Disruption of F-actin stimulates hypertonic activation of theBGT1 transporter i - 《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 ... 24 下一页

返回列表

查看: 396909\|回复: 235	go Disruption of F-actin stimulates hypertonic activation of theBGT1 transporter i [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 13:36 |只看该作者 |倒序浏览 |打印

作者：Jeremy L.Bricker, ShaoyouChu,  Stephen A.Kempson作者单位：Department of Cellular and Integrative Physiology, IndianaUniversity School of Medicine, Indianapolis, Indiana 46202-5120
   1 n1 v! V, V8 R/ ?3 J4 X8 l$ I3 W




         / Y$ o; m3 D# ], ]4 F
         8 n8 L1 H5 H7 |8 ~



      0 B/ p# z; r/ z, c
      ! {: R2 {9 o1 h# ^
      【摘要】
   Many membrane transport systems arealtered by changes in the state of the actin cytoskeleton. Although anintact microtubule network is required for hypertonic activation of thebetaine transporter (BGT1), the possible role of the actin cytoskeletonis unknown. BGT1 function in Madin-Darby canine kidney cell monolayerswas assessed as Na -dependent uptake of GABA, followingdisassembly of F-actin by cytochalasin D (1.0 µM) or latrunculin A(0.6 µM). Both drugs significantly increased ( P by 24-h hypertonicity (500 mosmol/kgH 2 O). In contrast, the hypertonic upregulation of Na -dependent alanine uptake remainedunaltered by cytochalasin D. Disruption of F-actin did not interferewith downregulation of BGT1 transport when cells were transferred fromhypertonic to isotonic medium. Immunofluorescence staining revealedcolocalization of BGT1 and F-actin at the plasma membrane of hypertoniccells. Surface biotinylation revealed no major change in BGT1 protein abundance after cytochalasin D action, suggesting that stimulation ofhypertonic activation of BGT1 transport is due to increased activity ofexisting BGT1 transporters. ( J' X0 y2 X. X
      【关键词】 cytochalasin D latrunculin A phalloidin osmotic stress alanine GABA
   INTRODUCTION( u  T2 I; b8 j4 i5 s6 s! }

WHEN EXPOSED TO HYPERTONIC stress, cells regulate their volume initially by ionaccumulation. During prolonged hypertonicity, this step is followed byamino acid accumulation and eventually by accumulation of organicsolutes or osmolytes. The intracellular accumulation of osmolytes doesnot disrupt metabolic pathways or interrupt protein synthesis, incontrast to the accumulation of ions and amino acids. Betaine is anosmolyte that enters cells via the betaine transporter (BGT1) that alsotransports GABA ( 21 ). BGT1 is found in many mammaliantissues, including the brain ( 5 ), and is abundant in cellsof the inner renal medulla where it is located in the basolateralmembrane ( 27, 36 ). Renal medullary cells are routinelyexposed to the high hypertonicity that is an important component of theurinary concentrating mechanism. The activity of BGT1 is upregulated inresponse to prolonged (24 h) hypertonic stress ( 26, 36 ),and this is an important part of the overall adaptation that allows themedullary cells to balance the osmotic stress across the plasmamembrane ( 6 ).

The activities of many membrane transporters have been shown to beaffected by changes in the state of the actin cytoskeleton ( 16 ), suggesting a possible regulatory role for actin( 24 ). For example, the open probability of thevoltage-dependent Na channel in rabbit cardiac myocytes isdecreased when the actin cytoskeleton was disrupted by cytochalasin D( 33 ), and in rat hippocampal neurons the Ca 2  influx by the Ca 2  transporter was decreased bycytochalasin D ( 13 ). In contrast, the conductance of thecystic fibrosis transmembrane regulator in 3T3 fibroblasts( 11 ), insertion of aquaporins into epithelial cell apicalmembranes ( 12 ), and the open probability of the epithelialNa channel ( 4 ) were increased by cytochalasinD. Previous studies in this laboratory showed that microtubuledisruption by nocodazole or colchicine prevented the typical hypertonicupregulation of BGT1 transport in Madin-Darby canine kidney (MDCK)cells ( 3 ). Microtubules are an important component of thetrafficking mechanism that delivers newly synthesized BGT1 transportersfrom intracellular compartments to the plasma membrane. The presentstudy extends these observations by focusing on the role of the actincytoskeleton in hypertonic upregulation of BGT1 in MDCK cells.

METHODS8 ?# g' ]+ R* g* }+ l+ K$ X+ n* d
9 y" T* `8 M, v' Y# f: g( C' N- T
Cell culture and transport measurement. MDCK cells (CCL-34, American Type Culture Collection, Rockville, MD)were used between passages 10 and 30 and were grown as monolayers in a 1:1 mixture of DMEM:Ham's F-12Kcontaining 10% bovine calf serum, 10 mM HEPES, 25 mMNaHCO 3 (pH 7.4), and penicillin G (100 U/ml), as inprevious studies ( 3, 18 ). Cultures were maintained at37°C in an atmosphere of 5% CO 2 in air. Cells were grownon glass coverslips for immunofluorescence studies and in six-wellplates for transport measurements.
! u) Q' H/ l9 Z4 S: F+ W4 w
Hypertonic stress was induced by replacing normal growth medium withgrowth medium made hypertonic by addition of sucrose to achieve a finalosmolality of 500 mosmol/kgH 2 O, as in previous studies( 3 ). The osmolality of all solutions used for transport was matched to the osmolality of the growth medium by addition ofsucrose where necessary. The transport function of endogenous BGT1 incell monolayers was determined by measuring cell uptake of[ 3 H]GABA, as described previously in detail ( 3, 18 ). Briefly, [ 3 H]GABA uptake was determined bothin medium containing Na and also in medium in whichNa was replaced by methyl- D -glucamine-HCl. Thedifference, which is referred to as the Na -dependentcomponent, represents transport specifically via BGT1.. t' h8 D1 [; }; C& m0 ~: E, i& A* b+ q
7 Z1 q- J8 ?$ E# A' o
Drugs were used from stock solutions that were diluted at least 1:1,000after addition to cell culture medium. Cytochalasin D (Sigma Aldrich,St. Louis, MO) and latrunculin A (Calbiochem, San Diego, CA) weredissolved in dimethylsulfoxide and were added to the growth medium whenthe cells were switched from isotonic to hypertonic conditions.Phalloidin (Sigma Aldrich) was dissolved in methanol, and treatment ofcells was always begun in isotonic medium for 24 h before a switchto hypertonic medium containing phalloidin. This allows adequate timefor phalloidin to accumulate within the cells ( 22, 23 )before onset of hypertonic stress.

Visualization of actin in cultured cells. The actin cytoskeleton was visualized by staining cells with Texasred-phalloidin (Molecular Probes, Eugene, OR), as described previously( 28 ). Cell monolayers on glass coverslips were rinsed inPBS of appropriate osmolality and were fixed for 15 min in 4%paraformaldehyde in PBS. The cells were rinsed three times, permeabilized by immersion for 2 min in 0.2% Triton X-100 in PBS, andwashed three times by 5-min immersion in PBS. After incubation for 30 min at 37°C in Texas red-phalloidin, diluted 1:400 in PBS, the cellswere rinsed three times by immersion in PBS and mounted influoromount-G (Southern Biotechnology, Birmingham, AL). Images wererecorded on CD-RW using an RT Color Spotdigital camera (Diagnostic Instruments, Sterling Heights, MI) mounted on an Optiphot-2 Nikon epifluorescent microscope with an oil-immersion ×60 lens. The imageswere imported into Adobe Photoshop 5.0 (Adobe Systems, San Jose, CA).

Colocalization of actin and BGT1 proteins. Cell monolayers were rinsed in PBS as above, and some were extractedwith a cytoskeleton-stabilizing buffer containing 1% Triton X-100, 300 mM sucrose, 100 mM NaCl, 3 mM MgCl 2, 1 mM EGTA, andprotease inhibitors (Halt kit, Pierce, Rockford, IL) for 15 min on ice( 34 ). All samples were fixed by immersion for 10 min incold methanol, as described previously ( 3 ). A 30-min preincubation in 2% gelatin in PBS was followed by a 2-h incubation inaffinity-purified anti-dog BGT1 rabbit antibody ( 3 ) (1:100 dilution) and an anti-actin mouse monoclonal antibody (clone AC-40, Sigma) (1:200) diluted in 1% gelatin/PBS. Primary antibodies were detected by a 1-h incubation in affinity-purified, FITC-conjugated goatanti-rabbit IgG (1:100) and TRITC-conjugated goat anti-mouse IgG (1:50)(Jackson ImmunoResearch, West Grove, PA) in 1% gelatin/PBS. Allincubations were at 37°C. Cells were mounted in fluoromount-G, andimages were acquired with a Zeiss LSM 510 confocal microscope using a×40 water-immersion lens with 1.2 numerical aperture and 22-µmworking distance. Excitation wavelengths were 488 (FITC) and 543 nm(TRITC), and emission was collected at 500-530 nm (FITC) and565-615 nm (TRITC). The instrument gain and offset were the samefor all samples.( b! q: B' i/ a5 a& q" g. {1 k
' p" j: v2 h' C# S! P+ N
Surface biotinylation. The procedure was similar to that described previously for MDCK( 27 ) and opossum kidney ( 37 ) cells. Cellmonolayers in 100-mm dishes were used 3 days after plating, just beforeconfluence, to allow access of reagents to the basolateral membranewhere BGT1 is primarily located ( 27 ). The cells werewashed four times in ice-cold PBS containing 0.1 mM CaCl 2 and 1.0 mM MgCl 2 (PBSCM), pH 7.4, and incubated with PBSCM(2 ml/dish) containing biotin-X-NHS (Calbiochem) at a finalconcentration of 1.0 mg/ml for 30 min at 4°C with constant rocking.After aspiration of the biotin solution, the cells were washed fivetimes in ice-cold PBSCM. When appropriate, all the preceding steps wereperformed with hypertonic solutions (500 mosmol/kgH 2 O). Thecells were lysed with 1 ml/dish of 1% Triton X-100 solution containing150 mM NaCl, 5 mM EDTA, 10% glycerol, 50 mM Tris, pH 7.4, and proteaseinhibitors (Halt cocktail, Pierce). Cells were collected by scraping,sonicated briefly (5 s) to disrupt DNA, incubated for 30 min at 4°Cwith rotation, and centrifuged at 13,000 g for 15 min. Thesupernatant was adjusted to a protein concentration of 1 mg/ml, andbiotinylated proteins were precipitated by addition of 100 µl of aslurry of streptavidin-coupled agarose (Pierce) followed by incubationwith rotation at room temperature for 1 h. To prevent nonspecificbinding, the agarose beads were previously incubated with PBSCMcontaining 2% bovine serum albumin, followed by four washeswith PBSCM. The Triton supernatant was centrifuged to pellet thebeads, and the supernatant was stored at 80°C. The beads werewashed four times in the Triton lysis buffer, mixed with an equalvolume of 2× electrophoresis sample buffer ( 3 ), heated at65°C for 15 min, and subjected to gel electrophoresis and Westernblotting, as described previously ( 3 ). Rabbit antibodiesto dog BGT1 and mouse E-cadherin, generously provided by Drs. Moo Kwon(Johns Hopkins University School of Medicine) and James Marrs (IndianaUniversity School of Medicine), were used at dilutions of 1:1,500 and1:10,000, respectively. Mouse monoclonal antibody to actin (Sigma) wasused at 1:1,000. Secondary antibodies conjugated to horseradishperoxidase and ECL reagents (Amersham) were used for detection. Shortexposures of the Western blots that were not saturating were used forquantitation. Films were scanned with a Bio-Rad GS-670 imagingdensitometer ( 3 ).& u1 f9 v& M. ~7 X

RESULTS& d$ g% `& z7 D

In MDCK cells maintained in isotonic medium, there was nosignificant Na -dependent uptake of GABA, indicating theabsence of BGT1 transport activity. However, as expected, after 24-htreatment of MDCK cells with hypertonic medium (500 mosmol/kgH 2 O), there was activation of BGT1 transport, asindicated by pronounced Na -dependent uptake of GABA (Fig. 1 ). When the 24-h hypertonic stress wasperformed in the presence of 1.0 µM cytochalasin D, there wassignificant stimulation (109 ± 45%) of the hypertonic activation of BGT1. In contrast, cytochalasin D did not change GABA uptake incells in isotonic medium. The final concentration of cytochalasin D(1.0 µM) was in the range shown previously to be effective for disrupting the actin cytoskeleton in epithelial and other cell types( 7, 23, 31, 35 ).
/ X* @3 n. i& _8 N
Fig. 1. Cytochalasin D (CD) stimulates hypertonic activation ofNa -dependent GABA uptake in Madin-Darby canine kidney(MDCK) cells. After incubation for 24 h in either hypertonic (500 mosmol/kgH 2 O) or isotonic growth medium, the 10-min uptakesof GABA were determined in the presence of Na (Na) or whenNa was replaced by methyl D -glucamine (MG) andin the presence or absence of 1.0 µM CD. Data are means ± SEfrom 3 separate experiments. * Significantly different( P t -test) compared withNa -dependent uptake in cells not treated with CD.

MDCK cells grown on glass filters were stained with Texasred-phalloidin to visualize the actin cytoskeleton after exposure tohypertonic medium in the presence and absence of cytochalasin Dfor 24 h. Filamentous actin in controls (Fig. 2 A ) was completely disrupted in cells treated with cytochalasin D (Fig. 2 B ). Most of the actin was clumped together around the cellperiphery, as shown previously ( 28 ). Additional studies(not shown) revealed that cytoskeletal disruption was establishedwithin 1 h after addition of the drug, as reported previously inmuscle cells ( 31 ).

Fig. 2. Filamentous actin of MDCK cells in hypertonic medium ( A )is reduced to clumps of globular actin during exposure to 1.0 µM CD( B ) for 24 h. This change was established within 1 h of drug addition (not shown). Actin was visualized by staining withTexas red-phalloidin. Original magnification ×1,000.% _7 u1 f: s: j' y& L

Latrunculin A also disrupts the actin cytoskeleton, but the mechanismis different from that for cytochalasin D. Latrunculin A sequestersG-actin to inhibit actin polymerization, whereas cytochalasin D changesthe polymerization and depolymerization rates of F-actin ( 22, 25 ). A final concentration of 0.6 µM latrunculin A was usedbecause an imaging study of the dose dependence showed that 0.6 µMdisrupted the actin cytoskeleton without causing rounding up and lossof cells from the culture dish. This is also within the concentrationrange used previously ( 17 ). Exposure of MDCK cells to 0.6 µM latrunculin A for 24 h produced a similar response in BGT1transport compared with cytochalasin D. Activation of BGT1 transport byhypertonic stress was more pronounced in the presence of latrunculin A,an increase of 52 ± 8% compared with hypertonicity alone (Fig. 3 ). Latrunculin A had no effect on GABAuptake in cells maintained in isotonic medium.

Fig. 3. Latrunculin A (LA) stimulates hypertonic activation ofNa -dependent GABA uptake in MDCK cells. Finalconcentration of LA was 0.6 µM. Other details as in Fig. 1. Data aremeans ± SE from 6 separate experiments. * P t -test).: Z9 h7 {. _: B- ?

In contrast to latrunculin A and cytochalasin D, phalloidin stabilizesthe actin cytoskeleton ( 22, 23 ). A dose-dependent studyconfirmed that 4 µM phalloidin was an effective concentration inhypertonic cells ( 23 ). This concentration was used tostudy the effect of actin stabilization on BGT1 transport in MDCKcells. Activation of BGT1 transport in cells exposed to hypertonicmedium containing 4 µM phalloidin for 24 h was not differentcompared with activation by hypertonic medium alone (Fig. 4 ).$ U/ _9 g4 _- E) L
% V2 _' t9 j( g0 S
Fig. 4. Phalloidin (PH) does not change hypertonic activation ofNa -dependent GABA uptake in MDCK cells. Finalconcentration of PH was 4.0 µM. Other details are as in Fig. 1. Dataare means ± SE from 3 separate experiments.) {1 e* P2 r. O% `( X8 o4 ]

Because cytochalasin B was shown previously to inhibit glucosetransporters ( 29 ), it was important to verify that thestimulation of BGT1 transport by cytochalasin D and latrunculin A wasdue primarily to actin destabilization. The possible direct action ofthese drugs on BGT1 transport was tested by including them in the GABAuptake medium instead of the cell growth medium. The hypertonicactivation of BGT1 transport remained unchanged during this relativelybrief 10-min exposure to either drug (Fig. 5 ), indicating that neither cytochalasinD nor latrunculin A had any direct effect on the BGT1 transporter.Phalloidin also had no effect on BGT1 transport when tested in this way(Fig. 5 ).

Fig. 5. Acute exposure of MDCK cells in hypertonic medium to CD(1.0 µM), LA (0.6 µM), or PH (4.0 µM) did not change the 10-minuptakes of GABA either in the presence or absence of Na  .Data are means ± SE of 3 separate experiments. Other details areas in Fig. 1.
' r! ~) q8 ]2 v+ V/ U* ?
To determine whether an intact cytoskeleton is a specific requirementfor hypertonic regulation of the BGT1 transporter, alanine uptake inMDCK cells was studied. We showed previously that amino acid system A,a major route for alanine uptake in MDCK cells, is upregulated within5 h after exposure to hypertonic stress ( 9 ). Cellswere treated with hypertonic cell growth medium containing 1.0 µMcytochalasin D for 5 h before determining alanine uptake. Comparedwith hypertonic medium alone, the presence of cytochalasin D had nosignificant effect on the upregulation of Na -dependentalanine uptake (Fig. 6 ). Similarly a 24-htreatment with cytochalasin D, corresponding to the time course of BGT1 upregulation, also did not change Na -dependent alaninetransport (not shown). This indicates that the stimulatory effect ofactin disassembly is selective for the hypertonic upregulation of theBGT1 transporter.* t! Y% R  ?! |4 ]# `# E, ?' T; n

Fig. 6. Stimulation of Na -dependent alanine uptakeby hypertonic stress for 5 h was not changed significantly by thepresence of CD (1.0 µM). Uptake data are means ± SE from 4 separate experiments. Uptakes are expressed relative to the alanineuptake determined in isotonic Na medium, which was0.89 ± 0.31 nmol · mg 1 · 10 min 1. Other details are as in Fig. 1.: I/ d/ w4 |* ^: ?5 D# R: Q0 `
' B) |8 I, x0 h  ^3 i- r
The action of cytochalasin D to augment hypertonic activation of BGT1transport could be due to increased insertion of new proteins or todecreased retrieval of existing proteins from the plasma membrane. Thelatter possibility was tested indirectly by monitoring downregulationof BGT1 transport after recovery in isotonic medium. MDCK cells werehypertonically stressed for 24 h, returned to isotonic medium, andthe downregulation of Na -dependent GABA uptake wasdetermined after recovery for 24 and 48 h in the presence orabsence of 1.0 µM cytochalasin D. Compared with the GABA uptake incells switched to isotonic medium in the absence of cytochalasin D, thepresence of cytochalasin D in the isotonic medium for either 24 or48 h had no effect on downregulation of BGT1 transport (Fig. 7 ). This strongly suggests that thestimulatory action of cytochalasin D on the hypertonic activation ofBGT1 transport is not due to blocking retrieval of BGT1 transporters from the membrane.- [( D* Q5 ^- j# M: q+ C$ v

Fig. 7. Presence of 1.0 µM CD does not inhibit thedownregulation of Na -dependent GABA transport duringisotonic recovery of MDCK cells for either 24 h ( A ) or48 h ( B ). GABA transport was first upregulated by 24-hexposure to hypertonic medium before switching to isotonic medium for24 or 48 h. Data are means ± SE of 3-5 separateexperiments. Uptakes are expressed relative to the Na uptake by CD-treated cells in isotonic medium. * P t -test) compared with Na uptakeafter 24 h in hypertonic medium.
  o* T) o0 k! m, G9 c% \
The next experiments focused on the upregulation of BGT1transport and used surface biotinylation to determine whethercytochalasin D increased the abundance of BGT1 protein in the plasmamembrane. As expected, no BGT1 protein was detected at the cell surface under isotonic conditions, but there was a marked increase in surfaceBGT1 after hypertonic stress (Fig. 8 ).Cytochalasin D produced no major change in surface BGT1 abundance undereither isotonic or hypertonic conditions. Blots were stripped andreprobed with antibody to E-cadherin, a loading control. E-cadherin was biotinylated under all conditions and its abundance did not change (Fig. 8 ). In each sample lane, the densities of the BGT1 and E-cadherin signals were quantitated by densitometric scans. On the basis of sixseparate experiments, the BGT1/cadherin ratio was 3.9 ± 0.6 (means ± SE) for hypertonic cells and 4.1 ± 1.1 forhypertonic cells treated with cytochalasin D ( P 0.05, paired t -test). These findings suggest that there wasno significant increase in surface abundance of BGT1 protein followingcytochalasin D treatment.
; _3 I$ W- E1 l+ t5 m) b; X
Fig. 8. Western blot showing the abundance of betaine transporter(BGT1) protein at the cell surface was not changed by CD under eitherisotonic (Iso) or hypertonic (Hyp) conditions. MDCK cells wereincubated for 24 h in either Iso or Hyp growth medium in thepresence or absence of 1.0 µM CD. Surface proteins were labeled withbiotin and recovered from cell lysates with streptavidin beads.E-cadherin was used as an internal loading control.
7 _1 R* d3 _1 z6 A( f( k* F4 S
The possibility that biotin may have access to an intracellular pool ofBGT1 in hypertonic or drug-treated cells was tested by probing foractin, an intracellular protein ( 37 ). Although actin wasvery abundant in the initial Triton lysate (before streptavidin treatment), almost no actin was detectable in the protein fractions recovered on streptavidin beads (Fig. 9 ).This indicates that biotinylation of intracellular proteins wasnegligible and was not affected by cytochalasin D. Thus data in Fig. 8 are not complicated by labeling of intracellular BGT1, which might maska change in surface labeling.# d* Y# k( H# _

Fig. 9. Western blot showing that actin in MDCK cells was notbiotinylated under Iso or Hyp conditions. Although abundant in thetotal cell lysate (Triton extract), almost no actin was detected in theprotein fraction pulled down by streptavidin beads.

The possibility of a physical interaction between F-actin and BGT1 inMDCK cells was tested by colocalization using dual antibody staining.Under isotonic conditions, there was little evidence of stress fibersin confluent cells, most of the F-actin stain was cytosolic (Fig. 10 A ). The same was true forBGT1 staining (Fig. 10 B ), and this was confirmed by themerged image (Fig. 10 C ). Triton extraction removed most ofthe cytosolic staining, only cortical F-actin remained (Fig. 10, D and E ), and there was little or no overlap(Fig. 10 F ). After 24 h of hypertonic stress, most ofthe F-actin was distributed around the cell periphery in the cortical pool (Fig. 10 G ). As expected, most of the BGT1 stain was inthe plasma membrane (Fig. 10 H ) where there was pronouncedcolocalization with cortical actin (Fig. 10 I ). Tritonextraction of hypertonic cells removed all the BGT1 without changingthe distribution of F-actin, and no overlapping signals were detected(Fig. 10, J - L ). This suggests that any directphysical interactions between F-actin and BGT1 are weak because theyare disrupted by 1% Triton. A similar conclusion is indicated byexperiments with cytochalasin D. Inclusion of cytochalasin D duringhypertonic stress disrupted the cortical F-actin and the cellboundaries were much less distinct (Fig. 10 M ). In contrast,there was no noticeable change in BGT1 distribution (Fig. 10 N ) resulting in a marked decrease in colocalization (Fig. 10 O ). As expected, the combination of cytochalasin D andTriton treatment removed more F-actin than either treatment alone (Fig. 10 P ) and all of the BGT1 (Fig. 10 Q ).
3 l4 H6 Z8 U" Z& \+ B0 V
Fig. 10. Colocalization ( right ) of the distribution of F-actin( left ) and BGT1 protein ( center ) in hypertonicMDCK cells is disrupted by CD or Triton extraction. Iso, control cellsin isotonic medium; TX, extracted with 1% Triton; Hyp, 24 h inhypertonic medium; CD, treated with 1 µM cytochalasin D. Bar = 20 µm, note the smaller scale in M - R. Confocalsettings were the same for all samples.: Z1 I" S/ P3 |2 r" A

Antibody staining of BGT1 in plasma membranes was heterogeneousbetween different cells in the same field of view (Fig. 10 H ) making quantitation difficult. Overall, there appeared to be no clear-cut differences in the intensity of the BGT1 fluorescence in theplasma membranes shown in Fig. 10 H (hypertonic) and Fig. 10 N (hypertonic cytochalasin D). This is at leastconsistent with the biotinylation data that provide a much betterquantitative comparison of surface BGT1 (Fig. 8 ).1 @+ b" B8 ]9 I* n' u' B

DISCUSSION0 z  A2 B7 d5 X7 z% o1 w$ U2 j

Cytochalasin D and latrunculin A disrupt the actin cytoskeletonthrough different mechanisms ( 22, 25 ). However, treatment of MDCK cells with either drug during hypertonic stress for 24 hproduced the same response, the stimulation of BGT1 transport activity.The low level of BGT1 transport under isotonic conditions remainedunchanged following drug treatment. The action of cytochalasin D on thecytoskeleton was established within 1 h and was maintained for upto 24 h. When included in the uptake medium, none of the testeddrugs had an acute effect on BGT1 transport under hypertonic conditions, which rules out a possible direct action on thetransporter. The response to cytochalasin D appeared to be selectivefor BGT1 because cytochalasin D did not stimulate the upregulation ofNa -dependent alanine uptake after 5 h. Alanine entersMDCK cells principally via system A (ATA2), which is also upregulatedby 5-6 h of hypertonic stress in both epithelial ( 9 )and endothelial cells ( 1, 2 ). The unchanged alaninetransport following cytochalasin D treatment also provides indirectevidence that the drug does not interfere with maintenance of thetransmembrane Na gradient. In summary, the stimulation ofBGT1 transport by cytochalasin D and latrunculin A is most likely dueto disruption of the cytoskeleton.4 ]6 V. R' E% K

Several studies suggest that the organization of the actin cytoskeletonis important for cell volume regulation and is altered by changes incell size ( 15 ). For example, the cytoskeleton of Chinesehamster ovary cells showed marked reorganization within 10 min afterincreasing medium osmolality to 500 mosmol/kgH 2 O ( 30 ). Because immunochemical staining revealed thepresence of intracellular BGT1 proteins under isotonic conditions( 3 ), the question arose that some of these preexistingproteins might gain access to the plasma membrane during the initialexposure to hypertonicity and while the cytoskeleton was beingreorganized. This initial response to hypertonicity may be independentof the presence of depolymerizing drugs. This possibility wasinvestigated by using the F-actin-stabilizing drug phalloidin. Thepresence of phalloidin did not inhibit the normal increase in BGT1transport activity in response to 24-h hypertonic stress (Fig. 4 ),suggesting that the increase in BGT1 transport due to 24-hhypertonicity is unlikely to be caused simply by the initial disruptionof the actin cytoskeleton. Induction of BGT1 transport duringhypertonicity occurs principally through activation of BGT1 genetranscription by the transcription factor tonicity-responsive enhancerbinding protein, as reported previously ( 14, 32 ).
7 B, \3 k% u) j( W
The mechanism by which the cytoskeleton is involved in upregulation ofBGT1 transport has not been established. It does not appear to berequired for downregulation of BGT1 transport when cells were returnedto isotonic medium after 24 h of hypertonicity (Fig. 7 ). There areat least two possible mechanisms to consider. First, the actincytoskeleton may act as a barrier or fence that restricts the access ofBGT1 proteins to the plasma membrane, as suggested for aquaporin-2( 12 ). Disruption of the actin cytoskeleton causes breaksin the fence, which may allow more BGT1 proteins to be inserted intothe plasma membrane during upregulation by hypertonic stress. Thismechanism seems very unlikely because it would lead to an increasedabundance of BGT1 in the plasma membrane. The biotinylation studiesrevealed no major change in surface expression of BGT1 proteinfollowing exposure to cytochalasin D. Furthermore, the upregulation ofNa -alanine cotransport was not stimulated by cytochalasinD, and it is difficult to understand why removal of a potential barrier would only affect upregulation of BGT1.1 @2 |3 A" }" w" F6 }0 A

The second, and more likely, explanation is that the actin cytoskeletonmay serve to inhibit BGT1 transport activity through directinteractions with the transporter protein. In MDCK cells, for example,the Na -K -ATPase in the basolateral membraneis linked to actin through ankyrin and spectrin ( 10, 19 ).The apical Na /H exchanger isoform NHE3 alsointeracts with the actin cytoskeleton ( 20 ). Disruption ofthe actin cytoskeleton by cytochalasin D may allow BGT1 proteinsalready in the plasma membrane to function at an increased rate.Although BGT1 protein colocalized with cortical F-actin in MDCK cellsafter hypertonic stress, the apparent association was completelydisrupted by treatment with either cytochalasin D or Triton X-100 (Fig. 10 ). This suggests that any direct interactions are weak ( 8, 38 ) or that only a small fraction of BGT1 interacts with F-actinthat cannot be detected by immunofluorescence. Actin and actin-bindingproteins may also regulate BGT1 by changing membrane fluidity or byinfluencing signaling pathways ( 19 ). Additional work willbe required to identify the specific mechanism of cytochalasin D actionon BGT1 upregulation and to determine if actin-BGT1 interactions have asignificant role in normal adaptation to hypertonic stress.! x( |! Q2 r2 t) z/ L
% }" w, c8 ^3 Z2 L6 c+ n
ACKNOWLEDGEMENTS) Q% |4 ~  e- Z4 Z; x4 j; ?; f
3 @. X. s% Z3 a! D* r
We are indebted to Dr. F. Pavalko and Dr. S. Norvell for continuedadvice and assistance." F& s2 V) Z, d) _9 B! I; o
      【参考文献】) G& ^- R1 ^5 E3 ]* J, b
1. Alfieri, RR,Cavazzoni A,Petronini PG,Bonelli MA,Caccamo AE,Borghetti AF,andWheeler KP. Compatible osmolytes modulate the response of porcine endothelial cells to hypertonicity and protect them from apoptosis. J Physiol 540:499-508,2002 .2 F# u1 S! M) I8 N( D3 f/ u

7 W/ ^% ?: q6 B& S9 y0 c) R

2. Alfieri, RR,Petronini PG,Bonelli MA,Caccamo AE,Cavazzoni A,Borghetti AF,andWheeler KP. Osmotic regulation of ATA2 mRNA expression and amino acid transport System A activity. Biochem Biophys Res Commun 283:174-178,2001  .2 q: \5 a6 u# N2 }5 N$ D
9 p* v) n: X7 ~

' H/ X$ t6 s7 i) p
3. Basham, JC,Chabrerie A,andKempson SA. Hypertonic activation of the renal betaine/GABA transporter is microtubule dependent. Kidney Int 59:2182-2191,2001  .% K2 H0 v7 l/ E# r' Q" M1 Z# R
2 X6 m5 b* z3 x- |: K

4. Berdiev, BK,Prat AG,Cantiello HF,Ausiello DA,Fuller CM,Jovov B,Benos DJ,andIsmailov, II Regulation of epithelial sodium channels by short actin filaments. J Biol Chem 271:17704-17710,1996 .
' |/ p' m/ F- i3 ^, o, o/ z# F

4 h* j# z# b' J: c, q3 c- i
5. Bitoun, M,andTappaz M. Gene expression of taurine transporter and taurine biosynthetic enzymes in brain of rats with acute or chronic hyperosmotic plasma. A comparative study. Brain Res Mol Brain Res 77:10-18,2000 .( g! f% z1 V6 x& c# R: J
& l5 q3 j/ |* ^! j& Q3 ?/ _
0 _; O1 J" ~  p& M
0 H- V. m8 a9 U
6. Burg, MB,Kwon HM,andKultz D. Osmotic regulation of gene expression. FASEB J 10:1598-1606,1996 .

7. Cantiello, HF. Role of the actin cytoskeleton on epithelial Na channel regulation. Kidney Int 48:907-984,1996.
1 B5 c! H3 `  S, j& d, C
3 E5 n+ b% x- ?- k, B  u/ T- U
% J( j/ E) ?4 [8 s. e, d
8. Cao, F,andBurke JM. Protein insolubility and late-stage morphogenesis in long-term postconfluent cultures of MDCK epithelial cells. Biochem Biophys Res Commun 234:719-728,1997  .6 j" @* w$ y7 d; h& @) k( l! B

* k# t8 e8 Y7 Q7 r" F

9. Chen, JG,Coe M,McAteer JA,andKempson SA. Hypertonic activation and recovery of system A amino acid transport in renal MDCK cells. Am J Physiol Renal Fluid Electrolyte Physiol 270:F419-F424,1996 .# c( Z& [6 z! ]4 y8 y* Q

! x' x6 \, N$ R$ T; ~9 N

10. Devarajan, P,Scaramuzzino DA,andMorrow JS. Ankyrin bins to two distinct cytoplasmic domains of Na,K/ATPase a subunit. Proc Natl Acad Sci USA 88:6971-6975,1994.
& J6 I+ W! _2 a& e( k
0 C# B. k) ?* \$ K

11. Fischer, H,Illek B,andMachen TE. Actin filament disrupter cytochalasin D activates the recombinant cystic fibrosis transmembrane conductance regualtor Cl channel in mouse 3T3 fibroblasts. J Physiol 489:745-754,1995 .0 |* ~+ D$ [; }) w

12. Franki, N,Ding G,Gao Y,andHays RM. The effect of cytochalasin D on the actin cytoskeleton of the toad bladder epithelial cell. Am J Physiol Cell Physiol 263:C995-C1000,1992 .% c% T2 ]5 f- C$ i& n$ q. t4 X

: O' X) i* G7 p" p4 O8 q4 P+ R9 e1 }
13. Furukawa, K,andMattson MP. Cytochalasins protect hippocampal neurons against amyloid -peptide toxicity: evidence that actin depolymerization suppresses Ca 2  influx. J Neurochem 65:1061-1068,1995  .

# n/ ]) V8 ]( F

14. Handler, JS,andKwon HM. Transcriptional regulation by changes in tonicity. Kidney Int 60:408-411,2001  .0 z: E$ z' q% N3 w0 Y

0 S% }4 j; x7 A5 l1 v4 `
2 Q* t1 |' N# p6 U: F3 f/ K4 e
15. Henson, JH. Relationships between the actin cytoskeleton and cell volume regulation. Microsc Res Tech 47:155-162,1999  .! S, X% B+ c6 }2 h' Z* b3 S

; r# S6 x1 Y) k! j7 r
16. Janmey, P. The cytoskeleton and cell signaling: component localization and mechanical coupling. Physiol Rev 78:763-781,1998 .. q$ @+ t3 t8 V) s, k5 H
6 P4 |" E  l0 L
# M6 Q9 Y8 {1 w. ?% E) F9 w

17. Jin, YP,Singh RP,Du ZY,Rajasekaran AK,Rozengurt E,andReed EF. Ligation of HLA class I molecules on endothelial cells induces phosphorylation of Src, paxillin, and focal adhesion kinase in an actin-dependent manner. J Immunol 168:5415-5423,2002 .6 |. z5 a2 W( g  B* _
' p& P5 _' n/ T, h" j' E4 ]0 Y+ h
& [: [$ K* B5 O8 f5 D
) D' A' [1 j! P: T& `4 c3 Z9 }1 w
18. Kempson, SA. Differential activation of system A and betaine/GABA transport in MDCK cell membranes by hypertonic stress. Biochim Biophys Acta 1372:117-123,1998 .3 K6 [# I/ d1 b, v9 V/ P7 G6 x2 t2 x

( d. ?1 b& l4 t+ r5 W8 B

19. Khurana, S. Role of actin cytoskeleton in regulation of ion transport: examples from epithelial cells. J Membr Biol 178:73-87,2000  .
2 `2 R% C' M/ O0 Y7 Y& M( w4 T2 T; {$ @
0 v: Z5 A7 x# K: q, O  E

20. Kurashima, K,D'Souza S,Szaszi K,Ramjeesingh R,Orlowski J,andGrinstein S. The apical Na /H exchanger isoform NHE3 is regulated by the actin cytoskeleton. J Biol Chem 274:29843-29849,1999 .
( n# V' N3 z: [& i/ b  @( W& S
" _/ X. G* C, G3 v* \% H
, X& g1 t! s7 V5 Z0 }
21. Matskevitch, I,Wagner CA,Stegen C,Broer S,Noll B,Risler T,Kwon HM,Handler JS,Waldegger S,Busch A,andLang F. Functional characterization of the betaine/GABA transporter BGT-1 expressed in Xenopus oocytes. J Biol Chem 274:16709-16716,1999 .

/ \, `7 d- f; k0 X
3 x5 e. ?; Y( d/ I+ x: |  k- T
22. Matthews, JB,Smith JA,andHrnjez BJ. Effects of F-actin stabilization or disassembly on epithelial Cl secretion and Na-K-2Cl cotransport. Am J Physiol Cell Physiol 272:C254-C262,1997 .

  Q& n9 N9 [" a, p" P$ x

23. Matthews, JB,Smith JA,Mun EC,andSicklick JK. Osmotic regulation of intestinal epithelial Na -K -Cl cotransport: role of Cl and F-actin. Am J Physiol Cell Physiol 274:C697-C706,1998 .- a2 E5 D5 {) O: c

! v( Y) S1 j! u) J
24. Mills, JW,andMandel LJ. Cytoskeletal regulation of membrane transport events. FASEB J 8:1161-1165,1994 .4 S* `2 c% _  G. N8 B

) K! u% D" n. }6 T6 l# Z; K
25. Morton, WM,Ayscough KR,andMcLaughlin PJ. Latrunculin alters the actin-monomer subunit interface to prevent polymerization. Nat Cell Biol 2:376-378,2000  .
5 I% p  N- R3 V- d/ J3 ^! u* E
  O$ D4 ~" s. b' m
/ `& \5 M+ e, m
26. Nakanishi, T,Turner RJ,andBurg MB. Osmoregulation of betaine transport in mammalian renal medullary cells. Am J Physiol Renal Fluid Electrolyte Physiol 258:F1061-F1067,1990 .
6 e. @- a& K0 J  C3 L4 A6 E" g
. l. U  F  |; M# q
6 Y, h1 C& ~# c5 E& |/ S3 E0 e* W
27. Perego, C,Bulbarelli A,Longhi R,Caimi M,Villa A,Caplan MJ,andPietrini G. Sorting of two polytopic proteins, the GABA and betaine transporters, in polarized epithelial cells. J Biol Chem 272:6584-6592,1997 .5 ], h9 I9 O6 C0 ^" i
$ G3 \/ g* h8 f2 U

28. Sawyer, SJ,Norvell SM,Ponik SM,andPavalko FM. Regulation of PGE 2 and PGI 2 release from human umbilical vein endothelial cells by actin cytoskeleton. Am J Physiol Cell Physiol 281:C1038-C1045,2001 .
6 B/ \7 m3 u/ p2 l# o. r/ L
9 H0 T2 Z$ k' t

29. Schuttert, JB,Fiedler GM,Grupp C,Blaschke S,andGrunewald RW. Sorbitol transport in rat renal inner medullary interstitial cells. Kidney Int 61:1407-1415,2002  .
+ Q* E+ f. W7 e4 @
' S1 c) _! |& Q9 L" z1 T# A  n

30. Szaszi, K,Grinstein S,Orlowski J,andKapus A. Regulation of the epithelial Na /H exchanger isoform by the cytoskeleton. Cell Physiol Biochem 10:265-272,2000  .
5 b1 J, g' C2 d6 l9 d2 m9 b. X

  x1 x9 Z7 q; \
31. Tsakiridis, T,Tong P,Matthews B,Tsiani E,Bilan PJ,Klip A,andDowney GP. Role of the actin cytoskeleton in insulin action. Microsc Res Tech 47:79-92,1999  .
) X6 [) L3 l3 ?

32. Uchida, S,Yamauchi A,Preston AS,Kwon HM,andHandler JS. Medium tonicity regulates expression of the Na and Cl dependent betaine transporter in MDCK cells by increasing transcription of the transporter gene. J Clin Invest 91:1604-1607,1993  .
1 M! l. r' }3 j- p" t
" k+ O, N; k& w) J8 y
/ l) j. j( v% `
33. Undrovinas, AI,Shander GF,andMakielski JC. Cytoskeleton modulates gating of voltage-dependent sodium channel in heart. Am J Physiol Heart Circ Physiol 268:H203-H214,1995.+ x, w1 y7 T$ n* ?
0 u$ _  A( x$ f; o$ t: T* ^0 ^0 h
0 V+ }( {. M; s$ @; p% j, i  `
4 W2 E: S+ T0 U  C2 \
34. Walss-Bass, C,Prassad V,Kreisberg KI,andLuduena RF. Interaction of the IV -tubulin isotype with actin stress fibers in cultured rat kidney mesangial cells. Cell Motil Cytoskeleton 49:200-207,2001  .+ K/ P# O; p7 F! b& q

35. Watson, AJM,Levine S,Donowitz M,andMontrose MH. Serum regulates Na /H exchange in Caco-2 cells by a mechanism which is dependent on F-actin. J Biol Chem 267:956-962,1992 .

2 R* ]9 E" u& w

36. Yamauchi, A,Kwon HM,Uchida S,Preston AS,andHandler J. Myo -inositol and betaine transporters regulated by tonicity are basolateral in MDCK cells. Am J Physiol Renal Fluid Electrolyte Physiol 261:F197-F202,1991 .

37. Yang, X,Amemiya M,Peng Y,Moe OW,Preisig PA,andAlpern RJ. Acid incubation causes exocytic insertion of NHE3 in OKP cells. Am J Physiol Cell Physiol 279:C410-C419,2000 .9 u3 e; s3 ~  C
% p) n' H2 \4 M
" E( B, N6 R9 l
5 t; l+ p! D' W( Y+ v2 `% j
38. Yuan, H,Takeuchi E,andSalant DJ. Podocyte slit-diaphragm protein nephrin is linked to the actin cytoskeleton. Am J Physiol Renal Physiol 282:F585-F591,2002 .

收藏0 分享0 顶0 踩0