Role of h 1 in activation of human mesangial BK channels by cGMP kinase - 《American Journal of Physiology》美国生理学杂志干细胞之家



免疫细胞治疗专区	欢迎关注干细胞微信公众号

12 3 4 5 6 7 8 9 10 ... 27 下一页

返回列表

查看: 469246\|回复: 264	go Role of h 1 in activation of human mesangial BK channels by cGMP kinase [复制链接]

轻羽

新手上路

Rank: 1

积分: 0
威望: 0
包包: 0

楼主

发表于 2009-4-21 13:42 |只看该作者 |倒序浏览 |打印

作者：Patrick E. Kudlacek, Jennifer L. Pluznick, Rong Ma, Babu Padanilam, and Steven C. Sansom作者单位：Department of Physiology and Biophysics, University of Nebraska MedicalCenter, Omaha, Nebraska 68198-4575 ' s9 v. x1 X5 I+ z; Q* a

   2 `( E8 N7 y" j: n0 X
      2 x4 L% I( G: v0 |



         - r4 ?8 ?# a3 e8 ?

                  % O8 Q3 p& J6 J2 v
        a! ]- v8 |: A

      1 t8 O3 J7 s6 x' x9 o; P
      【摘要】; Y9 I) R; Z/ d( S6 g7 n- ]
   In vascular smooth muscle and glomerular mesangial cells, relaxing agentssuch as nitric oxide and atrial natriuretic peptide activate large-conductanceCa 2 -activated K channels (BK) via the cGMPkinase pathway. BK are composed of pore-forming -subunits, encoded bythe slopoke gene ( Slo ), and one of four cell-specificaccessory -subunits (h 1-4 ). We used patch-clamp analysis to determine the influence of h 1,h 2, and h 4 on activation of human mesangialBK by cGMP kinase. We found that HEK 293 cells, coexpressing human (h) Slo with either h 1 or h 2,contained single BK currents activated by db-cGMP in cell-attached patches. However, recombinant BK were not activated by db-cGMP when h Slo was expressed alone or with h 4. DNA-RNAhybridization revealed that mesangial cells contained mRNA forh 1 but not h 2 or h 4. The BKresponse to db-cGMP was decreased when h 1 antisense but notscrambled oligonucleotides were incorporated into mesangial cells. Westernblot analysis showed that h 1 antisense oligonucleotideinhibited the amount of h 1 -V5 fusion protein expressed in HEK293 cells by 50%. These results show that mesangial cells containh 1, a BK accessory protein, which confers activation of BK bycGMP kinase.
. J- h) @! f: w! A0 T
: h* _, x9 B  Z# l1 }0 U+ ?

large-conductance calcium-activated potassium channels; maxi-potassium; human -subunit; antisense; patch clamp; guanosine 3',5'-cyclic monophosphate
      【关键词】 activation mesangial channels
   LARGE, CA 2  - ACTIVATED K CHANNELS (BK) arepresent in a variety of cell types and have been studied in the uterus( 35 ), brain ( 22, 24, 25 ), and smooth muscle cellsof various vascular beds ( 8, 19, 21, 26, 32, 34 ). Our laboratory haspreviously described BK in human mesangial cells( 29 ), smooth muscle-like cellsof the renal glomeruli that participate in regulating the rate of filtration( 3, 14, 15 ). In mesangial cells( 30, 31 ) and vascular smooth muscle( 2, 26, 34 ), BK are activated bynitric oxide and atrial natriuretic peptide via the cGMP kinase secondmessenger system. On activation of BK, the membrane potential hyperpolarizes,thereby reducing the entry of Ca 2 through voltage-gatedCa 2 channels. By preventing an influx ofCa 2 and lowering the concentration of intracellularCa 2  , smooth muscle cells are less responsive to acontractile agonist. Thus activation of BK by cGMP kinase contributes to anincreased glomerular filtration rate, vascular dilation, or the relaxation oftracheal and uterine smooth muscle.& L6 ^" A2 t- a* G" ~" P0 @
5 G* F8 L' |" C( B: y8 N. [
The structure of BK includes four identical pore-forming - and accessory -subunits. The slopoke gene ( Slo ) -subunits are present without -subunits in endothelial cells( 20 ) where they retain qualitatively typical Ca 2 /voltage-dependent gatingproperties. At least four -subunits associated with Slo have been described. It is thought that the variety of -subunits canexplain much of the well-described functional diversity of BK.
5 {' }) h7 k  ]9 P0 `" ~; |+ J
Type 1 of the human -subunit of the BK channel encoded by Slo (h 1 ) is found predominantly in smooth musclewhere it enhances Ca 2 and voltage sensitivity( 5 ). The physiologicalsignificance of the h 1 -subunit at the integrative level wasdemonstrated by Brenner et al.( 6 ), who developed a 1 knockout mouse that possessed a hypertensive phenotype.This hypertensive model supported the notion that BK is a negative feedbackregulator of smooth muscle contraction and is consistent with the role of 1 to enhance the Ca 2 sensitivity ofBK.
" j$ A8 n) R& D2 f% r6 @* a6 v
Although the h 2 -subunit contains an additionalinactivating protein extension, the h 1 - andh 2 -subunits confer similar functional properties toh Slo ( 5 ).Moreover, h 1 and h 2 are the most similar ofthe four -subunits, having 66% protein conservation and similarmembrane-spanning topology ( 5 ).The h 3 - and h 4 -subunits have similarmembrane-spanning topology to h 1 and h 2 butare structurally and functionally different. h 3, With fourknown isoforms, has a widespread tissue distribution, withh 3c and h 3d highly expressed in the pancreasand testes, respectively ( 33 ).h 4 Is primarily contained in nerve terminals where it decreases Ca 2 sensitivity of BK and enhancesneurotransmitter secretion ( 4 ).Interestingly, cGMP kinase activates BK in vascular smooth muscle andmesangial cells. However, cAMP kinase predominantly activates BK in rat brain( 25 ). Thus the specific BK -subunits may confer different regulatory properties onh Slo as demanded by the specialized function of the cell.
9 Z$ K& m7 q0 j' V
In the present study, we tested the hypothesis that the mesangial BKcontains the h 1 -subunit, which permits activation of BK bycGMP kinase. The association of a specific h -subunit with h Slo may partially explain the tissue-specific activation of BK by cGMP kinase.
8 G  P, N- d: s9 c$ |) b" j
METHODS
  W; w* z) V3 u- `4 q1 Y- m
Plasmid construction. The cDNA for h slo (accessionnos. U09384  and U02632  ), a generous gift from Richard Aldrich, was originally present in the pBluescript KS( /-) expression vector. The expressionvector chosen for most of these experiments was pEGFP-C1 (ClontechLaboratories, Palo Alto, CA), which encodes a green fluorescent protein (GFP)that had been optimized for expression in mammalian cells. GFP andh slo contain 239 and 1,142 amino acid residues, respectively.The h Slo /pBluescript plasmid and the pEGFP-C1 expressionvector were digested with Hin dIII and Bam HI restrictionenzymes. The desired Hin dIII/ Bam HI restrictionenzyme-digested nucleotide fragments of h Slo and pEGFP-C1 wereisolated and purified using a Prep-A-Gene DNA purification kit (Bio-RadLaboratories), and the h Slo cDNA was ligated in-frame intopEGFP-C1 in the presence of T4 DNA ligase. The resulting plasmids weretransformed into DH5 Escherichia coli and plated on rich brothagar with 30 mg/ml kanamycin. Isolated colonies were analyzed by restrictionenzyme analysis to confirm their sequence. The plasmids were proliferated in E. coli, the cells were lysed, and the plasmids were purified with ananion-exchange resin and alcohol precipitation (Qiagen, Valencia, CA).. M4 v% V  Q" N0 m0 H

The cDNAs of h -subunits were placed in similar expression vectors forcotransfection into HEK 293 cells. h 2 And h 4 were expressed in the absence of a fluorescent protein, whereash 1 was in the pTracer-CMV2 expression vector (Invitrogen,Carlsbad, CA) with the expressed GFP separated from the h 1 protein. In some experiments, h 1 cDNA was present in thepcDNA3.1/GeneStorm expression vector (Invitrogen) fused with a V5 epitope.

Transfection of plasmids into HEK 293 cells. HEK 293 cells were plated on 35-mm petri dishes in 10% FBS plus DMEM (pH = 7.2) supplemented withpenicillin (100 U/ml), L -glutamine (2.0 mM), sodium bicarbonate(0.375%), HEPES (10 mM), and pyruvic acid (1.0 mM) and incubated overnight at37°C in the presence of 5% CO 2. After a rinse with DMEM, asolution of 1.0 ml DMEM containing 1.0 µg of the plasmid and 1.7 µl of Lipofectamine Plus reagent with 2.5 µl Lipofectamine was applied to eachpetri dish of cells. The cells were incubated for 3 h at 37°C in thepresence of 5% CO 2. DMEM (1.0 ml) containing antibiotics and FBS(20%) were added to each dish, and the cells were incubated for an additional24-72 h.6 q3 G: D' _6 v/ s

Patch-clamp procedure. Single-channel analysis was performed at23°C using standard patch-clamp techniques( 11, 29 ). Experiments wereperformed with the pipette attached to the membrane (cell attached). Thepipette solution contained (in mM) 140 KCl, 1.0 CaCl 2, 2.0MgCl 2, 1.4 EGTA, and 10 mM HEPES (pH = 7.4) and the bath solutioncontained (in mM) 135 KCl, 5.0 KCl 2, 2.0 MgCl 2, 1.0CaCl 2, and 10 HEPES (pH = 7.4).
5 E& V7 _) N( e3 p
The patch pipette, partially filled with solution, was in contact with aAg-AgCl wire on a polycarbonate holder connected to the head stage of apatch-clamp apparatus (501A; Warner Instrument, Hamden, CT). The pipette waslowered on the cell membrane, 5G ) seal. The unitary current, defined as zero forthe closed state, was determined as the mean of the best-fit Gaussiandistribution of the amplitude histograms. Channels were considered in an openstate when the total current ( I ( n - ) I and * l0 J! ?, a; X' J1 X4 y8 Y

Mesangial cell cultures. As previously described( 12, 29 ), cultured human mesangialcells were subpassaged from generations 5-10 in DMEMsupplemented with 10 mM HEPES, 2.0 mM glutamine, 0.66 U/ml insulin, 1.0 mMsodium pyruvate, 0.1 mM nonessential amino acids, 100 U/ml penicillin, 100µg/ml streptomycin, and 20% FBS. On reaching confluency, cells were passedon 22 x 22 1-mm cover glasses (Fisher, Pittsburgh, PA), cultured at 37°C in 5.0% CO 2, and inserted in a perfusion chamber(23°C; Warner RC-2OH) for patch-clamp experiments.  R) c7 B) F- U$ x

DNA-RNA hybridization. Mesangial cell RNA was extracted by the guanidinium thiocyanate method previously described( 9 ). In brief, culturedmesangial cells were exposed to 1.0 ml TriReagent as detailed by themanufacturer (Molecular Research Center, Cincinnati, OH). The total RNA wasisolated, washed, and suspended in sterile H 2 O.7 s2 @' W8 e: c7 X1 C' k- P2 ]
& w. T: I4 Y' Q$ R6 @0 ?4 Z
PCR products were generated from h cDNAs by previously described methods ( 16 ) using thefollowing sequence-specific primers: h 1,CTTTGCCTGGGTGTAACCAT, CCAGGATGGACAGGTACTGG; h 2,GTTTATATGGACCAGTGGCCGG, CTATTGATCCGTTGGATCCTCTCAC; and h 4,GCTCCGGGTGGCTTACGAGTACACGGAAG, GTCCTCTGGTCTCTGATGCTG. PCR fragments ofh 1, h 2, and h 4 were blottedon a Zeta-Probe membrane using a Bio-Dot SF Microfiltration Apparatus (Bio-RadLaboratories, Hercules, CA). All PCR products were approximately the sameconcentration determined by ethidium bromide staining of an agarose gel. AfterDNA blotting, an AlkPhos Direct kit (Amersham Pharmacia Biotech, Piscataway, NJ) was used to prehybridize the blotted membrane, label a human mesangialcell RNA probe, hybridize the labeled probe, and wash the membrane. Along withthe experimental samples, the same solution in the absence of DNA was used asa control. CDP-Star was used to generate and detect a chemiluminescent signal(Amersham Pharmacia Biotech). The manufacturer protocols were followed withthe exception of performing all hybridization washes at 50°C. A signal wasdetected after exposure to film for 2 h.
- [/ k% s4 s8 k
Antisense procedures. In some experiments, cultured mesangial cells were incubated in 2.5 nM phosphorothioate-modified h 1 antisense mRNA primers (5'-CATCACCAGCTTCTTCACCAT) and a control scrambled sequence (5'-ATGTTCATCAAGGCCTACAGG) for 24-72 h beforeexperimentation. Because of the nucleotide sequence similarities betweenh 1 and h 2, we tested the efficiency of theantisense oligonucleotide with the h 1 cDNA fused to the nucleotide sequences encoding for a V5 epitope. This vector was transfectedinto HEK 293 cells and incubated in the presence or absence of theh 1 antisense primer (2.5 nM) for 72 h. The expression of theh 1 -V5 epitope fusion protein (in the pcDNA3.1 GeneStormexpression vector; Invitrogen) was detected by immunoblotting. The transfectedcells were collected and homogenized for 30 s with a Multi-Gen7 generator(ProScientific, Oxford, CT). Protein concentration was assayed using ProteinAssay Dye (Bio-Rad) according to the instructions of the manufacturer. Equalamounts of protein (20 µg) were boiled for 3 min in Laemmli sample buffer,applied to a 12.5% polyacrylamide gel (precast; Bio-Rad), electrophoresed, andthen transferred to a polyvinylidene difluoride nitrocellulose membrane. Themembrane was blocked overnight at 4°C in 5% milk and then exposedovernight at 4°C to a V5-specific antibody conjugated to alkalinephosphatase (1:2,000; Invitrogen). Bands were visualized using an Amplified Alkaline Phosphatase Immuno-Blot Assay Kit (Bio-Rad).
& {. G# m* u5 W; O
Data analysis. The effects of cGMP on recombinanth slo plus h and on mesangial cell BK were analyzed bycomparing the P o value before and after adding db-cGMPusing the paired t -test. Significance between values was establishedby a P value

RESULTS
/ J, ]( W2 u; A" X8 j
Influence of -subunits on activation of BK by cGMPkinase. The following experiments determined if one or more of three -subunits could confer activation of BK by db-cGMP. HEK 293 cells weretransfected with plasmids containing the coding sequence forGFP-h Slo - and h -subunits and then cultured for 24-72 h before patch-clamp analysis of the expressed BK in thecell-attached configuration. Transfection efficiency 90%,as detected by epifluorescence. The representative current tracings, whenh Slo was expressed alone or with h 1,h 2, and h 4 in HEK 293 cells, are shown in Fig. 1 A. The top set of tracings represents BK currents obtained from expressionof only h Slo [pipette potential ( V p ) =-20 mV]. The P o (1 channel) was initially 0.007 andremained low, at 0.001, after the addition of db-cGMP. The second set of tracings is representative of BK currents obtained by expression ofh Slo with h 1. The P o (4channels) was much greater (0.16), likely reflecting the enhancement of BKcurrent when h 1 associates with h Slo. Onaddition of db-cGMP, the P o increased to 0.34. As shown bythe third set of representative recordings, when h 2 wasexpressed with h Slo, the addition of db-cGMP increased the P o to 0.39. The bottom set of recordings isrepresentative of the currents obtained when h Slo wascoexpressed with h 4. In these experiments, the addition ofdb-cGMP did not affect the P o (from 0.01 to 0.005; 1channel) of BK.& g6 V; ^# L2 S9 f9 ^5 L, [
1 [9 W: S) d$ L; a4 |" P
Fig. 1. Representative current tracings ( A ) and summary bar graphs( B ) showing the effects of adding db-cGMP on the open probability( P o ) of large-conductanceCa 2 -activated K (BK) channels incell-attached patches of HEK 293 cells expressing the human (h) slopoke gene (h Slo ) alone and h Slo with h 1, h 2, and h 4. A : P o of BK increased in the presence of db-cGMPwhen h Slo was expressed with h 1 andh 2 ( middle ) but not when expressed alone( top ) or with h 4 ( bottom ). Arrows indicatethe closed state. B : summary of the db-cGMP-evoked mean change in P o ( P o ) whenh Slo ( n = 5 experiments),h Slo /h 1 ( n = 7),h Slo /h 2 ( n = 5), andh Slo /h 4 ( n = 7) were expressed.* P 4 ^6 @6 P" C* f% G

The bar graphs in Fig.1 B summarize the effects of db-cGMP on single BK currents(cell attached) in HEK 293 cells expressing h Slo alone orh Slo with h 1, h 2, andh 4. When h Slo alone was expressed, the P o decreased slightly, but not significantly [change in( ) P o = -0.06 ± 0.03; n = 5],on addition of db-cGMP. When h Slo was coexpressed withh 1 or h 2, the addition of db-cGMP increased P o significantly by 0.23 ± 0.11 ( n = 7)and 0.21 ± 0.03 ( n = 5), respectively. When h Slo was coexpressed with h 4, the P o of BK was not affected significantly (-0.01± 0.01, n = 7) by the addition of db-cGMP. These results showthat either h 1 or h 2, but not h 4, can confer activation of BK-h Slo bydb-cGMP.

Identification of the h -isoform in human mesangialcells. The previous experiments showed that db-cGMP activated BK current in HEK 293 cells only when h Slo was coexpressed withh 1 or h 2. We employed DNA-RNA hybridizationmethods to determine which -subunit is present in human mesangial cells.Specific DNA for h 1, h 2, andh 4 was generated using nucleotide primers ( Fig. 2 A ). As shown in Fig. 2 B, hybridizationof mesangial cell RNA to PCR-generated DNA (designed for different regions ofthe h -subunits) under optimized conditions confirmed the presence of RNAfor h 1 but not h 2 or h 4.These results suggest that h 1 is present and could beassociated with the h Slo component of the mesangial BK.6 B  ^6 i: _, g# @) O- z2 g0 K& L
4 r; B. B; w- o& |
Fig. 2. A : ethidium bromide-stained agarose gel of equal amounts ofPCR-generated products used for DNA-RNA hybridization experiments. Lane1, DNA standard; lane 2, h 1 PCR product; lane 3, h 2 PCR product; lane 4,h 4 PCR product. B : hybridization of human mesangialcell RNA to h 1 (but not h 2 andh 4 ) PCR-generated DNA on a Zeta-Probe membrane.
3 ^; ^* o2 q9 m1 U
Effect of h 1 antisenseoligonucleotides on activation of mesangial BK by cGMP. It was previouslyshown that db-cGMP activated BK in cell-attached patches of cultured humanmesangial cells ( 30, 31 ). Phosphorothioatedmodified antisense oligonucleotides, complimentary to the h 1 initiation coding sequence, were used to determine if h 1 wasnecessary for activation of BK by db-cGMP. As shown in Fig. 3 A, incubation ofmesangial cells with h 1 antisense oligonucleotides nearlyeliminated the activation of BK by db-cGMP (- V p =-20 mV). This effect is shown in the recordings of BK currents (cellattached) in a control cell (no antisense) in Fig. 3 A. The additionof db-cGMP to the bathing solution increased P o from 0.53to 0.83 ( P o = 0.30). As shown in Fig. 3 A, middle, in the presence of the specific anti-h 1 oligonucleotide, db-cGMP increased the P o from 0.14 toonly 0.25 ( P o = 0.11). In the presence of scrambled oligonucleotides, db-cGMP activated BK from 0.21 to 0.42( Fig. 3 A, bottom; P o = 0.21). As revealed in thetracings, there was considerable variability in the baseline P o values for BK. Therefore, the mean basal P o values (before cGMP) were not significantly differentbetween any of the treatment groups ( P 0.3), and no conclusionscould be made regarding differences in baseline activity. However, differenceswere apparent when changes in BK activity were determined (before and afterdb-cGMP addition). As shown in Fig.3 B, db-cGMP activated (control) mesangial BKsignificantly ( P o = 0.31 ± 0.02, n = 5). In the presence of the h 1 antisense oligonucleotides,db-cGMP did not significantly affect BK activity( P o = 0.06 ± 0.04, n = 4). However,db-cGMP activated BK in the presence of the scrambled oligonucleotides( P o = 0.21 ± 0.06, n = 5).

Fig. 3. A : representative current tracings (cell attached) demonstratingthe effects of db-cGMP on BK from mesangial cells cultured in normal growthmedia, media with h 1 antisense, or media containing scrambledoligonucleotide phosphothioate-modified primer. db-cGMP activated mesangial BKin normal media ( top ) and when mesangial cells were treated withscrambled oligonucleotides ( bottom ). h 1 Antisenseoligonucleotides effectively silenced the activation of BK by db-cGMP. Arrowsdenote the closed state. B : summary of db-cGMP-evoked mean P o when mesangial cells are incubated in normalmedia ( n = 5), h 1 antisense ( n = 4), andscrambled ( n = 5) oligonucleotide-treated media. * P 8 U8 P/ u: Q& J2 m
4 ]. }/ q& B8 v1 _5 z8 w: q& s
Western blot analysis was used to determine the effectiveness of theh 1 antisense oligonucleotides on the expression of recombinant h 1. Figure4 shows a representative Western blot demonstrating the effects ofh 1 antisense oligonucleotides (2.5 nM for 72 h) on theexpression of recombinant h 1 in HEK 293 cells. In thisexperiment, h 1 antisense oligonucleotides reduced theexpression of h 1 protein by 48%, as determined bydensitometry analysis. This experiment was repeated five times with a meanreduction in expression of 49.4 ± 7.6% ( n = 5) when the cellswere treated with h 1 antisense oligonucleotides.  [. b  C9 y2 j2 `4 N
7 K5 ~+ U  Y+ F& p9 R) d! \
Fig. 4. Effectiveness of h 1 antisense oligonucleotides asdetermined by immunoblot analysis of recombinant h 1 -V5 fusionprotein, isolated from the cytosolic fraction after 72 h of incubation withHEK 293 cells. Lane 1, low-molecular-weight prestained standard; lane 2, recombinant h 1 -V5 fusion protein at 30 kDa; lane 3, recombinant h 1 -V5 fusion protein incubatedin the presence of 2.5 nM h 1 antisense oligonucleotide.  u/ G2 w  K5 x1 f: @% j
" U5 j& D- v$ d
DISCUSSION

The BK channel is phenotypically varied partly because one of the fouraccessory -subunits influences the properties of the pore-forming -subunit. The goal of the present study was to determine if anaccessory -subunit was necessary for activation of BK by cGMP kinase inhuman mesangial cells. Both h 1 and h 2 effectively conferred activation of recombinant BK by db-cGMP in HEK 293cells. However, mesangial RNA hybridized only to h 1 -specificDNA and demonstrated no binding to h 2 - andh 4 -specific DNA. Moreover, h 1 antisenseoligonucleotides reduced the db-cGMP activation of BK. We conclude thath 1 is the predominant BK -subunit in human mesangialcells and has an essential role in the activation of BK by cGMP kinase.
; q& O0 @! ]3 K1 Q5 _
Role of -subunits in cGMP kinase activation of BK. The revelation that the cGMP kinase pathway activated BK in either thepresence of h 1 or h 2 but noth 4 is consistent with findings by Brenner et al.( 5 ), who demonstrated thath 1 and h 2 confer very similar functionalproperties to h Slo. The major difference between h 1 and h 2 is the presence of an additionalinactivation ball at the NH 2 terminus of the h 2 protein. However, these subunits have similar membrane-spanning topology and,with 65% amino acid homology, are the most closely related in primary protein structure among the h -subunits. On the other hand, h 4 has only 21 and 26% amino acid homology with h 1 andh 2, respectively ( 5 ). Theh 3 -subunit was not tested for the ability to confer activation of BK by cGMP kinase in this study. However, h 3 also contains a very different primary protein structure compared withh 1 and h 2. Although the expression studiesshowed that either h 1 or h 2 can confercGMP-mediated activation of BK, the mechanism is not understood. Theh -subunits may be phosphorylated in this reaction or could alter thephosphorylation/dephosphorylation of the -subunit or another proteinassociated with BK.9 S* }* \- |. Y+ V" ?# o
! X3 D. q0 n5 O% f
Alioua et al. ( 1 ) used 32 P-radiolabeled ATP to show that PKG directly phosphorylated theh Slo subunit. With the use of site-directed mutagenesis,studies have identified specific amino acid sites of PKG phosphorylation( 10, 18 ). One of these studiesdemonstrated PKG activation of BK expressed with h Slo in thepresence of h 1 ( 18 ). However, another studydemonstrated activation of Slo in Xenopus oocytes inthe absence of an expressed h -subunit ( 17 ). This result is seeminglyinconsistent with the present study, which could not demonstrate cGMP kinaseactivation when only h Slo was expressed in HEK 293 cells.However, although HEK 293 cells contain the necessary machinery for cGMPkinase activation of a substrate( 3, 13, 23 ), Xenopus oocytesrequire the coexpression with cGMP kinase. Overexpression of PKG in Xenopus oocytes in the absence of a -subunit may have resulted in (nonphysiological) phosphorylation of the Slo -subunit. This result would be consistent with the notion that the -subunit alters theconformation of the Slo -subunit to facilitate specific proteinkinase phosphorylation.

The absence of the phosphatase limb in the phosphorylation cycle may alsoexplain why both the - and 1 -subunits were required for activating BK in oocytes. Most kinase phosphorylation reactions arebalanced by dephosphorylation reactions that involve protein phosphatases (PP)and protein phosphatase inhibitors (PPI; see Ref. 28 ). BK are not only modulatedby PKG but also by PP, which dephosphorylates and inactivates mesangial BK( 7, 27 ). When cGMP activates BK, aPPI may also be activated, thereby suppressing the dephosphorylation limb ofthe cycle and ensuring maximal substrate phosphorylation. It is possible thatthe -subunit is necessary for the activation of PPI. In this scenario, if Xenopus oocytes do not contain PP and PPI, then merelyphosphorylating h Slo would enhance the P o of BK. However, HEK 293 cells, which contain the machinery for cGMP kinase(including a PPI and PP), may require the -subunit for cGMP kinaseactivation of PPI and inhibition of the PP.

Identification of h -subunit in mesangial BK. DNA-RNA hybridization revealed mRNA for the h 1 -subunit butnot the h 2 - and h 4 -subunits in humanmesangial cells. In a previous study, multiple tissue array expression usingradiolabeled cDNA revealed high expression of h 1 in tissuescontaining smooth muscle cells( 4 ). The same study showed thath 2 message was present mostly in endocrine tissue, andh 4 was predominantly in neural tissue( 4 ). It was interesting thath 1 was not highly expressed in the kidney. However, thisresult may reflect the fact that mesangial cells are a very minute componentof the kidney mass. The presence of mRNA for h 1 in mesangialcells reflects the phenotypic similarities between these cells and smoothmuscle.- V* N7 `& z6 H% h; X$ B  V$ n5 a
3 b% j. j; r: k
In this investigation, we have determined that the mRNA for h 1 is present in human mesangial cells. The results of this study suggest that h 1 associates with h Slo and confers cGMP activation of mesangial BK. In human mesangial cells, therole of the 1 -subunit may be to stabilize an intramolecularsite within the -subunit to facilitate the cGMP kinase-dependent phosphorylation of BK.
' P, g1 q: q$ m8 l% N3 G; W
DISCLOSURES

This work was supported by National Institute of Diabetes and Digestive andKidney Diseases Grants NIHRO1DK-49561 (to S. C. Sansom) and 1T32HL-0788 (to P.E. Kudlacek and J. L. Pluznick).

ACKNOWLEDGMENTS

We are grateful to Drs. Richard Aldrich and Robert Brenner (Stanford University) for providing cDNAs for h slo,h 1 -, h 2 -, andh 4 -subunits.+ q, E8 p/ |) J4 K1 L
      【参考文献】- k  ]; N9 M' B$ u( Q
Alioua A,Tanaka Y, Wallner M, Hofmann F, Ruth P, Meera P, and Toro L. The largeconductance, voltage-dependent, and calcium-sensitive K  channel, Hslo, is atarget of cGMP-dependent protein kinase phosphorylation in vivo. JBiol Chem 273:32950-32956, 1998.! Q7 J; i6 R" G8 }
9 y1 r# v/ B8 j; c, `

3 U7 ^/ Z+ W4 Y, |, U% C  H% \( l) B
Archer SL,Huang JMC, Hampl V, Nelson DP, and Shultz PJ. Nitric oxide and cGMP causevasorelaxation by activation of a charybdotoxin-sensitive K channel bycGMP-dependent protein kinase. Proc Natl Acad Sci USA 91: 7583-7587,1994.

; v/ I  {/ |8 @! F; ?' X9 |+ c- z

Arriba G,Barrio V, Olivera A, Rodriguez-Puyol D, and Lopez-Novoa JM. Atrialnatriuretic peptide inhibits angiotensin II-induced contraction of isolatedglomeruli and cultured glomerular mesangial cells of rat: the role of calcium. J Lab Clin Med 111:466-474, 1988.  b$ q+ F8 N) M) Z# A8 a
9 h+ A& j9 Y* Y( U6 j$ ^3 R# T

Behrens R,Nolting A, Reimann F, Schwarz M, Waldschütz R, and Pongs O. HKCNMB3and hKCNMB4, cloning and characterization of two members of thelarge-conductance calcium-activated potassium channel subunit family. FEBS Lett 474:99-106, 2000.* f0 b+ B0 B- I. \- G# d' ?) `

Brenner R,Jegla TJ, Wickenden A, Liu Y, and Aldrich RW. Cloning and functionalcharacterization of novel large conductance calcium-activated potassiumchannel beta subunits, hKCNMB3 and hKCNMB4. J BiolChem 275:6453-6461, 2000." X) a6 D5 F5 ^( n1 ?. s

6 t# z. ]5 f' y8 B$ {0 |
Brenner R,Perez GJ, Bonev AD, Eckman DM, Kosek JC, Wiler SW, Patterson AJ, Nelson MT,and Aldrich RW. Vasoregulation by the beta1 subunit of thecalcium-activated potassium channel. Nature 407: 870-876,2000.
+ a/ ^, x" m3 a( H& P" r  O3 ^# A0 f

7 \5 z7 X5 j! p
Carl A, KenyonJL, Uemura D, Fusetani N, and Sanders KM. Regulation ofCa 2 -activated K channels by protein kinaseA and phosphatase inhibitors. Am J Physiol CellPhysiol 261:C387-C392, 1991.1 n) V" I& E; P0 g4 d0 i
! w2 V( G5 \# ^' ^3 h, i' G; \
' Z8 F( d6 J2 H& U! v) ~

Carrier GO,Fuchs LC, Winecoff AP, Giulumian AD, and White RE. Nitrovasodilators relaxmesenteric microvessels by cGMP-induced stimulation of Ca-activated Kchannels. Am J Physiol Heart Circ Physiol 273: H76-H84,1997.
7 B$ v8 o9 _3 H0 V7 k# N; Q# Y6 h

Chomczynski P and Sacchi N. Single-step method of RNA isolation by acid guanidiniumthiocyanate-phenol-chloroform extraction. Anal Biochem 162: 156-159,1987.

: @0 V$ o' _1 d4 z5 v) Y3 E
3 j6 z: I* Z( p5 I
Fukao M, MasonHS, Britton FC, Kenyon JL, Horowitz B, and Keef KD. Cyclic GMP-dependentprotein kinase activates cloned BKCa channels expressed in mammalian cells bydirect phosphorylation at serine 1072. J Biol Chem 274: 10927-10935,1999.
8 ^" b3 ^& e0 A- Z0 ^- `* e

Hamill OP,Marty A, Neher E, Sackmann B, and Sigworth FJ. Improved patch-clamptechniques for high resolution current recording from cells and cell-freemembrane patches. Pflügers Arch 391: 85-100,1981.) Q% P( I4 _: r3 j  L' r% a/ j) v: s
* }* r6 y) m- `7 o& z9 X
( t5 C2 i, ?6 K, O5 b: r3 i1 [* z

Jaffer F,Saunders C, Shultz P, Throckmorton D, Weinshell E, and Abboud HE. Regulation of mesangial cell growth by polypeptide mitogens. Am JPathol 135:261-269, 1989.7 g3 _. q4 S6 S2 y

2 H3 n$ w+ W! ?# Q) d: W
Jockers R,Petit L, Lacroix I, deCoppet P, Barrett P, Morgan PJ, Guardiol B, DelagrangeP, Marullo S, and Strosberg AD. Novel isoforms of Mellc melatoninreceptors modulating intracellular cyclic guanosine3',5'-monophosphate levels. Mol Endocrinol 11: 1070-1081,1997.: i5 S5 _8 _7 p

Kriz W, ElgerM, Lemley K, and Sakai T. Structure of the glomerular mesangium: abiomechanical interpretation. Kidney Int 38: S2-S9,1990.; f% N$ h& ~4 H* P! n
8 D1 ]! H. n9 A( }$ P9 d, Y5 }; a; p

Kriz W, ElgerM, Mundel P, and Lemley KV. Structure-stabilizing forces in the glomerulartuft. J Am Soc Nephrol 5:1731-1739, 1995.

- _5 f, R! Y4 P, g( J

Mullis KB andFaloona FA. Specific synthesis of DNA in vitro via a polymerase-catalyzedchain reaction. Methods Enzymol 155: 335-350,1987.
$ s( n& _. K  A# A5 g& r# C3 b4 I
, {- x- G2 U% G# v1 i& y

Nara M,Dhulipala PD, Wang YX, and Kotlikoff MI. Reconstitution of beta-adrenergicmodulation of large conductance, calcium-activated potassium (maxi-K) channelsin Xenopus oocytes: identification of the cAMP-dependent proteinkinase phosphorylation site. J Biol Chem 273: 14920-14924,1998.
  p; T$ S7 ^8 I2 n. e

Nara M,Dhulipala PDK, Ji GJ, Kamasani UR, Wang YX, Matalon S, and Kotlikoff MI. Guanylyl cyclase stimulatory coupling to K Ca channels. Am J Physiol Cell Physiol 279:C1938-C1945, 2000.' k+ @9 t4 |. p
% ?- K8 ^) i+ Q$ c. Y
) f- ^$ ^3 r% i
$ Q* ~1 T% a1 }" U
Onoue H andKatusic ZS. Role of potassium channels in relaxations of canine middlecerebral arteries induced by nitric oxide donors. Stroke 28:1264-1270, 1997.) J. s$ a: i0 v) w8 \+ H3 H2 q0 W

! c' Z6 I  h. w& D$ K

Papassotiriou J, Köhler R, Prenen J, Krause H, Akbar M,Eggermont J, Paul M, Distler A, Nilius B, and Hoyer J. EndothelialK channel lacks the Ca 2 sensitivity-regulating subunit. FASEB J 14: 885-894,2000.

/ a4 l4 C8 m8 X+ b

Peng W, HoidalJR, and Farrukh IS. Regulation of Ca 2 -activatedK channels in pulmonary vascular smooth muscle cells: role ofnitric oxide. J Appl Physiol 81: 1264-1272,1996./ Q" p0 R  r0 Q( ?" D

. o0 w- U* {$ U7 [

Pozzo-Miller LD, Connor JA, and Andrews SB. Microheterogeneityof calcium signalling in dendrites. J Physiol 525: 53-61,2000.. W# X! L% ?: x' e
# m* O4 R# N. l9 [2 p; \, ?

5 L) M# d! G( ?% \! s2 }
Ramamoorthy S,Giovanetti E, Qian Y, and Blakely RD. Phosphorylation and regulation ofantidepressant-sensitive serotonin transporters. J BiolChem 273:2458-2466, 1998.
5 a( T$ ?: B! _+ S/ @7 N% u& Y& @5 k
- ~0 Z/ n+ G5 h9 d) w8 F
' w% z0 t  q7 B( q( d. t
Reinhart PH,Chung S, and Levitan IB. A family of calcium-dependent potassium channelsfrom rat brain. Neuron 2:1031-1041, 1989.8 R( b: T& ]8 B+ z4 J" B4 \* ]  l6 j

Reinhart PH,Chung S, Martin BL, Brautigan DL, and Levitan IB. Modulation ofcalcium-activated potassium channels from rat brain by protein kinase A andphosphatase 2A. J Neurosci 11:1627-1635, 1991.7 _4 U+ R; E% A8 a# ^

Robertson BE,Schubert R, Hescheler J, and Nelson MT. cGMP-dependent protein kinaseactivates Ca-activated K channels in cerebral artery smooth muscle cells. Am J Physiol Cell Physiol 265:C299-C303, 1993.
3 s7 ^1 e9 l% H$ J) C6 {0 f

Sansom SC,Stockand JD, Hall D, and Williams B. Regulation of large calcium-activatedpotassium channels by protein phosphatase 2A. J BiolChem 272:9902-9906, 1997.
& W" h+ }! o5 A# s5 c
- S9 i9 d$ P1 K, n& y
( x) A' ^$ F2 S) n' p, w
Shenolikar S and Ingebritsen TS. Protein (serine and threonine phosphate phosphatases).In: Methods in Enzymology, edited by Colowick SP andKaplan NO. New York: Academic, 1989, p.102-129.# b# E0 W' D) ^4 J% a# I, V
- r% k8 p$ ]" }; }& \" e
/ }8 l! ?% `3 @& m$ W

Stockand JD andSansom SC. Large Ca 2 -activated K channels responsive to angiotensin II in cultured human mesangial cells. Am J Physiol Cell Physiol 267:C1080-C1086, 1994.8 B1 I  H+ \2 j4 A0 C
8 C5 Q/ @9 Y/ A) `& h
2 t9 n- S& g+ i2 e

Stockand JD andSansom SC. Mechanism of activation by cGMP-dependent protein kinase oflarge Ca 2 -activated K channels in mesangialcells. Am J Physiol Cell Physiol 271: C1669-C1677,1996.- I% C/ A7 n% D. \
7 w1 E* S6 d5 K0 R. Z( U

3 T1 h& U6 d1 @( [# i
Stockand JD andSansom SC. Role of large, Ca-activated K channels in regulation bynitroprusside and ANP of mesangial contraction. Am J Physiol CellPhysiol 270:C1773-C1779, 1996.+ R" T) ~3 c: a$ Y6 \
3 N7 {1 J. D, S) }# t$ y

Tanaka Y,Igarashi T, Kaneko H, Yamaki F, Mochizuki Y, Aida M, Taniguchi H, Tanaka H,and Shigenobu K. NO-mediated MaxiK(Ca) channel activation producesrelaxation of guinea pig aorta independently of voltage-dependent L-typeCa 2 channels. Gen Pharmacol 34: 159-165,2000.

, W6 |- Y% x( h/ Y( T
Uebele VN,Lagrutta A, Wade T, Figueroa DJ, Liu Y, McKenna E, Austin CP, Bennett PB, andSwanson R. Cloning and functional expression of two families of -subunits of the large conductance calcium-activated K channel. J Biol Chem 275:23211-23218, 2000.
$ ?+ R7 J8 `7 N$ d/ M: f. ~) z
- v' o! \. a+ m1 K. o4 K
: |# l& i4 @; U
Williams DL Jr,Katz GM, Roy-Contancin L, and Reuben JP. Guanosine 5'-monophosphatemodulates gating of high-conductance Ca-activated K channels in vascularsmooth muscle cells. Proc Natl Acad Sci USA 85: 9360-9364,1988.
, @) W$ }% K" y1 y: I, {9 {
5 s- W8 N7 |0 q: ~% N

Zhou X,Schlossmann J, Hofmann F, Ruth P, and Korth M. Regulation of stablyexpressed and native BK channels from human myometrium by cGMP- andcAMP-dependent protein kinase. Pflügers Arch 436: 725-734,2000.

收藏0 分享0 顶0 踩0