| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
ARTICLES |
Departments of Obstetrics, Gynecology, and Reproductive Sciences (K.A.G., T.J.O., M.A., O.L.) and Physiology (K.A.G., T.P.F., J.S.H., P.A.W.), and the Center for Studies in Reproduction (K.A.G., J.S.H.), University of Maryland, Baltimore, Maryland 21201
Address all correspondence and requests for reprints to: Dr. Karen A. Gregerson, Department of Obstetrics, Gynecology, and Reproductive Sciences, 11–007 Bressler Research Building, University of Maryland, 655 West Baltimore Street, Baltimore, Maryland 21201.
 |
Abstract |
|---|
 Top Abstract IntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Functional analysis of the GIRK proteins was performed in the
heterologous expression system, Xenopus laevis oocytes.
Macroscopic K+ currents were examined in oocytes injected
with different combinations of Kir 3.0 complementary RNA
(cRNA) and G protein subunit (ß1
2) cRNA.
The current-voltage relationships demonstrated strong inward
rectification for each individual and pairwise combination of GIRK
channel subunits. Oocytes coinjected with any pair of GIRK subunit cRNA
exhibited significantly larger inward K+ currents than
oocytes injected with only one GIRK channel subtype. Ligand-dependent
activation of only one of the GIRK combinations (GIRK1 and GIRK4) was
observed when channel subunits were coexpressed with the D2
receptor in Xenopus oocytes. Dose-response data fit to a
Michaelis-Menten equation gave an apparent Kd similar to
that for DA binding in anterior pituitary tissue. GIRK1 and GIRK4
proteins were coimmunoprecipitated from anterior pituitary lysates,
confirming the presence of native GIRK1/GIRK4 oligomers in this tissue.
These data indicate that GIRK1 and GIRK4 are excellent candidate
subunits for the D2-activated, G protein-gated channel in
pituitary lactotropes, where they play a critical role in
excitation-secretion coupling.
|
Introduction |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
We have identified and characterized a voltage-independent, inwardly rectifying potassium channel activated by DA (KDA) that leads to hyperpolarization of the lactotrope membrane and cessation of calcium-dependent action potentials, the driving force for tonic PRL secretion (6, 7). Both the activation of this effector K+ channel and the inhibition of PRL release are mediated by D2 receptor activation and pertussis toxin (PTx)-sensitive G proteins. In vitro studies from our laboratory have demonstrated a critical role for this KDA channel in both the inhibition of PRL release by physiological (nanomolar) concentrations of DA and the stimulatory phase of PRL secretion elicited by DA withdrawal (5, 8).
The purpose of the present studies was to examine the molecular basis of this physiologically important channel by 1) employing a homology-based PCR approach to identify members of the inward rectifier K+ (IRK) channel family that are expressed in AP, and 2) performing functional analysis of candidate proteins using the Xenopus oocyte expression system. Using this approach we have for the first time cloned G protein-gated IRK channel transcripts from AP tissue. The present report also identifies two subunits, associated in pituitary tissue, which when expressed together exhibit D2 receptor-dependent activation similar to the KDA channel studied in primary lactotropes.
|
Materials and Methods |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Reverse hemolytic plaque assay (RHPA)
RHPA was used to identify lactotropes for electrophysiology and
for assessment of PRL release from individual cells. Briefly,
dissociated pituitary cells (2 x 105
cells/ml) were plated together with an excess of protein A-coated
erythrocytes on poly-L-lysine-coated glass coverslips in
modified Cunningham chambers (9). The chambers were placed
in a 95% air-5% CO2 atmosphere at 37 C for 45
min to allow cells to attach to the coverslip. Unattached cells and
excess erythrocytes were rinsed away with DMEM (without phenol red)
containing 0.1% BSA (DMEM-BSA). PRL antiserum (arPRL-86; final
dilution, 1:200) (10) was then introduced into the
chambers and allowed to incubate for 60 min. Chambers were again
rinsed, and areas of hemolysis (plaques) surrounding the PRL-secreting
cells were initiated with serum complement. When lactotropes were being
identified for subsequent electrophysiological studies, the source of
complement was the serum harvested from the donor of the pituitary
cells (8). This autologous serum was used at a final
concentration of 1:140 or 1:160 for 15–20 min. At the end of this
incubation, the chambers were dismantled while submerged in DMEM
containing 10% deactivated horse serum and 0.04 mg/ml gentamicin, and
coverslips with cells attached were transferred immediately to six-well
plates containing DMEM with 10% deactivated horse serum and 0.04 mg/ml
gentamicin. Cells were maintained in culture for 1–2 days until used
for patch-clamp studies.
For analysis of release from individual cells, plaque assays were performed that included 0.5 mM ascorbic acid (control) with or without 10-6 M DA in the incubation with the antiserum. These treatments were also performed in the presence of 10-6 M (+)butaclamol or after pretreatment (8 h) with PTx (1 µg/ml). Complete plaque development was accomplished by incubation with guinea pig serum complement at a final dilution of 1:50 for 30 min (37 C). Cells were then fixed with 2% glutaraldehyde for 10 min on ice, and the cells were stained with phloxine (2.5% in distilled water), followed by 0.05% azure II and 0.02% methylene blue in 0.05% sodium borate buffer. Stains were differentiated in a rosin-ethanol solution to produce bright blue nuclei and pale pink cytoplasm.
Plaque areas were quantitated by morphometric measurements using an image analysis system with JAVA imaging software (Jandel Scientific, Corta Madera, CA). The area of each plaque was calculated from two perpendicular measurements of diameter. Statistical significance was determined by two-way ANOVA followed by a t test between independent means.
Patch-clamp studies
Whole cell membrane potential recordings from plaque-identified
lactotropes were made according to the giga-ohm seal patch-clamp
technique. The perforated patch modification (11) was used
to avoid loss of soluble cytosolic factors. In this configuration,
electrical access to the cell interior is obtained through membrane
spanning pores formed by antibiotics contained in the patch pipette
rather than by disrupting the whole patch membrane.
Amphotericin stock solution (6 mg in 0.1 ml
dimethylsulfoxide) was made daily, sonicated, and kept on ice. This was
diluted in internal recording solution to a final concentration of 200
µg/ml. The tip of the recording pipette was filled with
amphotericin-free solution, then back-filled with the
amphotericin-containing solution. Access resistances ranged from 6–10
M
. The standard intracellular solution was comprised of 130
mM potassium aspartate, 20 mM KCl, 10
mM glucose, and 10 mM HEPES. Cells were bathed
in standard extracellular solution comprised of 145 mM
NaCl, 5 mM KCl, 2 mM
CaCl2, 1 mM
MgCl2, 10 mM glucose, and 10
mM HEPES. All solutions were adjusted to pH 7.3–7.4 and
295–305 mosmol.
Patch-clamp experiments were performed at room temperature in a Lexan
recording chamber mounted on the stage of a Nikon Diaphot
inverted microscope. A coverslip with the plaque-identified pituitary
cells attached comprised the floor of the recording chamber through
which external solutions were continuously perfused during the
experiment. Application of DA was accomplished by a U-tube device that
could be positioned next to a cell to apply and withdraw the test
solution rapidly while minimizing mechanical disturbance. The
D2 receptor antagonist, (+)butaclamol, was bath
applied. Whole cell voltage responses under current clamp were recorded
wideband (
50 kHz) on videotape using a digital audio processor
interface sampling at 44 kHz (PCM701, Sony, Park Ridge, NJ).
RNA isolation
APs harvested from proestrous female rats were frozen in liquid
nitrogen, pooled, weighed, and then homogenized at 5 C in 10 ml of a
solution containing 4 M guanidine thiocyanate, 25
mM sodium citrate (pH 7.0), 0.5% sodium lauryl sarcosine,
and 0.1 M ß-mercaptoethanol. Total RNA was subsequently
isolated using the extraction procedure described by Chomczynski and
Sacchi (12). Cerebellar and cardiac atrial RNA were
harvested in an identical manner. Both were then selected for
polyadenylated [poly(A)+] RNA by oligo(deoxythymidine) chromatography
using either spin columns (Pharmacia Biotech, Piscataway,
NJ) or the PolyAttract system (Promega Corp., Madison,
WI). Yields of messenger RNA (mRNA) were 1–2% of the total RNA with
OD 260/280 ratios greater than 1.80. RNA integrity was assessed by
denaturing agarose gel electrophoresis. In some studies isolation of AP
mRNA was performed directly from tissue homogenates using the FastTrack
2.0 system according to the manufacturer’s recommendations
(Invitrogen, San Diego, CA) with similar results.
RT-PCR
Rat pituitary gland mRNA (2–5 ng) was reverse transcribed using
oligo(deoxythymidine) and SuperScript reverse transcriptase (200 U) at
42 C for 50 min using the SuperScript RT kit (Life Technologies, Inc.). After the RT reaction, ribonuclease H was added to each
reaction tube (0.1 U/ml) and incubated at 37 C for 20 min. Negative
control reactions (RT-) were handled in an identical manner, except
that reverse transcriptase was excluded.
Nondegenerate primers corresponding to regions specific for IRK channels known to be G protein activated (GIRKs; gene subfamily Kir 3.0) were synthesized and used in the PCR with reverse transcribed female rat AP mRNA as the template. Each PCR reaction was carried out in 50 µl containing 1 µl AP RT reaction solution, 50 pM 5'- and 3'-primers, and reagents from the GeneAmp PCR Reaction system (Perkin-Elmer Corp., Palo Alto, CA) including 1 U AmpliTaq DNA polymerase. The hot start method was used to initiate the reaction. AmpliTaq DNA polymerase was withheld until the reaction temperature had reached 75 C. After the addition of the enzyme, the reaction was sequentially cycled 36 times for 1-min durations at each of the following temperatures: 94 C (denaturing); 54.5 C (Kir 3.2), 57 C (Kir 3.1 and Kir 3.4), or 62.5 C (Kir 3.3) (annealing); and 74 C (extending), using a thermal cycler MJ Research, Inc. (Cambridge, MA). All reactions were incubated at 72 C for 5 min after the last cycle for final extension, then placed on ice.
Kir 3.0 isoform-specific primer sets were used to
independently and specifically amplify Kir 3.0 complementary
DNAs (cDNAs) that are expressed in the AP. The subunit-specific primer
design shown with respect to the proposed primary structure of
Kir channels is illustrated in Fig. 4
. The primers used were
oligonucleotides corresponding to base pairs as follows: for
Kir 3.1, bp 378–401 (sense, 5'-ATGTCGGCAACTACACTCCCTGTG-3';
melting temperature (Tm) = 61.9 C) and bp 1115–1092 (antisense,
5'-CCTGCTCTTTCACACTGTACGGAG-3'; Tm = 61.9 C); for Kir
3.2, bp 569–586 (sense, 5'-CGGGGAGATATGGACCAC-3'; Tm = 49.5 C)
and bp 1123–1105 (antisense 5' CAGTTCCTCTTTAGGCAGC 3'; Tm = 46.6
C); for Kir 3.3, bp 511–528 (sense,
5'-TCGACCTGGAGCACCTGG-3'; Tm = 54 C) and bp 1313–1293 (antisense,
5'-GGGGATGGACCAGTAGAGATG-3'; Tm = 52.3 C); and for Kir
3.4, bp 522–541 (sense, 5'-TCCGAGGTGATCTGGACCAC-3'; Tm = 58.7 C)
and bp 1302–1282 (antisense, 5'-CCATTCCTCTTCATTTCTGCC-3'; Tm =
55.2 C). The specific sequences were chosen because these primers
correspond to regions that are not well conserved across all of the
known Kir channels (0–65% identity at the nucleotide
level).
|
Analysis of cycle-dependent amplification was used to determine the
relative differences in mRNA levels for the various GIRK isoforms
expressed in AP tissue. PCR reactions were carried out as described
above, except the number of cycles varied, ranging from 16–30. The
same AP RT+ reaction was used as the template for
amplification of all the unknown PCR reactions. To ensure that the
various primer sets amplified with similar efficiencies, identical
cycle-dependent PCR reactions were run using a known quantity of the
subcloned Kir 3.0 cDNAs as the templates for amplification
(Fig. 4B
). Each PCR reaction volume (50 µl) contained 0.5 pg of the
clone in the pCR-Script vector. The PCR products were fractionated by
electrophoresis in a 0.8% agarose gel, stained in ethidium bromide,
visualized with a UV transilluminator, and photographed using type 665
positive/negative film (Polaroid Corp., Cambridge, MA). The negative
images of the amplified products were analyzed by autoradiographic
scanning using a model 620 video densitometer (Bio-Rad Laboratories, Inc., Richmond, CA). The amount (band
intensity x area) of an amplified product was calculated as the
area under each sample band after baseline subtraction.
Northern blot analysis
Equal amounts (10 µM) of
poly(A)+ RNA extracted from APs and atria of
proestrous rats were separated on 1.2% agarose-formaldehyde gels and
transferred to nylon membrane. 32P-Labeled random
prime extension probes corresponding to the regions amplified in the
RT-PCR analyses (see Fig. 4) were generated for each Kir 3.0
isoform. The hybridization was performed in 50% formamide
hybridization buffer at 42 C. Final washes were of high stringency
[0.1 x SSC (standard saline citrate) and 0.1% SDS, 50 C].
Blots were exposed to Kodak XAR-5 film (Eastman Kodak Co., Rochester, NY) with an intensifying screen at
-70 C.
Complementary RNA (cRNA) synthesis
RNA was transcribed in vitro in the presence of
capping analog [m7G(5')ppp(5')G] using the
mMESSAGE mMACHINE kit (Ambion, Inc., Austin, TX). Briefly,
linearized cDNA was used as the template with the appropriate RNA
polymerase (T7 or SP6) used in two reactions. After deoxyribonuclease
treatment, cRNA was purified by phenol-chloroform extraction and
ammonium-acetate/ethanol precipitation. Yield and concentration were
quantified by spectrophotometry.
Oocyte injection
Standard protocols for the isolation and care of Xenopus
laevis oocytes were followed. Briefly, frogs were anesthetized by
immersion in 0.5% tricaine, and a partial oophorectomy was performed
through an abdominal incision. Oocyte aggregates were manually
dissected from the ovarian lobes and then incubated in
Ca2+-free ORII medium (82.5
mM NaCl, 2 mM KCl, 1
mM MgCl2, and 5
mM HEPES, pH 7.5) containing collagenase
(Sigma, type IA, 2 mg/ml) for approximately 2 h at
room temperature to remove the follicular layer. After washing the
oocytes extensively with collagenase-free ORII, they were placed in L15
medium (50% Leibovitz’s medium and 10 mM HEPES,
pH 7.5) and stored at 19 C. Twelve to 15 h after isolation,
healthy-looking Dumont stage V–VI oocytes were pneumatically injected
(PV 820 picopump, WPI, New Haven, CT) with 50 nl water
containing 0–5 ng cRNA. This range was chosen for optimal expression
based on our previous studies. The oocytes were then stored in L15
medium at 19 C, and channel activity was assessed 2–6 days
postinjection. X. laevis donors were selected for those
bearing oocytes with high translation activity and low endogenous
Kir channel activity (13).
Two-microelectrode voltage clamp
Whole cell oocyte currents were monitored using a
two-microelectrode voltage clamp equipped with a bath clamp circuit
(OC-725B, Warner, New Haven, CT). For these studies, oocytes were
placed in a small lucite chamber and continually superfused with
Ca-free ND88 (88 mM NaCl, 2 mM KCl, 1
mM MgCl2, and 5 mM HEPES,
pH 7.4) or KD-90 solution (90 mM KCl, 1 mM
MgCl2, and 5 mM HEPES, pH 7.4) at
room temperature (21-23 C). Voltage-sensing and current-injecting
electrodes had resistances of 0.5–1.5 M when back-filled with 3
M KCl. After attaining a stable impalement such that both
electrodes measured the same spontaneous membrane potential (±4 mV),
the pulse protocols shown below were conducted. Stimulation and data
acquisition were performed with a Macintosh Centris 650 computer using
an Instrutech ITC16 A/D, D/A converter and Pulse software. Data were
filtered at 2 kHz and digitized on-line to the hard disk using Pulse
for later analysis using Pulsefit.
Immunoprecipitation and Western blot analysis
APs, cerebellum, and cardiac atria harvested from proestrous
female rats were frozen in liquid nitrogen until used for protein
analysis. Tissues were homogenized in solubilization buffer [50
mM HEPES (pH 7.6), 150 mM NaCl, 1% Triton
X-100 with 1 mM EGTA, 10 µg/ml aprotinin, 10 µg/ml
leupeptin, and 1 mM phenylmethylsulfonylfluoride] at a
concentration of approximately 20 mg/ml. After 30 min on ice, lysate
was cleared by centrifugation (10 min at 10,000 x g),
and the supernatant was saved.
For Western blots, the cleared lysate was mixed 1:1 with 2 x Laemmli sample buffer and boiled for 5 min. After SDS-PAGE, protein was transferred to nitrocellulose (Schleicher & Schuell, Inc., Keene, NH) and blocked overnight with 2% nonfat milk/Tris-buffered saline. Blots were exposed to primary antibodies for 1 h. The anti-GIRK1 antibody (Alomone Laboratories, Jerusalem, Israel) was used at a final concentration of 0.4 ng/ml. The anti-GIRK4 antibody (aCIRN2) (14) was used at a final concentration of 1.2 ng/ml. Both of these antibodies have been extensively characterized and demonstrated to be specific. Indeed, the anti-GIRK4 antibody generated by Krapivinsky provided the critical reagent for identification of the GIRK4 protein in cardiac atrium. Chemiluminescent detection was performed with ECL (Amersham Pharmacia Biotech, Arlington Heights, IL).
For immunoprecipitations, the cleared lysate was generally diluted 1:1 with wash buffer (50 mM HEPES, 150 mM NaCl, and 0.1% Triton X-100). To 1 ml of this diluted lysate, 3–5 µl of primary antiserum were added. Samples were incubated at 4 C for 1 h with constant rocking. Fifty microliters of 50% protein A-agarose (Sigma) were added, and the incubation was continued for an additional hour. Samples were washed four times with 1 ml wash buffer (50 mM HEPES, 150 mM NaCl, and 0.1% Triton X-100) and boiled in 125 µl 1 x Laemmli sample buffer. Generally, 35 µl of this final sample were loaded in a lane for SDS-PAGE and processing for Western blots as described above.
|
Results |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
and summarized in
Fig. 2. During a 1-h incubation,
10-6 M DA was
able to suppress spontaneous PRL release, as represented by plaque
area, by approximately 67% (Figs. 1B and 2). Consistent with the
pharmacology of the D2 subtype of receptors, this
effect of DA was blocked by
10-6 M (+)butaclamol, a selective D2 antagonist (Figs. 1C and 2), whereas the inactive isomer (-)butaclamol
(10-6 M) did
not block DA inhibition of PRL release (data not shown). DA inhibition
of PRL release was also blocked by pretreatment of the pituitary cells
with pertussis toxin (an ADP-ribosylating agent that functionally
uncouples members of the
Gi/Go G protein family
(Figs. 1D and 2). As shown in Fig. 3, the
DA-activated hyperpolarization of individual lactotropes exhibits
identical pharmacology as the inhibitory effects on PRL release:
D2 receptor subtype and PTx sensitivity. The
D2 antagonist, (+)butaclamol, blocked the
DA-activated K+ current in 6 of 6 cells, whereas
application of (-)butaclamol (4 cells) was without effect. The
DA-induced hyperpolarization was absent in all lactotropes pretreated
with PTx (8 cells), whereas we routinely found a robust response in
greater than 85% of lactotropes (6) (11 of 12 cells in the present
study). Pretreatment of cells with cholera toxin (6 cells) was without
effect on the hyperpolarization elicited by DA (Fig. 3), demonstrating
that the activation of KDA is independent from
the effects of DA on cAMP in pituitary lactotropes.
|
|
|
). Figure 4A illustrates the
Kir 3.0 subunit- specific primer design and predicted
amplification product size with respect to the proposed primary
structure of Kir channels (shaded bar). The
filled boxes represent the two membrane-spanning domains
that flank the pore region of the channel. PCR products of predicted
sizes for Kir 3.1, 3.2, and 3.4 were observed by ethidium
bromide staining after 16 amplification cycles (n = 3 for each
transcript). A PCR product of predicted size for Kir 3.3 was
also obtained, but required many more amplification cycles to produce a
detectable product. In contrast, Kir 3.3 product was easily
amplified when reverse transcribed cerebellum mRNA from the same animal
was used as the template (not shown). The products for each
Kir 3.0 subunit obtained after 36 PCR amplification cycles
are shown in Fig. 4B. No products were observed in reactions using mRNA
rather than first strand cDNA as the template, ruling out spurious
genomic amplification (Fig. 4B). The identities of the PCR products
were confirmed by subcloning and sequencing several independent clones.
Sequences were identical to the corresponding regions published for
Kir 3.1, 3.2, 3.3, and 3.4.
The relative abundance of the four Kir 3.0 transcripts in AP
tissue were assessed by cycle dependence analysis. In these studies the
amount of PCR product produced from AP first strand cDNA (AP RT+) was
determined for a number of consecutive amplification cycles for each
transcript and compared with the cycle-dependent amplification of a
known amount of Kir 3.0 cDNA (Fig. 5A). These data fit the expected
characteristics of PCR amplification in which the amount of products
produced initially increases exponentially, but then the rate of
production slows and levels off (plateau effect) (15).
Because the cycle dependence of the cDNA amplifications is identical
for each isoform (Fig. 5B), the amplification efficiency of each
different primer set is not significantly different. Therefore,
differences in the cycle-dependent amplification of products from the
AP-RT+ template (upper series of photographs in Fig. 5A) can
be used to estimate relative differences in the expression levels of
the various Kir 3.0 transcripts in AP tissue. Although the
relative abundance of Kir 3.1, 3.2, and 3.4 in AP tissue is
quite similar, expression levels of Kir 3.3 are 2–3 orders
of magnitude less (Fig. 5C).
|
). The Kir 3.1 probe
hybridized to transcripts of 4.5 and 6.0 kb in both AP and cardiac
atrial mRNA. (Transcripts of 4.5 and 6.0 kb in atria also reported in
Refs. 14 and 16). The Kir 3.2
probe hybridized to a 2.4-kb transcript in AP only, consistent with
reports that this isoform is not expressed in heart (17).
A 2.4- to 2.8-kb transcript for Kir 3.2 in human insulinoma
cells has been reported (18). The Kir 3.4
probes hybridized to transcripts of 3.6 and 5.1 kb in both AP and
atria.
|
2) cRNA.
Representative families of K+ currents in oocytes
injected with GIRK1, GIRK2, GIRK4, or pairwise combinations of each are
shown in Fig. 7A. The current-voltage
relationships demonstrate strong inward rectification for each
individual and pairwise combination of GIRK channel subunit (Fig. 7B),
characteristic of the Kir channel family. Oocytes coinjected
with any pair of GIRK subunit cRNA exhibit significantly larger inward
K+ currents than oocytes injected with only one
GIRK channel subtype (Fig. 7C). These data are consistent with the
theory that the G protein-gated Kir channels are
heteromultimers.
|
). GIRK1 and GIRK4 were
coexpressed with either the long (D2L, Fig. 8) or the short
(D2S, Fig. 9) form
of the D2 receptor. Representative currents,
measured in the presence and absence of a maximal dose of DA (10
µM), are shown in Figs. 8A and 9A. The addition
of DA significantly increased channel activity (Figs. 8B and 9B),
although the presence of considerable ligand-independent current
suggests significant G protein-coupled receptor precoupling.
Dose-response data fit to the Michaelis-Menten equation (solid
line) gave an apparent Kd of 11.07 ±
2.14 nM for D2L (Fig. 8C)
and 17.58 ± 7.78 nM for
D2S (Fig. 9C). However, it should be noted that
most of the data points obtained were at the minimum or maximum
response level.
|
|
|
, lane 3), which represents heavily
glycosylated GIRK1 as also found in cardiac atrial membranes
(14). This mature form of the GIRK1 protein
coimmunoprecipitated with GIRK4 as shown in lane 4 of Fig. 10A and lane
6 of Fig. 10C. Likewise, GIRK4 coimmunoprecipitated with GIRK1, as
evidenced by the faint band at 45K recognized by the anti-GIRK4 (Fig. 10A, lane 1). This 45K GIRK4 is seen more clearly in lane 2 when
anti-GIRK4 was used for immunoprecipitation. The immunoprecipitation
with anti-GIRK1 was repeated using a lower concentration of detergent
(0.325% vs. 1% Triton X-100, final concentration), which
improved recovery of the associated GIRK4 protein (Fig. 10B, lane 4).
As shown in Fig. 10B, GIRK4 coimmunoprecipitated with GIRK1 from both
AP (lane 4) and atrial (lane 3) tissue, but not cerebellar (lane 5)
tissue where GIRK4 is not expressed. The 45K GIRK4 protein is also
clearly seen in Western blot analysis of whole cell lysates from AP and
atrium (Fig. 10B, lanes 1 and 2). The large band migrating at 50K in
the immunoprecipitated samples is the heavy chain of the rabbit
antibodies used for the immunoprecipitations. This is verified in the
controls presented in Fig. 10C. Staining of the heavy chain is shown in
lanes 1 and 5, which were only loaded with the anti-GIRK1 antiserum (no
cell lysate). Also contributing to this band is any contaminating
protein A Sepharose as demonstrated in lanes 2 and 8, which contained
cell lysate processed for immunoprecipitation with protein A-Sepharose,
but without any precipitating antibodies. This large band at 50K
(arrowheads) obscures the immature, nonglycosylated GIRK1
protein seen in whole cell lysate of AP tissue (arrow, lane
4, Fig. 10C) and has been previously reported in cardiac atria
(14). To further establish that the coimmunoprecipitations
of the GIRK1 and GIRK4 proteins truly reflect their association in AP
membranes and were not a procedural artifact, antiserum against the Kir
2.1 channel (Alomone Laboratories, Jerusalem, Israel) was used for
immunoprecipitation. This inwardly rectifying K+
channel does not assemble with GIRK proteins (19), and the
anti-Kir 2.1 did not coprecipitate either GIRK1 or GIRK4 (Fig. 10C, lanes 3 and 7, respectively) from AP membranes.
|
|
Discussion |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Molecular characterization of the G protein-gated IRKs in pituitary tissue is a necessary step in elucidating the physiological role of the DA-activated K+ channel in the regulation of PRL secretion. Dopaminergic inhibition of PRL release is mediated via the D2 subtype of DA receptors (22, 23). D2 receptors are both physically and functionally coupled to a PTx-sensitive G protein-coupled receptor (24, 25), and this receptor-G protein coupling plays a necessary role in the dopaminergic inhibition of secretion (26, 27, 28). In vitro studies demonstrate that a DA-activated K+ conductance, exhibiting identical pharmacological properties as these, plays a critical role in mediating the effects of DA on PRL release (5).
We have identified and characterized a voltage-independent K+ channel underlying this DA-activated conductance (KDA). Activation of this channel involves the D2 subtype of DA receptor and a PTx-sensitive G protein-coupled receptor and occurs over the same range of DA concentrations that inhibit PRL release (5–100 nM) (6). Single channel analysis has demonstrated that the KDA is a 40pS K+ channel that is voltage independent and inwardly rectifying (7). These are the two hallmarks of the IRK channel family (gene nomenclature Kir) (29, 30). The Kir 3.x subfamily of the IRK channels can be directly activated by G proteins (GIRKs) as is the KDA channel (31).
The DA-activated K+ conductance we have described is independent of Ca2+ and membrane potential and thus can be activated at or near the normal resting potential of the cell to impact on its physiology. This is consistent with experimental findings that a major action of DA in lactotropes is to activate a K+ conductance that is not normally manifested in the absence of DA and can support a long-lasting (at least many minutes) hyperpolarization (6, 7). Furthermore, these data indicate that members of the Kv channel family that are present in lactotropes in the absence of DA (32) and require membrane depolarization or Ca2+ for activation do not underlie the DA-activated responses.
Several experimental findings are consistent with the hypothesis that
D2 receptors in lactotropes are coupled to the
IRK channel directly and are not mediated by soluble cytoplasmic
factors. First, we found that the DA-activated hyperpolarizing response
is not lost over periods up to 1 h or more of whole cell dialysis
during electrophysiological studies. Second, the DA-activated
hyperpolarization is not diminished by buffering both external and
internal Ca2+ (with 10 mM EGTA and 10
mM BAPTA, respectively) (6). In addition, the
robust electrical responses are independent of D2
regulation of adenylate cyclase, as demonstrated in perforated patch
recordings from lactotropes treated with cAMP-elevating agents
(33). Finally, single D2
agonist-activated K+ channels have been recorded
in outside-out patches that are identical to those observed in
cell-attached patches in both conductance and gating properties
(31). These data suggest that the D2
receptor-K+ channel coupling in lactotropes is
mediated through a membrane-delimited pathway. Members of the G protein
gated subfamily of IRK channels (Kir 3.0) can be directly
activated by Gß dimers in a manner consistent with such
membrane-delimited activation of native channels (34, 35).
The expression of Kir 3.1, 3.2, and 3.4 in the AP make these proteins candidate subunits of the DA-activated, IRK channel in lactotropes. The abundance of these three transcripts in AP tissue was quite similar, and these transcripts were readily detected using Northern blot analysis of AP mRNA. Although we were able to amplify specific GIRK3 product from AP tissue, the low abundance of this transcript puts into question its physiological relevance.
The results of our functional analysis of the other three GIRK channel subunits are consistent with current theory that the G protein-gated Kir channels function as heteromultimers (14, 36). Oocytes injected with any pairwise combination of GIRK subunit cRNA always exhibited significantly larger inward K+ currents than oocytes injected with only one of the GIRK channel subtype pair. Injection of GIRK2 alone did produce K+ currents larger than either GIRK1 alone, GIRK4 alone, or the GIRK1/GIRK4 pair, as has been shown previously (36), and there is evidence that functional channels are formed from GIRK2 homomultimers (37).
Coexpression of the GIRK1/GIRK4 subunit pair with either the long (D2L) or the short (D2S) form of the D2 receptor resulted in ligand-dependent activation of the inward rectifying K+ current in Xenopus oocytes. The AP contains the mRNAs for both D2 receptor isoforms, although the expression levels of D2L are considerably greater than those of D2S (38, 39). DA dose-response data of the GIRK1/GIRK4 combination, fit to the Michaelis-Menten equation, yielded an apparent Kd of 11.07 ± 2.14 nM for D2L or 17.58 ± 7.78 nM for D2S, consistent with the Kd for ligand binding to native DA receptors in AP tissue (22) and within the concentration-dependent range of DA activation of the native KDA channel in identified lactotropes (6).
No ligand-dependent activation of either GIRK1/GIRK2 or GIRK2/GIRK4 was observed when these subunit pairs were coexpressed with either D2 receptor form. These findings suggest that the channels formed by these pairs are not functionally coupled to the D2 receptor. However, it is also possible that the very large ligand-independent currents produced by GIRK2 expression alone and in combination with other subunits masked any further increase in response to DA. Injecting smaller quantities of channel mRNA (to decrease channel and current density) did not reveal ligand activated currents in GIRK2 combinations (data not shown). Kuzhekandathil et al. (37) reported ligand-activated current when D2 receptors are coexpressed with GIRK2 homomultimers in CHO cells. Others (40, 41) have reported coupling between DA receptors and channels when only GIRK1 mRNA was injected. However, these channels were probably heteromultimers of the GIRK1 with the endogenous Kir channel found in Xenopus oocytes (XIR) (42, 43). In our studies we selected for oocytes expressing low endogenous Kir channel activity and did not examine D2 coupling with currents in oocytes injected with only one GIRK subunit mRNA. Further arguing against GIRK2 as a critical component of the lactotrope DA-activated channel is that PRL regulation appears normal in the weaver (wv) mutant mouse (44). The wv phenotype is associated with a missense mutation in the pore region of GIRK2 (45), rendering the channel constitutively active and nonselective in ion permeation. Homozygote mice have normal serum and pituitary contents of PRL, and the normal sex difference (PRL higher in females) is intact.
The presence of considerable ligand-independent GIRK1/GIRK4 channel activity in both D2L- and D2S-injected oocytes may be due to significant precoupling between the receptors and the GTP-binding proteins. This ligand-independent channel activity was consistently greater in the presence of the D2S receptor form. The two isoforms of the D2 receptor result from alternative splicing of the D2 DA receptor gene with the D2L isoform, including an insertion of 29 amino acids in the putative third cytoplasmic domain (38, 46, 47) that has been shown to be important for the coupling of seven-transmembrane receptors to GTP-binding proteins (48, 49) and differences in the efficiency of the two D2 isoforms to activate various effectors. For example, the D2S isoform has been demonstrated to be more efficient than D2L in inhibiting adenylate cyclase activity in JEG3 cells (50). The difference between the two receptor isoforms has been suggested to confer differential coupling to various G protein subtypes. Thus, the difference in ligand-independent GIRK1/GIRK4 current between D2L- and D2S-injected oocytes may reflect the specific G protein subtypes endogenously expressed in Xenopus oocytes. Studies to elucidate the apparent receptor-G protein precoupling are underway.
D2 receptor-dependent activation of GIRK1/GIRK4 is similar to the KDA channel in primary lactotropes. GIRK1 and GIRK4 have been demonstrated to be the subunits comprising the muscarinic-activated K+ channel in heart (14). Although many of our previous and present findings parallel those reported for the atrial KACh channel, important differences exist between the two native channels. First, the KDA channel in pituitary lactotropes exhibits weaker inward rectification in whole cell recordings, and although the atrial channel exhibits both rapid activation and rapid deactivation upon application and withdrawal of acetylcholine (51), the DA-induced hyperpolarization of lactotropes lasts considerably beyond agonist withdrawal (31). Such differences can reflect tissue-specific posttranslational processing of the subunit proteins, tissue-specific expression of different GTP-binding proteins, and/or differential involvement of additional signal transduction pathways that modulate the activity of the receptor-G protein-gated channel. In addition, the coexpression of other Kir subfamily members in the AP, which may interact with the GIRK proteins is possible. For example, Kir 2.3 has been cloned from GH3/B6 cells, a mammosomatotrope cell line, and demonstrated by Northern blot analysis to be expressed in AP tissue (52). Clearly, complete analysis of IRK subunits in AP tissue must be performed.
One must also consider the heterogeneity of cell types in this gland. Lactotropes are only one of six different secretory cell types in the AP. A recent report indicates that thyrotropes express GIRK1 (21), and GH-secreting adenoma cells isolated from acromegalic patients can express a DA-activated K+ current (53). In corticotropes, a constitutively active IRK current inhibited by CRH has been reported (54), as has evidence for a glybenclamide-sensitive component of proadrenomedullin N-terminal 20 peptide inhibition of ACTH release (55). However, in proestrous female glands, which were used in the present study, nearly half of the AP cells are lactotropes (10), and nearly 90% of lactotropes exhibit functional expression of the DA-activated K+ channel (6). Nevertheless, further molecular characterization of the DA-activated K+ channel in normal lactotropes will require cellular localization of specific Kir transcripts.
In conclusion, we demonstrate that three different G protein-gated K+ channel subunits are expressed in the AP gland: GIRK1, GIRK2, and GIRK4 (Kir 3.1, Kir 3.2, and Kir 3.4). Functional analyses of these Kir channel subunits in Xenopus oocytes demonstrate that the coexpression of GIRK1 and GIRK4 with the D2 receptor produces a DA-activated channel whose properties recapitulate many of the properties of the native KDA channel in pituitary lactotropes. Moreover, we found that native GIRK1/GIRK4 oligomers exist in AP membranes. Molecular characterization of the KDA channel is a critical step in elucidation of the pleiotropic actions of DA in the regulation of lactotrope function and PRL secretion.
|
Acknowledgments |
|---|
2. Thanks also to
Dr. Grigory Krapivinsky for his generous gift of anti-GIRK4
(aCIRN2). |
Footnotes |
|---|
2 Supported by a predoctoral fellowship on an interdepartmental
NIH Training Program in Integrative Membrane Biology
(T32-GM-08181).
Received December 27, 2000.
|
References |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
-subunit of GTP-binding proteins activate the
muscarinic K+ channel in heart. Nature 325:321–326[CrossRef][Medline]
subunits. Nature 370:143–146[CrossRef][Medline]
subunits and function as
heteromultimers. Proc Natl Acad Sci USA 92:6542–6546
G-protein subunits to inwardly rectifying K+
channels (GIRK1) in a Xenopus oocyte expression system:
selective antagonism by L-741,626 and L-745,870 respectively.
Neuropharmacology 37:983–987[CrossRef][Medline]
2-ß2 adrenergic receptors:
dilineation of domains involved in effector coupling and ligand binding
specificity. Science 240:1310–1316This article has been cited by other articles:
 |
|
 |
|
  A. E. Gonzalez-Iglesias, T. Murano, S. Li, M. Tomic, and S. S. Stojilkovic Dopamine Inhibits Basal Prolactin Release in Pituitary Lactotrophs through Pertussis Toxin-Sensitive and -Insensitive Signaling Pathways Endocrinology, April 1, 2008; 149(4): 1470 - 1479. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 A. E. Gonzalez-Iglesias, Y. Jiang, M. Tomic, K. Kretschmannova, S. A. Andric, H. Zemkova, and S. S. Stojilkovic Dependence of Electrical Activity and Calcium Influx-Controlled Prolactin Release on Adenylyl Cyclase Signaling Pathway in Pituitary Lactotrophs Mol. Endocrinol., September 1, 2006; 20(9): 2231 - 2246. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
D. K. Sarkar, K. Chaturvedi, S. Oomizu, N. I. Boyadjieva, and C. P. Chen Dopamine, Dopamine D2 Receptor Short Isoform, Transforming Growth Factor (TGF)-{beta}1, and TGF-{beta} Type II Receptor Interact to Inhibit the Growth of Pituitary Lactotropes Endocrinology, October 1, 2005; 146(10): 4179 - 4188. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 R. Xu, Y. Zhao, and C. Chen Growth hormone-releasing peptide-2 reduces inward rectifying K+ currents via a PKA-cAMP-mediated signalling pathway in ovine somatotropes J. Physiol., December 1, 2002; 545(2): 421 - 433. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
M. E. Cruz-Soto, M. D. Scheiber, K. A. Gregerson, G. P. Boivin, and N. D. Horseman Pituitary Tumorigenesis in Prolactin Gene-Disrupted Mice Endocrinology, November 1, 2002; 143(11): 4429 - 4436. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 M. Tomic', F. Van Goor, M.-L. He, D. Zivadinovic, and S. S. Stojilkovic Ca2+-Mobilizing Endothelin-A Receptors Inhibit Voltage-Gated Ca2+ Influx through Gi/o Signaling Pathway in Pituitary Lactotrophs Mol. Pharmacol., June 1, 2002; 61(6): 1329 - 1339. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| Endocrinology | Endocrine Reviews | J. Clin. End. & Metab. |
| Molecular Endocrinology | Recent Prog. Horm. Res. | All Endocrine Journals |
, The two membrane-spanning domains
that flank a putative pore region. The predicted sizes of the amplified
products are shown. B, RT-PCR products, after 36 amplification cycles,
using the Kir 3.0-specific primers and AP mRNA as a
template. Products were resolved by agarose gel electrophoresis and
ethidium bromide staining.
) or presence of
10-6 M DA (
