An increase in intracellular calcium concentration that is induced by basolatera - 《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: 337787\|回复: 234	go An increase in intracellular calcium concentration that is induced by basolatera [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 13:46 |只看该作者 |倒序浏览 |打印

作者：Patrice Bouyer, Yuehan Zhou,  Walter F. Boron作者单位：Department of Cellular and Molecular Physiology, Yale University Schoolof Medicine, New Haven, Connecticut 06520
   3 f9 e3 t+ [0 n# a7 O
   ' c+ ~% M( l  h$ H4 u5 [3 L6 M





         : u% {* z- k& D) T. [9 m


      ( w- C& O# }: J" O/ c' _
      4 V, x  p4 w% h7 s8 F
      【摘要】
   Working with isolated perfused S2 proximal tubules, we asked whether thebasolateral CO 2 sensor acts, in part, by raising intracellularCa 2 concentration([Ca 2 ] i ), monitored with the dye fura 2 (orfura-PE3). In paired experiments, adding 5% CO 2 /22 mM (constant pH 7.40) to the bath(basolateral) solution caused [Ca 2 ] i toincrease from 57 ± 3 to 97 ± 9nM( n = 8, P the same maneuver in the lumen had no effect. Intracellular pH(pH i ), measured with the dye BCECF, fell by 0.54 ± 0.08( n = 14) when we added to thelumen. In 14 tubules in which we added to thebath, pH i fell by 0.55 ± 0.11 in 9 with a high initial pH i, but rose by 0.28 ± 0.07 in the other 5 with a low initial pH i. Thus it cannot be a pH i change thattriggers the [Ca 2 ] i increase. Introducing tothe bath an out-of-equilibrium (OOE) solution containing 20% CO 2 /no caused[Ca 2 ] i to rise by 62 ± 17 nM( n = 10), whereas an OOE solution containing 0% CO 2 /22 mM causedonly a trivial increase. Removing Ca 2 from the lumenand bath, or adding 10 µM nifedipine (L- and T-typeCa 2 -channel blocker) or 2 µM thapsigargin[sarco-(endo) plasmic reticulum Ca 2 -ATPase inhibitor]or 4 µM rotenone (mitochondrial inhibitor) to the lumen and bath, failed toreduce the CO 2 -induced increase in[Ca 2 ] i. Adding 10 mM caffeine(ryanodine-receptor agonist) had no effect on[Ca 2 ] i. Thus basolateral CO 2,presumably via a basolateral sensor, triggers the release ofCa 2 from a nonconventional intracellular pool. " [0 X) C: T  S9 z% e9 C7 _: H
      【关键词】 intracellular pH carbon dioxide outofequilibrium solutions fura ions transport kidney
   A MAJOR ROLE OF THE KIDNEY is to maintain the pH of theextracellular fluid within normal limits. The proximal tubule activelyparticipates in this activity by reabsorbing 80% of the NaHCO 3 filtered at the glomeruli and also by generating "new" to neutralize non-volatile acidsgenerated by metabolism. Bicarbonate reabsorption ( J HCO3 )occurs as the apical Na/H exchanger and H pump secreteH  , and as this acid titrates luminal to CO 2 andH 2 O under the influence of apical carbonic anhydrase( 2, 20 ). The newly formedCO 2 and H 2 O then diffuse into the cells, where solublecarbonic anhydrase regenerates H and in the cytosol. Finally, theaforementioned H extruders recycle H to the lumen,while the basolateral Na-HCO 3 cotransporter moves to the blood( 8 ). The generation of new is similar to the reabsorption of except that the H secreted into the lumen titrates a buffer (e.g., ) other than, and the intracellularCO 2 and H 2 O derive from the blood rather than theluminal fluid.& C2 x' f& W8 \  a

The rate of H secretion by the proximal tubule, which is nearly identical to J HCO3, is under the control of severalhormones. For instance, angiotensin II( 36, 71 ) and nitric oxide( 70 ) increase J HCO3, whereas parathyroid hormone (PTH) has the opposite effect ( 21, 41 ). Another potent regulatorof J HCO3 is the acid-base status of blood. For example,respiratory acidosis {i.e., an increase in blood P CO2 that causes adecrease in blood pH and small increase in blood concentration( )} raises J HCO3 ( 1, 11, 22 ). To determine whether itis a change in P CO2, pH, or that is responsible for theincrease in J HCO3 during respiratory acidosis, thelaboratory developed a technique for making out-of-equilibrium (OOE) solutions. Using this approach it is possible to generate solutions havingphysiological levels of CO 2 concentration ([CO 2 ]) and pHbut virtually no (i.e., a"pure CO 2 " solution), or solutions having physiologicallevels of and pH but virtuallyno CO 2 (i.e., a "pure " solution).- Y) B, S  C1 Y7 h8 g' K6 U! [

OOE solutions were first used to study K-HCO 3 cotransport in squid giant axons ( 75 ). Morerecently, our laboratory adapted this technique to mammalian cells and foundthat removing from the basolateralor "bath" solution (pure CO 2 ) caused J HCO3 to increase, whereas removing CO 2 fromthe basolateral solution (pure )had the opposite effect ( 76 ).In other experiments, our laboratory used the OOE approach to vary basolateral[CO 2 ],, and pH one ata time, while holding the other two parameters constant. The most surprisingresult was that J HCO3 was totally insensitive to widechanges in basolateral pH, even though these changes in basolateral pH wereassociated with rather wide changes in intracellular pH (pH i ).Nevertheless, J HCO3 increased markedly in response toincreases in basolateral [CO 2 ]( 77 ). These results led to thehypothesis that renal proximal tubule cells have a mechanism at or near thebasolateral membrane for sensing CO 2 independently of pH. Thishypothesis is consistent with earlier work with equilibrated solutions that showed that adding to thebath, but not to the lumen, causes steady-state pH i to rise inproximal tubule cells ( 46 ) andstimulates luminal acid extruders( 17, 18 ).

In the present study, we investigated one of the potential intracellular signaling pathways of the basolateral CO 2 sensor by monitoring intracellular Ca 2 concentration([Ca 2 ] i ). Intracellular Ca 2 is a common second messenger for numerous stimuli( 19 ). For example, a rise in[Ca 2 ] i is a key step in the response of thechemosensitive cells in the carotid body to hypoxia, metabolic acidosis,respiratory acidosis, or isohydric hypercapnia ( 13 ). Based on these findings,we felt that a rise in [Ca 2 ] i was a goodcandidate as a signaling pathway for the CO 2 sensor. Some authorsworking on proximal tubule cells have reported that increases in[Ca 2 ] i raise J HCO3 ( 38 ), whereas others havereported that increases in [Ca 2 ] i lower J HCO3 ( 16 ). Here, using aCa 2 -sensitive fluorescent dye, we found that we couldtrigger a significant increase in [Ca 2 ] i byintroducing equilibrated to thebasolateral (but not luminal) side of the tubule, or by introducingbasolateral pure CO 2 (but not pure ). Also, we found that basolateralCO 2 does not increase [Ca 2 ] i bylowering pH i and that the source of the Ca 2 is a thapsigargin (Tg)-insensitive intracellular store. Our results are thusconsistent with the hypothesis that an increase in[Ca 2 ] i might be involved in the proximaltubule cell's response to basolateral CO 2.* N) W" z+ `; \. y
0 O5 s! n3 Y  s) `; ]' O9 a8 }
METHODS5 x3 f1 l; m# z2 t7 f* n
& @% h1 B! H$ U9 q6 L2 |
Biological Preparation* G9 B  A% d; r

All the experiments were carried out in "pathogen-free" female rabbits (New Zealand White, Elite, Covance, Denver, PA) weighing 1.4-2.0kg. The methods for preparing the animals, harvesting the kidneys, andperfusing the tubules were similar to those originally described by Burg etal. ( 14 ) and subsequently modified in our laboratory( 47, 76 ). The Yale Animal Care andUse Committee approved all the procedures. Briefly, an animal was euthanizedby intravenous injection of pentobarbital sodium; an incision of the abdominalwall was performed to expose the left kidney, which was rapidly removed. Thekidney was then cut into coronal slices and kept in cold (4°C) modifiedHanks' solution ( solution 1 in Table 1 ). The microdissection of the slice was carried out in the same solution under a dissecting microscope, using a pair of fine forceps to grasp a portion of a medullary rayand gently tear it from the rest of the slice, starting from the inner medullaand proceeding toward the cortex. Our initial landmark was the junctionbetween the thin descending limb of Henle's loop and the S3 segment (i.e., distal portion of the proximal straight tubule). We isolated a portion of theS2 segment that consisted of the distal-most 600-800 µm of theproximal convoluted tubules plus 200-300 µm of the proximal-most partof the proximal straight tubule. After transferring the tubule to a chamber (adapted for rapid mixing of OOE solutions; see below), we perfused thedistal-most 400-500 µm of the proximal convoluted tubule. Tubuleswere perfused and bathed at 37°C.) O( s9 |0 Y) C+ S

Table 1. Physiological solutions+ h1 V4 l$ I& C5 x& ~: X8 w

Solutions

The compositions of the solutions are given in Table 1. The HEPES-bufferedsolution ( solution 2 ), the equilibrated solution ( solution 3 ), the pure CO 2 OOE solution ( solution 4 ), and the pure OOE solution ( solution 5 )were adjusted to pH 7.40 at 37°C. The osmolalities, measured using avapor-pressure osmometer (model 5100C, Wescor, Logan UT), were adjusted to 300± 3 mosM. The method for generating OOE solutions was the same asoriginally described ( 75 ), asadapted for kidney tubules( 76 ). Briefly, we generated OOE solutions by exploiting the slow interconversion between CO 2 and to generate 20% pureCO 2 (i.e., virtually no ) and rapidly mixed solution 4 / part A and solution 4 / part B ( Table 1 ), each contained in a140-ml plastic syringe (140 ml, Monoject, Sherwood Medical Industries,Ballymoney, UK) driven by the same syringe pump (model 55-2222, HarvardApparatus, South Holliston, MA). The output of the syringe was connected to anarray of five-way valves (Eagle P/N E4-1PP-00-000, ClippardInstrument Laboratory, Cincinnati, OH) with Tygon tubing( in. ID, Norton Performance Plastics, Akron, OH), and then directed tostainless steel tubing surrounded by a water jacket to warm the solutionsufficiently so that the temperature in the chamber was 37°C. Shortlydownstream from the water jacket, the output of the stainless steel tubing wasconnected via Tygon tubing to a mixing "T," which in turn wasconnected to another length of Tygon tubing that was filled with nylon mesh topromote mixing. Finally, this Tygon tubing was connected to the chamber, whichconsisted of a straight canal that was 14-mm long x by 2.5-mm wide topromote a laminar flow. A comparable method was used to generate the pure OOE solution. All solutions flowedat 7 ml/min.

When using ionomycin for calibrating theCa 2 -sensitive dyes, or when using nigericin forcalibrating the pH-sensitive dye (see below), we introduced these agents intothe chamber via solution reservoirs, tubing, and an inlet port that werecompletely separate from those used for the physiological solutions. This precaution avoided contamination of the plumbing fixtures used for thephysiological solutions. After each experiment, we washed the chamberextensively with 70% ethanol in water to remove traces of ionomycin ornigericin ( 5 ).

For solutions containing 0.5 mM ATP (Sigma, St. Louis, MO), we increasedthe total concentration of CaCl 2 to 1.09 mM and the totalconcentration of MgSO 4 to 1.57 mM to compensate for the binding ofCa 2 and Mg 2 to ATP. Forsolutions containing 0.5 mM EGTA, we increased the total concentration ofMgSO 4 to 1.26 mM. For solutions containing 5 mM EGTA, the totalconcentration of MgCl 2 was increased to 1.82 mM to compensate forMg 2 binding to EGTA. We used BAD computer softwaredescribed by Brooks and Storey( 12 ) to compute the amount ofextra CaCl 2, MgSO 4, or MgCl 2 that we neededto add to maintain the free Ca 2 at 1 mM and the freeMg 2 at 1.2 mM.
" L0 T" d& K0 F5 B
Chemicals. 4-Bromo A-23187, ionomycin, and Tg, were obtained fromCalbiochem (Calbiochem-Novobiochem, La Jolla, CA). Rotenone was obtained fromICN (ICN Pharmaceuticals, Costa Mesa, CA). HEPES was obtained from USB (USB,Cleveland, OH). Nigericin, ATP, and the other chemicals in the physiologicalsolutions were obtained from Sigma.# h5 h7 e2 K0 n3 L

Fluorescence Measurements+ ~7 j: k6 C( f' H$ E
/ \- C/ r3 |- G, T& A  s9 r
Measurement of fluorescence ofCa 2 -sensitive dyes. Measurements of[Ca 2 ] i were performed by loading theisolated perfused tubule with 5 µM of either fura 2-AM (Molecular Probes, Eugene, OR) or fura-PE3-AM (TefLabs, Austin TX) in our HEPES-buffered solution( solution 2 in Table1 ) along with 0.5% (vol/vol) pluronic F-127 (Molecular Probes). Wedye-loaded the tubules at room temperature for 20-30 min for fura 2 and40-50 min for fura-PE3. We added the dye precursors as 2 mM stock solutions in DMSO and added the pluronic F-127 as a 20% wt/vol stock solutionin DMSO. Before the fluorescence recordings, we washed the tubule by flowing alarge volume of solution 2 through the chamber.
+ k! |- V6 v: @
In tubules loaded with fura 2, dye leakage led to a gradual loss offluorescence that often prevented us from performing 25 min), we used fura-PE3, which ismore resistant to dye leakage, has the same absorbance spectrum as fura 2( 69 ), and has been usedsuccessfully in proximal tubule cells by others( 53 ). However, we used fura 2in most of our experiments because fura-PE3 required a longer period of dyeloading, which reduced the number of experiments we could perform per rabbitand also increased our failure rate. Therefore, unless our experimental protocol required that we record [Ca 2 ] i fora lengthy period, we preferred fura 2 over fura-PE3.
8 U7 b5 h% b9 |
The microscope was an Olympus IX70 inverted microscope, equipped with a x 40 oil-immersion objective (1.35 numerical aperture, with a x 1.5magnification selector knob) and apparatus for epi-illumination. The lightsource was a 75-W xenon arc lamp. We generated light at two excitationwavelengths by using a filter wheel (Ludl Electronic Products, Hawthorne, NY)to alternate the placement of two filters, 340 ± 15 and 380 ± 15nm (Omega Optical, Brattleboro, VT), in the excitation light path. Appropriateneutral-density filters (Omega Optical) mounted on a second wheel were used toavoid overillumination of the specimen, which could cause photobleaching, andto equalize as nearly as possible the emitted fluorescent light intensities obtained while excitation occurred at the two wavelengths. The excitationlight was directed to the tubule via a 415-nm long-pass dichroic mirror (DM415, Omega Optical) and the aforementioned objective. The emitted light wascollected by the same objective and, via a band-pass filter (510 ± 40nm, Omega Optical), was directed to an intensified CCD camera (model 350F,Video Scope International, Dulles, VA).' I6 l* W6 [5 I( K
3 R. a6 |0 U& G, o4 A0 ^
The protocol for alternately exciting the tubule with wavelengths of light,and for subsequently acquiring the fluorescence images, was describedpreviously ( 76 ). Briefly, atypical data-acquisition cycle consisted of a 370-ms period ofillumination with 340-nm light, followed immediately by an identical periodwith 380-nm light. For each excitation wavelength, we averaged four successive video frames using an image-processing board (DT3155, Data Translation, Marlboro, MA) and thereby obtained the emitted light intensity for anexcitation of either 340 nm ( I 340 ) or 380 nm( I 380 ). This pair of excitations was repeated at intervalsranging from 2.5 to 8 s; between excitations cycles, a shutter on the filterwheel protected the tubule from the light. Software developed in ourlaboratory using the Optimas (Media Cybernetics, Silver Spring, MD) platformcontrolled data acquisition and analysis on an Intel-based computer runningWindows 98SE. We identified an area of interest that represented 30% ofthe tubule length. The sum of the I 340 values of thepixels in the area of interest, corrected for the background (see below), wasdivided by the sum of the corresponding background-subtracted I 380 values to yield the fluorescence excitation ratio( I 340 )/( I 380 ) or R 340/380,which strongly depends on [Ca 2 ] i but isrelatively insensitive to factors such as dye concentration. Because it wasour impression that sudden increases in the rate of dye loss were associatedwith sudden increases in( I 340 )/( I 380 ), we discardedexperiments in which I 340 and I 380 declined rapidly.
4 n" P+ N" m7 O8 c5 r( A
Intracellular calibration ofCa 2 -sensitive dyes. The generally accepted approach for converting R 340/380 values into[Ca 2 ] i values is that of Grynkiewicz et al.( 27 ), in which one determines R min (the minimum R 340/380 when[Ca 2  ] i 0) and R max (the maximum R 340/380 when [Ca 2  ] i ) foreach cell, and computes [Ca 2 ] i on theassumption that the dissociation of dye is the same inside the cell as it isin vitro+ P# l0 i. @, q
1 q; Y; V- ~1 w5 s+ W2 r, ]
( 1 )

where S f/b is the I 380 measured when[Ca 2 ] i 0 divided by I 380 when [Ca 2 ] i. R min is typically determined by exposing the cell to aCa 2 -free solution containing EGTA and aCa 2 ionophore. Similarly, R max is typicallydetermined by exposing the cell to a solution containing a high concentrationof Ca 2 plus the ionophore.
: v$ v$ J, b& @, C0 x  k  p7 }
Unfortunately, as reported by several groups, the above calibration approach is difficult to apply to isolated proximal tubules because ofproblems with dye leakage during the prolonged calibration procedure. Thusmost [Ca 2 ] i studies on proximal tubules, and the associated calibrations, have been done on collapsed tubules ( 9, 40, 74 ). Another group performedtheir physiological experiments in perfused tubules but obtained the valuesfor R min, R max, and S f/b byperforming calibrations on collapsed tubules( 4 ). We know only one study inwhich the authors calibrated a Ca 2 -sensitive dye (i.e.,fura-PE3) in a limited number of perfused proximal tubules( 53 ).) m; e: w5 S+ ?* z7 Z5 a

Despite various attempts to minimize dye loss and cell damage, we found itimpossible to perform physiological experiment and then routinely obtainR min, R max, and S f/b values on the same perfused tubule at 37°C. For example, although probenecid (aninhibitor of organic anion transporters) reduces the loss of fura 2 fromneurons ( 44 ), we did not findprobenecid (300-1,000 µM) useful in the proximal tubule. Similarly,neither lowering the ionomycin concentration to 1 µM, nor switching fromionomycin to 4-bromo A-23187 was helpful. Instead, we adopted the followingprocedure.& l9 c  s9 n0 t* d
# u7 j; Y1 `: ~6 I' H& D
First, we obtained a mean S f/b as well as mean,normalized values of R min and R max on a subset of 30tubules perfused at 37°C after we had performed physiological experiments on these tubules. At the end of the experiment, we switched successively tobath solutions containing 1 ) 0 mM Ca 2 plus 5mM EGTA and 5 µM of the Ca 2 ionophore ionomycin( solution 6 in Table1 ) 1, 2 ) 5 mM Ca 2 plus 5 µM ionomycin( solution 7 ) 2, and 3 ) 5 mM Mn 2 ( solution 8 ). This lastmaneuver allowed us to determine the autofluorescence of the tubule byquenching the fluorescence of the dye. We subtracted these quenched values of I 340 and I 380 from all respective I 340 and I 380 values in the experimentand used these background-subtracted values to compute R 340/380 values for each data point. Finally, we identified a segment of data at thebeginning of the experiment in which the R 340/380 values werestable with the HEPES-buffered solution ( solution 2 ) present in thelumen and bath, calculated the mean initial R 340/380 value, anddivided all R 340/380 values in the experiment by this mean initialR 340/380 value. The mean quotient during the calibration periodwith 0 mM Ca 2 was thus the normalized R min,and the mean quotient during the calibration period with 5 mMCa 2 was the normalized R max. In the 30tubules, R min was 0.63 ± 0.05, R max was 6.07 ± 0.65, and S f/b was 3.33 ± 0.51.; F: O% m9 T- r5 W5 o
% f$ r/ g( p4 v, n/ V; X- l# ~
Second, we used the above values of R min, R max, and S f/b to compute [Ca 2 ] i values in each of our experiments, including the 30 described above. In eachof these experiments, we normalized all R 340/380 values to the meaninitial R 340/380 value obtained with the HEPES-buffered solutionpresent in the lumen and bath (see above). We then used Eq. 1 tocompute [Ca 2 ] i values at each time point,employing the aforementioned mean value of S f/b, the meannormalized values of R min and R max, and a K d for fura 2 of 224 nM( 27 ) or a K d for fura-PE3 of 290 nM ( 69 ).
3 L5 q/ \8 p1 j* V
Measurement of pH i. The ratiometric optical techniqueused to measure pH i was similar to that described above for[Ca 2 ] i. Briefly, isolated microperfusedtubules were loaded with the acetoxymethyl ester of the pH-sensitive dyeBCECF-AM (Molecular Probes) at 10 µM final concentration, dissolved in the HEPES-buffered Ringer ( solution 2 in Table 1 ). The excitation band-pass filters were centered at 440 ± 5 and 495 ± 5 nm (OmegaOptical). We also used a 510-nm long-pass dichroic mirror and a 530-nmlong-pass emission filter (Omega Optical). We identified areas of interest asoutlined above for the [Ca 2 ] i measurements,subtracted the background ( 0.3% of the signal in BCECF-loaded tubules)from the I 440 and I 490 values as described previously ( 76 ), andcomputed the time course of I 490 / I 440.We discarded experiments in which the rate constant for the decrease in the I 440 signal (- k 440 ) exceeded 0.05 min - 1 ( 6 ).. C: f, |& J2 g. q0 C

We computed pH i values from the I 490 / I 440 ratios using a variation ofthe high-K /nigericin technique( 66 ), in which one performs aone-point calibration at pH i 7.00( 10 ). At the end of each experiment, we drove pH i toward 7.00 by introducing a pH-7.00 high-K /nigericin solution( 54 ) into the bath. Wenormalized the I 490 / I 440 ratios of theentire experiment by dividing them by the I 490 / I 440 ratio obtained atpH i 7.00 and then used the following equation( 10 ) to calculatepH i
* Z9 z, D6 n7 G, j8 k
( 2 )
8 n% d7 E" `! I9 q
From a separate series of 64 fluorescence measurements in a total of 10tubules, we obtained values for p K and b by using anigericin-containing solution to alter pH i, as described elsewhere( 54 ). We used a nonlinearleast-squares method to fit the parameters in the above equation, which forcesthe best-fit curve to pass through unity at pH i = 7.00, to the calibration data. The best-fit values were p K = 7.24 ± (SD)0.01 and b = 1.79 ± (SD)0.02.
1 A4 j' Q. m. \1 U& [3 d
Data Analysis and Statistics+ P3 k( K+ K5 `# F# |3 A
! p5 w0 A9 {, {& g* {
Except for the curve fitting discussed above, all the values are means± SE, with n being the number of observations. The statisticalsignificance of the data was assessed by two-tailed Student's t -testson paired or unpaired data as indicated, using the Analysis Toolpack ofMicrosoft Excel. Mean steady-state [Ca 2 ] i values were obtained by averaging [Ca 2 ] i values over a period of 1 min.
* @( m, \$ h! A) j
RESULTS
, o9 m/ C9 c( z. ~9 G8 W7 a# G; p4 N3 D
In 30 tubules in which we obtained individual R min,R max, and S f/b values in each tubule and usedthese values to compute [Ca 2 ] i in eachtubule, the mean, steady-state [Ca 2 ] i was 63± 10 nM. As discussed in METHODS, we also used the data fromthe above 30 tubules to compute mean values for R min, R max, and S f/b and then used these meancalibration parameters to compute initial[Ca 2 ] i values in a total of 131 tubules, including the 30 noted above. The average steady-state[Ca 2 ] i for these 131 tubules was 61 ±1 nM ( n = 131), which is not significantly different from the meanvalue for the 30 tubules ( P = 0.8, unpaired t -test).

Effect on [Ca 2 ] i of Applying Unilaterally! a0 C- ^( Y1 M& P+ k

Our first approach in studying the effect of solutions on [Ca 2 ] i was to measure[Ca 2 ] i while exposing either the luminal orthe basolateral side of the tubule, but not both, to 5% CO 2 /22 mM at a fixed extracellular pH of7.40. A typical recording is shown in Fig.1 A. At the beginning of the experiment, we bilaterallyperfused the tubule with a solution buffered to pH 7.40 with HEPES( solution 2 ). After we switched the luminal solution from onebuffered with HEPES to one buffered with 5%CO 2 /22 mM ( solution 3 ),[Ca 2 ] i slowly drifted upward by a smallamount ( segment ab ). On the other hand, after we removed the fromthe lumen ( bc ) and then introduced the -buffered solution to the bath, [Ca 2 ] i increased to a new and substantially higher steady-state value ( cd ). Switching back to the bath solution caused [Ca 2 ] i to return closeto baseline ( de ). Figure1 B shows that we obtained the same result when we madethe luminal and basolateral solution changes in the opposite order.

Fig. 1. Effect on intracellular Ca 2 concentration([Ca 2 ] i ) of basolateral vs. luminal exposureto 5% CO 2 /22 mM. A and B: representative recordings of[Ca 2 ] i. An S2 proximal tubule loaded withthe Ca 2 -sensitive dye fura 2 was initially perfused inlumen and bath with a HEPES-buffered, -freesolution ( solution 2, Table1 ). At the indicated times, the lumen and the bath solution wereswitched to a similar solution buffered with ( solution 3 ). A and B differ only in the order ofthe solution changes. C : data summary. Filled bar, the mean change in[Ca 2 ] i ( [Ca 2 i ]) caused by the luminal switchfrom the HEPES buffer to;stippled bar, comparable [Ca 2 ] i forthe corresponding basolateral solution change. The statistical analysessummarized in the figure are the results of paired 2-tailed Student's t -tests. The mean initial [Ca 2 ] i values with the HEPES-buffered solutions in the lumen and bath were 61± 4 nM just before the luminal switch to the -buffered solutions, and 57 ± 3 nM just before the correspondingbasolateral switch; the difference between these values is not statisticallysignificant ( P = 0.5, paired 2-tailed Student's t -tests).% ~; \, Y) n! M: Y- T' _

The histogram in Fig.1 C represents the mean paired changes in[Ca 2 ] i ( [Ca 2 ] i ) elicited in eight tubules byswitching the solution in either the lumen (filled bar, corresponding tosegment ab in Fig.1 A and cd in Fig. 1 B ) or the bath (stippled bar, corresponding to segment cd in Fig. 1 A and ab in Fig.1 B ) from HEPES-buffer to 5% CO 2 /22 mM. [Ca 2 ] i was not statisticallysignificant when we applied to thelumen ( P = 0.8) but was significant when we applied to thebath ( P

Effect on pH i of Applying Unilaterally

To test the hypothesis that the increase in[Ca 2 ] i was caused by a change inpH i, we repeated the above protocol while measuring pH i in a total of 14 different tubules. Because we included lactate in our luminalsolutions to mimic the conditions in other parallel experiments in ourlaboratory, we anticipated that the tubules would have a high initialpH i. Previous work has shown that adding lactate to the lumen ofthe salamander proximal tubule, or adding acetate to the lumen of the rabbit S3 segment, raises pH i by 0.2 due to the coupled apical entry of Na and monocarboxylate followed by the coupled exit of H and lactate (or lactate/OH exchange) across the basolateral membrane ( 45, 59 ). Indeed, in 9 of the 14tubules, the initial pH i was relatively high (averaging 7.54± 0.08). However, for unknown reasons, in the other five tubules, theinitial pH i in HEPES was lower (averaging 7.23 ±0.07). 3 Regardless ofwhether the initial pH i in HEPES was high or low, introducing 5%CO 2 /22 mM to the lumencaused a sustained decrease inpH i. 4 Onthe other hand, the initial pH i in HEPES had a major impact on thepH i response when we added to thebath. In the nine tubules with a high initial pH i, introducing 5%CO 2 /22 mM to the bathcaused a sustained acidification, whereas in the five other tubules with alower initial pH i, introducing 5% CO 2 /22 mM induced an alkalinization. Theresults are summarized in the Table2. As noted in the DISCUSSION, the divergent responseto the addition of basolateral isconsistent with observations made in other preparations.( y  q$ c0 F+ y2 E7 x% d3 ~

Table 2. Effect of 5% CO 2 /22 mM onintracellular pH i+ m" q% T: M2 p6 E! `
0 D" s1 K! H' q- N
Thus in all of the experiments in which we monitored pH i,introducing luminal causedan acidification; in all experiments in which we monitored[Ca 2 ] i, introducing luminal had noeffect. The relationship between pH i and [Ca 2 ] i was just the opposite intwo-thirds of the experiments in which we added to thebath. In 9 of 14 tubules in which we monitored pH i, introducingbasolateral causedan acidification, just as if we had added to thelumen. However, in all experiments in which we monitored[Ca 2 ] i, introducing basolateral causedan increase in [Ca 2 ] i. Therefore, a changein pH i cannot be the cause of the[Ca 2 ] i increase elicited by basolateral.9 g/ f6 F4 u" X( n; K5 p; r8 M0 }
) B$ @7 @5 ]# b2 j' Y: Q
Effect on [Ca 2 ] i of Applying Pure orPure CO 2 Basolaterally0 B7 V8 Q( Q! H" b& t
; v4 E/ j; d- H9 [) t
The above experiments ruled out a role for pH i in the increase of [Ca 2 ] i elicited by bath butdid not discriminate between bath CO 2 and bath. Next, we used OOE solutions toinvestigate separately the effect of pure CO 2 ( solution 4 )and pure ( solution 5 ) on[Ca 2 ] i. We chose to use 20% CO 2 (nominally no, pH 7.40) becausethis basolateral P CO2 causes a substantially larger stimulation of J HCO3 in the S2 proximal tubule than does 5%CO 2 ( 77 ). AHEPES-buffered solution continuously perfused the lumen. As shown in Fig. 2 A, introducing22 mM pure to the bath caused, atmost, a trivial increase in [Ca 2 ] i ( segment ab ), whereas introducing 20% pure CO 2 alwayscaused a substantial and sustained increase in [Ca 2 ] i ( Fig. 2 A, segmentcde ). Removing the pure CO 2 solution caused[Ca 2 ] i to decrease rapidly, but not all theway to the baseline. In a total of 10 similar experiments( Fig. 2 B ), pure elicited a mean [Ca 2 ] i of 7 ± 2 nM ( n = 10; P 2 ] i of 76 ± 3 ( n = 10).This small [Ca 2 ] i increase could be theresult of a small CO 2 contamination in our pure solutions. On the other hand,measurements with a CO 2 electrode did not detect CO 2 inthe pure solutions exiting themixing T of our OOE apparatus. In the same tubules, pure CO 2 elicited a much larger [Ca 2 ] i = 62 ± 17 nM ( n = 10; P [Ca 2 ] i of 78 ± 2 ( n =10). These results support the hypothesis that it is basolateralCO 2, not, that isresponsible for increasing [Ca 2 ] i in theproximal tubule.% m8 u/ s% B2 I$ z4 b7 I" k0 g
% ?0 X/ ?$ G& |; j
Fig. 2. Effect on [Ca 2 ] i of basolateral exposureto a 22 mM "pure " vs.a 20% "pure CO 2 " solution. A: representativerecording of [Ca 2 ] i. The lumen and bath wereinitially perfused with a HEPES-buffered, -freesolution ( solution 2, Table1 ). At the indicated times, the bath solution was switched toeither a 22 mM pure solution( solution 5 ) or to a 20% pure CO 2 solution ( solution4 ). B: data summary. Hatched bar, mean [Ca 2 ] i caused by the basolateralswitch from the HEPES buffer to pure; open bar, comparable [Ca 2 ] i for the corresponding switchto pure CO 2 solution. The statistical analyses summarized in thefigure are the results of paired 2-tailed Student's t -tests. The meaninitial [Ca 2 ] i values with theHEPES-buffered solutions in the lumen and bath were 76 ± 3 nM justbefore the basolateral switch to the pure solution, and 78 ± 2 nMjust before the corresponding switch to the pure CO 2 solution; thedifference between these values is not statistically significant ( P =0.6, paired 2-tailed Student's t -tests).& }* K0 y* s! F
/ m+ t; Z4 F- B
One should recall that in Fig.2 A, [Ca 2 ] i did notfully return to its baseline value after removal of bath pure CO 2 In a series of 9 tubules distinct from the 10 discussed above, we exposed thebasolateral side of tubules to twin pulses of pure CO 2, with adelay of 5 min between pulses. The mean steady-state [Ca 2 ] i before the first pulse was 57± 1 nM. Because [Ca 2 ] i often did notreturn to the initial baseline, the mean steady-state[Ca 2 ] i before the second pulse wassignificantly higher, 107 ± 19 nM ( P [Ca 2 ] i elicited by the first pureCO 2 pulse was 85 ± 27 nM, whereas the [Ca 2 ] i elicited by the second pureCO 2 pulse (starting from a higher baseline) was only 48 ±14, a difference that is on the verge of statistical significance ( P = 0.052).: w6 r! s/ @" d- G: D! n6 ^
" w7 N8 W; W1 w6 R" L9 t
In six other experiments, we measured pH i while switching the basolateral solution from HEPES ( solution 2 ) to pure CO 2 ( solution 4 ). The mean pH i in bilateral HEPES was 7.23± 0.02, whereas the mean pH i during bath exposure to pureCO 2 was 6.87 ± 0.08, a mean difference of 0.36 ±0.09. Thus, even though the pH i decrease elicited by basolateralpure CO 2 was substantially less than that elicited by luminal (0.36vs. the values of 0.55 and 0.53 shown in Table 2 ), basolateral pureCO 2 triggered an increase in[Ca 2 ] i, whereas luminal didnot. This result thus provides additional support for the hypothesis that itis CO 2 itself, and not the change in pH i, that isresponsible for the [Ca 2 ] i increase in ourexperiments.
' R9 w, W3 v/ f7 ?5 V0 r: W+ S
A technical question that arises is whether the large decrease inpH i elicited by luminal mayhave affected the ability of fura 2 to report[Ca 2 ] i. Although in their original paperGrynkiewicz et al. ( 27 )reported that fura 2 is poorly pH sensitive, others have reported thatlowering the pH causes the K d of fura 2 to increase( 34, 39 ). In our experiments, wedid not attempt to correct for this pH sensitivity of the K d because we did not simultaneously measurepH i and Ca 2  . Thus we probably underestimatedthe rise in [Ca 2 ] i induced by the pureCO 2 solution in tubules with a relatively high initialpH i.

Mechanism of the [Ca 2 ] i IncreaseInduced by Basolateral CO 2- Q: i7 D4 J: R$ v4 a! o% z" S

Effect of bilateral Ca 2 -free solutionson the CO 2 -induced [Ca 2 ] i increase. We nextinvestigated the source of Ca 2 responsible for theCO 2 -induced increase in [Ca 2 ] i.Our first approach was to expose the tubule briefly to basolateral 20% pureCO 2, as in the second half of Fig. 2 A, first in thepresence and then in the absence of Ca 2  . Figure 3 A shows suchan experiment. Initially, the lumen and bath contained a HEPES-bufferedsolution ( solution 2 ). A control pulse of 20% pure CO 2 elicited a [Ca 2 ] i increase ( segmentab ) that averaged 16 ± 2 nM ( n = 8) and was partiallyreversed in this experiment by removing the CO 2 ( bc ). Wethen switched the luminal solution to a variant of solution 2 inwhich we omitted the Ca 2 and added 0.5 mM EGTA tochelate trace amounts of Ca 2  . This removal ( point c ) reversed the slow upward drift in[Ca 2 ] i and caused[Ca 2 ] i to begin to decrease slowly. When wethen similarly removed Ca 2 from the bath ( pointd ), [Ca 2 ] i fell more rapidly( de ). Because exposing tubules to Ca 2 -freesolutions for long periods ( 10 min) interfered with tubule integrity, wechallenged the tubule with a second CO 2 pulse even as[Ca 2 ] i continued to decline. We found thatthe second 20% pure CO 2 pulse, in the continued bilateral absenceof Ca 2  , caused a[Ca 2 ] i increase ( ef ) that averaged13 ± 2 nM ( n = 8) and was indistinguishable from the first( P = 0.12). On removal of the bath pure CO 2 solution,[Ca 2 ] i fell ( fg ) to a value thatwas lower than the value prevailing before we applied the CO 2.Reintroducing Ca 2 to the lumen and bath restored[Ca 2 ] i to its initial level ( gh ). Figure 3 B summarizesthe mean [Ca 2 ] i values in thepresence and absence of extracellular Ca 2  .; u, Q0 P$ z# {$ j

Fig. 3. Effect of Ca 2 removal ( EGTA) on the[Ca 2 ] i increase elicited by 20% pureCO 2. A: representative recording of[Ca 2 ] i. The lumen and bath were initiallyperfused with a HEPES-buffered, -freesolution ( solution 2, Table1 ). At the indicated times, the bath solution was twice switchedto a 20% pure CO 2 solution ( solution 4 ), the second timeafter Ca 2 was removed from both the lumen and bath. B: data summary. Open bar, mean [Ca 2 ] i caused by the basolateralswitch from the HEPES buffer to the pure CO 2 solution in thepresence of Ca 2  ; checkered bar, comparable [Ca 2 ] i in the absence ofCa 2  . The statistical analyses summarized in the figureare the results of paired 2-tailed Student's t -tests. The meaninitial [Ca 2 ] i values with theHEPES-buffered solutions in the lumen and bath were 68 ± 3 nM justbefore the basolateral switch to the pure CO 2 solution in thepresence of Ca 2  , and 62 ± 3 nM just before thecorresponding switch in the absence of Ca 2  ; thedifference between these values is statistically significant ( P t -tests).

Effect of bilateral nifedipine on the CO 2 -induced[Ca 2 ] i increase. Inexperiments similar to that shown in Fig.3 A, we examined the effect of adding 10 µM nifedipine,which blocks dihydropyridine-sensitive (L- and T-type)Ca 2 channels( 62 ), to both the lumen and the bath (not shown). We found that control 20% pure CO 2 pulses elicited a mean [Ca 2 ] i of 10 ±3 nM ( n = 6), a value that was not significantly different from the [Ca 2 ] i of 13 ± 1 nM elicitedby 20% pure CO 2 in the presence of bilateral nifedipine ( P = 0.4). The results of the experiments in this and the previous paragraphindicate that an influx of extracellular Ca 2 is notdirectly responsible for the CO 2 -induced increase in[Ca 2 ] i.

Effect of Tg on CO 2 -induced[Ca 2 ] i increase. IfCO 2 causes the release of Ca 2 from anintracellular store, then blocking the reuptake of Ca 2 into this store ought to deplete the store and reduce the size of theCO 2 -induced increase [Ca 2 ] i. Tg is a well-known inhibitor of SERCA, the Ca 2 pumpresponsible for the uptake of Ca 2 into the sarco- andendoplasmic reticulum ( 52, 65 ). Figure 4 A shows anexperiment in which we tested the effect of Tg on the CO 2 -inducedincrease in [Ca 2 ] i. As a control, we firstexposed the basolateral side of the tubule to 20% pure CO 2,observing a reversible increase in [Ca 2 ] i ( abc ). Adding 2 µM Tg to the lumen and bath caused a transient rise in [Ca 2 ] i ( point c ), probablydue to the decrease in Ca 2 reuptake into the stores, ashas been observed for other cell types( 48, 61, 65 ). Subsequently exposing thetubule to 20% pure CO 2 in the continued presence of Tg induced arise in [Ca 2 ] i that was actually somewhatgreater than in the absence of the drug. As summarized in Fig. 4 C for a total of10 experiments, Tg produced a small but statistically significant increase in the CO 2 -induced increase in[Ca 2 ] i.

Fig. 4. Effect of thapsigargin (Tg) on the [Ca 2 ] i increase elicited by either 20% pure CO 2 or ATP. A: representative recording of [Ca 2 ] i during abath pure CO 2 pulse. The lumen and bath were initially perfusedwith a HEPES-buffered, -freesolution ( solution 2, Table1 ). At the indicated times, the bath solution was twice switchedto a 20% pure CO 2 solution ( solution 4 ), the second timein the continued presence of 2 µM Tg. B: representative recordingof [Ca 2 ] i during a bath ATP pulse. Theprotocol is the same as in A, except that we twice pulsed with ATP( solution 2 containing 0.5 mM ATP). C: data summary. Openbar, mean [Ca 2 ] i caused by thebasolateral switch from the HEPES buffer to the pure CO 2 solutionin the absence of Tg; checkered bar, comparable [Ca 2 ] i in the presence of the drug;wavy and horizontally striped bars, comparable data for the ATP pulses. Thestatistical analyses summarized in the figure are the results of paired2-tailed Student's t -tests. For the above 4 data sets, the meaninitial [Ca 2 ] i values with theHEPES-buffered solutions in the lumen and bath, just before the basolateralswitch to the pure CO 2 solution and ATP, were 1 ) 75± 4 nM and 2 ) 76 ± 4 nM, and the mean values inpresence of Tg just before the switch to the pure CO 2 solution andATP were 3 ) 68 ± 13 nM and and 4 ) 66 ± 20 nM.Neither the difference between 1 and 3 ( P = 0.5)nor the difference between 2 and 4 ( P = 0.6) isstatistically significant.
. x1 l2 N; f( s- q/ Z
To verify that Tg was indeed blocking the sarco-endoplasmic Ca 2 pump, we performed a positive control experiment inwhich we used extracellular ATP to activate the P2Y purinergic receptor andthereby release Ca 2 from Tg-sensitive stores( 9, 74 ). As shown in Fig. 4 B, basolateralATP (0.5 mM) caused a very large but transient rise in[Ca 2 ] i, and Tg virtually eliminated thiseffect. As summarized in Fig.4 C for a total of eight similar experiments, theinhibition by Tg was statistically significant.. T! `$ G; V, g5 I; R: P+ C
) N/ R6 Y/ a5 w, Z1 o0 q( ~- u
One might argue that a desensitization of the P2Y receptor may have beenresponsible for the absence of a [Ca 2 ] i increase during the second ATP pulse in Fig. 4 B, an effectthat would have led us to overestimate the blockade by Tg. We therefore performed a separate series of experiments (not shown) in which we exposedtubules to two ATP pulses ( 5 min apart) in the absence of inhibitors. Thefirst exposure of the basolateral side of the tubule to 0.5 mM ATP caused amean [Ca 2 ] i of 99 ± 20 nM,whereas the second induced a mean [Ca 2 ] i of 110 ± 22 nM( n = 7); this difference is not statistically significant ( P = 0.6)., _3 m. c1 @8 V% g* U9 i6 y
2 y5 b; B* H8 o; R- e1 Y. ]1 z* \
Finally, we also performed two experiments (not shown) similar to the onein Fig. 4 B, but inwhich, in the presence of Tg, we first pulsed the tubule with 0.5 mM ATP andthen with 20% pure CO 2. Even though ATP had a minimal effect,CO 2 still elicited an increase in[Ca 2 ] i. The results of these three series of Tg experiments thus indicate that CO 2 does not cause the release ofCa 2 from Tg-sensitive Ca 2 stores.4 W& b; ^" Z4 C% |$ j* A' _7 }5 d

Effect of caffeine. To explore the possibility that a ryanodine receptor might be involved in the CO 2 -induced increase in[Ca 2 ] i, we assessed the ability of caffeine,a well-known agonist of this receptor( 32, 78 ), to raise[Ca 2 ] i in proximal tubule cells. In a totalof four experiments similar to the one shown in Fig. 5, we exposed the proximaltubule to 10 mM caffeine for 2 min. The mean[Ca 2 ] i value measured before application ofcaffeine was 57 ± 1 nM; adding caffeine caused a mean [Ca 2 ] i of 1 ± 2 nM, a valuenot statistically different from the baseline value ( P = 0.8). On theother hand, applying ATP always caused a transient increase in[Ca 2 ] i. In the same four experiments, from amean baseline [Ca 2 ] i of 58 ± 1 nM,adding ATP caused a mean [Ca 2 ] i of128 ± 29 nM. We conclude from these experiments that S2 proximal tubules have no demonstrable ryanodine receptor activity and that it isunlikely that these receptors play a role in the CO 2 -inducedincrease in [Ca 2 ] i.6 U# T% y9 z+ W9 B( D2 j

Fig. 5. Representative recording of [Ca 2 ] i duringsequential bath exposures to 10 mM caffeine and 0.5 mM ATP. The tubule lumenwas continuously perfused with a HEPES-buffered, -freesolution ( solution 2, Table1 ). At the indicated times, the bath solution was temporarilyreplaced twice, first with solution 2 supplemented with 10 mMcaffeine and the second time with solution 2 supplemented with 0.5 mMATP. In a total of 4 such experiments, caffeine elicited a mean [Ca 2 ] i (compared with paired baseline[Ca 2 ] i value just before the application ofcaffeine) of 1 ± 2 nM ( P = 0.8, paired t -test),whereas ATP elicited a mean [Ca 2 ] i of128 ± 29 nM ( P t -test).

Effect of rotenone on CO 2 -induced[Ca 2 ] i. To explore the possibility that CO 2 causes the release ofCa 2 from the mitochondria, we examined the effect ofrotenone on the CO 2 -induced[Ca 2 ] i increase. Our protocol was the sameas for Tg (see Fig.4 A ). Because rotenone blocks electron transport, we wouldexpect that rotenone would cause Ca 2 to leak out of themitochondria. Indeed, applying 4 µM rotenone caused the baseline[Ca 2 ] i to increase from 100 ± 14 to155 ± 24 nM ( P 0.02, n = 5). Nevertheless, assummarized in Fig. 6, pulsing with 20% pure CO 2 produced, if anything, a larger[Ca 2 ] i increase in the presence of rotenonethan in its absence, although the difference was not statistically significant( P = 0.08, n = 5). An unavoidable complication in theseexperiments is that rotenone undoubtedly disturbed cellular energy metabolism. If these changes in energy metabolism did not affect the mechanism by whichCO 2 releases Ca 2 from internal stores, wewould conclude that the mitochondria are not the source of theCa 2 released in response to CO 2.% a+ K6 W/ g; u/ W- Y5 k3 t: h
1 B" U: y7 O: s8 F
Fig. 6. Effect of rotenone on the [Ca 2 ] i increaseelicited by 20% pure CO 2. Open bar, mean [Ca 2 ] i caused by the basolateralswitch from the HEPES buffer to the pure CO 2 solution in theabsence of rotenone; filled bar, comparable [Ca 2 ] i in the presence of 4 µMrotenone. See the text for details. The statistical analysis summarized in thefigure is the result of a paired 2-tailed Student's t -test.

DISCUSSION& |3 k  w: r; G, C2 ?

By increasing the rate at which they transport into the blood, the kidneys playan important role in the response to respiratory acidosis. Although work fromour laboratory indicates that the trigger for increased transport is basolateralCO 2 per se, rather than the accompanying acidosis, the underlyingintracellular signals have not yet been resolved. We performed the presentexperiments to evaluate whether a rise in[Ca 2 ] i might be an element in one of the signaling pathways by which basolateral CO 2 acts on renal proximal tubule cells.- E  ^# E* i1 M5 k- ^

Influence of Initial pH i in HEPES on the pH i Response to Bath, P0 a5 U* ]7 [. T3 j0 s( B
7 b4 U) n* y3 F/ i% |/ z
As noted in RESULTS and summarized in Table 2, 9 of the 14 tubulesthat we tested had a relatively high initial pH i in HEPES andunderwent a sustained acidification when we introduced to thebath. The other five tubules had a relatively low initial pH i andunderwent an alkalinization when we introduced to thebath. This dependence on the initial pH i is consistent with threeprevious observations made using other preparations.
7 i' \) d) |& J: N! V# ~; V
First, in the rabbit S3 segment, perfusing the lumen with a monocarboxylate-free solution results in a relatively low initial pH i. Under these conditions, adding basolateral causesa transient pH i fall followed by a large and sustained rise( 46 ), reflecting a three- tofourfold stimulation of apical Na/H exchangers and H pumps( 17, 18 ) that overcomes theacidifying influence of the basolateral Na-HCO 3 cotransporter. However, when the tubule lumen is perfused with acetate, the initialpH i is relatively high, and adding basolateral causesa large and sustained fall in pH i ( 46 ). 5
# j, v5 b$ k9 S" E
Finally, in both hippocampal neurons( 58 ) and hippocampal astrocytes ( 7 ), the effect ofadding on steady-state pH i critically depends on the initialpH i. The induced alkalinization is greatest at the lowest initialpH i values and gradually falls off (or even reverses in the case ofthe astrocytes) at progressively higher initial pH i values. Ageneral explanation for all three cases is that relatively high pH i values stimulate acid loading but inhibit acid extrusion.

Basolateral CO 2 Directly Triggers an Increase in[Ca 2 ] i
* x2 h& J' S1 `- [+ o5 T
Ten years ago, Nakhoul et al.( 46 ), working with the rabbit S3 proximal tubule (which always had a relatively low initial pH i under the conditions of their experiments), showed that adding to thelumen causes a sustained pH i decrease that is presumably due to 1 ) the rapid diffusion of CO 2 into the cell, followed by 2 ) the formation of H and and the sustained basolateral exitof. On the other hand, they foundthat adding to thebath induces only a transient pH i decrease, followed by a sustainedincrease that is presumably due to an increase in net acid extrusion from thecell. Indeed, Chen and Boron( 17, 18 ) showed that addingequilibrated to thelumen had no effect on rates of apical Na/H exchange and H pumping, whereas adding the same solution to the bath increased these rates bytwo- to fourfold. Here, we report a parallel observation: adding to thelumen never elicits a significant rise in[Ca 2 ] i, whereas adding to thebath always triggers an increase in [Ca 2 ] i.Thus it appears that basolateral, but not luminal, produces several unique effects: 1 ) an increase in steady-statepH i when the initial pH i is low, 2 ) an increasein apical H extrusion, and 3 ) an increase in[Ca 2 ] i.
% v: g' V9 t( S4 A; C
One key question is whether it is CO 2 or that causes the above threeeffects. Other work from our laboratory shows that it is specificallybasolateral CO 2, and not basolateral, that increases J HCO3 ( 76, 77 ). In the present study, wehave made an additional parallel observation: a 20% pure CO 2 solution in the bath can elicit a substantial rise in[Ca 2 ] i, whereas a 22 mM pure solution in the bath cannot. IfCO 2, and not, is indeedthe trigger, why is it that CO 2 added to the lumen does not diffuseacross the apical membrane and through the cytoplasm to exert a measurableeffect at the basolateral membrane? We presume that, under the conditions ofsuch an experiment, the CO 2 concentration near the basolateralmembrane is too low to produce a measurable stimulation of some sort of aCO 2 sensor.
/ d9 w# P0 P  |( e. u6 n4 `/ [
The observation that it is basolateral CO 2 and not that triggers the increase in[Ca 2 ] i does not distinguish between thepossibilities that basolateral CO 2 1 ) acts directly on thetubule to raise [Ca 2 ] i or 2 ) acts indirectly by lowering pH i, which in turn leads to a rise in [Ca 2 ] i. A precedent for the latterhypothesis is that cytosolic acidification causes[Ca 2 ] i to rise in gastric parietal cells ( 67 ), platelets( 67 ), and cultured collectingduct cells ( 60 ). On the otherhand, cytosolic acidification causes [Ca 2 ] i to fall in squid giant axons( 3 ), and cytosolicalkalinization causes [Ca 2 ] i to rise in bothHT 29 cells ( 48 ) and ratpancreatic acinar cells( 61 ).5 V2 c3 t1 r, Y, B: K1 S

Did pH i indirectly control[Ca 2 ] i in our experiments? Although anisolated increase in bath  causespH i to increase( 76 ), we found that switchingto a pure solution causes only atrivial increase in [Ca 2 ] i ( Fig. 2 ). Thus, ifpH i controls [Ca 2 ] i, it wouldhave to be a pH i decrease that causes[Ca 2 ] i to rise. Indeed, switching to a pureCO 2 solution in the bath caused pH i to fall by 0.35 and consistently caused [Ca 2 ] i to increase,apparently supporting thepH i -[Ca 2 ] i hypothesis. However,we found that adding to thelumen always causes pH i to fall 0.5( Table 2 ) but has no effect on[Ca 2 ] i ( Fig. 1 ), ruling out thepH- i [Ca 2 ] i hypothesis. Finally,as noted in the presentation of Table2 in RESULTS, introducing equilibrated intothe bath caused pH i to decrease by 0.3 in 9 of 14 tubules(i.e., the high-pH i tubules) but caused [Ca 2 ] i to rise in 8 consecutive tubules. Thechance of randomly choosing eight consecutive high-pH i tubules isonly 3%. We conclude that a change in pH i is not theintermediary through which CO 2 raises[Ca 2 ] i. This conclusion represents a third parallelism between the CO 2 -induced increase in[Ca 2 ] i and CO 2 -induced changes inacid-base transport: In the proximal tubule, the CO 2 -inducedincrease in J HCO3 does not occur via a decrease inpH i ( 77 )./ H6 Q( R0 M& T# i6 }
) d, s( F; m9 a2 c8 {- c# U8 O
Ca 2 Originates From an As Yet UnidentifiedIntracellular Pool

Two pieces of evidence indicate that the immediate source of theCa 2 for the CO 2 -induced increase in[Ca 2 ] i is an intracellular store. First, theCO 2 -induced increase in [Ca 2 ] i occurs even when Ca 2 is absent from the lumen and bath( Fig. 3 ). Second, thedihydropyridine derivative nifedipine fails to attenuate theCO 2 -induced increase in [Ca 2 ] i.We chose nifedipine because the proximal tubule has dihydropyridine-sensitiveCa 2 channels that mediate Ca 2 influx during volume regulation after a hypotonic shock( 40 ), in response to PTH( 63 ), or during hypoxia( 49 ).

One of the classic types of Ca 2 stores in cells isthe Tg-sensitive store, which often is triggered by inositol1,4,5-trisphosphate (IP 3 ). Indeed, the P2Y purinergic receptor onthe basolateral membrane of the proximal tubule releasesCa 2 from a Tg-sensitive pool( 9, 74 ). Although we confirmedthat adding Tg blocks the rise in [Ca 2 ] i stimulated by extracellular ATP ( Fig. 4 B ), we found the drug to be ineffective in reducing themagnitude of the [Ca 2 ] i increase elicited bybasolateral pure CO 2 ( Fig. 4 A ). In fact, in the presence of Tg, a pureCO 2 pulse elicits a greater[Ca 2 ] i increase than a matched pulse in theabsence of the drug ( Fig.4 C ). It is possible that, with Tg preventing the loadingof Tg-sensitive stores, Tg-insensitive stores may accumulate extraCa 2 that they release in response to CO 2, resulting in a larger-than-normal CO 2 -induced increase in[Ca 2 ] i.
. b3 e& L* v* K( S( O4 g- v) Q( F
Ca 2 pools released by the ryanodine receptor areusually also Tg sensitive. However, we ruled out the possibility thatryanodine receptors are involved in the CO 2 -induced release ofCa 2 by demonstrating that millimolar concentrations ofcaffeine, which lead to a Ca 2 -independent activation ofthe ryanodine channel ( 32, 78 ), do not elicit a rise in[Ca 2 ] i in the proximal tubule.
# k$ V, c" w1 \2 t6 L) }9 Q2 x
One Tg-insensitive Ca 2 pool is the mitochondria( 23, 26, 29 ). However, our rotenonedata are not consistent with the hypothesis that CO 2 causes therelease of Ca 2 from mitochondria. Thus our data areconsistent with the hypothesis that, via a CO 2 sensor at or nearthe basolateral membrane, CO 2 triggers the release ofCa 2 from a nonconventional intracellular store.
. G: y6 {: a; Y1 J) V9 X; Z1 p
Other investigators have demonstrated that multiple nonmitochondrial Ca 2 stores, functionally and spatially distinct, maycoexist in the same cell ( 25, 43, 50, 52 ) and have in particular demonstrated the presence of Tg-insensitive pools. For example, a variety ofcell lines have a nonmitochondrial pool that can take upCa 2 after maximal inhibition by Tg( 51, 64 ). In goldfish somatotrophs,GnRH causes a release of Ca 2 from a Tg-insensitivestore ( 30 ). Moreover, in seaurchin eggs, the second messenger nicotinic acid adenine dinucleotide causes the release of Ca 2 from a Tg-insensitive store that isdistinct from that triggered by either IP 3 or cADP-ribose( 24, 35 ). TheCa 2 pumps responsible for accumulatingCa 2 in the Golgi apparatus are Tg insensitive. Certainagonists (e.g., arginine vasopressin, histamine) coupled to the generation ofIP 3 can partially release Ca 2 from this pool( 43, 50 ). Thus several pools arecandidates for the CO 2 -induced release ofCa 2  .

Potential Roles of the CO 2 -Induced Increase in[Ca 2 ] i5 n+ _5 m5 W5 u  \% B

Previous work has established conflicting precedents for the effects thatincreases in [Ca 2 ] i have on acid-basetransport in the proximal tubule. Four lines of evidence suggest that anincrease in [Ca 2 ] i is associated with anincrease in acid-base transport and/or J HCO3. First, inexperiments on in vivo microperfused proximal tubules, raising[Ca 2 ] i by the luminal addition of theCa 2 ionophore A-23187 increases J HCO3 in a dose-dependent manner( 38 ). Second, addingangiotensin II to the basolateral side of a proximal tubule leads to increasesin both J HCO3 ( 37 ) and[Ca 2 ] i ( 31 ). Third, carbacholtriggers an increase in [Ca 2 ] i ( 42, 56 ) and stimulates theNa-HCO 3 cotransporter; conversely, the Ca 2 chelator BAPTA prevents the stimulation of the cotransporter( 56 ). Fourth and finally,CO 2 causes insertion of vesicles containing H pumpsinto the apical membrane of the proximal tubule( 57 ). In the turtle bladder,the application of CO 2 triggers a rise inCa 2 ( 15 ), and this rise in[Ca 2 ] i is required for the apical insertionof vesicles ( 68 ). A similar process may be at work in the rabbit outer medullary collecting duct( 28 ).* d/ ~. V% ~& u. ]2 m1 l- I8 v

Three lines of evidence suggest that an increase in[Ca 2 ] i is associated with a decrease inacid-base transport and/or J HCO3 in the proximal tubule.First, increasing [Ca 2 ] i by adding ionomycinto the bath leads to a decrease in J HCO3 ( 16 ). Second, PTH, a potentinhibitor of J HCO3 ( 20, 21 ), also increases[Ca 2 ] i ( 63 ). And third, a rise in[Ca 2 ] i inhibits the apical Na/H exchanger( 72, 73 )./ g+ w9 j! F0 ?+ c6 Q' e

One explanation for the apparently divergent data discussed above is thatthe relevant changes in [Ca 2 ] i occur within microdomains, and local changes in [Ca 2 ] i are more important than global ones( 25, 33, 55 ). Another explanation forthese divergent effects is that they are the consequence of different frequencies of Ca 2 spikes or waves. In the context ofthese possibilities, it is difficult to predict the role that[Ca 2 ] i plays in the response of the proximaltubule to basolateral CO 2. We propose that CO 2 binds toa CO 2 sensor at or near the basolateral membrane and, independentlyof a change in pH i, triggers the release ofCa 2 from a nonmitochondrial intracellular store that isinsensitive to Tg. The released Ca 2 might 1 ) modulate cellular processes not directly related to J HCO3, 2 ) be part of a signal-transduction pathway that results in anincrease in J HCO3, or 3 ) be part of a brakingmechanism that helps prevent runaway J HCO3 duringCO 2 stimulation.
" C& ]/ g) E* O
DISCLOSURES. @# a( J: d' H4 Z. d
, G  A2 a" ^" d! `% m
This work was supported by National Institute of Diabetes and Digestive andKidney Diseases Program Project Grant PO1-DK-17433. P. Bouyer was supported byBourse Lavoisier du Ministère des Affaires EtrangèreFrançaise and the National Kidney Foundation.
1 a! \# {/ j) X  V  R
ACKNOWLEDGMENTS" d% l; ^6 K. \, A' z

The authors thank Duncan Wong for computer programming and for providinginformation-technology assistance and Dr. Barbara Ehrlich for helpfuldiscussions.6 W  ~1 J/ H5 ~6 E8 H
      【参考文献】
Adrogué HJ and Madias NE. Renal acidification duringchronic hypercapnia in the conscious dog. PflügersArch 406:520-528, 1986.
2 |" [( O( M1 A: W
0 o7 r$ G7 [# A( V! P( x. L
2 r" [# ^8 o3 t7 n
Aronson PS. Mechanisms of active H secretion in the proximal tubule. Am J Physiol Renal Fluid Electrolyte Physiol 245: F647-F659,1983.' |, T" d. e0 v
+ m# D* E9 V1 d' }; }2 h

Baker PF andHonerjager P. Influence of carbon dioxide on level of ionised calcium insquid axons. Nature 273:160-161, 1978.' ]5 ~5 P! |8 T. s8 h) {# t' l

4 X( V1 ?8 b3 d( l- g' T3 d0 ~
Beck JS, BretonS, Laprade R, and Giebisch G. Volume regulation and intracellular calciumin the rabbit proximal convoluted tubule. Am J Physiol Renal FluidElectrolyte Physiol 260:F861-F867, 1991.

Bevensee MO,Bashi E, and Boron WF. Effect of trace levels of nigericin onintracellular pH and acid-base transport in rat renal mesangial cells. J Membr Biol 169:131-139, 1999.

Bevensee MO,Schwiening CJ, and Boron WF. Use of BCECF and propidium iodide to assessmembrane integrity of acutely isolated CA1 neurons from rat hippocampus. J Neurosci Methods 58:61-75, 1995.
: D0 k6 p7 k5 o8 X
4 T6 b3 E1 E6 C( p) K* J

Bevensee MO,Weed RA, and Boron WF. Intracellular pH regulation in cultured astrocytesfrom rat hippocampus. I. Role of. J Gen Physiol 110:453-465, 1997.

3 a+ A. t8 j7 R  `) @# d  A0 {
Boron WF andBoulpaep EL. Intracellular pH regulation in the renal proximal tubule ofthe salamander: basolateral transport. J Gen Physiol 81:53-94, 1983.
" h% ~. m" |9 Y" ~/ t8 W; ?: {

Bouyer P,Paulais M, Cougnon M, Hulin P, Anagnostopoulos T, and Planelles G. Extracellular ATP raises cytosolic calcium and activates basolateral chlorideconductance in Necturus proximal tubule. JPhysiol 510:535-548, 1998.
; J" P7 @$ t( N& `* w$ g4 J( Z$ W

Boyarsky G,Ganz MB, Sterzel B, and Boron WF. pH regulation in single glomerularmesangial cells. I. Acid extrusion in absence and presence of. Am J Physiol CellPhysiol 255:C844-C856, 1988.
( a% z) s4 L4 u
1 V! a; L  T5 n# F# i# C1 V/ _

Brazeau P andGilman A. Effect of plasma CO 2 tension on renal tubularreabsorption of bicarbonate. Am J Physiol 175: 33-38,1953.

9 F) @% Z) m& `+ D. m" T
Brooks SP andStorey KB. Bound and determined: a computer program for making buffers ofdefined ion concentrations. Anal Biochem 201: 119-126,1992.

$ w) `; ]0 g! ]9 h8 H* G' K
Buckler KJ andVaughan-Jones RD. Effects of acidic stimuli on intracellular calcium inisolated type I cells of the neonatal rat carotid body. Pflügers Arch 425:22-27, 1993.
  z0 b5 x4 X( L. K$ S8 U

Burg M,Grantham J, Abramow M, and Orloff J. Preparation and study of fragments ofsingle rabbit nephrons. Am J Physiol 210: 1293-1298,1966.: K. {" N9 B& J+ f& ^. n
" B; a8 x8 u; z% f$ P
/ i4 ]$ c$ Y& C( f) C: @  D
' D, R3 N" @- o2 z- d
Cannon C, VanAdelsberg J, Kelly S, and Al Awqati Q. Carbon-dioxide-induced exocytoticinsertion of H pumps in turtle-bladder luminal membrane: role ofcell pH and calcium. Nature 314: 443-446,1985.
- c/ j, C+ k) G. r

9 @3 a$ f4 l* S( V5 `& f
Chatsudthipong V and Chan YL. Inhibitory effect of angiotensinII on renal tubular transport. Am J Physiol Renal Fluid ElectrolytePhysiol 260:F340-F346, 1991.- t3 F; f. N7 ~- k! \5 r0 v/ E# n

: r1 s  r1 S( U
Chen LK andBoron WF. Acid extrusion in S3 segment of rabbit proximal tubule. I.Effect of bilateral. Am J Physiol Renal Fluid Electrolyte Physiol 268: F179-F192,1995.9 ?& A; E  |' S+ a( W) f  U6 L% I

Chen LK andBoron WF. Acid extrusion in S3 segment of rabbit proximal tubule. II.Effect of basolateral. Am J Physiol Renal Fluid Electrolyte Physiol 268: F193-F203,1995.. f* y3 w# p$ `1 z$ d$ a
/ V4 N! ]# v/ \" N# ~$ Z! v6 u

Clapham DE. Calcium signaling. Cell 80:259-268, 1995.
7 ?4 }! _( i5 E8 F1 k

; N% O6 ^2 ~6 j: P7 w2 S2 A5 ?) d
Cogan MG andAlpern RJ. Regulation of proximal bicarbonate reabsorption. AmJ Physiol Renal Fluid Electrolyte Physiol 247: F387-F395,1984.

Dennis VW. Influence of bicarbonate on parathyroid-induced changes in fluid absorption bythe proximal tubule. Kidney Int 10: 373-380,1976.. T; `6 F7 N" h2 I, ^
& x# `: ^5 K4 V+ h  L" l0 C4 l
7 e$ ^# d' v1 _5 j2 n, h! I- M
$ W5 y: \1 a4 R2 q: W4 B
Dorman PJ,Sullivan WJ, and Pitts RF. The renal response to acute respiratoryacidosis. J Clin Invest 33:82-90, 1954." e. y5 s, d& h

Duchen MR. Contributions of mitochondria to animal physiology: from homeostatic sensor tocalcium signalling and cell death. J Physiol 516: 1-17,1999.

Genazzani AA and Galione A. Nicotinic acid-adenine dinucleotide phosphate mobilizesCa 2 from a thapsigargin-insensitive pool. Biochem J 315:721-725, 1996.

6 F% Z0 |( C& v5 f$ K
" c+ l, A5 F6 k3 `  K
Golovina VA andBlaustein MP. Spatially and functionally distinctCa 2 stores in sarcoplasmic and endoplasmic reticulum. Science 275:1643-1648, 1997.* L4 S& ^5 t# g! H. F, P: V- \

González A, Schulz I, and Schmid A. Agonist-evokedmitochondrial Ca 2 signals in mouse pancreatic acinarcells. J Biol Chem 275:38680-38686, 2000.

7 ~4 N' r3 X6 m, x2 X  i

Grynkiewicz G,Poenie M, and Tsien RY. A new generation of Ca 2 indicators with greatly improved fluorescence properties. J BiolChem 260:3440-3450, 1985.

$ k$ ~7 J4 |& }- @6 |# ?

Hays SR andAlpern RJ. Inhibition of Na -independent H pump byNa -induced changes in cell Ca 2  . J Gen Physiol 98:791-813, 1991.1 B! [5 J: i/ J& ~% r
! U- k* n( a- Y+ n9 m9 L
) n" e" y2 R  C* N6 u9 t

Jaconi M, BonyC, Richards SM, Terzic A, Arnaudeau S, Vassort G, and Pucéat M. Inositol 1,4,5-trisphosphate directs Ca 2 flow betweenmitochondria and the endoplasmic/sarcoplasmic reticulum: a role in regulatingcardiac autonomic Ca 2 spiking. Mol BiolCell 11:1845-1858, 2000.

Johnson JD andChang JP. Novel, thapsigargin-insensitive intracellularCa 2 stores control growth hormone release from goldfishpituitary cells. Mol Cell Endocrinol 165: 139-150,2000.
  I) ^: w1 W2 l9 e
3 c& l& ~; y3 F: @
0 {, h: K- `, N8 o  N1 g; B
Jung KY andEndou H. Biphasic increasing effect of angiotensin II on intracellularfree calcium in isolated rat early proximal tubule. Biochem BiophysRes Commun 165:1221-1228, 1989.

' Y  H+ q4 u8 h: d0 z

Koulen P andThrower EC. Pharmacological modulation of intracellularCa 2 channels at the single-channel level. Mol Neurobiol 24:65-86, 2001.

' j  J, q+ U" T: Z
Krause E, GobelA, and Schulz I. Cell side-specific sensitivities of intracellularCa 2 stores for inositol 1,4,5-trisphosphate, cyclicADP-ribose, and nicotinic acid adenine dinucleotide phosphate in permeabilizedpancreatic acinar cells from mouse. J Biol Chem 277: 11696-11702,2002.
( L9 {3 P( y5 }& _4 P" ?. d
( p  n' `! ?7 K0 Y6 p$ x8 o! n. d$ ~

Lattanzio FA Jr. The effects of pH and temperature on fluorescent calcium indicators asdetermined with chelex-100 and EDTA buffer systems. Biochem BiophysRes Commun 171:102-108, 1990.

" \1 P: \9 L2 L/ y; i2 E
% W" V% ^% X5 e- i
Lee HC andAarhus R. A derivative of NADP mobilizes calcium stores insensitive toinositol trisphosphate and cyclic ADP-ribose. J BiolChem 270:2152-2157, 1995.+ X/ U5 L1 ], i6 D

) y/ \6 Z( C! E" V  v8 d
. b6 V& |0 H1 N  ?7 f$ E- e- o
Liu FY andCogan MG. Angiotensin II: a potent regulator of acidification in the ratearly proximal convoluted tubule. J Clin Invest 80: 272-275,1987.

$ i: i- Q, l3 Z3 f8 y' l( O* C# K
Liu FY andCogan MG. Angiotensin II stimulates early proximal bicarbonate absorptionin the rat by decreasing cyclic adenosine monophosphate. J ClinInvest 84:83-91, 1989.3 i% z2 X0 J# ]
( y* u: `+ q' ]) i, x1 H
2 \$ Q; V: Y6 L* V3 k, }

Liu FY andCogan MG. Effects of intracellular calcium on proximal bicarbonateabsorption. Am J Physiol Renal Fluid ElectrolytePhysiol 259:F451-F457, 1990.
# b" w! h4 G/ v) O* u4 O

, u+ A: Z- x! Y$ S# ]
Martínez-Zaguilán R, Martínez GM, Lattanzio F, and GilliesRJ. Simultaneous measurement of intracellular pH andCa 2 using the fluorescence of SNARF-1 and fura 2. Am J Physiol Cell Physiol 260:C297-C307, 1991.7 O! r& e7 Y) ^# e* }
) ^0 U0 @' ]  H2 V
! C  G1 E4 N8 H) {2 p% C
4 |3 z% _" c8 l
McCarty NA andO'Neil RG. Dihydropyridine-sensitive cell volume regulation in proximaltubule: the calcium window. Am J Physiol Renal Fluid ElectrolytePhysiol 259:F950-F960, 1990.
# [/ [& R6 ~+ e8 b
+ @6 g, A9 W9 L! L0 t; U- k2 O
" m, a$ _  V  t& q+ m; ^
McKinney TD andMyers P. PTH inhibition of bicarbonate transport by proximal convolutedtubules. Am J Physiol Renal Fluid Electrolyte Physiol 239: F127-F134,1980.
9 _- B! @) I' N( n
- }. h' C  @* x

Meneton P,Imbert-Teboul M, Bloch-Faure M, and Rajerison RM. Cholinergic agonistsincrease phosphoinositide metabolism and cell calcium in isolated rat renalproximal tubule. Am J Physiol Renal Fluid ElectrolytePhysiol 271:F382-F390, 1996.

. _7 U4 R8 V5 V, A4 V$ C
/ m4 L7 D: ~, o3 \5 h- J
Missiaen L,Vanoevelen J, Parys JB, Raeymaekers L, De Smedt H, Callewaert G, Erneux C, andWuytack F. Ca 2 uptake and release properties of athapsigargin-insensitive nonmitochondrial Ca 2 store inA7r5 and 16HBE14o-cells. J Biol Chem 277: 6898-6902,2002." [# H# `  u( c$ y: E- p8 M* @

* N6 X" N/ O  A- c& Y/ i
Munsch T andDeitmer JW. Maintenance of fura-2 fluorescence in glial cells and neuronsof the leech central nervous system. J NeurosciMethods 57:195-204, 1995.
" @1 R3 S  X# c$ E" J
  v: r3 V9 D' B2 {

Nakhoul NL andBoron WF. Acetate transport in the S3 segment of the rabbit proximaltubule and its effect on intracellular pH. J GenPhysiol 92:395-412, 1988./ V! H2 y( w& K' ]- ^- ^7 }$ _) |

) R% m, E& p0 J; o
% G9 j3 K! I7 ^
Nakhoul NL,Chen LK, and Boron WF. Effect of basolateral onintracellular pH regulation in the rabbit S3 proximal tubule. J GenPhysiol 102:1171-1205, 1993.
4 C5 M. X9 a, K* m- f! ~

Nakhoul NL,Lopes AG, Chaillet JR, and Boron WF. Intracellular pH regulation in the S3segment of the rabbit proximal tubule in -free solutions. J GenPhysiol 92:369-393, 1988.- f& ]9 Z$ X) W8 p

Nitschke R,Riedel A, Ricken S, Leipziger J, Benning N, Fischer KG, and Greger R. Theeffect of intracellular pH on cytosolic Ca 2 inHT 29 cells. Pflügers Arch 433: 98-108,1996." R0 I  X6 y7 c* @2 G9 g; f+ _

# b4 \3 R; \' ?: S3 A2 F$ K
Peters SM,Tijsen MJ, van Os CH, Wetzels JF, and Bindels RJ. Hypoxia decreasescalcium influx into rat proximal tubules. Kidney Int 53: 703-708,1998.
2 M# [# j$ j  D: A$ E8 w  T

Pinton P,Pozzan T, and Rizzuto R. The Golgi apparatus is an inositol1,4,5-trisphosphate-sensitive Ca 2 store, withfunctional properties distinct from those of the endoplasmic reticulum. EMBO J 17:5298-5308, 1998.7 m, ^# W# X, r, s2 ?- h: @

% `5 |) ~" c; h4 @

Pizzo P,Fasolato C, and Pozzan T. Dynamic properties of an inositol1,4,5-trisphosphate- and thapsigargin-insensitive calcium pool in mammaliancell lines. J Cell Biol 136:355-366, 1997.
3 {! I! ?( i  h5 c

( i) t- B; J; K# N9 v6 a! Z  l
Pozzan T,Rizzuto R, Volpe P, and Meldolesi J. Molecular and cellular physiology ofintracellular calcium stores. Physiol Rev 74: 595-636,1994.

" O- R0 S* t- N' o
Quigley R,Kennerly DA, Sheu JN, and Baum M. Stimulation of proximal convolutedtubule phosphate transport by epidermal growth factor: signal transduction. Am J Physiol Renal Fluid Electrolyte Physiol 269: F339-F344,1995.

Ravesloot JH,Eisen T, Baron R, and Boron WF. Role of Na-H exchangers and vacuolarH pumps in intracellular pH regulation in neonatal ratosteoclasts. J Gen Physiol 105:177-208, 1995.

5 p, ^  r$ U- r5 G6 G' M- ]- ^% a

Rizzuto R,Brini M, Murgia M, and Pozzan T. Microdomains with highCa 2 close to IP 3 -sensitive channels that aresensed by neighboring mitochondria. Science 262: 744-747,1993.

1 |* K/ g1 c9 \3 ^; L. x
8 K9 {* V9 c0 c$ Z/ {0 r
Ruiz OS, QiuYY, Cardoso LR, and Arruda JA. Regulation of the renal Na-HCO 3 cotransporter. VII. Mechanism of the cholinergic stimulation. Kidney Int 51:1069-1077, 1997.

Schwartz GJ andAl Awqati Q. Carbon dioxide causes exocytosis of vesicles containingH pumps in isolated perfused proximal and collecting tubules. J Clin Invest 75:1638-1644, 1985.

Schwiening CJ and Boron WF. Regulation of intracellular pH in pyramidal neurons from therat hippocampus by Na -dependent exchange. J Physiol 475:59-67, 1994.
6 [- Y4 g* k0 M* n2 u2 w% L

; z6 e* G! _8 S. S; ^
Siebens AW andBoron WF. Effect of electroneutral luminal and basolateral lactatetransport on intracellular pH in salamander proximal tubules. J GenPhysiol 90:799-831, 1987.2 ]( h5 `5 w! K' ^

2 d( F) a2 n) Q8 q: l! s' w# `

Slotki I,Schwartz JH, and Alexander EA. Interrelationship between cell pH and cellcalcium in rat inner medullary collecting duct cells. Am J PhysiolCell Physiol 265:C432-C438, 1993.

Speake T andElliott AC. Modulation of calcium signals by intracellular pH in isolatedrat pancreatic acinar cells. J Physiol 506: 415-430,1998.

/ v0 \- U- b* {2 N7 V" Q) Q

Spedding M andPaoletti R. Classification of calcium channels and the sites of action ofdrugs modifying channel function. Pharmacol Rev 44: 363-376,1992.

( ?2 o/ P* Q1 [3 ^
! L+ Z5 r0 N) U6 A9 @/ `
Tanaka H,Smogorzewski M, Koss M, and Massry SG. Pathways involved in PTH-inducedrise in cytosolic Ca 2 concentration of rat renalproximal tubule. Am J Physiol Renal Fluid ElectrolytePhysiol 268:F330-F337, 1995.

Tanaka Y andTashjian AH Jr. Functional identification and quantitation of threeintracellular calcium pools in GH 4 C 1 cells: evidencethat the caffeine-responsive pool is coupled to a thapsigargin-resistant,ATP-dependent process. Biochemistry 32: 12062-12073,1993.
7 w5 \$ j* H* [6 i, I1 T0 Q

1 w8 Q1 i4 z; @4 h5 K  r
Thastrup O,Cullen PJ, Drøbak BK, Hanley MR, and Dawson AP. Thapsigargin, atumor promoter, discharges intracellular Ca 2 stores byspecific inhibition of the endoplasmic reticulumCa 2 -ATPase. Proc Natl Acad SciUSA 87:2466-2470, 1990.) u! m# R& P* x3 ?# l  {+ O

Thomas JA,Buchsbaum RN, Zimniak A, and Racker E. Intracellular pH measurements inEhrlich ascites tumor cells utilizing spectroscopic probes generated in situ. Biochemistry 18:2210-2218, 1979., B9 k7 d% }# B: u
, e& N& F5 j7 _. N0 P  b

0 J; D) _5 V9 O" ^
Tsunoda Y,Matsuno K, and Tashiro Y. Cytosolic acidification leads toCa 2 mobilization from intracellular stores in singleand populational parietal cells and platelets. Exp CellRes 193:356-363, 1991.2 z4 q3 P# h% D( F; X

Van Adelsberg J and Al Awqati Q. Regulation of cell pH by exocytotic insertion ofH -ATPases. J Cell Biol 102: 1638-1645,1986.9 r8 u$ W1 o# T; K2 q/ H

. `/ N2 B* H: n2 d7 P( i" z
Vorndran C,Minta A, and Poenie M. New fluorescent calcium indicators designed forcytosolic retention or measuring calcium near membranes. BiophysJ 69:2112-2124, 1995." W* c. D, G2 E' @. z2 ]8 i% ~

Wang T. Nitric oxide regulates andNa transport by a cGMP-mediated mechanism in the kidney proximaltubule. Am J Physiol Renal Physiol 272: F242-F248,1997.7 N0 y$ `- z1 J# @

% z3 W+ T% _: p$ ^9 x- M* _
# t0 S/ R1 ?! w( d+ R1 r! R
Wang T and ChanYL. Mechanism of angiotensin II action on proximal tubular transport. J Pharmacol Exp Ther 252:689-695, 1990.! M# n4 l  \0 h  Z4 H5 U6 y
; [  T1 a2 J8 ^/ _& i/ ?! C0 R

7 U  k! p8 G+ S7 l2 s3 s# G
Weinman EJ,Dubinsky WP, Fisher K, Steplock D, Dinh Q, Chang L, and Shenolikar S. Regulation of reconstituted renal Na /H exchanger bycalcium-dependent protein kinases. J Membr Biol 103: 237-244,1988.2 ]/ B9 q: h1 L; N

Yamada H, SekiG, Taniguchi S, Uwatoko S, Nosaka K, Suzuki K, and Kurokawa K. Roles ofCa 2 and PKC in regulation of acid/base transport inisolated proximal tubules. Am J Physiol Renal Fluid ElectrolytePhysiol 271:F1068-F1076, 1996.4 ^1 f1 G1 N  v. m

Yamada H, SekiG, Taniguchi S, Uwatoko S, Suzuki K, and Kurokawa K. Mechanism of[Ca 2 ] i increase by extracellular ATP inisolated rabbit renal proximal tubules. Am J Physiol CellPhysiol 270:C1096-C1104, 1996.  R5 W+ o2 u! T1 p" s- `0 n  {

Zhao J, HoganEM, Bevensee MO, and Boron WF. Out-of-equilibrium solutions and their use in characterizing a new K/HCO 3 cotransporter. Nature 374:636-639, 1995.) P& M, f& n: N, U

; c! E) a: N' ?9 H' S7 D$ Q

Zhao J, Zhou Y,and Boron WF. Effect of the isolated removal of either basolateral or basolateral CO 2 on reabsorption by the rabbit S2proximal tubule. Am J Physiol Renal Physiol 285: F359-F369,2003.

1 ~% s$ ~- v0 W* }
, }# d: T$ h& a& k/ b2 O6 r
Zhou Y, Zhao J,Bouyer P, and Boron WF. Regulation of bicarbonate reabsorption by S2proximal tubule: evidence for a CO 2 -sensor (Abstract). FASEB J 15:A142, 2001., o* i* A, {0 P
1 F% c+ T% X5 }2 h

" i- P) \! W; ]* U
Zucchi R andRonca-Testoni S. The sarcoplasmic reticulum Ca 2 channel/ryanodine receptor: modulation by endogenous effectors, drugs anddisease states. Pharmacol Rev 49: 1-51,1997.

收藏0 分享0 顶0 踩0