-29, and unc-43 showed greatly increased
transcript levels in the male; this increase was abolished in
slo-1(rg432) worms (Figure 3, C and D). Thus, in response
to the rg432 allele, cells appear to downregulate other mol-
ecules, along with the SLO-1 channel, involved in regulating
In addition to genes enriched in excitable cells, we also
looked at the transcript levels of genes with more general
expression patterns: let-363 (Tor-like kinase), sir-2.1 (his-
tone deacetylase), cdc-42 (RHO GTPase), and pmp-3 [ATP-
binding cassette (ABC) transporter] (Chen et al. 1993;
McKim et al. 1993; Frye 2000; Hoogewijs et al. 2008). Rel-
ative to hermaphrodites, male transcript levels were in-
creased for let-363 and sir-2.1 (Figure 3, C and D). mRNA
levels for let-363, sir-2.1, and pmp-3, but not cdc-42, drop-
ped in slo-1(rg432) males; however, the differences were
not as pronounced as compared to transcripts enriched in
excitable cells (Figure 3, C and D). In addition, transcript
levels for let-363, sir-2.1, and pmp-3 were increased in slo-1
(rg432) hermaphrodites, in contrast to the lower transcript
levels of the analyzed excitable cell-enriched genes. Taken
together, these data suggest that the slo-1(rg432) allele
broadly affects the levels of mRNA encoded by genes that
regulate cell excitability more so than genes with more gen-
eral cell functions.
The excitability of neurons and muscles need to be fine-tuned
for the optimal performance of complex behaviors. C. elegans
male mating is a demanding behavior requiring coordinated
feedback mechanisms to promote the multiple steps of mat-
ing. Failure of any one step leads to unsuccessful mating. For
example, full spicule protraction occurs when the spicules
penetrate the vulval slit and is inhibited until the correct cues
are received from all inputs. A breakdown in regulation leads
to promiscuous protraction at ectopic areas, thus reducing
successful sperm transfer. Spicule protraction depends on
the integration of signaling from neurons and muscles, includ-
ing the SPC motor neuron and the dorsal and ventral pro-
tractor muscles. The SPC releases acetylcholine (ACh) onto
the protractors, initiating contraction. Gap junctions between
the protractors and other muscles and neurons in the circuit
help reduce the excitability threshold of the protractors, so
they can respond to the ACh released by the SPC neurons
(Garcia et al. 2001; Liu et al. 2011). BK/SLO-1, ERG-like/
UNC-103, and EAG/EGL-2 K+channels attenuate circuit ac-
tivity until sufficient cues have been received (Figure 5A).
An important contributor to regulating male mating
circuit excitability is BK channel/slo-1. slo-1 mutants display
ectopic spicule protraction, which can be rescued with spe-
cific slo-1 isoforms. Full rescue is achieved by shortening the
distance between the two regulator of K+channel conduc-
tance (RCK) sites. RCK sites are proposed to mediate calci-
um’s gating of the channel (Lee and Cui 2010; Yuan et al.
2010), a modification that is common across species (Fodor
and Aldrich 2009). Similarly, isoform-specific rescue of flight
was shown in Drosophila (Brenner et al. 2000). Across dif-
ferent species, age, cell type, and stress impact which iso-
forms of slo-1 are expressed (Tseng-Crank et al. 1994; Xie
and McCobb 1998; MacDonald et al. 2006; Kim et al. 2010).
In C. elegans SLO-1, inclusion of the exon at splice site C1
results in channels with slower activation when expressed in
Xenopus oocytes (Johnson et al. 2011). Our data suggest
a functional consequence for modifying this linker region.
The subtle changes in slo-1 that promote the channel’s abil-
ity to inhibit precocious spicule protraction highlight the
large impact that small changes in channel structure have
In addition to identifying an isoform change that pro-
motes reduction of sex muscle excitability, we identified an
intronic point mutation [slo-1(rg432)] that significantly
reduces slo-1 transcript levels and sex muscle excitability.
slo-1(rg432) also reduces the transcript levels of genes in-
volved in regulating cell excitability. This suggests that the
compensatory response to reduced slo-1 levels is to diminish
the levels of other genes involved in cell depolarization.
Many human diseases are the result of intronic point muta-
tions (Watanabe et al. 2000; Lamba et al. 2002; Keren et al.
Overlapping K+Channels in Mating
2010); identifying ectopic genetic responses provides poten-
tial therapeutic routes to alleviate the effects of such
SLO-1 does not regulate male mating circuit excitability
alone but functions with the EAG family K+channels EAG/
EGL-2 and ERG/UNC-103. By examining how these K+
channels impact one another, we elucidated their coor-
dinated effort to regulate cell excitability and behavior.
Approximately 70–80% of males lacking slo-1 display pre-
mature spicule protraction; this fraction does not change
when both egl-2 and unc-103 are removed. However, remov-
ing only one results in males with reduced spontaneous
muscle spasms. Thus, egl-2 or unc-103 can partially compen-
sate for slo-1 deficiency. These three channels have also been
shown to mediate synapse formation in Drosophila (Budnik
et al. 1990; Berke et al. 2006; Lee and Wu 2010), highlight-
ing the conserved mechanisms used to set up cell excitability
and produce behavior.
Studying transcript levels in different K+
mutants indicates egl-2 stabilizes unc-103 and slo-1 mRNA.
In egl-2(0) males, unc-103 and slo-1 mRNA levels are lower
than wild type. Genetic interactions indicate egl-2 allows slo-
1 to compensate for the loss of unc-103 function. unc-103
(0); egl-2(0) males display a higher instance of abnormal
spicule protraction than do unc-103(0) males. The differ-
ence is likely due to an upregulation of slo-1 in an unc-103
(0) background. slo-1 mRNA stability increases in an unc-
103(0) background, as does expression of a reporter gene
from the slo-1 promoter. If the increase in transcription pro-
motes protein synthesis, more EGL-2 and SLO-1 in the mem-
brane could be compensating for UNC-103 deficiency.
Additionally, the effect of the unc-103(sy557) neomorphic
allele on egl-2 and slo-1 function provides further evidence
of an interaction between the three K+channels. unc-103
(sy557) induces spicule protraction at a level similar to unc-
103(0); egl-2(0) slo-1(lf) and the increased levels of slo-1
and egl-2 transcription seen in unc-103(0) males are re-
versed in unc-103(sy557) mutants. These data suggest that
unc-103(sy557) negatively impacts egl-2 and slo-1, and con-
versely, these three K+channels function interdependently
in wild-type circuits. Finally, egl-2 needs to be absent for unc-
103 to compensate for slo-1 deficiency: the instance of spon-
taneous spicule protraction drops 40% from slo-1(lf) males
compared to egl-2(0) slo-1(lf) males. These results suggest
SLO-1 activity is dependent upon EGL-2, while EGL-2
attenuates UNC-103 compensatory function.
In C. elegans, EAG and BK channels also interact to down-
regulate sex muscle excitability when the young adult males
undergo a period of starvation, presumably to direct the
males’ behavior from copulation and toward food acquisi-
tion (Gruninger et al. 2006; LeBoeuf et al. 2011). We pre-
viously reported that EAG K+channels play a prominent
role during food deprivation (LeBoeuf et al. 2007, 2011);
here, we show that BK channels are also important. When
slo-1(lf) males are food deprived, spontaneous spicule pro-
traction is partially suppressed in an EAG K+channel-de-
pendent manner. Modifying EAG K+channel function
through a neomorphic genetic mutation enhances the effect
of food deprivation to suppress slo-1(lf)–induced muscle
Modifying K+channel function is one way for the circuit
to regulate excitability; adjusting calcium levels is another.
Calcium is a main transducer of excitatory signals and a cir-
cuit must maintain tight control of intracellular calcium lev-
els. The Ca2+-sensitive BK channels provide one method of
regulatory response; Ca2+-activated kinases are another. A
candidate is CaMKII, which acts as a repressor of excitability
in the male mating circuit (LeBoeuf et al. 2007). K+chan-
nels in the circuit set up the delicate balance necessary for
reproduction, and CaMKII may influence their transcription
(Figure 5B). CaMKII’s role in regulating transcription is well
Figure 5 K+channel regulation of sex muscle excitability. (A) Abbreviated
cartoon of the male tail. Sex muscles are indicated in gray. The ERG, EAG,
and BK K+channels inhibit spicule protraction until the male has pene-
trated the hermaphrodite vulva. (B) Diagram of part of the male mating
circuit. The SPC motor neuron innervates the protractor muscles and
releases constant low levels of acetylcholine (ACh) to help maintain sex
muscle excitability. ACh activates muscarinic ACh receptors, releasing
Gaq that results in intracellular increase in Ca2+(Liu et al. 2007b). This
in turn activates CaMKII, which increases the transcription of K+channel
genes to help lower the sex muscle excitability.
B. LeBoeuf and L. R. Garcia
documented (Li et al. 2010; Ely et al. 2011; Oruganti et al.
2011), especially in cardiac muscle (Zhang et al. 2004;
Backs et al. 2006; Little et al. 2007). In C. elegans, it is
possible CaMKII is regulating the transcription of slo-1,
egl-2, and unc-103, since all three transcript levels are de-
creased in a CaMKII/unc-43(lf) mutant. CaMKII also acts
through EAG K+channels during starvation and both EAG
and ERG K+channels under standard conditions (LeBoeuf
et al. 2007, 2011). However, there is genetic evidence
to suggest that CaMKII acts through additional K+chan-
nels in the male. Nearly 100% of unc-103(0); unc-43(lf)
(LeBoeuf et al. 2007) and unc-43(lf); slo-1(lf) double
mutant males (this article) spontaneously protract their
spicules. However, only 79% of triple K+channel mutant
males display abnormal behavior. This indicates that there
are additional unidentified molecules through which
CaMKII is acting to regulate male sex muscle excitability.
We thank LaShundra Rodgers for technical assistance and
Benjamin Russo, Daisy Gualberto, Liusuo Zhang, and James
Midkiff for critical reading of the manuscript. C. elegans
strains were provided by the Caenorhabditis Genetic Center,
which is funded by the National Center for Research Resour-
ces, National Institutes of Health. This work was supported
by the Howard Hughes Medical Institute.
Anderson, P., 1995
Biology, Caenorhabditis elegans: Modern Biological Analysis of
an Organism, edited by H. F. Epstein and D. C. Shakes. Aca-
demic Press, San Diego.
Backs, J., K. Song, S. Bezprozvannaya, S. Chang, and E. N. Olson,
2006 CaM kinase II selectively signals to histone deacetylase 4
during cardiomyocyte hypertrophy. J. Clin. Invest. 116: 1853–
Bargmann, C. I., 1998 Neurobiology of the Caenorhabditis elegans
genome. Science 282: 2028–2033.
Berke, B. A., J. Lee, I. F. Peng, and C. F. Wu, 2006
Ca2+ dynamics affected by voltage- and Ca2+-gated K+ chan-
nels: regulation of the soma-growth cone disparity and the qui-
escent state in Drosophila neurons. Neuroscience 142: 629–644.
Brenner, R., J. Y. Yu, K. Srinivasan, L. Brewer, J. L. Larimer et al.,
2000 Complementation of physiological and behavioral de-
fects by a slowpoke Ca(2+) -activated K(+) channel transgene.
J. Neurochem. 75: 1310–1319.
Brenner, S., 1974 The genetics of Caenorhabditis elegans. Genetics
Budnik, V., Y. Zhong, and C. F. Wu, 1990
of motor axons in Drosophila mutants with altered excitability.
J. Neurosci. 10: 3754–3768.
Carre-Pierrat, M., K. Grisoni, K. Gieseler, M. C. Mariol, E. Martin et al.,
2006The SLO-1 BK channel of Caenorhabditis elegans is critical
for muscle function and is involved in dystrophin-dependent mus-
cle dystrophy. J. Mol. Biol. 358: 387–395.
Chen, W., H. H. Lim, and L. Lim, 1993
from Caenorhabditis elegans. Complementation of yeast muta-
tion. J. Biol. Chem. 268: 13280–13285.
Mutagenesis, pp. 31–58 in Methods in Cell
The CDC42 homologue
Davies, A. G., J. T. Pierce-Shimomura, H. Kim, M. K. VanHoven,
T. R. Thiele et al., 2003 A central role of the BK potassium
channel in behavioral responses to ethanol in C. elegans. Cell
Du, W., J. F. Bautista, H. Yang, A. Diez-Sampedro, S. A. You et al.,
2005Calcium-sensitive potassium channelopathy in human
epilepsy and paroxysmal movement disorder. Nat. Genet. 37:
Ely, H. A., P. L. Mellon, and D. Coss, 2011
gene via phosphorylation of SRF by the calcium/calmodulin
kinase II pathway. Mol. Endocrinol. 25: 669–680.
Fleming, J. T., M. D. Squire, T. M. Barnes, C. Tornoe, K. Matsuda
et al., 1997
Caenorhabditis elegans levamisole resistance genes
lev-1, unc-29, and unc-38 encode functional nicotinic acetylcho-
line receptor subunits. J. Neurosci. 17: 5843–5857.
Fodor, A. A., and R. W. Aldrich, 2009
alternative splices at domain boundaries of the BK channel.
Annu. Rev. Physiol. 71: 19–36.
Frye, R. A., 2000Phylogenetic classification of prokaryotic and
eukaryotic Sir2-like proteins. Biochem. Biophys. Res. Commun.
Garcia, L. R., and P. W. Sternberg, 2003
UNC-103 ERG-like potassium channel regulates contractile be-
haviors of sex muscles in males before and during mating.
J. Neurosci. 23: 2696–2705.
Garcia, L. R., P. Mehta, and P. W. Sternberg, 2001
distinct muscle behaviors controls the C. elegans male’s copula-
tory spicules during mating. Cell 107: 777–788.
Glauser, D. A., B. E. Johnson, R. W. Aldrich, and M. B. Goodman,
2011Intragenic alternative splicing coordination is essential
for Caenorhabditis elegans slo-1 gene function. Proc. Natl. Acad.
Sci. USA 108: 20790–20795.
Granato, M., H. Schnabel, and R. Schnabel, 1994
able marker for gene transfer in C. elegans. Nucleic Acids Res.
Grimm, P. R., and S. C. Sansom, 2010
form of hypertension. Kidney Int. 78: 956–962.
Gruninger, T. R., D. G. Gualberto, B. LeBoeuf, and L. R. Garcia,
2006 Integration of male mating and feeding behaviors in
Caenorhabditis elegans. J. Neurosci. 26: 169–179.
Hawasli, A. H., O. Saifee, C. Liu, M. L. Nonet, and C. M. Crowder,
2004 Resistance to volatile anesthetics by mutations enhanc-
ing excitatory neurotransmitter release in Caenorhabditis ele-
gans. Genetics 168: 831–843.
Hodgkin, J. A., H. R. Horvitz, and S. Brenner, 1979
disjunction mutants of the nematode Caenorhabditis elegans.
Genetics 91: 67–94.
Hodgkin, J., A. Papp, R. Pulak, V. Ambros, and P. Anderson,
1989A new kind of informational suppression in the nema-
tode Caenorhabditis elegans. Genetics 123: 301–313.
Hoogewijs, D., K. Houthoofd, F. Matthijssens, J. Vandesompele,
and J. Vanfleteren, 2008 Selection and validation of a set of
reliable reference genes for quantitative sod gene expression
analysis in C. elegans. BMC Mol. Biol. 9: 9.
Hughes, K.,S. Edin,A. Antonsson,
2001 Calmodulin-dependent kinase II mediates T cell recep-
tor/CD3- and phorbol ester-induced activation of IkappaB ki-
nase. J. Biol. Chem. 276: 36008–36013.
Hwang, J. M., D. J. Chang, U. S. Kim, Y. S. Lee, Y. S. Park et al.,
1999Cloning and functional characterization of a Caenorhab-
ditis elegans muscarinic acetylcholine receptor. Receptors Chan-
nels 6: 415–424.
Johnson, B. E., D. A. Glauser, E. S. Dan-Glauser, D. B. Halling, R. W.
Aldrich et al., 2011 Alternatively spliced domains interact to
regulate BK potassium channel gating. Proc. Natl. Acad. Sci.
USA 108: 20784–20789.
GnRH induces the c-Fos
Convergent evolution of
pha-1, a select-
BK channels and a new
Overlapping K+Channels in Mating
Kapiloff, M. S., J. M. Mathis, C. A. Nelson, C. R. Lin, and M. G.
Rosenfeld, 1991 Calcium/calmodulin-dependent protein ki-
nase mediates a pathway for transcriptional regulation. Proc.
Natl. Acad. Sci. USA 88: 3710–3714.
Keren, H., M. Donyo, D. Zeevi, C. Maayan, T. Pupko et al.,
2010 Phosphatidylserine increases IKBKAP levels in familial
dysautonomia cells. PLoS ONE 5: e15884.
Kim, H., J. T. Pierce-Shimomura, H. J. Oh, B. E. Johnson, M. B.
Goodman et al., 2009The dystrophin complex controls BK
channel localization and muscle activity in Caenorhabditis ele-
gans. PLoS Genet. 5: e1000780.
Kim, J. M., R. Beyer, M. Morales, S. Chen, L. Q. Liu et al.,
2010 Expression of BK-type calcium-activated potassium chan-
nel splice variants during chick cochlear development. J. Comp.
Neurol. 518: 2554–2569.
Lamba, J. K., Y. S. Lin, E. G. Schuetz, and K. E. Thummel,
2002 Genetic contribution to variable human CYP3A-medi-
ated metabolism. Adv. Drug Deliv. Rev. 54: 1271–1294.
LeBoeuf, B., T. R. Gruninger, and L. R. Garcia, 2007
vation attenuates seizures through CaMKII and EAG K+ Chan-
nels. PLoS Genet. 3: e156.
LeBoeuf, B., X. Guo, and L. R. Garcia, 2011
sient starvation persist through direct interactions between
CaMKII and ether-a-go-go K+ channels in C. elegans males.
Neuroscience 175: 1–17.
Lee, J., and C. F. Wu, 2010Orchestration of stepwise synaptic
growth by K+ and Ca2+ channels in Drosophila. J. Neurosci.
Lee, U. S., and J. Cui, 2010 BK channel activation: structural and
functional insights. Trends Neurosci. 33: 415–423.
Li, H., W. Li, A. K. Gupta, P. J. Mohler, M. E. Anderson et al.,
2010 Calmodulin kinase II is required for angiotensin II-medi-
ated vascular smooth muscle hypertrophy. Am. J. Physiol. Heart
Circ. Physiol. 298: H688–H698.
Lints, R., and S. W. Emmons, 1999
neurotransmitter identity among Caenorhabditis elegans ray sen-
sory neurons by a TGFbeta family signaling pathway and a Hox
gene. Development 126: 5819–5831.
Little, G. H., Y. Bai, T. Williams, and C. Poizat, 2007
calcium/calmodulin-dependent protein kinase IIdelta preferen-
tially transmits signals to histone deacetylase 4 in cardiac cells.
J. Biol. Chem. 282: 7219–7231.
Liu, J., M. Asuncion-Chin, P. Liu, and A. M. Dopico, 2006
kinase II phosphorylation of slo Thr107 regulates activity and
ethanol responses of BK channels. Nat. Neurosci. 9: 41–49.
Liu, K. S., and P. W. Sternberg, 1995
mating behavior in Caenorhabditis elegans. Neuron 14: 79–89.
Liu, Q., B. Chen, Q. Ge, and Z. W. Wang, 2007a
+/calmodulin-dependent protein kinase II modulates neuro-
transmitter release by activating BK channels at Caenorhabditis
elegans neuromuscular junction. J. Neurosci. 27: 10404–10413.
Liu, Y., B. LeBoeuf, and L. R. Garcia, 2007b
muscarinic acetylcholine receptors enhance nicotinic acetylcho-
line receptor signaling in Caenorhabditis elegans mating behav-
ior. J. Neurosci. 27: 1411–1421.
Liu, Y., B. LeBeouf, X. Guo, P. A. Correa, D. G. Gualberto et al.,
2011 A cholinergic-regulated circuit coordinates the main-
tenance and bi-stable states of a sensory-motor behavior dur-
ing Caenorhabditis elegans male copulation. PLoS Genet. 7:
Livak, K. J., and T. D. Schmittgen, 2001
expression data using real-time quantitative PCR and the 2
(-Delta Delta C(T)). Method. Methods 25: 402–408.
Longman, D., R. H. Plasterk, I. L. Johnstone, and J. F. Caceres,
2007 Mechanistic insights and identification of two novel fac-
tors in the C. elegans NMD pathway. Genes Dev. 21: 1075–1085.
The effects of tran-
Patterning of dopaminergic
Sensory regulation of male
Analysis of relative gene
MacDonald, S. H., P. Ruth, H. G. Knaus, and M. J. Shipston,
2006 Increased large conductance calcium-activated potas-
sium (BK) channel expression accompanied by STREX variant
downregulation in the developing mouse CNS. BMC Dev. Biol.
Maquat, L. E., 2004Nonsense-mediated mRNA decay: splicing,
translation and mRNP dynamics. Nat. Rev. Mol. Cell Biol. 5:
McKim, K. S., K. Peters, and A. M. Rose, 1993
required for meiotic chromosome pairing in Caenorhabditis ele-
gans. Genetics 134: 749–768.
Mello, C. C., J. M. Kramer, D. Stinchcomb, and V. Ambros,
1991 Efficient gene transfer in C.elegans: extrachromosomal
maintenance and integration of transforming sequences. EMBO
J. 10: 3959–3970.
Mongan, N. P., A. K. Jones, G. R. Smith, M. S. Sansom, and D. B.
Sattelle, 2002 Novel alpha7-like nicotinic acetylcholine recep-
tor subunits in the nematode Caenorhabditis elegans. Protein Sci.
Oruganti, S. R., S. Edin, C. Grundstrom, and T. Grundstrom,
2011 CaMKII targets Bcl10 in T-cell receptor induced activa-
tion of NF-kB. Mol. Immunol. 48: 1448–1460.
Park, E. C., and H. R. Horvitz, 1986
effects on the behavior and morphology of the nematode Cae-
norhabditis elegans. Genetics 113: 821–852.
Ramirez, M. T., X. L. Zhao, H. Schulman, and J. H. Brown,
1997The nuclear deltaB isoform of Ca2+/calmodulin-depen-
dent protein kinase II regulates atrial natriuretic factor gene
expression in ventricular myocytes. J. Biol. Chem. 272:
Reiner, D. J., D. Weinshenker, and J. H. Thomas, 1995
dominant mutations affecting muscle excitation in Caenorhabdi-
tis elegans. Genetics 141: 961–976.
Reiner, D. J., E. M. Newton, H. Tian, and J. H. Thomas,
1999 Diverse behavioural defects caused by mutations in Cae-
norhabditis elegans unc-43 CaM kinase II. Nature 402: 199–203.
Reiner, D. J., D. Weinshenker, H. Tian, J. H. Thomas, K. Nishiwaki
et al., 2006 Behavioral genetics of Caenorhabditis elegans unc-
103-encoded erg-like K(+) channel. J. Neurogenet. 20: 41–66.
Ronkainen, J. J., S. L. Hanninen, T. Korhonen, J. T. Koivumaki, R.
Skoumal et al., 2011 Ca2+-calmodulin-dependent protein ki-
nase II represses cardiac transcription of the L-type calcium
channel 1C-subunit gene (Cacna1c) by DREAM translocation.
J. Physiol. 589: 2669–2686.
Salkoff, L., A. Butler, G. Ferreira, C. Santi, and A. Wei, 2006
conductance potassium channels of the SLO family. Nat. Rev.
Neurosci. 7: 921–931.
Sanguinetti, M. C., C. Jiang, M. E. Curran, and M. T. Keating,
1995 A mechanistic link between an inherited and an acquired
cardiac arrhythmia: HERG encodes the IKr potassium channel.
Cell 81: 299–307.
Santi, C. M., A. Yuan, G. Fawcett, Z. W. Wang, A. Butler et al.,
2003Dissection of K+ currents in Caenorhabditis elegans mus-
cle cells by genetics and RNA interference. Proc. Natl. Acad. Sci.
USA 100: 14391–14396.
Schmitt, N., M. Schwarz, A. Peretz, I. Abitbol, B. Attali et al.,
2000A recessive C-terminal Jervell and Lange-Nielsen muta-
tion of the KCNQ1 channel impairs subunit assembly. EMBO J.
Schnabel, H., and R. Schnabel, 1990
ation gene, pha-1, from Caenorhabditis elegans. Science 250:
Sulston, J., M. Dew, and S. Brenner, 1975
in the nematode Caenorhabditis elegans. J. Comp. Neurol. 163:
Two types of sites
Mutations with dominant
An organ-specific differenti-
B. LeBoeuf and L. R. Garcia
Sulston, J. E., D. G. Albertson, and J. N. Thomas, 1980
norhabditis elegans male: postembryonic development of non-
gonadal structures. Dev. Biol. 78: 542–576.
Sulston, J. E., E. Schierenberg, J. G. White, and J. N. Thomson,
1983 The embryonic cell lineage of the nematode Caenorhab-
ditis elegans. Dev. Biol. 100: 64–119.
Sze, J. Y., M. Victor, C. Loer, Y. Shi, and G. Ruvkun, 2000
and metabolic signalling defects in a Caenorhabditis elegans
serotonin-synthesis mutant. Nature 403: 560–564.
Terskikh, A., A. Fradkov, G. Ermakova, A. Zaraisky, P. Tan et al.,
“Fluorescent timer”: protein that changes color with
time. Science 290: 1585–1588.
Torres, Y. P., F. J. Morera, I. Carvacho, and R. Latorre, 2007
marriage of convenience: beta-subunits and voltage-dependent
K+ channels. J. Biol. Chem. 282: 24485–24489.
Tseng-Crank, J., C. D. Foster, J. D. Krause, R. Mertz, N. Godinot
et al., 1994 Cloning, expression, and distribution of function-
ally distinct Ca(2+)-activated K+ channel isoforms from human
brain. Neuron 13: 1315–1330.
Vandesompele, J., K. De Preter, F. Pattyn, B. Poppe, N. Van Roy
et al., 2002Accurate normalization of real-time quantitative
RT-PCR data by geometric averaging of multiple internal control
genes. Genome Biol. 3: RESEARCH0034.
Wang, Z. W., O. Saifee, M. L. Nonet, and L. Salkoff, 2001
potassium channels control quantal content of neurotransmitter
release at the C. elegans neuromuscular junction. Neuron 32:
Watanabe, Y., S. Kato, Y. Adachi, and K. Nakashima, 2000
shift, nonsense and non amino acid altering mutations in SOD1
in familial ALS: report of a Japanese pedigree and literature
review. Amyotroph. Lateral Scler. Other Motor Neuron Disord.
Weinshenker, D., G. Garriga, and J. H. Thomas, 1995
pharmacological analysis of neurotransmitters controlling egg
laying in C. elegans. J. Neurosci. 15: 6975–6985.
Weinshenker, D., A. Wei, L. Salkoff, and J. H. Thomas, 1999
of an ether-a-go-go-like K(+) channel by imipramine rescues egl-
2 excitation defects in Caenorhabditis elegans. J. Neurosci. 19:
Werner, M. E., P. Zvara, A. L. Meredith, R. W. Aldrich, and M. T.
Nelson, 2005Erectile dysfunction in mice lacking the large-
conductance calcium-activated potassium (BK) channel. J. Phys-
iol. 567: 545–556.
Xie, J., and D. P. McCobb, 1998
potassium channels by stress hormones. Science 280: 443–446.
Yuan, P., M. D. Leonetti, A. R. Pico, Y. Hsiung, and R. MacKinnon,
2010Structure of the human BK channel Ca2+-activation ap-
paratus at 3.0 Å resolution. Science 329: 182–186.
Zhang, T., S. Miyamoto, and J. H. Brown, 2004
calcium and calcium/calmodulin-dependent protein kinase II:
friends or foes? Recent Prog. Horm. Res. 59: 141–168.
Control of alternative splicing of
Communicating editor: R. Anholt
Overlapping K+Channels in Mating