Tetanus toxin abolishes exocytosis of ROMK1 induced byinhibition of protein tyr - 《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 ... 23 下一页

返回列表

查看: 481010\|回复: 227	go Tetanus toxin abolishes exocytosis of ROMK1 induced byinhibition of protein tyr [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 13:50 |只看该作者 |倒序浏览 |打印

作者：HyacinthSterling, Dao-HongLin, YuanWei,  Wen-HuiWang作者单位：Department of Pharmacology, New York Medical College,Valhalla, New York 10595
   ! v" Z3 |3 u  s; ]9 S1 P& ]
   4 Y9 P2 F# |4 O; _8 `, j
      - W8 F0 K# a4 @0 v4 D8 ^. \/ O

         $ Q$ r& o0 \3 f' j* p
         3 M8 R+ w0 h2 m' c  R+ W3 a

         ( K' }& l4 \& D$ I) u1 J7 s

      ( {, D3 B( t% w  v# d


      【摘要】
   We used confocal microscopy,patch-clamp, and biotin-labeling techniques to examine the role ofsoluble N -ethylmaleimide-sensitive factor attachment proteinreceptor (SNARE) proteins in mediating the effect of inhibition of PTKon ROMK1 trafficking in HEK-293 cells transfected with c-Src and greenfluorescent protein (GFP)-ROMK1. Inhibition of c-Src with herbimycin Asignificantly decreased the tyrosine phosphorylation level of ROMK1.Patch-clamp studies demonstrated that addition of herbimycin Aincreased the activity of ROMK1 in cell-attached patches. Confocalmicroscopic imaging showed that herbimycin A decreased theintracellular intensity of GFP-ROMK1. The biotin-labeling techniquedemonstrated that the inhibition of c-Src increased surface ROMK1 by110%. In contrast, inhibition of c-Src did not increase the K channelnumber in HEK cells transfected with R1Y337A, a ROMK1 mutant in whichtyrosine residue 337 was mutated to alanine. This suggests thattyrosine residue 337 is essential for the herbimycin A-induced increase in surface ROMK1 channels. To determine whether SNARE proteins areinvolved in mediating exocytosis of ROMK1 induced by the inhibition ofc-Src, we examined the effect of herbimycin A on ROMK1 trafficking incells treated with tetanus toxin. The incubation of cells in a mediumcontaining tetanus toxin abolished the herbimycin A-induced increase inthe number of surface ROMK1. In contrast, inhibition of c-Src stillincreased the numbers of surface ROMK1 in cells treated with boiledtetanus toxin. We conclude that tyrosine dephosphorylation enhances theexocytosis of ROMK1 and that SNARE proteins are required for exocytosisinduced by inhibition of PTK. 0 p2 Q  S& _$ ]# @8 d' J$ L
      【关键词】 cSrc soluble N ethylmaleimidesensitive factorattachment protein receptor proteins protein tyrosinephosphatase
   INTRODUCTION

ROMK 1 IS AN INWARDLY RECTIFYING K channel located in the apical membrane ofthe cortical collecting duct (CCD) ( 9, 33 ) and isresponsible for renal K secretion ( 5, 28 ). K secretion isa two-step process: K enters the cell via the basolateral Na-K-ATPase and is secreted into the lumen across the apical membrane through the Kchannels. Although two types of K channels have been identified in theapical membrane of the CCD ( 2, 3 ), it is generally believed that the small-conductance K (SK) channel is a maincontributor to apical K conductance ( 14, 20, 21, 27 ). Alarge body of evidence indicates that ROMK1 is a key component of theSK channel identified in native tissue ( 4, 14, 19, 27 ).# I' |5 r+ }3 u

Several studies have demonstrated that the number of functionalROMK-like SK channels in the apical membrane of the CCD varies anddepends on dietary K intake: an increase in K intake augments thenumber of SK channels in the CCD ( 13, 29, 31 ). Incontrast, low-K intake reduces the number of ROMK channels in the cellmembrane in the renal cortex and outer medulla ( 8 ). Ourprevious studies demonstrated that PTK and protein tyrosine phosphatase(PTP) play an important role in mediating the effect of low-K intake onthe number of the ROMK-like SK channels in the CCD ( 28, 30, 31 ). We have reported that inhibition of PTK increases theactivity of the SK channels in the apical membrane of the rat CCD( 28 ). The effect of inhibition of PTK is possibly mediatedby increasing SK channel insertion because this depends on the intactcytoskeleton ( 32 ). This speculation has further beensuggested by confocal microscopy, demonstrating that inhibition ofc-Src increased the membrane density of green fluorescent protein(GFP)-tagged ROMK1 in oocytes injected with ROMK1 and c-Src( 10 ). However, confocal microscopic imaging of oocytescannot clearly separate the ROMK1 channels located in the cell membranefrom those in the submembrane. Moreover, the mechanism of membraneinsertion of ROMK1 may be different between oocytes and mammaliancells. Finally, it is not known whether soluble N -ethylmaleimide-sensitive factor attachment proteinreceptor (SNARE) proteins are required for exocytosis of ROMK1 inducedby inhibition of c-Src and other types of PTKs. In the presentexperiments, these questions were investigated by examining the effectof inhibition of PTK on the insertion of ROMK1 in HEK-293 cellstransfected with ROMK1 and c-Src.
$ n+ M1 N3 N/ c% [: c& Q3 Z
METHODS

Construction of GFP-ROMK and c-Src. ROMK1 or mutant ROMK1 (R1Y337A) was cloned into the Eco R1and Bam H1 sites of PEGFPC1 (Clontech, Palo Alto, CA) usingprimers 5'-TGGGCCTAAAAGAATTCAGCTGCTGTGCAGACAAC (sense, nt81-115) and 5'-TTGTAGGTGGAAGGATCCCTGCTACATCTGGGTGTCG(antisense, nt 1310-1346) to amplify PCR fragments of ROMK1 or themutant that was subcloned into PCDNA3.1. The coding sequence of c-Srcwas cut from pGEM vector with Hin DIII and Eco RIand ligated into pCDNA3.1 expression vector (Invitrogen). All vectorsequences were confirmed by automated DNA sequencing at the WilliamKeck Biotechnology Laboratory at Yale University.
2 j  g( ?% O6 L. j+ q- R+ C
Transfection of HEK-293 cells. HEK-293 cells were plated in 35-mm dishes and transfected with 1 µgof ROMK1 or R1Y337A and 1 µg of c-Src using 7 µl of LT1 reagent(PanVera, Madison, WI) according to instructions provided by themanufacturer. To study the effect of tetanus toxin on ROMK1 trafficking, 30 nM tetanus toxin or boiled toxin was added at the sametime as the transfection. Tetanus toxin was obtained from Calbiochem(San Diego, CA). The experiments were carried out 2 days aftertransfection. The successful rate for cell transfection was between 60 and 70%.

Confocal microscopy. The method using the confocal microscope has been previously described( 24 ). Briefly, a Bio-Rad MRC 1000 confocal microscope wasused in the study. GFP fluorescence was excited at 488 nm with an argonlaser beam and viewed with an Olympus microscope equipped with a ×60oil lens. All images were acquired, processed, and printed withidentical parameters before and after herbimycin A treatment.
6 r  q  v2 W6 n
Biotinylation, immunoprecipitation, and Western blot analysis. Changes in surface ROMK after herbimycin A treatment were quantitatedby labeling cells with cell-impermeant sulfo-NHS-biotin (Pierce)according to the protocol provided by the manufacturer. Afterbiotinylation, the cells were washed 2× with PBS and trypsinized withtrypsin-EDTA. They were pelleted by centrifugation for 5 min at 10,000 rpm, washed 2× with PBS, and lysed with cold RIPA buffer (1× PBS, 1%Igepal CA-630, 0.1% SDS, 0.5% deoxycholate) supplemented with 1 mMsodium molybdate, 1 mM sodium fluoride, 1 µM PMSF, and 100 µl ofprotease inhibitor cocktail/ml (Sigma) of lysis buffer. Afterclarification, total protein concentrations were determined with aBio-Rad (Bio-Rad Laboratories, Richmond, CA) protein assay kit, andaliquots of lysates containing equal amounts of protein wereimmunoprecipitated overnight with 1 µg of anti-GFP (Clontech)monoclonal antibody and 20 µl of protein A/G-agarose (Santa CruzBiotechnology, Santa Cruz, CA). After collection of the immune complexby centrifugation and a washing 2× with PBS, proteins were resolved byelectrophoresis on 10% SDS gels and transferred to PVDF membranes(Bio-Rad). The membranes were blocked with 5% milk in Tris-bufferedsaline, and the biotin-labeled GFP-ROMK1 proteins were detected usingNeutrAvidin horseradish peroxidase (Pierce). To determine the effect ofherbimycin A on the tyrosine phosphorylation level of ROMK1, PY20antibody (Santa Cruz Biotechnology), which reacts with thetyrosine-phosphorylated proteins, was used to detect the phosphorylatedROMK1. Changes in biotin-labeled surface ROMK1 proteins or thetyrosine-phosphorylated ROMK1 levels were normalized with correspondingtotal ROMK1 protein, which was determined with ROMK antibody (AlomoneLaboratories, Jerusalem, Israel). The density of the band wasdetermined using Alpha DigiDoc 1000 (Alpha Innotech, San Leandro, CA)., g' J. v  T4 K& D+ q# ]: ^! `6 u% O! O
9 n* O: `; V' z+ e
Patch-clamp technique. An Axon 200A patch-clamp amplifier was used to record channel current.The current was low-pass filtered at 1 kHz by an eight-pole Besselfilter (902LPF, Frequency Devices, Haverhill, MA) and digitized by anAxon interface (Digitada 1200). Data were acquired by an IBM-compatiblePentium computer (Gateway 2000) at a rate of 4 kHz and analyzed usingthe pClamp software system 6.04 (Axon Instruments, Burlingame, CA).Channel activity was defined as NP o, which was calculated from data samples of 30-s duration in the steady state asfollows o =

∑ (1t 1 +2t 2 +… it i2 p4 B: [* V& |4 d# j6 ?$ Z6 V: K: A. }

( 1 )# P- J5 C0 B' `2 G! \4 v
. p! D8 E" e( Y, I( Z1 n
where t i is the fractional open timespent at each of the observed current levels. The experiments wereperformed at cell-attached patches, and the bath temperature wasmaintained at 37°C.

Experimental solution and statistics. To examine the role of SNARE proteins in the regulation of ROMK1exocytosis, 30 nM tetanus toxin or boiled toxin was added to theincubation media at the same time the cells were transfected with ROMK1and c-Src.

The bath solution for the patch-clamp study was composed of (in mM) 140 NaCl, 5 KCl, 1.8 MgCl 2, 1.8 CaCl 2, and 10 HEPES(pH 7.4). The pipette solution was composed of (in mM) 140 KCl, 1.8 MgCl 2, and 10 HEPES (pH 7.4). Herbimycin A was purchasedfrom Sigma (St. Louis, MO) and added directly to the bath to reach thefinal concentration. We present data as means ± SE. Student's t -test was used to determine significance.
/ a1 `1 i) ^4 W% F( N  p% p
RESULTS0 _; R- V5 s7 r) E+ n0 o/ \1 B) }

We used HEK-293 cells transfected with the GFP-tagged ROMK1 andc-Src to study the effect of herbimycin A, an inhibitor of PTK, onROMK1 trafficking. A previous study demonstrated that the GFP-taggedROMK1 had the same biophysical properties of ROMK1 ( 10, 24 ). Figure 1, A - C, is a representative confocal image from14 experiments showing the effect of herbimycin A on ROMK1 distributionin living HEK-293 cells transfected with GFP-ROMK1 and c-Src. Undercontrol conditions, a large number of GFP-ROMK1 channels accumulated inthe perinuclear region. Addition of 1 µM herbimycin A decreased theintracellular location of ROMK1 channels and increased the number ofROMK1 channels in the cell membrane. This is evidenced by theobservation that the cell diameter appears to be larger in the presenceof the PTK inhibitor than before addition of herbimycin A. The effectof herbimycin A on ROMK1 channel trafficking was specific because itdid not have a significant effect on ROMK1 channel distribution inHEK-293 cells transfected with GFP-ROMK1 alone (Fig. 1, D - F ). Also, there is no significant changein ROMK1 location within 15 min in the absence of herbimycin A in cellstransfected with ROMK1 and c-Src (data not shown).
' C# e& S% V& p$ K! G& \
Fig. 1. Confocal microscopy showing the effect of 1 µM herbimycin Atreatment on the redistribution of ROMK1 channels in living HEK cellstransfected with green fluorescent protein (GFP)-ROMK and c-Src( A - C ). A : cell image taken beforethe addition of herbimycin A. B and C : channellocation at 5 and 10 min after addition of the PTK inhibitor,respectively. Cell images were taken at the same focal plane. Bar = 10 µM. D - F : effect of herbimycin A onROMK1 location in HEK-293 cells transfected with GFP-ROMK1alone.
+ f5 @) a  A# F7 |9 U2 m: d
After finding that the inhibition of c-Src lowered the distribution ofROMK1 in the intracellular compartment, we used the biotin-labelingtechnique to examine whether herbimycin A increases the number of ROMK1channels in the cell membrane. Figure 2 is a Western blot analysis showing the effect of herbimycin A treatment on the surface localization of ROMK1 in HEK-293 cells transfected withGFP-ROMK1 and c-Src. The ROMK1 channels were harvested by immunoprecipitation of the cell lysate with GFP antibody after a 15-minincubation of cells with 1 µM herbimycin A. The surface-localized ROMK1 (71 kDa) labeled with biotin was detected by neutravidin ( A ), and total ROMK proteins were recognized by ROMKantibody ( B ). The level of surface-located ROMK1 channelswas normalized compared with total ROMK protein. It was calculated thatinhibition of c-Src significantly augmented the biotin-labeled fractionof ROMK1 by 110 ± 20% ( n = 7) compared withthose from untreated cells. In addition to the 71-kDa band, biotin alsolabeled a low-molecular-mass band (60-65 kDa). However, the natureof the low-molecular-mass band is not clear. This band cannot be aglycosylated ROMK1 because the molecular mass of glycosylated ROMK1should be greater than 71 kDa. This low-molecular-mass proteinmay be a ROMK1-associated membrane protein or a degraded ROMK1 proteinthat cannot be recognized by ROMK antibody. Further study is clearlyrequired to determine the nature of the protein.

Fig. 2. Western blot analysis showing the effect of herbimycin A(Herb) treatment on the surface localization of ROMK1. HEK-293 cellstransfected with GFP-ROMK1 and c-Src were treated with herbimycin A orvehicle for 15 min. The ROMK1 channels were harvested byimmunoprecipitation (IP) of the cell lysate with GFP antibody. Thesurface-localized ROMK1 was detected by biotin labeling, followed byWestern blot analysis with neutravidin ( A ). Total ROMK isrecognized by ROMK antibody ( B ). The level of ROMK in themembrane is normalized compared with the relative amount of total ROMKprotein. Arrow, biotin-labeled ROMK1. A second band can also beobserved in the gel stained with avidin. It may be a degraded productof ROMK1 protein generated during the preparation of the tissue. IB,immunoblotted.

The notion that the inhibition of c-Src increased the ROMK1 channelnumber in the cell membrane is also supported by experiments in whichthe patch-clamp technique was used to test the effect of herbimycin Aon ROMK1 channel activity. Figure 3 is achannel recording demonstrating that the inhibition of c-Src increased the activity of ROMK1 channels. From an inspection of Fig. 3, it isapparent that there was no K channel activity before addition ofherbimycin A. The inhibition of c-Src caused a sharp increase in thenumber of ROMK1 channels. This increase was most likely the result offusion of a vesicle containing ROMK1 into the cell membrane because weoften observed that more than two channels were simultaneously open. In10 experiments, blocking c-Src increased NP o from 0.5 ± 0.1 to 3.0 ± 0.3. Because herbimycin A did not increase the channel open probability (data not shown), the effect ofherbimycin A must result from an increase in channel number. Thisfinding is consistent with a previous report that herbimycin Aincreased the activity of ROMK1-like channels in the CCD obtained fromrats on a K-deficient diet from 0.7 ± 0.1 to 3.1 ± 0.3 ( 29 ). The effect of herbimycin A results from theinhibition of c-Src because herbimycin A had no effect on channelactivity in cells transfected with ROMK1 alone (data not shown). Also,herbimycin A significantly decreased the tyrosine-phosphorylated ROMK1channel population. Figure 4 is a Westernblot analysis illustrating the effect of herbimycin A on the tyrosinephosphorylation of ROMK1. The cells were treated with herbimycin A orvehicle for 15 min. The ROMK1 channels were harvested byimmunoprecipitation of the cell lysate with GFP antibody;phosphorylated protein was detected with py20 antibody (Fig. 4 A ), and total ROMK1 was recognized by ROMK antibody (Fig. 4 B ). Protein phosphorylation was normalized compared withtotal ROMK1 proteins. Inhibition of c-Src reduced thetyrosine-phosphorylated ROMK1 by 48 ± 5% ( n = 5).
1 }" w! J8 H# j' v7 S5 D
Fig. 3. A channel recording shows the effect of 1 µM herbimycinA on the activity of ROMK1 in HEK-293 cells transfected with GFP-ROMK1and c-Src. The pipette holding potential was 0 mV, and the experimentwas performed in a cell-attached patch. C, channel-closed level. The top trace shows the time course of the experiment, and 3 parts of the data ( 1-3 ) are extended to show the fasttime course. The interruption in the top of the trace is dueto the saturation of the amplifier induced by a sharp increase inchannel activity.* [% ^2 L, J/ I( j

Fig. 4. Western blot analysis illustrating the effect ofherbimycin A on the tyrosine phosphorylation of ROMK1. The cellstransfected with ROMK1 and c-Src were treated with herbimycin A orvehicle for 15 min and the ROMK1 channels were harvested by IP of thecell lysate with GFP antibody. Phosphorylated protein was detected withpy20 antibody ( A ), and total ROMK1 is recognized by ROMKantibody ( B ). Protein phosphorylation was normalizedcompared with the relative amount of ROMK1 protein. Arrow, position ofGFP-ROMK1.

We have previously demonstrated that tyrosine residue 337 is a majorsite for PTK-induced phosphorylation ( 7 ). To examine therole of tyrosine residue 337 in mediating the effect of inhibiting PTKon the ROMK1, we used confocal microscopy to examine the effect ofherbimycin A on ROMK1 mutant R1Y337A. Figure 5 is a confocal image from nineexperiments showing the effect of herbimycin A on the distribution ofR1Y337A in cells cotransfected with c-Src. Clearly, inhibition of c-Srcdid not have a significant effect on R1Y337A distribution becauseintracellular R1Y337A location was not altered by 15-min treatment withherbimycin A. This suggests that the effect of herbimycin A on ROMK1results from the stimulation of tyrosine dephosphorylation of ROMK1.Because R1Y337A cannot be phosphorylated by PTK, the population of thetyrosine-phosphorylated ROMK1 channels was absent. Accordingly,inhibition of PTK did not have an effect on ROMK1 membrane insertion.This hypothesis is also supported by experiments in which the effect ofherbimycin A on ROMK1 surface number was investigated with thebiotin-labeling technique.

Fig. 5. Aconfocal image showing the effect of herbimycin A on the distributionof R1Y337A in HEK-293 cells transfected with R1Y337A and c-Src. Thecell image ( A ) taken before the addition of herbimycin Aserves as a control. B - D : channel locationat 5, 10, and 15 min after addition of the inhibitor, respectively.Bar = 10 µM.2 {! q* B6 x7 b" Z

Figure 6 is a Western blot analysisshowing the effect of herbimycin A on the surface localization ofR1Y337A in HEK-293 cells transfected with GFP-R1Y337A and c-Src. TheROMK1 channels were harvested by immunoprecipitation of the cell lysatewith GFP antibody and detected with ROMK antibody (Fig. 6 B ),and the surface-localized ROMK1 was identified by biotin labeling (Fig. 6 A ). The results clearly demonstrated that inhibition ofc-Src did not increase the intensity of the avidin-recognized 71-kDaband (94 ± 11% of the control value, n = 5).: }2 P! o  E) L7 I! t/ S+ q
1 X- H5 c* g6 @2 x7 F
Fig. 6. Western blot analysis showing the effect of herbimycin Atreatment on the surface localization of R1Y337A detected bybiotin-labeling. HEK-293 cells transfected with GFP-R1Y337A and c-Srcwere treated with herbimycin A or vehicle for 15 min, and the mutantROMK1 channels were harvested by IP of the cell lysate with GFPantibody. The surface-localized ROMK1 was detected by biotin labeling( A ), and total R1Y337A protein was identified with ROMKantibody ( B ). The level of R1Y337A in the membrane wasnormalized compared with the relative amount of total R1Y337Aprotein.* \  `, ]7 e% R

After establishing that the inhibition of c-Src stimulates theinsertion of ROMK1, we investigated whether exocytosis induced byinhibiting c-Src depended on SNARE proteins, using confocal microscopyand biotin-labeling techniques. HEK cells were transfected withGFP-ROMK1 and c-Src and incubated in media containing either tetanustoxin (30 nM) or boiled tetanus toxin. Figure 7, A - C, is aconfocal image from nine experiments showing the effect of herbimycin Aon the distribution of ROMK1 channels in cells incubated with tetanustoxin (30 nM). It is apparent that treatment of cells with tetanustoxin abolished the effect of herbimycin A on ROMK1 trafficking becausefluorescence intensity in the intracellular compartment, an indicationof intracellularly located ROMK1, was not significantly decreased byherbimycin A. Also, the fluorescence intensity in the cell membrane didnot increase. In contrast, we observed that addition of herbimycin Asignificantly decreased the ROMK1 number in the intracellularcompartment in the cells treated with boiled toxin (Fig. 7, D - F ). The notion that the inhibition ofSNARE proteins abolishes the effect of inhibition of c-Src is alsosupported by results obtained from biotin-labeling experiments. Figure 8 is a typical Western blot demonstratingthe effect of herbimycin A on biotin-labeled surface ROMK1 intensity inthe presence of boiled toxin (Fig. 8 A ) and unboiled toxin(Fig. 8 B ). As expected, inhibition of c-Src can stillsignificantly increase the intensity of the avidin-recognized 71-kDaband corresponding to the surface-located ROMK1 (herbimycin Atreatment, 175 ± 20% of the control value) in cells treated withboiled toxin. In contrast, unboiled toxin completely abolished theeffect of herbimycin A because it did not increase the surface-locatedROMK1 channels (92 ± 11% of the control value, n = 7). This suggests that the inhibition of c-Src-induced exocytosisdepends on SNARE proteins.- _7 H8 X1 o5 ]6 V7 \9 X
1 i$ L0 ]1 P7 ^8 K; N( O* E: V! ^& \; r
Fig. 7. Aconfocal image showing the effect of herbimycin A on the relocation ofROMK1 channels in cells transfected with GFP-ROMK1 and cSrc afterincubation of cells with tetanus toxin ( A - C )or boiled toxin ( D - F ). The cell images weretaken before addition of herbimycin A ( A ) and 5 ( B ) and 15 min ( C ) after the PTK inhibitor,respectively. Bar = 10 µM. D - F :confocal images demonstrating the effect of herbimycin A on ROMK1location at 0 ( D ), 5 ( E ), and 10 min( F ) in cells incubated with boiled toxin.
; E) k5 I/ L: k$ o
Fig. 8. Western blot analysis demonstrating the effect ofherbimycin A on biotin-labeled surface ROMK1 in the presence of boiled( A ) or unboiled tetanus toxin ( B ). Cells weretransfected with GFP-ROMK1 and cSrc followed by incubation in a mediacontaining either boiled or unboiled tetanus toxin (30 nM). ROMK1channels were harvested by IP of the cell lysate with GFP antibody. Thedata were normalized according to total ROMK1 protein. Arrow,biotin-labeled ROMK1.

DISCUSSION& x' L2 T* Y' a. ]3 r
6 y; Z: T) q) Q( y. z6 L" |6 e
In the present study, we have demonstrated that the inhibition ofPTK decreases the level of tyrosine-phosphorylated ROMK1 and stimulatesmembrane insertion of ROMK1. Because it is possible that otherpotential subunits such as the CFTR or sulfonylurea receptor may berequired to form native SK channels in the CCD, the term ROMK1 meansROMK1 subunit in the present study. Two lines of evidence suggest thata decrease in tyrosine phosphorylated ROMK1 proteins is required forinitiating the insertion of ROMK1 channels into the cell membrane.First, in vitro phosphorylation experiments have demonstrated thatROMK1 is a substrate of PTK. Second, an increase in ROMK1 insertioninduced by inhibiting PTK was absent in cells transfected with c-Srcand R1Y337A.

In a previous study, we have shown that tyrosine residue 337 of ROMK1is the main site for the PTK-induced phosphorylation, and mutation oftyrosine residue 337 to alanine almost completely abolished thec-Src-induced tyrosine phosphorylation of ROMK1 ( 7 ). Weused herbimycin A as a tool to reduce the level of the tyrosinephosphorylation of ROMK1. Three pieces of evidence strongly suggestthat the effect of herbimycin A results from the inhibition of PTK: 1 ) tyrosine-phosphorylated ROMK1 was diminished when cellswere treated with herbimycin A; 2 ) mutation of tyrosine residue 337 to alanine abolished the effect of herbimycin A on ROMK1redistribution; and 3 ) a previous study showed that the effect of herbimycin A on ROMK1 channel activity was absent in thepresence of the PTP inhibitor ( 30 ). This suggests that the effect of herbimycin A on ROMK1 results from an indirect potentiation of PTP function.) B, Z, a' c* ]0 V5 e
& ^: o2 t+ p& E
In the present study, we employed three different techniques,biotin-labeling, confocal microscopy, and a patch clamp, to demonstratethat inhibition of PTK in cells transfected with ROMK1 and c-Srcstimulates the insertion of ROMK1 in the cell membrane. First, confocalmicroscopy showed that inhibition of PTK dramatically diminished ROMK1located in the intracellular compartment. Second, application of thePTK inhibitor caused a sharp increase in ROMK1 channel activity. Thisobservation is consistent with the previous finding that herbimycin Aincreased K current in oocytes injected with ROMK1 and c-Src. Finally,biotin labeling revealed that inhibition of PTK significantly increasedthe number of surface ROMK1 channels. Therefore, the data haveunambiguously indicated that a decrease in the tyrosine phosphorylationlevel of ROMK1 enhances the insertion of ROMK1 channels. This notionis also supported by the observation that inhibition ofmicrotubule assembly abolished the effect of inhibition of PTK( 32 ) because the microtubule is critically involved inmediating vesicle transportation and fusion.
3 G( E2 p2 x5 d
In contrast, several lines of evidence indicate that the stimulation oftyrosine phosphorylation of ROMK1 caused the internalization of ROMK1( 24 ). First, inhibition of PTP augmented the level of thetyrosine-phosphorylated ROMK1 in HEK-293 cells transfected with ROMK1and c-Src and reduced the biotin-labeled surface ROMK1. Second,confocal microscopic imaging demonstrated that GFP-tagged ROMK1 wasaccumulated in the intracellular compartment. Third, the patch-clampexperiments showed that inhibition of PTP diminished the activity ofROMK1. The effect of inhibition of PTP on ROMK1 channels has also beenobserved in native isolated CCD, in which inhibition of PTP decreasedthe ROMK-like SK channels and the effect of the PTP inhibitor wasblocked by a hypertonic bath solution ( 30 ).$ w8 \. |& U# b; b  p8 l8 t3 T
. g0 c$ U; m- `# Z+ c. _" M
Two lines of evidence strongly suggest that the interaction betweentarget membrane (t)-SNAREs and vesicle (v)-SNAREs is involved inmediating the effect of the inhibition of PTK on ROMK1 insertion: 1 ) inhibition of PTK did not decrease ROMK1 content in theintracellular compartment in cells incubated with tetanus toxin; and 2 ) toxin treatment abolished the herbimycin A-inducedincrease in biotin-labeled ROMK1 channels. The effect of tetanus toxinwas specific because inhibition of c-Src could still increase thenumber of surface ROMK1 in cells treated with boiled tetanus toxin.Tetanus toxin has also been shown to inhibit exocytosis of the protonpump in the male reproductive tract ( 1 ).' j; b0 [0 @# u1 Q/ w! }( c

A large body of evidence indicates that SNARE proteins involved insynaptic vesicle translocation, docking, and membrane fusion ( 18 ) are also required for the regulation of membranefusion in epithelial tissues ( 1, 6, 12, 17, 22, 23 ).Relevant to this conclusion is the finding that insulin-inducedinsertion of the glucose transporter GLUT4 depends on SNARE proteins( 25 ). According to the present concept regarding the roleof SNARE proteins in the regulation of vesicle fusion process, thereare two types of SNAREs: t-SNAREs, which serve as the docking pointsfor vesicles at the plasma membrane, and v-SNAREs, which areresponsible for the interaction with t-SNAREs. The interaction betweent-SNAREs and v-SNAREs causes vesicle docking at the plasma membrane.The mechanism by which tetanus toxin blocks vesicle docking and fusion is that the toxin cleaves v-SNAREs and abolishes the interaction between t-SNAREs and v-SNAREs ( 16 ).: z% K3 S. I0 N$ \

SNARE proteins are involved in the regulation of membrane transporterprotein trafficking ( 6, 12, 17, 22, 23 ). It has been shownthat syntaxin 1A stoichometrically binds to the NH 2 terminus of CFTR and regulates Cl currents of CFTR ( 12 ). Moreover, syntaxin 1A and other members of the SNARE family are involved in the control of cell surface number of epithelial Na channels ( 17, 22 ). In the CCD, SNAP-23 has beendemonstrated to be colocalized with aquaporin-2 and play an importantrole in mediating the effect of vasopressin on aquaporin-2 insertion ( 23 ). Although the observation that tetanus toxinabolished the effect of herbimycin A on ROMK1 trafficking stronglysuggests that SNARE proteins are involved in the regulation of ROMK1trafficking, it is not known which syntaxin and SNAP-25 homologues arespecifically responsible for ROMK1-containing vesicle fusion. However,several investigations have reported that SNAP homologues and syntaxin are expressed in the CCD ( 6, 23 ). We need furtherexperiments to determine which member of the SNARE family is involvedin the regulation of ROMK1 insertion.& K( ^( |( v8 S2 R! S. i' X, }2 e

ROMK1 is responsible for K secretion in the CCD under physiologicalconditions ( 4 ). We and others have demonstrated that ahigh-K intake increases the number of ROMK-like SK channels in the CCD( 13, 29 ). Although channel activity in the CCD from ratson a low-K diet is not significantly lower than those on a normal-Kdiet ( 15, 31 ), it is possible that channel activity in theCCDs obtained from rats on a normal-K diet is underestimated. This isbecause the patch-clamp experiments were performed in split-opentubules, a preparation that is different from in vivo conditions. Forinstance, an increase in Na delivery can stimulate the basolateralNa-K-ATPase, and activity of Na-K-ATPase is closely coupled to apicalROMK-like SK channel activity ( 11, 26 ). Because theactivity of Na-K-ATPase in the split-open CCD may be lower than that inin vivo conditions, the SK channel activity observed in ourexperimental conditions may not represent "true" channel activityin vivo. Therefore, it is conceivable that the activity of SK channelsin CCD from rats on a normal-K diet is higher than that from rats on alow-K diet. Moreover, it is possible that there are fewer "silent"ROMK1 channels in the apical membrane of CCD from K-depleted rats thanthose of rats on a normal-K diet. This speculation is supported by theobservation that ROMK expression in the cell membrane fraction wassignificantly lower in K-depleted rats than that in rats on a normal-Kdiet ( 8 ). We hypothesize that there are two populations ofROMK1 in the cell membrane of the CCD: active and inactive ROMK1channels. Inactive ROMK1 channels serve as a reserved K channel pool,which can become active in response to a variety of stimuli such ashormones and other local factors. However, the pool size of inactiveROMK1 diminishes when dietary K intake is low and PTK activity isstimulated. Accordingly, ROMK1 channels are internalized and theresponse of ROMK1 channels to a given hormone is abolished. This canfunction as a K-saving mechanism. When dietary K intake is normal orhigh, the activity of PTK is suppressed. Consequently, ROMK1 channelsare tyrosine dephosphorylated, reinserted into the apical membrane, andready to respond to a particular factor that stimulates K secretion. This notion is supported by the observation that thetyrosine-phosphorylated ROMK1 channel level is significantly less inkidneys from rats on a high-K diet than those from animals on aK-deficient diet ( 7 ).
+ m# P4 s% R. w1 T9 N! T. k% ~7 T
We conclude that inhibition of PTK stimulates thetyrosine-dephosphorylated ROMK1 channels and increases the insertion of ROMK1 in the cell membrane by a SNARE protein-dependent mechanism.

ACKNOWLEDGEMENTS
, k0 P9 w% I, K' ]
This work was supported by National Institute of Diabetes andDigestive and Kidney Diseases Grants DK-47402 and DK-54983." h2 S4 D8 J+ K* `% y# e3 ]
      【参考文献】1 ~, @  q! S+ G' {4 x7 y# w
1. Breton, S,Nsumu NN,Galli T,Sabolic I,Smith PJS,andBrown D. Tetanus toxin-mediated cleavage of cellubrevin inhibits proton secretion in the male reproductive tract. Am J Physiol Renal Physiol 278:F717-F725,2000 .
3 K# N3 \& }4 v' b
8 q& G8 }2 N' I3 J0 H4 ^; x3 W3 `
+ z" E; X7 b" J+ F
2. Frindt, G,andPalmer LG. Ca-activated K channels in apical membrane of mammalian CCT, and their role in K secretion. Am J Physiol Renal Fluid Electrolyte Physiol 252:F458-F467,1987 .0 f# j, M1 H. D  g7 x: ^
9 O" X8 o* G+ Y5 X; {

+ ]* _: v! ]  b  w# }- H
3. Frindt, G,andPalmer LG. Low-conductance K channels in apical membrane of rat cortical collecting tubule. Am J Physiol Renal Fluid Electrolyte Physiol 256:F143-F151,1989 .' x$ c5 r6 M) i  x$ Y
( W& v& ]' m, M- T9 F0 i4 H
: J/ i4 C# ^& W
' B/ @5 M& D" c9 i2 w- K
4. Giebisch, G. Renal potassium transport: mechanisms and regulation. Am J Physiol Renal Physiol 274:F817-F833,1998 .

5. Ho, K,Nichols CG,Lederer WJ,Lytton J,Vassilev PM,Kanazirska MV,andHebert SC. Cloning and expression of an inwardly rectifying ATP-regulated potassium channel. Nature 362:31-38,1993 .
  F7 h4 t: D! I9 `

6. Inoue, T,Nielsen S,Mandon B,Terris J,Kishore BK,andKnepper MA. SNAP-23 in rat kidney: colocalization with aquaporin-2 in collecting duct vesicles. Am J Physiol Renal Physiol 275:F752-F760,1998 .
( t% Z8 w; X8 t* p8 T0 _% e/ ^" i( L
0 V7 `. {* D7 D/ S8 Z: @! ?
6 r" e& e" n7 U+ q8 k
7. Lin, DH,Sterling H,Lerea KM,Welling P,Jin L,Giebisch G,andWang W-H. K depletion increases the protein tyrosine-mediated phosphorylation of ROMK. Am J Physiol Renal Physiol 283:F671-F677,2002 ., A# ], j: `5 ~& `# o
7 r7 N+ v( V' N6 w& s3 O
. h0 f  x$ }& I: N& C% R& C
/ T4 n* l/ t5 N  P; v2 R6 ]9 C# S
8. Mennitt, PA,Frindt G,Silver RB,andPalmer LG. Potassium restriction downregulates ROMK expression in rat kidney. Am J Physiol Renal Physiol 278:F916-F924,1999 .
5 v4 ~3 g0 R6 B5 A
. i) ]9 `+ \# |
, F9 D1 W3 D6 X- Q6 l3 I  u
9. Mennitt, PA,Wade JB,Ecelbarger CA,Palmer LG,andFrindt G. Localization of ROMK channels in the rat kidney. J Am Soc Nephrol 8:1823-1830,1997 .2 ]' B! v% Z. n8 t1 h9 f7 h0 S. M9 S

10. Moral, Z,Dong K,Wei Y,Sterling H,Deng H,Ali S,Gu RM,Huang XY,Hebert SC,Giebisch G,andWang WH. Regulation of ROMK1 channels by protein tyrosine kinase and tyrosine phosphatase. J Bio Chem 276:7156-7163,2001 .* P1 p% _5 J$ ^9 B+ B

9 g! z5 U9 ^  U& F
7 F7 \8 l. D. h0 P, d" z
11. Muto, S,Asano M,Seldin D,andGiebisch G. Basolateral Na pump modulates apical Na and K conductance in rabbit cortical collecting ducts. Am J Physiol Renal Physiol 276:F143-F158,1999 ./ B, Z! ^- e- u! e

: |; R0 K4 @# }
12. Naren, AP,Di A,Cormet-Boyaka E,Boyaka PN,McGhee JR,Zhou W,Akagawa K,Fujiwara T,Thome U,Engelhardt JF,Nelson DJ,andKirk KL. Syntaxin 1A is expressed in airway epithelial cells, where it modulates CFTR Cl currents. J Clin Invest 105:377-386,2000  .

6 @0 P) Z( V0 S% M& Q8 d9 e, U

13. Palmer, LG,Antonian L,andFrindt G. Regulation of apical K and Na channels and Na/K pumps in rat cortical collecting tubule by dietary K. J Gen Physiol 105:693-710,1994./ x) L+ e. G& L4 }
+ l) A* b+ V( \3 A* b" v

14. Palmer, LG,Choe H,andFrindt G. Is the secretory K channel in the rat CCT ROMK? Am J Physiol Renal Physiol 273:F404-F410,1997 .
- j; }# k( f  i& _0 a

6 j7 l0 Q( i# _
15. Palmer, LG,andFrindt G. Regulation of apical K channels in the rat cortical collecting tubule during changes in dietary K intake. Am J Physiol Renal Physiol 277:F805-F812,1999 .
% i5 G: A, b: X2 ^9 P

16. Pellizzar, R,Rossetto O,Lozzi L,Giovedi S,Johnson E,Shone CC,andMontecucco C. Structural determinants of the specificity for synaptic vesicle-associated membrane protein/synaptobrevin of tetanus and botulinum type B and G neurotoxins. J Biol Chem 271:20353-20358,1996 .
+ ~% G& q! |7 Z* _8 k) E: V
. ~( P% L$ e3 K, |) t( w3 n+ D6 J
) a- H8 ~9 V+ A( G
17. Qi, J,Peters KW,Liu C,Wang JM,Edinger RS,Johnson JP,Watkins SC,andFrizzell RA. Regulation of the amiloride-sensitive epithelial sodium channel by syntaxin 1A. J Biol Chem 274:30345-30348,1999 .

18. Robinson, LJ,andMartin TFJ Docking and fusion in neuroscience. Curr Opin Cell Biol 10:483-492,1998  .
+ T) M  g. W8 l% }- w9 A; s5 \0 p
+ ^2 X# w/ Z, P- g. X

19. Ruknudin, A,Schulze DH,Sullivan SK,Lederer WJ,andWelling P. Novel subunit composition of a renal epithelial K ATP channel. J Biol Chem 273:14165-14171,1998 .
) m+ {, w# l$ Q# F9 W' W

( C9 s" K' }6 m: `8 b& D
20. Satlin, LM. Postnatal maturation of potassium transport in rabbit cortical collecting duct. Am J Physiol Renal Fluid Electrolyte Physiol 266:F57-F65,1994 .$ x% B7 G. T5 m7 E
' k; A% ?+ l( V% z& `

1 U7 z, a$ }! T
21. Satlin, LM,andPalmer LG. Apical K conductance in maturing rabbit principal cell. Am J Physiol Renal Physiol 272:F397-F404,1997 .
, k$ r1 P4 n8 X$ J+ h% @; \
7 K* o0 `3 k2 T; v2 X
2 X/ }0 J5 p' g5 `
22. Saxena, S,Quick MW,Tousson A,Oh Y,andWarnock DG. Interaction of syntaxins with the amiloride-sensitive epithelial sodium channel. J Biol Chem 274:20812-20817,1999 .
/ }" C) Y: d  n

23. Shukla, A,Hager H,Corydon TJ,Bean AJ,Dahl R,Vajda Z,Li H,Hoffmann HJ,andNielsen S. SNAP-25 associated Hrs-2 protein colocalizes with AQP2 in rat kidney collecting duct principal cells. Am J Physiol Renal Physiol 281:F546-F556,2001 .
. l. _* C2 M, O( m, n+ x. O" Y1 R2 J

24. Sterling, H,Lin DH,Gu RM,Dong K,Hebert SC,andWang WH. Inhibition of protein-tyrosine phosphatase stimulates the dynamin-dependent endocytosis of ROMK1. J Biol Chem 277:4317-4323,2002 .( e! ?+ @% d- h+ b6 h" g

25. Tong, P,Khayat ZA,Huang C,Patel N,Ueyama A,andKlip A. Insulin-induced cortical actin remodeling promotes GLUT4 insertion at muscle cell membrane ruffles. J Clin Invest 108:371-381,2001  .
2 h- ~6 g" ]* {, k3 \2 |1 j

, l( D# G! `, t4 c
26. Wang, WH,Geibel J,andGiebisch G. Mechanism of apical K channel modulation in principal renal tubule cells. Effect of inhibition of basolateral Na -K -ATPase. J Gen Physiol 101:673-694,1993 .1 s3 G7 h- O6 ]( w. }4 P
( s  ?) m& s1 n8 E: x

27. Wang, WH,Hebert SC,andGiebisch G. Renal K channels: structure and function. Annu Rev Physiol 59:413-436,1997  ./ G% U/ y/ b6 H: Y
) p2 Y. G( Q# i% @& K% x
5 f, }6 `7 w% h
; p/ }; A$ P; i1 z8 K/ b/ U
28. Wang, W-H,Lerea KM,Chan M,andGiebisch G. Protein tyrosine kinase regulates the number of renal secretory K channels. Am J Physiol Renal Physiol 278:F165-F171,2000 .0 Q! a( G; f% W% B

% J  V# D# u5 o& j+ V  Y# R
' F# M9 O: n. M. E/ e0 v
29. Wang, W-H,Schwab A,andGiebisch G. Regulation of small-conductance K channel in apical membrane of rat cortical collecting tubule. Am J Physiol Renal Fluid Electrolyte Physiol 259:F494-F502,1990 .

& ]7 {: Z* j1 O
30. Wei, Y,Bloom P,Gu RM,andWang W-H. Protein-tyrosine phosphatase reduces the number of apical small conductance K channels in the rat cortical collecting duct. J Biol Chem 275:20502-20507,2000 .

# t- Q4 ~" v# s
31. Wei, Y,Bloom P,Lin DH,Gu RM,andWang W-H. Effect of dietary K intake on the apical small-conductance K channel in the CCD: role of protein tyrosine kinase. Am J Physiol Renal Physiol 281:F206-F212,2001 .7 N+ O4 g8 z2 C" v3 j$ B2 ?

6 q( R+ ^& n9 x& j9 X

32. Wei, Y,andWang W-H. The role of cytoskeleton in mediating the effect of vasopressin and herbimycin A on the secretory K channels in the CCD. Am J Physiol Renal Physiol 282:F680-F686,2002 .

, Z; W9 i0 l. Q, `# s& k* _- c
33. Xu, JZ,Hall AE,Peterson LN,Bienkowski MJ,Eessalu TE,andHebert SC. Localization of the ROMK protein on apical membranes of rat kidney nephron segments. Am J Physiol Renal Physiol 273:F739-F748,1997 .

收藏0 分享0 顶0 踩0