Molecular Cloning and Characterization of LKv1, a
Novel Voltage-Gated Potassium Channel in Leech
Jamie L. Weiss,1 Jay Yang,2 Chunfa Jie,1 Diana L. Walker,1 Shawn Ahmed,1
Yunxia Zhu,1 Yueqiao Huang,1 Kristen M. Johansen,1 Jørgen Johansen1
1
Department of Zoology and Genetics, 3156 Molecular Biology Bldg., Iowa State University,
Ames, Iowa 50011
2
Department of Anesthesiology and Pharmacology, University of Rochester Medical Center,
Rochester, New York 14624
Received 1 May 1998; accepted 11 August 1998
ABSTRACT:
We have cloned a novel voltage-gated
K channel, LKv1, in two species of leech. The properties
of LKv1 expressed in transiently transfected HEK293
cells is that of a delayed rectifier current. LKv1 may be
a major modulator of excitability in leech neurons, since
antibody localization studies show that LKv1 is expressed in the soma and axons of all neurons in both the
central and peripheral nervous systems. Comparison of
the biophysical and pharmacological properties of LKv1
with native voltage-gated conductances in leech neurons
suggests that LKv1 may correspond to the previously
characterized delayed rectifier current, IK. Phylogenetic
analysis of LKv1 shows that it is related to the Shaker
subfamily of voltage-gated K channels although it occupies a separate branch from that of the monophyletic
Shaker clade composed of the flatworm, Aplysia, Drosophila, and mammalian Shaker homologs as well as
from that of two recently identified Shaker-related K
channels in jellyfish. Thus, this analysis indicates that
this group of voltage-gated K channels contains several
evolutionarily divergent lineages. © 1999 John Wiley & Sons,
A particularly advantageous organism in which to
address questions of how the distribution and properties of ion channels relate to the biological function of
neurons can be found in the leech (Johansen, 1991).
The leech central nervous system (CNS) is very simple, consisting of a fixed number of 33 ganglia: 21
similar midbody ganglia, four ganglia fused into a
head brain, and seven fused into a tail brain, and a
supraesophageal ganglion (Muller et al., 1981). Each
ganglion contains very few neurons, ⬍400, most of
which are bilaterally symmetrically arranged, leaving
no more than about 200 –300 types of different neurons per ganglion (Muller et al., 1981; Macagno,
1980). The ganglia are to a high degree segmentally
reiterated, further limiting the total number of different neurons within the CNS. These neurons are relatively large, ranging in diameter from 20 to 100 ␮m.
The neurons surround a central neuropil forming a
single layer, making each individual neuron easily
accessible for electrophysiological recording. For
these reasons, many leech neurons are functionally
identified (Muller et al., 1981), and several neuronal
networks mediating a number of behaviors—for example, swimming (Stent et al., 1978), control of heartbeat (Stent et al., 1979), and the bending and withdrawal reflexes (Kristan, 1982)— have been described
in great detail. Furthermore, applications of voltageclamp techniques have allowed the identification and
Correspondence to: J. Johansen
Contract grant sponsor: Iowa State Biotechnology
Contract grant sponsor: Iowa State University
Contract grant sponsor: NIH; contract grant number: NS28857
Contract grant sponsor: NSF; contract grant number:
DIR9113595
Contract grant sponsor: Hatch Act
Contract grant sponsor: State of Iowa
© 1999 John Wiley & Sons, Inc. CCC 0022-3034/99/020287-13
Inc. J Neurobiol 38: 287–299, 1999
Keywords: Shaker; leech; immunocytochemistry; K
channels; phylogeny
287
288
Weiss et al.
characterization of many of the conductances underlying the regulation of excitability in these neurons
(Johansen and Kleinhaus, 1986, 1990; Johansen et al.,
1987; Arbas and Calabrese, 1987; Stewart et al.,
1989). As a beginning to analyze these currents on the
molecular level, we here describe the cloning of a
leech voltage-gated K channel, LKv1, and its expression in transiently transfected HEK293 cells. The
properties of LKv1 expressed in these cells are that of
a delayed rectifier current. Antibody localization studies show that LKv1 is expressed in the soma and
axons of all known neurons in the CNS and peripheral
nervous system (PNS), suggesting that this current
may constitute a major regulator of excitability in
leech neurons. Although phylogenetic analysis of
LKv1 shows that it is related to the Shaker subfamily
of voltage-gated K channels, its sequence is highly
divergent with many unusual features. Consequently,
LKv1 is placed on a branch that is separate both from
that of the monophyletic Shaker clade composed of
the flatworm (Kim et al., 1995), Aplysia (Pfaffinger et
al., 1991), Drosophila (Papazian et al., 1987), and
mammalian Shaker homologs (Tempel et al., 1988;
Pongs, 1993), as well as from that of the two Shakerrelated voltage-gated K channels recently described in
jellyfish (Jegla et al., 1995). Thus, this group of voltage-gated K channels appear to be heterogeneous and
to contain several evolutionarily divergent lineages.
MATERIALS AND METHODS
Experimental Preparations
For the present experiments, we used the three hirudinid
leech species: Hirudo medicinalis, Haemopis marmorata,
and Macrobdella decora. The leeches were either captured
in the wild or purchased from commercial sources. Dissections of nervous tissue and embryos were performed in
leech saline solutions with the following composition (in
millimoles): 110 NaCl, 4 KCl, 2 CaCl2, 10 glucose, 10
Hepes, pH. 7.4. In some cases, 8% ethanol was added and
the saline solution cooled to 4°C to inhibit muscle contractions. Breeding, maintenance, and staging of leech embryos
at 22–25°C were as previously described (Fernández and
Stent, 1982; Macagno et al., 1983; Jellies et al., 1987).
Embryonic day 10 (E10) was characterized by the first sign
of a tail sucker, while E30 is the termination of embryogenesis.
Molecular Cloning and Sequence
Analysis
Degenerate universal primers to regions conserved in the
four subfamilies of voltage-gated K⫹ channels: Shaker,
Shal, Shab, and Shaw NEYFFD (T1 assembly domain) and
TMTTVG (pore-lining region) were used in a polymerase
chain reaction (PCR)-based screen. Haemopis genomic
DNA was amplified using low-stringency annealing conditions (37°C) for 30 cycles using the Perkin-Elmer GeneAmp
kit according to the manufacturer’s instructions. The degenerate oligonucleotide primers were synthesized on the basis
of conserved sequences in Drosophila and included EcoRI
and SalI restriction linker sequences for cloning purposes:
5⬘-AGTCTAGAATTCICC A / G TAICCIA C / G IGTNGTCAT-3⬘ and 5⬘-GACTGCAGTCGAC AAT/CGAT/CTAT/
C
C C
CTT /TTTCTGA /T /AG-3⬘ (the primers were the generous
gift of Dr. A. Wei and Dr. L. Salkoff). The PCR reaction
products were digested with EcoRI and SalI, treated with
calf intestinal alkaline phosphatase to prevent multimerization, and subcloned into M13mp19 according to standard
protocols (Sambrook et al., 1989) to generate a minilibrary
of PCR reaction products. This minilibrary was plated,
transferred to nitrocellulose filters, and probed using lowstringency conditions (Sambrook et al., 1989). A mixed
Drosophila K⫹ channel probe was made by PCR amplification, as described above, of Drosophila Shaker, Shab,
Shal, and Shaw K⫹ channel clones (generous gift of Dr. A.
Wei and Dr. L. Salkoff). The probe was radioactively labeled using the Prime-A-Gene kit (Promega) according to
the manufacturer’s protocol. A partial LKv1 genomic sequence was identified in this screen and subsequently used
to identify clones from Haemopis genomic and cDNA libraries as well as from a Hirudo CNS-enriched cDNA
library. The Haemopis genomic library was constructed in
BamHI-cut ␭DASH (Stratagene) using size-selected Sau 3A
partial digestion products from high-molecular-weight
genomic DNA according to standard protocols (Ausubel et
al., 1987; Stratagene ␭DASH protocols). The Haemopis
cDNA library (Johansen and Johansen, 1995) and the
Hirudo CNS-enriched cDNA library (Huang et al., 1997)
have been previously described. Positive clones were either
subcloned or excised into pBluescript (Stratagene). From
these screens, a full-length Hirudo LKv1 cDNA was obtained, but the largest Haemopis cDNA extended only to the
S1 transmembrane domain. The missing 5⬘ coding region
was obtained by mapping and sequencing the 5⬘ end of a
15-kb Haemopis genomic clone, and exonic sequences were
identified which contained the full 5⬘ coding region.
DNA sequence was performed using a DNA Sequencer
377A (Applied Biosystems) at the Iowa State University
Nucleic Acid Sequencing Facility, using commercially
available universal and reverse sequencing primers (Stratagene) or specific primers synthesized by the facility based
on known LKv1 sequences. The nucleotide and predicted
amino acid sequences were analyzed using the Genetics
Computer Group (Wisconsin package, version 9.1) suite of
programs (Devereux et al., 1984). LKv1 sequences were
compared with known and predicted proteins in the
SwissProt and GenBank databases using the FASTA and
TFASTA programs within the GCG package. In addition,
BLAST searches were performed using the NCBI BLAST
e-mail server (Altschul et al., 1990) comparing the LKv1
sequences with SwissProt, PIR, and GenPept databases.
Transmembrane domains were determined both by align-
LKv1, a Novel K Channel in Leech
ment to other voltage-gated K channel protein sequences
and by hydropathy analysis using the program TopPred II.
Phylogenetic Analysis
Alignments used to produce maximum parsimony trees
were generated by GCG’s Pileup program and initially
encompassed the entire amino acid sequence of the voltagegated channels analyzed. However, any gaps in the resulting
alignments were removed by deleting residues corresponding to the gaps. This method left mostly, but not exclusively,
sequences from the T1 assembly domain, the six transmembrane domains, and the pore region in the analysis. Trees
were constructed by maximum parsimony using the PAUP
computer program, version 3.1.1 (Swofford, 1993) on a
Power Macintosh 8100. All trees were generated by heuristic searches and the resulting tree topology was verified by
PAUP’s branch and bound mode. Bootstrap values in percentages (Felsenstein, 1985) are indicated on the bootstrap
consensus tree.
Antibody Production and
Immunocytochemistry
The coding sequence for the first 75 residues of Haemopis
LKv1 was amplified using PCR with exact-match primers
that had EcoRI and SalI restriction sites incorporated. The
amplified fragment was isolated and subcloned into a Pharmacia pGEX4T-2 vector. The correct orientation and reading frame of the insert were verified by sequencing. The
resulting fusion protein was expressed in XL1-Blue cells
(Stratagene) and purified over a glutathione agarose column
using standard protocols. The purified LKv1 fusion protein
was used to generate polyclonal antibodies in rabbits. Two
rabbits were injected with 200 – 400 ␮g of fusion protein
and then boosted at 21-day intervals, as described in Harlow
and Lane (1988). After the second boost, serum samples
were collected 7 and 10 days after injection (Harlow and
Lane, 1988). The sera were analyzed for specificity by
comparing the staining obtained with the antisera and the
preimmune sera on nitrocellulose filters spotted with purified fusion protein and the expressed pGEX4T-2 vector. The
sera were titrated from undiluted to a 1:5000 dilution in
Blotto [0.5% Carnation nonfat dry milk in Tris-buffered
saline (TBS)].
For affinity purification of the LKv1 antiserum, the insert
from the pGEX4T-2 LKv1 GST expression clone was transferred to the PinPoint Xa-3 vector (Promega) and induced
(as per the manufacturer’s instructions) to produce a LKv1
fusion protein with a 13-kD biotinylation tag. The biotinylated LKv1 fusion protein, which contained no GST sequences, was subsequently bound to 1.5 mL tetrameric
avidin–agarose resin (Sigma) by incubation overnight at
4°C. The resin was washed extensively in phosphate-buffered saline (PBS) and used to construct an affinity column.
Polyclonal LKv1 antisera (1 mL) was loaded onto the
column and allowed to incubate for 2 h before continuing
column flow. The column was washed with 10 bed volumes
of PBS followed by three elutions with 1.5 mL of 4 M
289
guanidine HCl. Eluted antibody was dialyzed against PBS,
concentrated, and tested for immunoreactivity on spot tests
using both GST-LKv1 fusion protein and GST only, to
confirm affinity purification. The affinity-purified LKv1 antibody had no detectable immunoreactivity to GST in these
tests.
For immunocytochemistry, dissected leech embryos
were fixed overnight at 4°C in 4% paraformaldehyde in 0.1
M phosphate buffer, pH 7.4, incubated overnight at room
temperature in LKv1 antiserum (1:1000) or affinity-purified
antibody (1:50) containing 0.4% Triton X-100, washed in
PBS with 0.4% Triton X-100, and incubated with horseradish peroxidase (HRP)– conjugated goat anti-rabbit antibody
(BioRad; 1:200 dilution). After washing in PBS, the HRPconjugated antibody complex was visualized by reaction in
3,3⬘-diaminobenzidine (0.03%) and H2O2 (0.01%) for 10
min. In some experiments, to test for the specificity of the
polyclonal antisera, 30 ␮g of the LKv1 fusion protein was
added to the diluted antisera before staining. The final
preparations were dehydrated in alcohol, cleared in xylene,
and embedded as whole mounts in Depex mountant. Adult
ganglia were fixed in 4% paraformaldehyde and the connective capsules opened with fine forceps before they were
processed for antibody labeling in a similar way to the
embryos. The labeled preparations were photographed on a
Zeiss Axioskop using Ektachrome 64T film. The color
positives were digitized using Adobe Photoshop and a Nikon Coolscan slide scanner. In Photoshop, the images were
converted to black and white and image processed before
being imported into Freehand (Macromedia) for composition and labeling.
Sodium Dodecyl Sulfate–Polyacrylamide
Gel Electrophoresis (PAGE) and Western
Blotting
Sodium dodecyl sulfate–polyacrylamide gel electrophoresis
was performed according to standard procedures (Laemmli,
1970). Electroblot transfer was performed as in Towbin et
al. (1979). For these experiments we used the BioRad
minigel system, electroblotting to 0.2 ␮m nitrocellulose,
and using HRP-conjugated secondary antibody (1:3000) for
visualization of primary antibody diluted 1:2000 in Blotto
(1:50 for affinity-purified antibody) in immunoblot analysis.
The signal was developed with 3,3⬘-diaminobenzidine (0.1
mg/mL) and H2O2 (0.03%) and enhanced with 0.008%
NiCl2. In some experiments, to test for the specificity of the
polyclonal antisera, 30 ␮g of the LKv1 fusion protein was
added to the diluted antisera before staining. The Western
blots were digitized using the NIH-image software, a cooled
high-resolution CCD camera (Paultek), and a PixelBuffer
framegrabber (Perceptics) or an Arcus II scanner (AGFA).
Electrophysiological Analysis of LKv1 in
HEK293 Cells
A full-length Hirudo LKv1 cDNA, including 150 nt of
upstream and 345 nt of downstream sequence, was subcloned into the EcoRI multiple cloning site of an eukaryotic
290
Weiss et al.
expression vector pCI-neo (Promega). Two micrograms of
Hirudo LKv1-pCI-neo plasmid prepared by Qiagen Maxiprep was transiently transfected into a 35-mm tissue culture
dish containing HEK293 cells using 10 ␮L Transfectamine
(Gibco) following the manufacturer’s recommended protocol. In addition, 200 ng of green fluorescent protein cDNA
subcloned into pCI-neo was cotransfected to allow detection
of successfully transfected cells at the time of electrophysiological recording. The HEK293 cells were propagated in
a high-glucose DMEM media (Gibco) supplemented with
10% fetal calf serum (Hyclone) and antibiotics. Electrophysiological recordings were obtained 48 –72 h after transfection.
Whole-cell patch-clamp recordings were obtained from
fluorescent cells visually identified under an inverted epifluorescent microscope. External solution consisted of (in
millimoles): 140 NaCl, 5.8 KCl, 0.8 MgCl2, 3.0 CaCl2, 10
Hepes, 10 glucose, and pH was titrated to 7.4 with 1N
NaOH. In some experiments, 20 mM tetraethylammonium
(TEA) or 4 mM 4-AP was added to the external solution.
The patch electrodes fashioned from a 1.2-mm fiber-filled
borosilicate glass capillary (WPI) were fire-polished and
filled with an internal solution consisting of (in millimoles):
140 KCl, 4 NaCl, 4 MgCl2, 10 Hepes, 10 EGTA, and pH
was titrated to 7.4 with 1N KOH. The estimated K-reversal
potential under these ionic concentrations is ⫺80 mV. The
junction potential was nulled with the electrode in the bath
prior to contacting the cell and the bath was grounded to the
amplifier (AI 200; Axon Instruments) via an Ag/AgCl agar–
KCl bridge. In some experiments, 20 mM TEA and 4 mM
4-AP were added to the external solution and the bathing
solution changed during the experiment using a computercontrolled perfusion system. Data were selected for cells
showing precompensation series resistance of ⬍15 Mohms
after establishing the whole-cell recording configuration and
which allowed analog series resistance compensation of
⬎70%. Command voltage and analog data sampling were
accomplished using Pclamp v5.1 (Axon Instruments) running on a DX2-486 computer. Current magnitudes directly
read in Clampfit were further analyzed in SigmaPlot v2.0
(Jandel). Model parameters for the two-state Boltzman distribution for activation data were obtained by a minimal
sum-squared error fit of conductance at a given voltage g(V)
to the equation: g(V) ⫽ gmax/(1 ⫹ exp([V50 ⫺ V)/a]), where
gmax is the maximal conductance, V50 is the half-maximal
activation voltage (mV), and a is a slope factor (mV/e-fold).
The current activation kinetics were determined by a monoexponential fit to the rising phase of the current in Clampfit.
RESULTS
Cloning and Characterization of LKv1 in
Haemopis and Hirudo
Degenerate universal primers to regions conserved in
the four voltage-gated K channel subfamilies, Shaker,
Shal, Shab, and Shaw in Drosophila (Butler et al.,
1989), were used for PCR amplification of Haemopis
genomic DNA. The resulting PCR products were subcloned into M13 vector and screened at low stringency with mixed K channel probe amplified by PCR
with the universal primers of cDNAs from the four
Drosophila K channel subfamilies. From this screen,
a single partial 1.5-kb genomic LKv1 clone was isolated and sequenced. Probes based on the LKv1 sequence were subsequently used to screen a Haemopis
oligo(dT)-primed cDNA library, a Haemopis genomic
library, as well as a CNS-specific random primed
Hirudo cDNA library. From these screens, we obtained several partial Haemopis LKv1 cDNA clones
containing 3⬘ coding region and the poly A⫹ tail, a
15-kb Haemopis genomic clone, in addition to a
Hirudo cDNA clone containing a 1356-nucleotide
open-reading frame terminated by a TGA stop codon.
This open-reading frame is likely to be full length
since it has a 5⬘ ATG initiation codon with a favorable
A⫺3/G⫹4 context of initiation (Kozak, 1986) just
downstream from an in-frame TAA stop codon (data
not shown). We were unable to identify a Haemopis
cDNA containing the entire 5⬘ coding region; therefore, the complete sequence for Haemopis LKv1 is
partly based on sequence derived from the 15-kb
genomic clone (Fig. 1). The Haemopis genomic coding region contains only a single 0.7-kb intron that
intervenes the sequence coding for the S2 transmembrane domain. The predicted amino acid sequence of
the open-reading frame for LKv1 extends over 452
amino acids for Hirudo and 453 amino acids for
Haemopis (Fig.1), with calculated molecular weights
of 51.4 and 51.3 kD, respectively. The Haemopis and
Hirudo LKv1 sequences share 92% identity on the
amino acid level.
Figure 2 shows the sequence comparison of LKv1
(Hirudo) with the four most homologous sequences in
the data bank, all of which belong to voltage-gated K
channels of the Shaker subfamily found in mouse,
Drosophila, Aplysia, and flatworm. While the N-terminal amino acid sequence of LKv1 appears unique,
extensive conservation between LKv1 and these
Shaker homologs is present in the T1 tetramerization
domain (Li et al., 1992; Shen et al., 1993), in the six
transmembrane domains (S1–S6), and in the K⫹ ion–
selective pore region (P). However, there are also
several significant differences. For example LKv1 has
only five positively charged amino acids, compared to
seven for the other Shaker homologs in the S4 transmembrane domain, a domain that has been implicated
in voltage sensing (Papazian et al., 1991). This is
partly because LKv1 is missing a three–amino acid
turn of the ␣-helix containing one of the positive
charges. There is also a three–amino acid deletion in
the S5 transmembrane ␣-helix, which in this case is
compensated for by a three–amino acid insert (CML)
LKv1, a Novel K Channel in Leech
Figure 1 Alignment of the predicted amino acid sequence
for Hirudo and Haemopis LKv1. The Hirudo amino acid
sequence was derived from a full-length cDNA, whereas the
Haemopis sequence is compiled from both genomic and
cDNA transcripts. The amino acid sequence in Haemopis
based on only genomic DNA extends from the starting
methionine to the point indicated by an arrow. The Hirudo
LKv1 protein consists of 452 amino acids with a predicted
molecular weight of 51.4 kD, whereas the Haemopis LKv1
protein has 453 amino acids and a molecular weight of 51.3
kD. The two proteins share 92% identity, as indicated by
lines. Conservative amino acid substitutions are indicated
by dots. These sequence data and the corresponding nucleotide sequences are available from EMBL/GenBank/DDBJ
under Accession Nos. AF039834 (Hirudo LKv1) and
AF082189/AF082190 (Haemopis LKv1).
unique to LKv1 (Fig. 2). In contrast to other Shaker
homologs LKv1 has a very short C-terminal domain
consisting of only 17 amino acids. Furthermore, in
pairwise comparisons of the conserved core regions
(T1, S1–S6, and P) LKv1 shows only 50 –57% amino
acid identity to the flatworm, Aplysia, Drosophila,
and mouse Shaker homologs, the level of identity
between which ranges from 78% to 89% (Table 1).
The amino acid identity to the core sequences of the
two jellyfish K channels, jShak1 and jShak2, is 48%
and 52%, approximately the same as the amino acid
identity to the corresponding sequences in Drosophila
Shab, Shaw, and Shal K channels. Thus, to determine
the evolutionary relationship between LKv1 and other
members of the voltage-gated K channel subfamilies,
we constructed phylogenetic trees based on maximum
parsimony. Figure 3 shows a consensus tree based on
291
Shaker, Shab, Shal, and Shaw K channel core sequences from different organisms. The tree is rooted
using sequences from the first homology domain of
rat Na channel II (Noda et al., 1986) as an outgroup;
however, the same topology was obtained for unrooted trees. LKv1 is clearly grouped together with
the Shaker-related K channels, which, in addition, all
share what may be the Shaker subfamily signature
sequence of SxxxLRV at the beginning of the S4
domain. Interestingly, however, in nearly all the trees
we generated, LKv1 occupied a separate branch from
that of the Shaker homologs from flatworm, Aplysia,
Drosophila, and mouse which form a monophyletic
clade that is supported by a bootstrap value of 99%.
Flatworms are acoelomats and clearly less evolutionarily advanced than leeches which, like Aplysia, Drosophila, and mouse, belong to the coelomats. Thus,
these results suggest that LKv1 diverged from a common ancestor prior to the origin of the group of
monophyletic Shaker homologs. Furthermore, the two
jellyfish K channels, jShak1 and jShak2, are also
placed separately, indicating that the Shaker-related
group of voltage-gated K channels contains several
evolutionarily divergent lineages.
Immunocytochemical Localization
of LKv1
For studying the tissue expression and localization of
LKv1, we generated polyclonal antibodies in rabbits
to a GST-fusion protein containing the first 75 residues of the N-terminus of Haemopis LKv1 sequence.
On immunoblots of Haemopis CNS extracts LKv1
antiserum, as well as affinity-purified antiserum, recognizes a single 54-kD band not present in the preimmune serum (Fig. 4). Furthermore, preabsorption
of the LKv1 antiserum with LKv1-GST fusion protein
abolishes staining by the LKv1 antiserum of the
54-kD protein (Fig. 4). The molecular weight of this
band is in close correspondence to the predicted molecular weight of LKv1 of 51 kD. In addition to
Haemopis, the LKv1 antiserum recognizes an identical 54-kD band in CNS extracts from Hirudo and
Macrobdella (data not shown).
Figure 5(A) shows an adult midbody ganglion
from Haemopis labeled with LKv1 antiserum. All
central neuronal somata are antibody positive in contrast to glial and muscle cells which are not labeled by
the antibody. In control experiments, preimmune serum as well as LKv1 antiserum preabsorbed with
LKv1-GST fusion protein did not show immunocytochemical labeling (data not shown). To address
whether LKv1 is expressed by all neurons and
whether expression is truly specific to the nervous
system, we used LKv1 antiserum to label-staged leech
292
Weiss et al.
Figure 2 Alignment of the deduced amino acid sequence of Hirudo LKv1 with Shaker voltagegated K channels. The sequence from the T1 subunit assembly domain to the C-terminal end is
compared to the mouse Shaker Homolog Kv1.1 (Tempel et al., 1988), Drosophila Shaker B (Timpe
et al., 1988), the Aplysia Shaker homolog AK01a (Pfaffinger et al., 1991), and the flatworm Shaker
homolog SKv1.1 (Kim et al., 1994). Shared amino acids of these Shaker homologs with LKv1 are
in white typeface outlined in black. The T1 assembly domain, the six transmembrane domains
(S1–S6), and the pore-forming region (P) are underlined.
embryos. In embryos, it is possible to assay all components of the nervous system including the CNS,
PNS, and all other tissues. Figure 5(C) shows the
labeling of peripheral neurons located in the body
wall of a Macrobdella E12 embryo hemisegment. The
results show that all known peripheral neurons are
LKv1, a Novel K Channel in Leech
Table 1 Percent Conservation of Amino Acids
between LKv1 and Four Shaker K Channel Homologs
LKv1
LKv1
SKv1.1
AK01a
Shaker B
SKv1.1
AK01a
Shaker B
Kv1.1
53%
57%
84%
53%
78%
88%
50%
80%
87%
89%
Percent amino acid identities are shown for pairwise alignments
between LKv1 and Shaker homologs from flatworm (SKv1.1),
Aplysia (AK01a), Drosophila (Shaker B), and mouse (Kv1.1). Only
the identities in the core sequences of the T1 assembly domain, the
S1–S6 transmembrane domains, and the pore region were measured.
labeled by the antibody including sensillar neurons
(Johansen et al., 1992), the HO cells which are putative stretch receptors (Blackshaw et al., 1982), and the
NNC cell which innervates the nephridium (Wenning,
1983). In addition, all the axons constituting the peripheral nerves express LKv1. In contrast, muscle
cells and other nonneuronal tissues do not show labeling with LKv1 antiserum. Thus, LKv1 appears to
be a neuron-specific K conductance that is expressed
in somata and axons by all known neurons in both the
CNS and PNS. The earliest detection of LKv1 by
antibody labeling is in E8 –E9 embryos [Fig. 5(B)],
suggesting that the channel is expressed shortly after
the neurons differentiate.
Functional Expression of LKv1 in
Transiently Transfected HEK293 Cells
To characterize the electrophysiological properties of
LKv1, we transiently expressed the current in
HEK293 cells. Untransfected HEK293 control cells
did not have detectable endogenous voltage-gated K
currents. A recombinant plasmid containing the entire
open-reading frame of the Hirudo LKv1 cDNA, including 150 nt of upstream and 345 nt of downstream
sequence, was used for transfection. Figure 6(A)
shows an example of whole-cell voltage-clamp currents obtained in response to depolarizing steps from
a holding potential of ⫺80 mV in successfully LKv1transfected HEK293 cells. The biophysical and pharmacological properties of these currents are summarized in Table 2. LKv1 currents expressed in HEK293
cells have the properties of a delayed rectifier [Fig.
6(A)] and do not inactivate fully even during prolonged depolarizations (data not shown). However,
LKv1 does exhibit a limited degree of voltage-dependent inactivation relative to the peak current at positive potentials [Fig. 6(C)]. These properties of LKv1
are in contrast to the previously described currents of
Shaker-related invertebrate K channels which are all
293
transient A-type currents with rapid activation and
inactivation kinetics (Papazian et al., 1987; Pfaffinger
et al., 1991; Kim et al., 1994; Jegla et al., 1995). The
LKv1 current activated at relatively positive membrane potentials [Fig. 6(B)] with a V50 of 8.3 mV, as
determined from Boltzman fits of g(V) (Table 2).
Analysis of the reversal potential for tail currents of
LKv1 [Fig. 6(C,D)] shows that it is close to the
calculated EK of ⫺80 mV under these recording conditions, suggesting that LKv1 is selective mainly for
K⫹ ions. Pharmacologically, LKv1 is sensitive to
TEA with more than 90% of the current blocked by 20
mM, whereas 4 mM 4-AP reduces the current by only
10 –15% [Fig. 6(E,F); Table 2]. The sensitivity of
LKv1 to TEA correlates well with the presence of a
phenylalanine at position 406 in the pore region. It has
previously been reported that the presence of an aromatic amino acid at this position is responsible for
conferring sensitivity to external TEA (Heginbotham
and MacKinnon, 1992).
In previous studies using a two-electrode voltageclamp, it was demonstrated that two voltage-dependent K currents, a delayed rectifier type current IK and
a transient A-type current IA, could be identified in
isolated soma of leech neurons (Johansen and Kleinhaus, 1986, 1990). A comparison of the properties of
these two currents with LKv1 (Fig. 7 and Table 2)
suggests that LKv1 may correspond to IK in these
cells. Both LKv1 and IK do not inactivate; they have
similar half-maximal activation voltages V50 of 8.3
and 5.2 mV, respectively; and they are both relatively
insensitive to 4-AP, whereas they are more than 90%
reduced by 20 –25 mM TEA (Table 2). This is in
contrast to IA, which rapidly inactivates with a time
constant of 90 ms, which has a V50 of ⫺17 mV, and
which is completely blocked by 3 mM 4-AP and has
no demonstrable sensitivity to TEA (Table 2). However, the properties of IK and LKv1 are not completely
congruent, as the activation kinetics of LKv1 expressed in HEK293 cells are more than twice as slow
as those determined for IK at comparable voltages.
DISCUSSION
In this study, we have cloned a novel voltage-gated K
channel, LKv1, in two species of leech. Phylogenetic
analysis shows that while LKv1 is clearly related to
the Shaker subfamily of K channels and contains what
may be this subfamily’s signature sequence of SxxxLRV at the beginning of the S4 domain: It occupies a
separate branch from that of Shaker homologs from
flatworm, Aplysia, Drosophila, and vertebrates, which
form a distinct monophyletic clade. LKv1 is also
placed separately from two recently cloned jellyfish K
294
Weiss et al.
Figure 3 Consensus maximum parsimony tree derived
from Hirudo LKv1 and 13 other voltage-gated K channels
representing the Shaker, Shal, Shab, and Shaw subfamilies.
The tree is rooted using sequences from the first homology
domain of rat Na channel II as an outgroup (RatNaII,
X03639) (Noda et al., 1986). The strict consensus of 100
maximum parsimony trees is depicted with associated bootstrap support values. The consistency index for each of the
individual most parsimonious trees used to construct the
consensus tree was 0.78. The following voltage-gated K
channel sequences were used: Drosophila Shaker (DroShak, P08512) (Papazian et al., 1987); mouse Shaker
(MShak, P16388) (Tempel et al., 1988); Aplysia Shaker
(AplShak, M95914) (Pfaffinger et al., 1991); flatworm
Shaker (FwormShak, L26968) (Kim et al., 1994); jellyfish
Shaker 1 and 2 (JShak1/JShak2, U32922/U32923) (Jegla et
al., 1995); Drosophila Shab, Shal, and Shaw (DroShab/
DroShal/DroShaw, M32659/M32660/M32661) (Butler et
al., 1989); mouse Shaw (MShaw, S69381) (Goldman-Wohl
et al., 1994); mouse Shal (MShal, M64226) (Pak et al.,
1991a); mouse Shab (MShab, M64228) (Pak et al., 1991b);
and Aplysia Shab (AplShab, S68356) (Quattrocki et al.,
1994).
channels, jShak1 and jShak2 (Jegla et al., 1995). From
this phylogenetic analysis, it is clear that LKv1 and
jShak1 and jShak2 cannot be true homologs of the K
channels comprising the monophyletic Shaker clade,
but rather are orthologs. Thus, the Shaker subfamily
as defined by containing the SxxxLRV motif contains
several divergent lineages of voltage-gated K channels, underscoring the evolutionary diversity of K
channels (Salkoff and Jegla, 1995). However, whether
LKv1 and the jellyfish Shakers are unique to their
respective organisms and fulfill a specialized function
in these species, or whether they define new subgroups with representatives in other organisms remains to be determined.
The functional properties of LKv1 when expressed
in HEK293 cells are that of a noninactivating delayed
rectifier current. This is the first example of a Shakerrelated ␣-subunit in invertebrates with these properties. All previously described invertebrate Shakers
have been of the rapidly activating and inactivating
A-type currents; however, in mammals Shaker genes
have been found to code for proteins giving rise to
both types of conductances (Stuhmer et al., 1989).
LKv1 is activated at relatively positive potentials with
a V50 of 8.3 mV. This is more positive than that for the
higher metazoan Shakers, which typically have V50
values around ⫺15 mV, but not as extreme as the
positive shift in voltage dependence of the Shakerrelated jellyfish K channels, which have V50 values at
approximately 20 mV (Jegla et al., 1995). Although
LKv1 has only five positively charged amino acids in
the S4 transmembrane domain, its reduced voltage
sensitivity may not necessarily reflect a difference in
Figure 4 Polyclonal antiserum identifies LKv1 as a 54-kD
protein on immunoblots of Haemopis CNS extracts. Lane 2
shows labeling of a 54-kD band which is not present in the
preimmune serum (lane 1). Lane 3 demonstrates that preabsorption of LKv1 antiserum with LKv1 fusion protein
abolishes staining of the 54-kD protein. Lane 4 shows
labeling of the 54-kD protein by affinity-purified LKv1
antibody.
LKv1, a Novel K Channel in Leech
295
Figure 5 Expression of LKv1 in leech nervous system. (A) LKv1 antiserum labels central
neuronal somata (n) in adult Haemopis ganglia, in contrast to glia and muscle cells, which are not
labeled by the antibody. Anterior is at the top; scale bar ⫽ 150 ␮m. (B) Labeling of all developing
CNS neurons by LKv1 antiserum (three representative neurons are indicated by arrows) in a
whole-mount E9 embryonic ganglion. Anterior is at the top; scale bar ⫽ 70 ␮m. (C) Labeling by
LKv1 antiserum of peripheral neurons and nerves in the bodywall of a E12 Macrobdella embryo
hemisegment. Among the identified peripheral neurons labeled are the nephridial nerve cell (NNC),
the putative stretch receptor HO cells (arrows), and slightly out-of-focus four groups of sensillar
neurons (S2–S5). The axons of peripheral and central neurons project to and from the CNS (G)
through the AA, MA, DP, and PP nerves. Anterior is to the left and dorsal is at the top; scale bar
⫽ 75 ␮m.
gating charges. The jellyfish K channel, jShak2, while
being shifted to even more positive activation potentials than LKv1, still has seven positive charges in the
S4 region, as do all higher metazoan Shakers. Apart
from fewer positive charges in the S4 domain, LKv1
has several other significant differences in its amino
acid sequence from that of the other Shaker-related
channels. For example, there is a three–amino acid
deletion in the S5 domain which is compensated for
by a three–amino acid insert, CML, unique to LKv1
and found in both the Haemopis and Hirudo sequence. In addition, LKv1 has a very short C-terminal
domain consisting of only 17 amino acids. As for
Shakers in mammals (Chandy et al., 1990) and for the
Shaker-related K channels in the jellyfish Polyorchis
(Jegla et al., 1995), alternative splicing is not likely to
play a major role in generating functional diversity
because only a single intron is present in the coding
region for LKv1 of Haemopis genomic DNA. This
differs from Drosophila, in which extensive alternative splicing from a single Shaker gene gives rise to a
wide range of functional properties (Jan and Jan,
1990).
Antibody localization studies show that LKv1 is
296
Weiss et al.
Figure 6 Biophysical and pharmacological properties of
Hirudo LKv1 expressed in HEK293 cells. (A) Family of
outward currents in a HEK293 cell transiently transfected
with LKv1 cDNA. The traces were obtained by whole-cell
patch-clamp recording from a holding potential of ⫺80 mV
with voltage steps to 34 mV in 12-mV increments. (B) Peak
conductance–voltage relationship of the current traces
shown in (A). The solid line represents a Boltzman fit of the
data (see Materials and Methods). For this current family,
V50 was 9.9 mV and the slope factor a was 9.1 mV/e. (C)
Tail-current reversal of LKv1 currents. The currents were
elicited by a depolarizing step to 34 mV from a holding
potential of ⫺80 mV. After the depolarization, the holding
potential were varied in 30-mV increments from ⫺120 to
⫺30 mV. The four traces are superimposed. (D) Plot of the
relationship between peak tail-current amplitude and voltage. The reversal potential of the tail currents is close to -80
mV which is the calculated EK. (E) LKv1 current is more
than 90% blocked by 20 mM TEA. The left panel shows a
control current, the middle panel shows the current in the
presence of 20 mM TEA, and the right panel is after wash,
showing that the effect of TEA is reversible. The currents
were elicited from a holding potential of ⫺80 mV by a
depolarizing step to 34 mV. (F) The response of LKv1
current to 4 mM 4-AP. The left panel show the control
expressed by all neuronal somata and axons in both
the CNS and PNS. Thus, LKv1 may be the major
delayed rectifier current modulating excitability in
leech neurons. This is in contrast to Drosophila neurons and muscle, where the main delayed rectifier
current is encoded by Shab (Tsunoda and Salkoff,
1995a), and where Shaker currents appear to be expressed mainly in muscles and photoreceptors
(Tsunoda and Salkoff, 1995b). In leech, LKv1 is not
only expressed in mature neurons but also in the
newly differentiated neurons in ganglia of E8 –E9
embryos, suggesting a developmental role for LKv1
as excitability of the neurons is gradually being established.
Comparison of the biophysical and pharmacological properties of LKv1 as expressed in HEK293 cells
with those of IK and IA, two previously identified
native voltage-gated K channel currents in the soma
of leech neurons (Johansen and Kleinhaus, 1986,
1990), suggests that LKv1 corresponds to IK. LKv1
and IK are both delayed rectifier-type currents which
do not inactivate; they have similar voltage dependence for activation; and they are both relatively
insensitive to 4-AP and are more than 90% blocked by
20 –25 mM TEA. Furthermore, IK has been demonstrated to be present in every central neuron examined
in several leech species including medial nociceptive
neurons and touch cells, which do not possess the
transient inactivating A-type K current, IA (Johansen
and Kleinhaus, 1990). Thus, the known distribution of
IK correlates well with the finding from the antibody
localization studies that LKv1 is present in the soma
of all neurons. The only difference in the properties
measured for both IK and LKv1 is that the activation
kinetics of LKv1 in HEK293 cells are more than twice
as slow as those determined for IK in situ at comparable membrane potentials. However, several cellspecific factors due to the use of a heterologous expression system may account for the observed
differences in activation and/or inactivation kinetics
of native and expressed channels (Baro et al., 1996).
First, distinct cellular milieus between HEK293 cells
and leech neurons could lead to cell-specific differences in posttranslational modifications that could
alter the properties of the currents. Second, the gating
properties of channels can be affected by their phosphorylation state owing to the action of cell specific
protein kinases or phosphatases (Levitan, 1994; Drain
et al., 1994). Third, the presence or absence of ␤-sub-
current, the middle panel is the current in the presence of 4
mM 4-AP, and the right panel is the current after wash. The
currents were elicited from a holding potential of ⫺80 mV
by a depolarizing step to 34 mV.
LKv1, a Novel K Channel in Leech
297
Table 2 Biophysical and Pharmacological Properties of LKv1 and Comparison with Two Native Voltage-Gated K
Currents, IA and IK, Found in Leech Neuronal Somata
Activation
V 50 (mV)
Slope a (mV/e)
Time constant at 10 mV (ms)
Inactivation
Time constant at 10 mV
Pharmacology
Sensitivity to TEA†
Sensitivity to 4-AP†
LKv1 in HEK293 Cells
IK in Leech
Retzius Cells*
IA in Leech
Retzius Cells*
8.3 ⫾ 0.3(n ⫽ 8)
11.2 ⫾ 0.4(n ⫽ 8)
18.5 ⫾ 0.9(n ⫽ 6)
5.2
7.8
8
⫺17.0
9.0
2
⬎1 s
⬎1 s
90 ms
⬎90% (20 mM; n ⫽ 4)
10–15% (4 mM; n ⫽ 4)
⬎95% (25 mM)
10–15% (3 mM)
Not sensitive
100% (3 mM)
* The biophysical and pharmacological data for IK and IA were obtained from Johansen and Kleinhaus (1986).
†
The sensitivity to TEA and 4-AP is given as the reduction in percentage of peak current in the presence of the drug.
units interacting with the ␣-subunit of the channel
(Pongs, 1995) may also account for differences in the
biophysical properties of ␣-subunits expressed in heterologous cells. Thus, the majority of the evidence,
and especially the pharmacological data, strongly suggest that LKv1 is the ␣-subunit responsible for the IK
current in leech neurons.
The further cloning and molecular and biophysical characterization of different leech conductances in combination with the particularly advantageous features of the leech CNS promise to make
possible a more comprehensive understanding of
how neuronal networks underlying specific behaviors are organized, how functionally different cells
within these networks express different combinations of ion conductances, and how the specific
properties of the conductances correlate with a neuron’s particular role in behavior.
The authors thank Dr. A. Wei and Dr. L. Salkoff for
generously providing primers, laboratory facilities, and help
with the initial identification of partial sequence for LKv1.
They also thank Anna Yeung for expert technical assistance,
Roberta Allen for constructing the Hirudo cDNA library,
and Dr. J. F. Wendel and Dr. G. Naylor for help and advice
on the phylogenetic analysis. This work was supported by
an Iowa State Biotechnology Grant, an Iowa State University Grant, NIH Grant NS28857, and NSF Training Grant
Figure 7 Comparison of LKv1 expressed in HEK293 cells with the two native voltage-gated K
currents, IK and IA, in the soma of leech neurons. (A) Family of outward currents in a HEK293 cell
transiently transfected with LKv1 cRNA. The traces were obtained by whole-cell patch-clamp
recording from a holding potential of ⫺80 mV with voltage steps to 34 mV in 12-mV increments.
(B) Family of delayed rectifier currents, IK, obtained by two-electrode voltage clamp in an acutely
isolated leech Retzius cell. The currents were elicited from a holding potential of ⫺40 mV in
response to depolarizing steps to 30 mV in 10-mV increments (modified from Fig. 3 in Johansen and
Kleinhaus, 1986). (C) Family of transient K currents, IA, obtained by two-electrode voltage clamp
in an acutely isolated leech lateral nociceptive cell. The currents were elicited from a holding
potential of ⫺70 mV in response to depolarizing steps to 10 mV in 10-mV increments (modified
from Fig. 1 in Johansen and Kleinhaus, 1990). All three families of K currents are shown on the
same time scale of 100 ms.
298
Weiss et al.
DIR9113595 graduate fellowships (JLW and DLW). This is
Journal Paper No. 17815 of the Iowa Agriculture and Home
Economics Experiment Station, Ames, Iowa, Project No.
3371, and is supported by the Hatch Act and State of Iowa
funds.
REFERENCES
Altschul SF, Gish W, Miller W, Myers EW, Lipman DJ.
1990. Basic local alignment search tool. J Mol Biol
215:403– 410.
Arbas EA, Calabrese RL. 1987. Ionic conductances underlying the activity of interneurons that control heartbeat in
the medicinal leech. J Neurosci 7:3945-–3952.
Ausubel FM, Brent R, Kingston RE, Moore DD, Seidman
JG, Smith JA, Struhl K. 1987. Current protocols in molecular biology. New York: Wiley.
Baro DJ, Coniglio LM, Cole CL, Rodriguez HE, Lubell JK,
Kim MT, Harris-Warrich RM. 1996. Lobster shal: comparison with Drosophila shal and native potassium currents in identified neurons. J Neurosci 16:1689 –1701.
Blackshaw SE, Nicholls JG, Parnas I. 1982. Physiological
responses, receptive fields and terminal arborizations of
nociceptive cells in the leech. J Physiol 326:251–260.
Butler A, Wei A, Baker K, Salkoff L. 1989. A family of
putative potassium channel genes in Drosophila. Science
243:943–947.
Chandy KG, Williams CB, Spencer RH, Aguilar BA, Ghanshani S, Tempel BL, Gutman GA. 1990. A family of
three mouse potassium channel genes with intronless
coding regions. Science 247:973–975.
Devereux J, Haekerli P, Smithies O. 1984. A comprehensive set of sequence programs for the VAX. Nucl Acids
Res 12:387–395.
Drain P, Dupin AE, Aldrich RW. 1994. Regulation of
Shaker K⫹ channel inactivation gating by the cAMPdependent protein kinase. Neuron 12:1097–1109.
Felsenstein J. 1985. Confidence limits on phylogenies: an
approach using the bootstrap. Evolution 39:783–791.
Fernandez J, Stent, GS. 1982. Embryonic development of
the hirudinid leech Hirudo medicinalis: structure, development and segmentation of the germinal plate. J Embryol Exp Morphol 72:71–96.
Goldman-Wohl DS, Chan E, Baird D, Heintz N. 1994.
Kv3.3b: a novel Shaw type potassium channel expressed
in terminally differentiated cerebellar Purkinje cells and
deep cerebellar nuclei. J Neurosci 14:511–522.
Harlow E, Lane E. 1988. Antibodies: a laboratory manual.
Cold Spring Harbor, NY: Cold Spring Harbor Laboratory.
Heginbotham L, MacKinnon R. 1992. The aromatic binding
site for tetraethylammonium ion on potassium channels.
Neuron 8:483– 491.
Huang Y, Jellies J, Johansen KM, Johansen J. 1997. Differential glycosylation of Tractin and LeechCAM, two
novel Ig-superfamily members, regulates neurite extension and fascicle formation. J Cell Biol 138:143–157.
Jan LY, Jan YN. 1990. How might the diversity of potas-
sium channels be generated? Trends Neurosci 13:415–
419.
Jegla T, Grigoriev N, Callin WJ, Salkoff L, Spencer AN.
1995. Multiple Shaker potassium channels in a primitive
metazoan. J Neurosci 15:7989 –7999.
Jellies J, Loer CM, Kristan WB Jr. 1987. Morphological
changes in leech Retzius neurons after target contact
during embryogenesis. J Neurosci 7:2618 –2629.
Johansen J. 1991. Ion conductances in identified leech neurons. Comp Biochem Physiol 100A:33– 40.
Johansen J, Kleinhaus AL. 1986. Transient and delayed
potassium currents in the Retzius cell of the leech, Macrobdella decora. J Neurophysiol 56:812– 822.
Johansen J, Kleinhaus AL. 1990. Ionic conductances in two
types of sensory neurons in the leech, Macrobdella
decora. Comp Biochem Physiol 97A:577–582.
Johansen J, Yang J, Kleinhaus AL. 1987. Voltage-clamp
analysis of the ionic conductances in a leech neuron with
a purely calcium-dependent action potential. J Neurophysiol 58:1468 –1484.
Johansen KM, Johansen J. 1995. Filarin, a novel invertebrate intermediate filament protein present in axons and
perikarya of developing and mature leech neurons. J Neurobiol 27:227–239.
Johansen KM, Kopp DM, Jellies J, Johansen J. 1992. Tract
formation and axon fasciculation of molecularly distinct
peripheral neuron subpopulations during leech embryogenesis. Neuron 8:559 –572.
Kim E, Day TA, Bennett JL, Pax RA. 1995. Cloning,
characterization and functional expression of a Shakerrelated voltage-gated potassium channel gene from Schistosoma mansoni (Trematoda: Digenea). Parasitology
110:171–180.
Kozak M. 1986. Point mutations define a sequence flanking
the AUG initiator codon that modulates translation by
eukaryotic ribosomes. Cell 44:283–292.
Kristan WB. 1982. Sensory and motor neurons responsible
for the local bending responses in leeches. J Exp Biol
96:161–180.
Laemmli UK. 1970. Cleavage of structural proteins during
assembly of the head of bacteriophage T4. Nature 227:
680 – 685.
Levitan IB. 1994. Modulation of ion channels by protein
phosphorylation and dephosphorylation. Annu Rev
Physiol 56:193–212.
Li M, Jan YN, Jan LY. 1992. Specification of subunit
assembly by the hydrophilic amino-terminal domain of
Shaker potassium channel. Science 257:1225–1230.
Macagno ER. 1980. Number and distribution of neurons in
leech segmental ganglia. J Comp Neurol 190:283–302.
Macagno ER, Stewart RR, Zipser B. 1983. The expression
of antigens by embryonic neurons and glia in segmental
ganglia of the leech Haemopis marmorata. J. Neurosci
3:1746 –1759.
Muller KJ, Nicholls JG, Stent GS. 1981. Neurobiology of
the leech. Cold Spring Harbor, NY: Cold Spring Harbor
Laboratory. 320 p.
Noda M, Ikeda T, Kayano T, Suzuki H, Takeshima H,
Kurasaki M, Takahashi H, Numa S. 1986. Existence of
LKv1, a Novel K Channel in Leech
distinct sodium channel messenger RNAs in rat brain.
Nature 320:188 –192.
Pak MD, Covarrubias M, Butler A, Ratcliffe A, Salkoff L.
1991a. mShal, a subfamily of A-type K⫹ channel cloned
from mammalian brain. Proc Natl Acad Sci USA 88:
4386 – 4390.
Pak MD, Covarrubias M, Ratcliffe A, Salkoff L. 1991b. A
mouse brain homoloque of the Drosophila Shab K⫹
channel with conserved delayed rectifier properties.
J Neurosci 11:869 – 880.
Papazian DM, Schwartz TL, Tempel BL, Jan YN, Jan LY.
1987. Cloning of genomic and complementary DNA
from Shaker, a putative potassium channel gene from
Drosophila. Science 237:749 –753.
Papazian DM, Timpe L, Jan YN, Jan LY. 1991. Alteration
of voltage-dependence of Shaker potassium channel by
mutations in the S4 sequence. Nature 349:305–310.
Pfaffinger PJ, Furukawa Y, Zhao B, Dugan D, Kandel ER.
1991.Cloning and expression of an Aplysia K⫹ channel
and comparison with native Aplysia K⫹ currents. J Neurosci 11:918 –927.
Pongs O. 1993. Shaker-related K⫹ channels. Semin Neurosci 5:93–100.
Pongs O. 1995. Regulation of the activity of voltage-gated
potassium channels by ␤-subunits. Semin Neurosci
7:137–146.
Quattrocki EH, Marshal J, Kaczmarek LK. 1994. A Shab
potassium channel contributes to action potential broadening in peptidergic neurons. Neuron 12:73– 86.
Salkoff L, Jegla T. 1995. Surfing the DNA databases for K⫹
channels nets yet more diversity. Neuron 15:489 – 492.
Sambrook J, Fritsch EF, Maniatis T. 1989. Molecular cloning: a laboratory manual. 2nd ed. Cold Spring Harbor,
NY: Cold Spring Harbor Laboratory.
Shen NV, Chen X, Boyer MM, Pfaffinger PJ. 1993. Deletion analysis of K⫹ channel assembly. Neuron 11:67–76.
299
Stent GS, Kristan WB, Friesen WO, Ort CA, Poo M, Calabrese RL. 1978. Neuronal generation of the leech swimming movement. Science 200:1348 –1357.
Stent GS, Thompson WJ, Calabrese RL. 1979. Neuronal
control of heartbeat in the leech and in some other invertebrates. Physiol Rev 59:101–136.
Stuhmer W, Ruppersberg JP, Schroter KH, Sakman B,
Stacker M, Giese KP, Perschke A, Baumann A, Pongs O.
1989. Molecular basis of functional diversity of voltagegated potassium channels in mammalian brain. EMBO J
8:3235–3244.
Stewart RR, Nicholls JG, Adams WB. 1989. Na⫹, K⫹, and
Ca2⫹ currents in identified leech neurones in culture. J
Exp Biol 141:1–20.
Swofford DL. 1993. Phylogenetic analysis using parsimony. Champaign, IL: Illinois Natural History Survey.
Tempel BL, Jan YN, Jan LY. 1988. Cloning of a probable
potassium channel gene from mouse brain. Nature 332:
837– 839.
Timpe LC, Schwarz LC, Tempel BL, Papazian DM, Jan
YN, Jan LY. 1988. Expression of functional potassium
channels from Shaker cDNA in Xenopus oocytes. Nature
331:143–145.
Towbin H, Staehelin T, Gordon J. 1979. Electrophoretic
transfer of proteins from polyacrylamide gels to nitrocellulose sheets: procedure and some applications. Proc Natl
Acad Sci USA 76:4350 – 4354.
Tsunoda S, Salkoff L. 1995a. The major delayed rectifier in
both Drosophila neurons and muscle is encoded by Shab.
J Neurosci 15:5209 –5221.
Tsunoda S, Salkoff L. 1995b. Genetic analysis of Drosophila neurons: Shal, Shaw, and Shab encode most embryonic potassium currents. J Neurosci 15:1741–1754.
Wenning A. 1983. A sensory neuron associated with the
nephridia of the leech Hirudo medicinalis L. J Comp
Physiol 152:455– 458.