NBCn1 (slc4a7) Mediates the Na+-Dependent Bicarbonate Transport Important for Regulation of Intracellular pH in Mouse Vascular Smooth Muscle
http://www.100md.com
《循环研究杂志》
The Water and Salt Research Center (E.B., J.P., C.A.)
Department of Physiology (E.B., C.A.), Institute of Physiology and Biophysics
Institute of Anatomy (J.P.), University of Aarhus, Denmark.
Abstract
The contribution of sodium-dependent bicarbonate transport to intracellular pH (pHi) regulation in vascular smooth muscle cells is controversial, partly because the molecular identity of the transporter(s) responsible has not been identified. Here, using the pH-sensitive fluorophore 2',7'-bis-(2-carboxyethyl)-5-(and-6)-carboxyfluorescein (BCECF), we show that smooth muscle cells of intact mouse mesenteric, coronary, and cerebral small arteries all display a sodium- and bicarbonate-dependent pHi recovery after an NH4+-prepulse. The sodium-dependent bicarbonate flux was largely 4,4'-diisothiocyanatostilbene-2,2'-disulphonic acid (DIDS) sensitive (56% to 91%) and of a magnitude similar to the amiloride-sensitive flux. Additionally, steady-state pHi was lower (0.2 to 0.4 pH units magnitude) in all 3 vascular beds when CO2/bicarbonate was omitted. RT-PCR analyses showed that NBCn1 (slc4a7) is the only Na+-dependent bicarbonate transporter of the slc4 family detectable at the mRNA level in all 3 vascular beds investigated. Whole-mount immunolabeling and immunogold electron microscopy confirmed the presence of NBCn1 protein in the sarcolemma of mouse mesenteric small arterial smooth muscle cells. Intact mouse mesenteric small arteries were electropermeated to facilitate transfection with small interfering RNA targeting NBCn1, which resulted in an approximate 43% decrease in the ratio of NBCn1 to glyceraldehyde-3-phosphate dehydrogenase mRNA. After knock-down, we found a decreased steady-state pHi (0.21±0.08 pH units) as well as a 68±10% decrease in the net Na+-dependent, amiloride-insensitive base influx after acid load. Finally, omission of CO2/bicarbonate resulted in a decreased contractile response to norepinephrine after sustained exposure to the agonist, underlining the importance of CO2/bicarbonate for vascular contractility. We conclude that NBCn1 mediates the Na+-dependent bicarbonate transport important for pHi regulation in smooth muscle cells of mouse mesenteric, coronary, and cerebral small arteries.
Key Words: vascular smooth muscle cells bicarbonate intracellular pH siRNA NBC
Introduction
Intracellular pH (pHi) of vascular smooth muscle cells (VSMC) is tightly related to fundamental vascular functions. In resistance arteries acute intracellular acidification leads to contraction,1eC3 and during force development, pHi of VSMC has a tendency to decline.4,5 Intracellular pH is reported to modulate VSMC growth,6 and in general pH affects the conformation and function of many proteins, such as KATP and BK channels, present in VSMC.7,8 Consequently, the control of pHi is essential to maintain normal vascular function.
Transmembrane movement of acid/base equivalents is important for pHi regulation in VSMC and 2 well-characterized transporters are known to contribute to the regulation. The first discovered of these is the anion exchanger, which mediates an electroneutral exchange of CleC for HCO3eC. In VSMC, this transporter is involved in maintaining a high intracellular chloride concentration,9 and during intracellular alkalinization it is activated to participate in the regulation of pHi.10 The first anion exchanger (AE1) was cloned 20 years ago from erythrocytes,11 and in VSMC the transporter is also well defined at the molecular level: AE2 is predominant together with smaller amounts of AE3.12
The other well-described transporter is the Na+-H+ exchanger, which extrudes protons in exchange for sodium. This transporter is activated during intracellular acidification and participates in the recovery of pHi under such conditions.10,13,14 In VSMC, the transporter is well defined at the molecular level as NHE1.15 Interestingly, Na+-H+ exchangers have been reported to be implicated in hypertension, further underlining the importance of pHi regulation for normal cardiovascular function.16
A third transport system may be important for pHi regulation in VSMC. A Na+-dependent bicarbonate transport seems to be present in rat mesenteric small arteries,4,17 but its importance in the vasculature as a whole has been questioned.18 Furthermore, at present, no molecular information exists about the transporter(s) responsible. Considering the importance of pHi regulation in VSMC, this reveals a major gap in our current knowledge.
The aim of the present study was to examine the general importance of Na+-dependent bicarbonate transport in VSMC and determine the molecular identity of the transporter(s) responsible.
Materials and Methods
Mesenteric, coronary, and cerebral small arteries were isolated from male Naval Medical Research Institute (NMRI) mice (7 to 10 weeks old), mounted for isobaric or isometric investigation and loaded with the pH-sensitive fluorophore 2',7'-bis-(2-carboxyethyl)-5-(and-6)-carboxyfluorescein (BCECF) or the Ca2+-sensitive fluorophore fura-2. The animals were bred in the institute’s own breeding facility. The care and use of the animals are described in the online data supplement, available at http://circres.ahajournals.org.
Arteries were experimentally acidified with the NH4+-prepulse technique, and the rate of pHi recovery was measured in the presence and absence of CO2/bicarbonate, Na+, amiloride (600 eol/L), and 4,4'-diisothiocyanatostilbene-2,2'-disulphonic acid (DIDS) (200 eol/L).
The expression of known and putative Na+,HCO3eC cotransporters of the slc4 family was investigated by RT-PCR (primer sequences in Table 1). Gel electrophoresis of PCR products was performed and products of the expected sizes were sequenced to confirm specificity.
The expression of NBCn1 at the protein level was investigated by whole-mount immunolabeling and immunogold electron microscopy using a newly developed N-terminal NBCn1 primary antibody19 in combination with an Alexa488- or gold-labeled secondary antibody.
Intact isolated mouse mesenteric small arteries were transfected with small interfering RNA (siRNA) using electropermeation. A modification of the protocol previously described for DNA plasmid transfection20 was used. Arteries were kept in electroporation buffer containing 50 nmol/L siRNA (targeting NBCn1 or scrambled) and 8 "square-wave" pulses (125 V/cm, 10-ms duration and 100-ms interpulse interval) were applied. Arteries were subsequently cultured for 3 days at 37°C before the effect on pHi regulation was investigated. The degree of knock-down was assessed by TaqMan quantitative PCR (qPCR). NBCn1 mRNA levels were normalized to glyceraldehyde-3-phosphate dehydrogenase (GAPDH) mRNA levels and expressed either as the percent decrease or as the difference between the cycle number at which NBCn1 and GAPDH came up.
Finally, the effect of long-term changes in pHi (seen after omission of CO2/bicarbonate) on acute and sustained agonist-induced contractions was investigated. The latter was performed by initially constricting mouse mesenteric small arteries with 10 eol/L norepinephrine for 5 minutes followed by 2 hours of exposure to 10 eol/L norepinephrine. Five minutes after washout of norepinephrine, arteries were finally reexposed to 10 eol/L norepinephrine for 5 minutes. The initial and final contractions were compared, and the change in contractile function calculated as:
where dmax, dinitial, and dfinal denote the fully dilated diameter and the diameter during the initial and final exposure to norepinephrine, respectively.
The decrease in Ca2+-concentration was estimated on the basis of the fura-2 ratio (R) and quantified as:
where baseline denotes values measured immediately before the contractions.
An expanded Materials and Methods section can be found in the online data supplement.
Results
pHi Recovery From Acid Load
BCECF-loaded arteries were acid loaded with the NH4+-prepulse technique (Figure 1 shows original trace). Washout of NH4Cl into a Na+-free solution containing 600 eol/L amiloride was performed to quantify the Na+-independent, amiloride-insensitive pHi recovery.
Wash-in of Na+ in the continued presence of amiloride allowed the Na+-dependent, amiloride-insensitive increase in pHi recovery rate to be measured. Assuming that Na+ affects pHi solely through membrane transporters, this increase in pHi recovery rate is a direct measure of the contribution of Na+-dependent, amiloride-insensitive membrane transport to the pHi recovery from an acid load.
In the nominal absence of CO2/bicarbonate, essentially no increase in pHi recovery rate took place after wash-in of Na+ in the continued presence of amiloride (Figure 2A through 2C). In contrast, when CO2/bicarbonate was present, the pHi recovery rate increased considerably as Na+ was washed in (Figure 2D through 2F), reflecting the action of a Na+- and HCO3eC-dependent, amiloride-insensitive transporter. This increase in pHi recovery rate was present in all 3 vascular beds investigated, demonstrating that Na+-dependent bicarbonate transport is an important pHi regulatory mechanism not only in mouse mesenteric small arteries but also in coronary and cerebral small arteries.
To allow comparison of transport rates in the presence and absence of CO2/bicarbonate and at different pHi values, the buffering capacity must be taken into account. The net base influx (or equivalently net acid efflux) equals the product of the pHi recovery rate and the buffering capacity: J=dpHi/dt · t
The increase in net base influx resulting from wash-in of Na+ was calculated by subtracting the Na+-independent, amiloride-insensitive flux (dpHi/dt measured over the last 2 minutes before addition of Na+) from the flux measured after wash-in of sodium in the continued presence of amiloride (dpHi/dt measured over the first 3 minutes after addition of Na+). A significantly larger Na+-dependent, amiloride-insensitive base influx was seen in the presence of CO2/bicarbonate compared with in its absence (Figure 2G through 2I).
In the presence of CO2/bicarbonate, the Na+-dependent increase in flux was largely inhibited by the presence of 200 eol/L DIDS. This inhibition was found to be 56±10% in mesenteric arteries, 82±20% in cerebral arteries, and 91±10% in coronary arteries (Figure 2G through 2I).
Washout of amiloride activated an additional transport as the pHi recovery rate increased instantaneously (Figure 2A through 2F). This amiloride-sensitive transport is widely accepted to reflect Na+-H+ exchange (NHE). Quantification of the relative contribution of the Na+-dependent bicarbonate transport and the NHE for the pHi recovery after an acid load (Figure 2G through 2I) revealed some variability between the vascular beds, but overall the Na+-dependent bicarbonate transport seems quantitatively as important as NHE in VSMC of mouse small arteries.
Intracellular pH modulates the activity of acid/base transporters,10,21 and consequently only fluxes measured at similar pHi levels should be compared. In the aforementioned comparisons, this condition is fulfilled with the exception that pHi levels were generally lower in bicarbonate-free solution. As acid-extrusion and base-uptake mechanisms are activated when pHi decreases10,21 proton fluxes with bicarbonate are somewhat lower than what one would expect if fluxes with bicarbonate had been measured at an equally low pH as fluxes without bicarbonate. Our estimates are thus somewhat conservative with respect to the importance of bicarbonate-dependent transport.
Steady-State pHi
Our results show that Na+-dependent bicarbonate transport is important not only during intracellular acidification but also in the regulation of pHi during steady-state conditions in relaxed VSMC.
In all 3 vascular beds, we found pHi to be significantly lower in the nominal absence of CO2/bicarbonate than in its presence (Table 2).
Reverse-Transcription PCR
After having found a considerable Na+-dependent bicarbonate transport in resistance arteries from several vascular beds, we went on to further investigate the molecular identity of the transporter responsible.
Relevant molecular candidates for Na+-dependent bicarbonate transport are found in the slc4 family including 2 electrogenic Na+,HCO3eC cotransporters (NBCe1 and NBCe2), 1 electroneutral Na+,HCO3eC cotransporter (NBCn1), and 2 Na+-dependent CleC/HCO3eC exchangers (NCBE and NDCBE). In addition, the putative bicarbonate transporters AE4 and BTR1 are included in the family based on sequence homology with the other members.22
RT-PCR for each of these transporters showed that NBCn1 was the only Na+-dependent member of the slc4 family detectable at the mRNA level (Figure 3). All primer sets were tested on a positive control tissue (Figure 3) to ensure the assay was working and bands of the correct size were sequenced to verify specificity.
Whole-Mount Immunolabeling and Immunogold Electron Microscopy
Whole-mount immunolabeling was performed to confirm the presence of NBCn1 protein and determine its subcellular localization in VSMC of mouse mesenteric small arteries. These experiments showed predominant staining of VSMC plasma membranes (Figure 4A). The staining did not appear when the primary antibody was omitted or preincubated with the immunizing peptide confirming the specificity of the labeling (Figure 4B and 4C).
Additionally, immunogold electron microscopy was performed and as can be seen in Figure 4E gold particles appear most abundantly at the VSMC plasma membranes. Few gold particles seemed associated with the outer nuclear membrane, whereas no other organelles were labeled. No staining was seen when the primary antibody was omitted (not shown).
These experiments, taken together, confirm the expression of NBCn1 protein in mouse mesenteric small arteries and its localization to the plasma membrane of VSMC.
siRNA Transfections
At this point, we had demonstrated a Na+-dependent bicarbonate transport in VSMC of mouse small arteries and found NBCn1 to be the most likely molecular candidate. To directly associate NBCn1 expression with the Na+-dependent bicarbonate transport, isolated intact mouse mesenteric small arteries were transfected with siRNA to achieve specific degradation of NBCn1 mRNA.
TaqMan qPCR was performed to confirm knock-down of NBCn1. Two groups of control arteries were used: 1 where arteries were electroporated without siRNA present (nontransfected), the other where arteries were transfected with a scrambled siRNA. No significant difference (P=0.85) was found in the ratio between NBCn1 and GAPDH mRNA levels among nontransfected arteries (cycle difference=8.86±0.20; n=7) and arteries transfected with scrambled siRNA (cycle difference=8.78±0.40; n=6), and these data were therefore pooled.
As shown in Figure 5A, the ratio between NBCn1 and GAPDH mRNA levels showed a significant decrease (P<0.05) of approximately 43% in arteries transfected with siRNA against NBCn1 (n=6) compared with control arteries (n=13).
Importantly, no significant difference (P=0.86) was found between the cycle number at which GAPDH came up in control arteries and in arteries transfected with siRNA against NBCn1.
Steady-State pHi After siRNA Transfection
Steady-state pHi of relaxed VSMC in nontransfected arteries (pHi=7.21±0.06; n=7) and arteries transfected with scrambled siRNA (pHi=7.26±0.07; n=8) were not significantly different (P=0.64), and consequently these data were pooled and used as controls.
In the presence of CO2/bicarbonate, steady-state pHi of relaxed VSMC was 7.24±0.04 (n=15) in controls compared with 7.03±0.07 (n=7) in arteries transfected with siRNA against NBCn1 (P<0.05). This demonstrates the importance of NBCn1 for the inwardly directed bicarbonate transport present at steady-state pHi.
pHi Recovery From Acid Load After siRNA Transfection
No significant difference (P=0.39) in Na+-dependent, amiloride-insensitive base influx was found between nontransfected arteries (J=7.45±1.25 mmol/L per minute; n=5) and arteries transfected with scrambled siRNA (J=9.13±1.26 mmol/L per minute; n=8). These data were therefore pooled and used as controls.
Arteries transfected with siRNA against NBCn1 recovered much slower from an acid load than the control arteries (Figure 5B). Thus, in arteries transfected with siRNA against NBCn1, the Na+-dependent, amiloride-insensitive base influx in bicarbonate-containing solution (J=2.76±0.78 mmol/L per minute; n=6) was significantly decreased (P<0.001) compared with control arteries (J=8.48±0.91 mmol/L per minute; n=13). The knock-down of NBCn1 resulted in a 68±10% decrease in the Na+-dependent, amiloride-insensitive base influx (Figure 5C).
The base influx found in control arteries was not significantly different from the base influx found in freshly isolated mesenteric small arteries (P=0.91).
Effect of CO2/Bicarbonate on Contractility
An important reason for investigating the regulation of pHi in VSMC is the potential effect of changes in pHi on vascular contractility. Previous studies have focused primarily on acute changes in pHi,23 which are, however, unlikely to predominate in vivo. We focused on the effect of long-term changes in pHi by testing the effect of CO2/bicarbonate omission on acute and sustained contractile responses to norepinephrine.
Acute Agonist-Induced Contractions
The acute contractile response to norepinephrine was investigated in mouse mesenteric small arteries, and we found that pHi decreased equally in the presence (0.08±0.04 pH units; n=3) and in the absence (0.08±0.02 pH units; n=3) of CO2/bicarbonate. Consequently, the pHi difference observed in the relaxed arteries persisted during acute contractions; however, no significant difference in the contractile response to norepinephrine was observed (Figure 6A). This was the case also when arteries were mounted for isometric contractions (Figure 6B).
Sustained Norepinephrine-Induced Contractions
When mouse mesenteric small arteries were exposed to norepinephrine for 2 hours in the presence of CO2/bicarbonate, pHi did not change significantly (pHi=0.04±0.05; n=3). In the absence of CO2/bicarbonate, pHi fell significantly (P<0.05) during the 2 hours (pHi=eC0.11±0.03; n=3).
As seen in Figure 6C and 6D, sustained exposure to norepinephrine resulted in diminished responsiveness to the agonist. The loss of contractile response was significantly larger (P<0.01) for arteries kept in bicarbonate-free solution (31±4%) compared with arteries kept in bicarbonate-containing solution (14±4%). The difference was quantified at the end of the 5-minute contraction and could not be explained by differences in the calcium response. A general decrease in cytosolic Ca2+ concentration (Figure 6E and 6F) was apparent in the presence as well as in the absence of CO2/bicarbonate, but no significant difference (P=0.70) was observed between arteries kept in bicarbonate-free (36±12%) and bicarbonate-containing solution (42±8%). When arteries were kept for 2 hours in bicarbonate-free solution without norepinephrine present, the contractility of the arteries did not change (loss of contractile response=0±3%; n=3).
The relative diameter before the final exposure to norepinephrine was significantly smaller for arteries kept in bicarbonate-containing solution than for arteries kept in bicarbonate-free solution (Figure 6C), indicating increased myogenic tone. In the presence of CO2/bicarbonate, the relative diameter was also significantly smaller than before the initial contraction. This increase in basal tone took place despite unchanged basal [Ca2+]i as the basal fura-2 ratio after the 2-hour exposure was 1.00±0.09 compared with 0.96±0.09 before the 2-hour exposure (P=0.59).
At the end of the experiments, the response to 5 mU/mL vasopressin was tested, and no significant difference (P=0.19) between arteries kept in the presence (relative diameter=0.60±0.02; n=9) and nominal absence (relative diameter=0.64±0.03; n=11) of CO2/bicarbonate was found at the end of the 5-minute contraction.
Discussion
Regulation of pHi in VSMC depends on several acid/base transporters in the sarcolemma. In the present study, we investigated bicarbonate-dependent regulation of pHi in VSMC of mesenteric small arteries to build on previous work4,17 and extend the investigations to include coronary and cerebral small arteries because (1) these are clinically relevant arteries; (2) nothing is known about Na+,HCO3eC cotransport (NBC) in these arteries; and (3) it allows us to evaluate the general importance of NBC in the vasculature. Previous work has shown that during intracellular acidification, NHE plays an important role,10,24,25 and NHE1 is the only NHE isoform found in VSMC.15 In rat mesenteric small arteries, evidence has been provided for an electroneutral Na+- and HCO3eC-dependent transporter that seems to take particular importance during contraction.4,17
We demonstrate here the importance of NBC in mouse small arteries and show that during intracellular acidification, this transport has a quantitative importance of similar magnitude as the NHE. In mesenteric, coronary, and cerebral small arteries, the NBC-mediated transport after an NH4+ prepulse (Figure 2G through 2I) represented 127%, 189%, and 64% of the NHE-mediated transport, respectively, suggesting that Na+-dependent bicarbonate transport is of general importance in the vasculature.
For the first time, we show that NBCn1 in VSMC is the only known candidate transporter detectable at the mRNA level, that NBCn1 protein is present in the plasma membranes of VSMC in mouse mesenteric small arteries, and that knock-down of NBCn1 in mouse mesenteric small arteries leads to a dramatic decrease in the Na+-dependent base influx and in steady-state pHi. Taken together, this provides strong evidence for NBCn1 as the quantitatively important Na+-dependent bicarbonate transporter in mouse small arteries.
A particular strength of the current evidence is the direct link between the investigated protein and its function. Certain functions of intact tissues have in the past often been ascribed to specific proteins based on indirect evidence, but the development of new techniques (siRNA, antisense, dominant-negative proteins, among others) enables us to draw much firmer conclusions. With the siRNA technique, we managed for the first time to show a direct link between a membrane acid/base transporter and the pHi regulatory function.
In this study, we found that relaxed VSMC in the wall of mouse mesenteric, coronary, and cerebral small arteries present with much lower steady-state pHi when CO2/bicarbonate is omitted from the bath solution. The observation that knock-down of NBCn1 results in lower steady-state pHi is consistent with a role for Na+-dependent bicarbonate transport in this phenomenon.
Previous studies on rat mesenteric arteries4 and guinea pig femoral arteries18 reported no difference in steady-state pHi between relaxed VSMC in the presence and absence of CO2/bicarbonate. Lower steady-state pHi was, however, found in contracting rat mesenteric small arteries when CO2/bicarbonate was omitted.4
Studies on isolated VSMC and VSMC-derived cell cultures have given variable results,18 with different studies showing either increased pHi,25,26 decreased pHi,27,28 or no effect on pHi when CO2/bicarbonate was omitted from the solution.
One likely explanation for the different effects of CO2/bicarbonate omission is differences in net fluxes of bicarbonate. In relaxed VSMC of mouse small arteries, the increased pHi in the presence of CO2/bicarbonate would then suggest net bicarbonate influx, which in steady-state is accompanied by additional net CO2 efflux. Although other explanations may also exist, the results stress the importance of performing in vitro experiments in CO2/bicarbonate-buffered solutions.
Although the data strongly point to NBCn1 as the only important Na+-dependent bicarbonate transporter in VSMC, 1 interesting observation remains unexplained. The DIDS sensitivity of NBCn1 (and its human ortholog NBC3) has previously been investigated with the transporter heterologously expressed in Xenopus oocytes and HEK293 cells. In these experiments, even high concentrations of DIDS (500 eol/L to 1 mmol/L) had little effect on the NBC.29eC31 Similar results have since been found in isolated perfused tubules (medullary thick ascending limb)32 and choroid plexus,33 where the Na+-dependent bicarbonate transport proposed to be mediated by NBCn1 was found to be virtually insensitive to DIDS.
In mouse small arteries, we find a Na+-dependent bicarbonate transport that is largely inhibited by DIDS. This is consistent with earlier studies on rat mesenteric small arteries, where the transport was inhibited approximately 75% by DIDS.17 In the present study, the Na+-dependent transport is moreover greatly reduced when NBCn1 is knocked down by siRNA transfection. Intriguingly, this provides evidence for tissue-specific differences in NBCn1 DIDS sensitivity, which could potentially be a result of tissue-specific splice variation, posttranslational modification, or interaction with tissue specific proteins.
At the extracellular end of putative TM5, the DIDS-sensitive members of the slc4 family have a putative DIDS-binding site (KXXK). In NBCn1 this site is disrupted (KLFH) which has been proposed to account for the DIDS insensitivity.22 We amplified this part of the NBCn1 mRNA from mouse mesenteric small arteries and confirmed the disruption by sequencing (unpublished observation, 2004).
The precise magnitude of NBCn1 DIDS sensitivity in VSMC is difficult to determine, as neither the addition of DIDS nor the knock-down of NBCn1 leads to a complete abolition of the Na+-dependent bicarbonate transport.
Previous studies on VSMC in primary culture from dog femoral arteries suggested the presence of a Na+-dependent CleC/HCO3eC exchanger,24 whereas NBCn1 was found to be independent of chloride when heterologously expressed in Xenopus oocytes.29
In the present study, it was not possible, by RT-PCR, to find evidence for any of the known Na+-dependent CleC/HCO3eC exchangers, and in earlier work on freshly isolated intact rat mesenteric small arteries, the Na+-dependent bicarbonate transport was demonstrated to be independent of chloride.17 Our results in small arterial VSMC are consistent with this result. The differences observed may be explained by species- or tissue-specific differences or by the fact that expression of proteins, and therefore the phenotype of cells, may change during culture.
The marked effect of CO2/bicarbonate on pHi of VSMC revealed a unique opportunity for investigating the effect of isolated changes in pHi and hence the potential importance of NBCn1 for VSMC contractility. Earlier studies have focused on the effect of acute changes in pHi, and in those experiments, the general finding has been an increase in the agonist-induced tone when pHi was experimentally lowered.23 In the present study, pHi was changed for at least 1 hour before the agonist-induced response was investigated, and we were unable to detect any effect of pHi on acute norepinephrine-induced contractions. One possible explanation is that the increase in vascular tone following acute intracellular acidification is a result of the change in pHi rather than the sustained low pHi.
In contrast, however, we find a pronounced effect on norepinephrine-induced contractions after long-term acidification (or omission of CO2/bicarbonate) in the presence of norepinephrine. Arteries are under a constant sympathetic tone, and with the importance of arterial diameter for vascular resistance, the decline in contractile response could have a significant impact on the distribution of blood flow during sustained intracellular acidosis.
The different contractile responses in the presence and nominal absence of CO2/bicarbonate could not be explained by different Ca2+ responses. This makes an effect via receptor downregulation unlikely and may suggest an interaction with the Ca2+-sensitizing machinery. Arteries kept in bicarbonate-containing solution apparently increase their Ca2+ sensitivity during sustained exposure to norepinephrine and, in this way, may compensate for the substantial decrease in the Ca2+ response and minimize the effect on the contraction. Under bicarbonate-free conditions, this sensitization seems to be largely abolished, as the contractile response attenuates much faster despite more or less similar declines in the cytosolic Ca2+ concentration.
An increase in Ca2+ sensitivity after sustained exposure to norepinephrine in the presence of CO2/bicarbonate could also explain why these arteries developed augmented myogenic tone, a phenomenon that did not appear when arteries were exposed to norepinephrine in the nominal absence of CO2/bicarbonate.
More work is needed to clarify the underlying mechanism, but 1 important purpose of bicarbonate-dependent pHi regulation in VSMC may be the maintenance of contractile ability during sustained agonist exposure.
In conclusion, we show that NBCn1 (slc4a7) mediates the Na+-dependent bicarbonate transport important for pHi recovery after acid load as well as for regulation of steady-state pHi in VSMC of mouse small arteries. This transport is largely inhibited by DIDS and is of an overall magnitude similar to the amiloride-sensitive transport.
Considering the importance of pHi for protein function in general and for VSMC growth and contractility in particular, this work significantly increases our understanding of vascular physiology. This understanding may also be important for the comprehension of pathological conditions associated with pH abnormalities.
Acknowledgments
The Water and Salt Research Center at the University of Aarhus was established and is supported by the Danish National Research Foundation (Danmarks Grundforskningsfond).
This work was supported by the Danish Medical Research Council (grant no. 22-04-0044). We thank Per Larsen and Finn Skou Pedersen, The siRNA Delivery Center, University of Aarhus, for discussions about the siRNA-technique. We thank Joergen Andresen, Susie Mogensen, and Edith Bjoern Moeller for technical assistance.
References
Aalkjaer C, Mulvany MJ. Effect of changes in intracellular pH on the contractility of rat resistance vessels. Prog Biochem Pharmacol. 1988; 23: 150eC158.
Tian R, Vogel P, Lassen NA, Mulvany MJ, Andreasen F, Aalkjaer C. Role of extracellular and intracellular acidosis for hypercapnia-induced inhibition of tension of isolated rat cerebral arteries. Circ Res. 1995; 76: 269eC275.
Apkon M, Weed RA, Boron WF. Motor responses of cultured rat cerebral vascular smooth muscle cells to intra- and extracellular pH changes. Am J Physiol. 1997; 273: H434eCH445.
Aalkjaer C, Cragoe EJ. Intracellular pH regulation in resting and contracting segments of rat mesenteric resistance vessels. J Physiol. 1988; 402: 391eC410.
Naderali EK, Buttell N, Taggart MJ, Bullock AJ, Eisner DA, Wray S. The role of the sarcolemmal Ca2+-ATPase in the pH transients associated with contraction in rat smooth muscle. J Physiol. 1997; 505: 329eC336.
Berk BC, Vallega G, Muslin AJ, Gordon HM, Canessa M, Alexander RW. Spontaneously hypertensive rat vascular smooth muscle cells in culture exhibit increased growth and Na+/H+ exchange. J Clin Invest. 1989; 83: 822eC829.
Schubert R, Krien U, Gagov H. Protons inhibit the BKCa channel of rat small artery smooth muscle cells. J Vasc Res. 2001; 38: 30eC38.
Wang X, Wu J, Li L, Chen F, Wang R, Jiang C. Hypercapnic acidosis activates KATP channels in vascular smooth muscles. Circ Res. 2003; 92: 1225eC1232.
Davis JP, Chien PF, Chipperfield AR, Gordon A, Harper AA. The three mechanisms of intracellular chloride accumulation in vascular smooth muscle of human umbilical and placental arteries. Pflugers Arch. 2000; 441: 150eC154.
Kahn AM, Cragoe EJ Jr, Allen JC, Seidel CL, Shelat H. Effects of pHi on Na+-H+, Na+-dependent, and Na+-independent CleC-HCO3eC exchangers in vascular smooth muscle. Am J Physiol. 1991; 261: C837eCC844.
Kopito RR, Lodish HF. Primary structure and transmembrane orientation of the murine anion exchange protein. Nature. 1985; 316: 234eC238.
Brosius FC 3rd, Pisoni RL, Cao X, Deshmukh G, Yannoukakos D, Stuart-Tilley AK, Haller C, Alper SL. AE anion exchanger mRNA and protein expression in vascular smooth muscle cells, aorta, and renal microvessels. Am J Physiol. 1997; 273: F1039eCF1047.
Fliegel L. The Na+/H+ exchanger isoform 1. Int J Biochem Cell Biol. 2005; 37: 33eC37.
Boyarsky G, Ganz MB, Cragoe EJ Jr, Boron WF. Intracellular-pH dependence of Na-H exchange and acid loading in quiescent and arginine vasopressin-activated mesangial cells. Proc Natl Acad Sci U S A. 1990; 87: 5921eC5924.
Kalaria RN, Premkumar DR, Lin CW, Kroon SN, Bae JY, Sayre LM, LaManna JC. Identification and expression of the Na+/H+ exchanger in mammalian cerebrovascular and choroidal tissues: characterization by amiloride-sensitive [3H]MIA binding and RT-PCR analysis. Brain Res Mol Brain Res. 1998; 58: 178eC187.
Bobulescu IA, Di Sole F, Moe OW. Na+/H+ exchangers: physiology and link to hypertension and organ ischemia. Curr Opin Nephrol Hypertens. 2005; 14: 485eC494.
Aalkjaer C, Hughes A. Chloride and bicarbonate transport in rat resistance arteries. J Physiol. 1991; 436: 57eC73.
Aickin CC. Regulation of intracellular pH in smooth muscle cells of the guinea-pig femoral artery. J Physiol. 1994; 479: 331eC340.
Damkier HH, Nielsen S, Praetorius J. An Anti-N-terminal antibody localizes the NBCn1 to heart endothelia, skeletal and vascular smooth muscle cells. Am J Physiol Heart Circ Physiol. 2006; 290: H172eCH180.
Martin JB, Young JL, Benoit JN, Dean DA. Gene transfer to intact mesenteric arteries by electroporation. J Vasc Res. 2000; 37: 372eC380.
Leem CH, Lagadic-Gossmann D, Vaughan-Jones RD. Characterization of intracellular pH regulation in the guinea-pig ventricular myocyte. J Physiol. 1999; 517: 159eC180.
Romero MF, Fulton CM, Boron WF. The SLC4 family of HCO3eC transporters. Pflugers Arch. 2004; 447: 495eC509.
Smith GL, Austin C, Crichton C, Wray S. A review of the actions and control of intracellular pH in vascular smooth muscle. Cardiovasc Res. 1998; 38: 316eC331.
Kahn AM, Cragoe EJ Jr, Allen JC, Halligan RD, Shelat H. Na+-H+ and Na+-dependent CleC-HCO3eC exchange control pHi in vascular smooth muscle. Am J Physiol. 1990; 259: C134eCC143.
Neylon CB, Little PJ, Cragoe EJ Jr, Bobik A. Intracellular pH in human arterial smooth muscle. Regulation by Na+/H+ exchange and a novel 5-(N-ethyl-N-isopropyl)amiloride-sensitive Na+- and HCO3eC-dependent mechanism. Circ Res. 1990; 67: 814eC825.
Vigne P, Breittmayer JP, Frelin C, Lazdunski M. Dual control of the intracellular pH in aortic smooth muscle cells by a cAMP-sensitive HCO3eC/CleC antiporter and a protein kinase C-sensitive Na+/H+ antiporter. J Biol Chem. 1988; 263: 18023eC18029.
Korbmacher C, Helbig H, Stahl F, Wiederholt M. Evidence for Na/H exchange and Cl/HCO3 exchange in A10 vascular smooth muscle cells. Pflugers Arch. 1988; 412: 29eC36.
Putnam RW, Grubbs RD. Steady-state pHi, buffering power, and effect of CO2 in a smooth muscle-like cell line. Am J Physiol. 1990; 258: C461eCC469.
Choi I, Aalkjaer C, Boulpaep EL, Boron WF. An electroneutral sodium/bicarbonate cotransporter NBCn1 and associated sodium channel. Nature. 2000; 405: 571eC575.
Pushkin A, Abuladze N, Lee I, Newman D, Hwang J, Kurtz I. Cloning, tissue distribution, genomic organization, and functional characterization of NBC3, a new member of the sodium bicarbonate cotransporter family. J Biol Chem. 1999; 274: 16569eC16575.
Park M, Ko SB, Choi JY, Muallem G, Thomas PJ, Pushkin A, Lee MS, Kim JY, Lee MG, Muallem S, Kurtz I. The cystic fibrosis transmembrane conductance regulator interacts with and regulates the activity of the HCO3eC salvage transporter human Na+-HCO3eC cotransport isoform 3. J Biol Chem. 2002; 277: 50503eC50509.
Odgaard E, Jakobsen JK, Frische S, Praetorius J, Nielsen S, Aalkjaer C, Leipziger J. Basolateral Na+-dependent HCO3eC transporter NBCn1-mediated HCO3eC influx in rat medullary thick ascending limb. J Physiol. 2004; 555: 205eC218.
Bouzinova EV, Praetorius J, Virkki LV, Nielsen S, Boron WF, Aalkjaer C. The Na+ dependent HCO3eC uptake into the rat choroid plexus epithelium is partially DIDS-sensitive. Am J Physiol Cell Physiol. 2005; 289: C1448eCC1456.(Ebbe Boedtkjer, Jeppe Pra)
Department of Physiology (E.B., C.A.), Institute of Physiology and Biophysics
Institute of Anatomy (J.P.), University of Aarhus, Denmark.
Abstract
The contribution of sodium-dependent bicarbonate transport to intracellular pH (pHi) regulation in vascular smooth muscle cells is controversial, partly because the molecular identity of the transporter(s) responsible has not been identified. Here, using the pH-sensitive fluorophore 2',7'-bis-(2-carboxyethyl)-5-(and-6)-carboxyfluorescein (BCECF), we show that smooth muscle cells of intact mouse mesenteric, coronary, and cerebral small arteries all display a sodium- and bicarbonate-dependent pHi recovery after an NH4+-prepulse. The sodium-dependent bicarbonate flux was largely 4,4'-diisothiocyanatostilbene-2,2'-disulphonic acid (DIDS) sensitive (56% to 91%) and of a magnitude similar to the amiloride-sensitive flux. Additionally, steady-state pHi was lower (0.2 to 0.4 pH units magnitude) in all 3 vascular beds when CO2/bicarbonate was omitted. RT-PCR analyses showed that NBCn1 (slc4a7) is the only Na+-dependent bicarbonate transporter of the slc4 family detectable at the mRNA level in all 3 vascular beds investigated. Whole-mount immunolabeling and immunogold electron microscopy confirmed the presence of NBCn1 protein in the sarcolemma of mouse mesenteric small arterial smooth muscle cells. Intact mouse mesenteric small arteries were electropermeated to facilitate transfection with small interfering RNA targeting NBCn1, which resulted in an approximate 43% decrease in the ratio of NBCn1 to glyceraldehyde-3-phosphate dehydrogenase mRNA. After knock-down, we found a decreased steady-state pHi (0.21±0.08 pH units) as well as a 68±10% decrease in the net Na+-dependent, amiloride-insensitive base influx after acid load. Finally, omission of CO2/bicarbonate resulted in a decreased contractile response to norepinephrine after sustained exposure to the agonist, underlining the importance of CO2/bicarbonate for vascular contractility. We conclude that NBCn1 mediates the Na+-dependent bicarbonate transport important for pHi regulation in smooth muscle cells of mouse mesenteric, coronary, and cerebral small arteries.
Key Words: vascular smooth muscle cells bicarbonate intracellular pH siRNA NBC
Introduction
Intracellular pH (pHi) of vascular smooth muscle cells (VSMC) is tightly related to fundamental vascular functions. In resistance arteries acute intracellular acidification leads to contraction,1eC3 and during force development, pHi of VSMC has a tendency to decline.4,5 Intracellular pH is reported to modulate VSMC growth,6 and in general pH affects the conformation and function of many proteins, such as KATP and BK channels, present in VSMC.7,8 Consequently, the control of pHi is essential to maintain normal vascular function.
Transmembrane movement of acid/base equivalents is important for pHi regulation in VSMC and 2 well-characterized transporters are known to contribute to the regulation. The first discovered of these is the anion exchanger, which mediates an electroneutral exchange of CleC for HCO3eC. In VSMC, this transporter is involved in maintaining a high intracellular chloride concentration,9 and during intracellular alkalinization it is activated to participate in the regulation of pHi.10 The first anion exchanger (AE1) was cloned 20 years ago from erythrocytes,11 and in VSMC the transporter is also well defined at the molecular level: AE2 is predominant together with smaller amounts of AE3.12
The other well-described transporter is the Na+-H+ exchanger, which extrudes protons in exchange for sodium. This transporter is activated during intracellular acidification and participates in the recovery of pHi under such conditions.10,13,14 In VSMC, the transporter is well defined at the molecular level as NHE1.15 Interestingly, Na+-H+ exchangers have been reported to be implicated in hypertension, further underlining the importance of pHi regulation for normal cardiovascular function.16
A third transport system may be important for pHi regulation in VSMC. A Na+-dependent bicarbonate transport seems to be present in rat mesenteric small arteries,4,17 but its importance in the vasculature as a whole has been questioned.18 Furthermore, at present, no molecular information exists about the transporter(s) responsible. Considering the importance of pHi regulation in VSMC, this reveals a major gap in our current knowledge.
The aim of the present study was to examine the general importance of Na+-dependent bicarbonate transport in VSMC and determine the molecular identity of the transporter(s) responsible.
Materials and Methods
Mesenteric, coronary, and cerebral small arteries were isolated from male Naval Medical Research Institute (NMRI) mice (7 to 10 weeks old), mounted for isobaric or isometric investigation and loaded with the pH-sensitive fluorophore 2',7'-bis-(2-carboxyethyl)-5-(and-6)-carboxyfluorescein (BCECF) or the Ca2+-sensitive fluorophore fura-2. The animals were bred in the institute’s own breeding facility. The care and use of the animals are described in the online data supplement, available at http://circres.ahajournals.org.
Arteries were experimentally acidified with the NH4+-prepulse technique, and the rate of pHi recovery was measured in the presence and absence of CO2/bicarbonate, Na+, amiloride (600 eol/L), and 4,4'-diisothiocyanatostilbene-2,2'-disulphonic acid (DIDS) (200 eol/L).
The expression of known and putative Na+,HCO3eC cotransporters of the slc4 family was investigated by RT-PCR (primer sequences in Table 1). Gel electrophoresis of PCR products was performed and products of the expected sizes were sequenced to confirm specificity.
The expression of NBCn1 at the protein level was investigated by whole-mount immunolabeling and immunogold electron microscopy using a newly developed N-terminal NBCn1 primary antibody19 in combination with an Alexa488- or gold-labeled secondary antibody.
Intact isolated mouse mesenteric small arteries were transfected with small interfering RNA (siRNA) using electropermeation. A modification of the protocol previously described for DNA plasmid transfection20 was used. Arteries were kept in electroporation buffer containing 50 nmol/L siRNA (targeting NBCn1 or scrambled) and 8 "square-wave" pulses (125 V/cm, 10-ms duration and 100-ms interpulse interval) were applied. Arteries were subsequently cultured for 3 days at 37°C before the effect on pHi regulation was investigated. The degree of knock-down was assessed by TaqMan quantitative PCR (qPCR). NBCn1 mRNA levels were normalized to glyceraldehyde-3-phosphate dehydrogenase (GAPDH) mRNA levels and expressed either as the percent decrease or as the difference between the cycle number at which NBCn1 and GAPDH came up.
Finally, the effect of long-term changes in pHi (seen after omission of CO2/bicarbonate) on acute and sustained agonist-induced contractions was investigated. The latter was performed by initially constricting mouse mesenteric small arteries with 10 eol/L norepinephrine for 5 minutes followed by 2 hours of exposure to 10 eol/L norepinephrine. Five minutes after washout of norepinephrine, arteries were finally reexposed to 10 eol/L norepinephrine for 5 minutes. The initial and final contractions were compared, and the change in contractile function calculated as:
where dmax, dinitial, and dfinal denote the fully dilated diameter and the diameter during the initial and final exposure to norepinephrine, respectively.
The decrease in Ca2+-concentration was estimated on the basis of the fura-2 ratio (R) and quantified as:
where baseline denotes values measured immediately before the contractions.
An expanded Materials and Methods section can be found in the online data supplement.
Results
pHi Recovery From Acid Load
BCECF-loaded arteries were acid loaded with the NH4+-prepulse technique (Figure 1 shows original trace). Washout of NH4Cl into a Na+-free solution containing 600 eol/L amiloride was performed to quantify the Na+-independent, amiloride-insensitive pHi recovery.
Wash-in of Na+ in the continued presence of amiloride allowed the Na+-dependent, amiloride-insensitive increase in pHi recovery rate to be measured. Assuming that Na+ affects pHi solely through membrane transporters, this increase in pHi recovery rate is a direct measure of the contribution of Na+-dependent, amiloride-insensitive membrane transport to the pHi recovery from an acid load.
In the nominal absence of CO2/bicarbonate, essentially no increase in pHi recovery rate took place after wash-in of Na+ in the continued presence of amiloride (Figure 2A through 2C). In contrast, when CO2/bicarbonate was present, the pHi recovery rate increased considerably as Na+ was washed in (Figure 2D through 2F), reflecting the action of a Na+- and HCO3eC-dependent, amiloride-insensitive transporter. This increase in pHi recovery rate was present in all 3 vascular beds investigated, demonstrating that Na+-dependent bicarbonate transport is an important pHi regulatory mechanism not only in mouse mesenteric small arteries but also in coronary and cerebral small arteries.
To allow comparison of transport rates in the presence and absence of CO2/bicarbonate and at different pHi values, the buffering capacity must be taken into account. The net base influx (or equivalently net acid efflux) equals the product of the pHi recovery rate and the buffering capacity: J=dpHi/dt · t
The increase in net base influx resulting from wash-in of Na+ was calculated by subtracting the Na+-independent, amiloride-insensitive flux (dpHi/dt measured over the last 2 minutes before addition of Na+) from the flux measured after wash-in of sodium in the continued presence of amiloride (dpHi/dt measured over the first 3 minutes after addition of Na+). A significantly larger Na+-dependent, amiloride-insensitive base influx was seen in the presence of CO2/bicarbonate compared with in its absence (Figure 2G through 2I).
In the presence of CO2/bicarbonate, the Na+-dependent increase in flux was largely inhibited by the presence of 200 eol/L DIDS. This inhibition was found to be 56±10% in mesenteric arteries, 82±20% in cerebral arteries, and 91±10% in coronary arteries (Figure 2G through 2I).
Washout of amiloride activated an additional transport as the pHi recovery rate increased instantaneously (Figure 2A through 2F). This amiloride-sensitive transport is widely accepted to reflect Na+-H+ exchange (NHE). Quantification of the relative contribution of the Na+-dependent bicarbonate transport and the NHE for the pHi recovery after an acid load (Figure 2G through 2I) revealed some variability between the vascular beds, but overall the Na+-dependent bicarbonate transport seems quantitatively as important as NHE in VSMC of mouse small arteries.
Intracellular pH modulates the activity of acid/base transporters,10,21 and consequently only fluxes measured at similar pHi levels should be compared. In the aforementioned comparisons, this condition is fulfilled with the exception that pHi levels were generally lower in bicarbonate-free solution. As acid-extrusion and base-uptake mechanisms are activated when pHi decreases10,21 proton fluxes with bicarbonate are somewhat lower than what one would expect if fluxes with bicarbonate had been measured at an equally low pH as fluxes without bicarbonate. Our estimates are thus somewhat conservative with respect to the importance of bicarbonate-dependent transport.
Steady-State pHi
Our results show that Na+-dependent bicarbonate transport is important not only during intracellular acidification but also in the regulation of pHi during steady-state conditions in relaxed VSMC.
In all 3 vascular beds, we found pHi to be significantly lower in the nominal absence of CO2/bicarbonate than in its presence (Table 2).
Reverse-Transcription PCR
After having found a considerable Na+-dependent bicarbonate transport in resistance arteries from several vascular beds, we went on to further investigate the molecular identity of the transporter responsible.
Relevant molecular candidates for Na+-dependent bicarbonate transport are found in the slc4 family including 2 electrogenic Na+,HCO3eC cotransporters (NBCe1 and NBCe2), 1 electroneutral Na+,HCO3eC cotransporter (NBCn1), and 2 Na+-dependent CleC/HCO3eC exchangers (NCBE and NDCBE). In addition, the putative bicarbonate transporters AE4 and BTR1 are included in the family based on sequence homology with the other members.22
RT-PCR for each of these transporters showed that NBCn1 was the only Na+-dependent member of the slc4 family detectable at the mRNA level (Figure 3). All primer sets were tested on a positive control tissue (Figure 3) to ensure the assay was working and bands of the correct size were sequenced to verify specificity.
Whole-Mount Immunolabeling and Immunogold Electron Microscopy
Whole-mount immunolabeling was performed to confirm the presence of NBCn1 protein and determine its subcellular localization in VSMC of mouse mesenteric small arteries. These experiments showed predominant staining of VSMC plasma membranes (Figure 4A). The staining did not appear when the primary antibody was omitted or preincubated with the immunizing peptide confirming the specificity of the labeling (Figure 4B and 4C).
Additionally, immunogold electron microscopy was performed and as can be seen in Figure 4E gold particles appear most abundantly at the VSMC plasma membranes. Few gold particles seemed associated with the outer nuclear membrane, whereas no other organelles were labeled. No staining was seen when the primary antibody was omitted (not shown).
These experiments, taken together, confirm the expression of NBCn1 protein in mouse mesenteric small arteries and its localization to the plasma membrane of VSMC.
siRNA Transfections
At this point, we had demonstrated a Na+-dependent bicarbonate transport in VSMC of mouse small arteries and found NBCn1 to be the most likely molecular candidate. To directly associate NBCn1 expression with the Na+-dependent bicarbonate transport, isolated intact mouse mesenteric small arteries were transfected with siRNA to achieve specific degradation of NBCn1 mRNA.
TaqMan qPCR was performed to confirm knock-down of NBCn1. Two groups of control arteries were used: 1 where arteries were electroporated without siRNA present (nontransfected), the other where arteries were transfected with a scrambled siRNA. No significant difference (P=0.85) was found in the ratio between NBCn1 and GAPDH mRNA levels among nontransfected arteries (cycle difference=8.86±0.20; n=7) and arteries transfected with scrambled siRNA (cycle difference=8.78±0.40; n=6), and these data were therefore pooled.
As shown in Figure 5A, the ratio between NBCn1 and GAPDH mRNA levels showed a significant decrease (P<0.05) of approximately 43% in arteries transfected with siRNA against NBCn1 (n=6) compared with control arteries (n=13).
Importantly, no significant difference (P=0.86) was found between the cycle number at which GAPDH came up in control arteries and in arteries transfected with siRNA against NBCn1.
Steady-State pHi After siRNA Transfection
Steady-state pHi of relaxed VSMC in nontransfected arteries (pHi=7.21±0.06; n=7) and arteries transfected with scrambled siRNA (pHi=7.26±0.07; n=8) were not significantly different (P=0.64), and consequently these data were pooled and used as controls.
In the presence of CO2/bicarbonate, steady-state pHi of relaxed VSMC was 7.24±0.04 (n=15) in controls compared with 7.03±0.07 (n=7) in arteries transfected with siRNA against NBCn1 (P<0.05). This demonstrates the importance of NBCn1 for the inwardly directed bicarbonate transport present at steady-state pHi.
pHi Recovery From Acid Load After siRNA Transfection
No significant difference (P=0.39) in Na+-dependent, amiloride-insensitive base influx was found between nontransfected arteries (J=7.45±1.25 mmol/L per minute; n=5) and arteries transfected with scrambled siRNA (J=9.13±1.26 mmol/L per minute; n=8). These data were therefore pooled and used as controls.
Arteries transfected with siRNA against NBCn1 recovered much slower from an acid load than the control arteries (Figure 5B). Thus, in arteries transfected with siRNA against NBCn1, the Na+-dependent, amiloride-insensitive base influx in bicarbonate-containing solution (J=2.76±0.78 mmol/L per minute; n=6) was significantly decreased (P<0.001) compared with control arteries (J=8.48±0.91 mmol/L per minute; n=13). The knock-down of NBCn1 resulted in a 68±10% decrease in the Na+-dependent, amiloride-insensitive base influx (Figure 5C).
The base influx found in control arteries was not significantly different from the base influx found in freshly isolated mesenteric small arteries (P=0.91).
Effect of CO2/Bicarbonate on Contractility
An important reason for investigating the regulation of pHi in VSMC is the potential effect of changes in pHi on vascular contractility. Previous studies have focused primarily on acute changes in pHi,23 which are, however, unlikely to predominate in vivo. We focused on the effect of long-term changes in pHi by testing the effect of CO2/bicarbonate omission on acute and sustained contractile responses to norepinephrine.
Acute Agonist-Induced Contractions
The acute contractile response to norepinephrine was investigated in mouse mesenteric small arteries, and we found that pHi decreased equally in the presence (0.08±0.04 pH units; n=3) and in the absence (0.08±0.02 pH units; n=3) of CO2/bicarbonate. Consequently, the pHi difference observed in the relaxed arteries persisted during acute contractions; however, no significant difference in the contractile response to norepinephrine was observed (Figure 6A). This was the case also when arteries were mounted for isometric contractions (Figure 6B).
Sustained Norepinephrine-Induced Contractions
When mouse mesenteric small arteries were exposed to norepinephrine for 2 hours in the presence of CO2/bicarbonate, pHi did not change significantly (pHi=0.04±0.05; n=3). In the absence of CO2/bicarbonate, pHi fell significantly (P<0.05) during the 2 hours (pHi=eC0.11±0.03; n=3).
As seen in Figure 6C and 6D, sustained exposure to norepinephrine resulted in diminished responsiveness to the agonist. The loss of contractile response was significantly larger (P<0.01) for arteries kept in bicarbonate-free solution (31±4%) compared with arteries kept in bicarbonate-containing solution (14±4%). The difference was quantified at the end of the 5-minute contraction and could not be explained by differences in the calcium response. A general decrease in cytosolic Ca2+ concentration (Figure 6E and 6F) was apparent in the presence as well as in the absence of CO2/bicarbonate, but no significant difference (P=0.70) was observed between arteries kept in bicarbonate-free (36±12%) and bicarbonate-containing solution (42±8%). When arteries were kept for 2 hours in bicarbonate-free solution without norepinephrine present, the contractility of the arteries did not change (loss of contractile response=0±3%; n=3).
The relative diameter before the final exposure to norepinephrine was significantly smaller for arteries kept in bicarbonate-containing solution than for arteries kept in bicarbonate-free solution (Figure 6C), indicating increased myogenic tone. In the presence of CO2/bicarbonate, the relative diameter was also significantly smaller than before the initial contraction. This increase in basal tone took place despite unchanged basal [Ca2+]i as the basal fura-2 ratio after the 2-hour exposure was 1.00±0.09 compared with 0.96±0.09 before the 2-hour exposure (P=0.59).
At the end of the experiments, the response to 5 mU/mL vasopressin was tested, and no significant difference (P=0.19) between arteries kept in the presence (relative diameter=0.60±0.02; n=9) and nominal absence (relative diameter=0.64±0.03; n=11) of CO2/bicarbonate was found at the end of the 5-minute contraction.
Discussion
Regulation of pHi in VSMC depends on several acid/base transporters in the sarcolemma. In the present study, we investigated bicarbonate-dependent regulation of pHi in VSMC of mesenteric small arteries to build on previous work4,17 and extend the investigations to include coronary and cerebral small arteries because (1) these are clinically relevant arteries; (2) nothing is known about Na+,HCO3eC cotransport (NBC) in these arteries; and (3) it allows us to evaluate the general importance of NBC in the vasculature. Previous work has shown that during intracellular acidification, NHE plays an important role,10,24,25 and NHE1 is the only NHE isoform found in VSMC.15 In rat mesenteric small arteries, evidence has been provided for an electroneutral Na+- and HCO3eC-dependent transporter that seems to take particular importance during contraction.4,17
We demonstrate here the importance of NBC in mouse small arteries and show that during intracellular acidification, this transport has a quantitative importance of similar magnitude as the NHE. In mesenteric, coronary, and cerebral small arteries, the NBC-mediated transport after an NH4+ prepulse (Figure 2G through 2I) represented 127%, 189%, and 64% of the NHE-mediated transport, respectively, suggesting that Na+-dependent bicarbonate transport is of general importance in the vasculature.
For the first time, we show that NBCn1 in VSMC is the only known candidate transporter detectable at the mRNA level, that NBCn1 protein is present in the plasma membranes of VSMC in mouse mesenteric small arteries, and that knock-down of NBCn1 in mouse mesenteric small arteries leads to a dramatic decrease in the Na+-dependent base influx and in steady-state pHi. Taken together, this provides strong evidence for NBCn1 as the quantitatively important Na+-dependent bicarbonate transporter in mouse small arteries.
A particular strength of the current evidence is the direct link between the investigated protein and its function. Certain functions of intact tissues have in the past often been ascribed to specific proteins based on indirect evidence, but the development of new techniques (siRNA, antisense, dominant-negative proteins, among others) enables us to draw much firmer conclusions. With the siRNA technique, we managed for the first time to show a direct link between a membrane acid/base transporter and the pHi regulatory function.
In this study, we found that relaxed VSMC in the wall of mouse mesenteric, coronary, and cerebral small arteries present with much lower steady-state pHi when CO2/bicarbonate is omitted from the bath solution. The observation that knock-down of NBCn1 results in lower steady-state pHi is consistent with a role for Na+-dependent bicarbonate transport in this phenomenon.
Previous studies on rat mesenteric arteries4 and guinea pig femoral arteries18 reported no difference in steady-state pHi between relaxed VSMC in the presence and absence of CO2/bicarbonate. Lower steady-state pHi was, however, found in contracting rat mesenteric small arteries when CO2/bicarbonate was omitted.4
Studies on isolated VSMC and VSMC-derived cell cultures have given variable results,18 with different studies showing either increased pHi,25,26 decreased pHi,27,28 or no effect on pHi when CO2/bicarbonate was omitted from the solution.
One likely explanation for the different effects of CO2/bicarbonate omission is differences in net fluxes of bicarbonate. In relaxed VSMC of mouse small arteries, the increased pHi in the presence of CO2/bicarbonate would then suggest net bicarbonate influx, which in steady-state is accompanied by additional net CO2 efflux. Although other explanations may also exist, the results stress the importance of performing in vitro experiments in CO2/bicarbonate-buffered solutions.
Although the data strongly point to NBCn1 as the only important Na+-dependent bicarbonate transporter in VSMC, 1 interesting observation remains unexplained. The DIDS sensitivity of NBCn1 (and its human ortholog NBC3) has previously been investigated with the transporter heterologously expressed in Xenopus oocytes and HEK293 cells. In these experiments, even high concentrations of DIDS (500 eol/L to 1 mmol/L) had little effect on the NBC.29eC31 Similar results have since been found in isolated perfused tubules (medullary thick ascending limb)32 and choroid plexus,33 where the Na+-dependent bicarbonate transport proposed to be mediated by NBCn1 was found to be virtually insensitive to DIDS.
In mouse small arteries, we find a Na+-dependent bicarbonate transport that is largely inhibited by DIDS. This is consistent with earlier studies on rat mesenteric small arteries, where the transport was inhibited approximately 75% by DIDS.17 In the present study, the Na+-dependent transport is moreover greatly reduced when NBCn1 is knocked down by siRNA transfection. Intriguingly, this provides evidence for tissue-specific differences in NBCn1 DIDS sensitivity, which could potentially be a result of tissue-specific splice variation, posttranslational modification, or interaction with tissue specific proteins.
At the extracellular end of putative TM5, the DIDS-sensitive members of the slc4 family have a putative DIDS-binding site (KXXK). In NBCn1 this site is disrupted (KLFH) which has been proposed to account for the DIDS insensitivity.22 We amplified this part of the NBCn1 mRNA from mouse mesenteric small arteries and confirmed the disruption by sequencing (unpublished observation, 2004).
The precise magnitude of NBCn1 DIDS sensitivity in VSMC is difficult to determine, as neither the addition of DIDS nor the knock-down of NBCn1 leads to a complete abolition of the Na+-dependent bicarbonate transport.
Previous studies on VSMC in primary culture from dog femoral arteries suggested the presence of a Na+-dependent CleC/HCO3eC exchanger,24 whereas NBCn1 was found to be independent of chloride when heterologously expressed in Xenopus oocytes.29
In the present study, it was not possible, by RT-PCR, to find evidence for any of the known Na+-dependent CleC/HCO3eC exchangers, and in earlier work on freshly isolated intact rat mesenteric small arteries, the Na+-dependent bicarbonate transport was demonstrated to be independent of chloride.17 Our results in small arterial VSMC are consistent with this result. The differences observed may be explained by species- or tissue-specific differences or by the fact that expression of proteins, and therefore the phenotype of cells, may change during culture.
The marked effect of CO2/bicarbonate on pHi of VSMC revealed a unique opportunity for investigating the effect of isolated changes in pHi and hence the potential importance of NBCn1 for VSMC contractility. Earlier studies have focused on the effect of acute changes in pHi, and in those experiments, the general finding has been an increase in the agonist-induced tone when pHi was experimentally lowered.23 In the present study, pHi was changed for at least 1 hour before the agonist-induced response was investigated, and we were unable to detect any effect of pHi on acute norepinephrine-induced contractions. One possible explanation is that the increase in vascular tone following acute intracellular acidification is a result of the change in pHi rather than the sustained low pHi.
In contrast, however, we find a pronounced effect on norepinephrine-induced contractions after long-term acidification (or omission of CO2/bicarbonate) in the presence of norepinephrine. Arteries are under a constant sympathetic tone, and with the importance of arterial diameter for vascular resistance, the decline in contractile response could have a significant impact on the distribution of blood flow during sustained intracellular acidosis.
The different contractile responses in the presence and nominal absence of CO2/bicarbonate could not be explained by different Ca2+ responses. This makes an effect via receptor downregulation unlikely and may suggest an interaction with the Ca2+-sensitizing machinery. Arteries kept in bicarbonate-containing solution apparently increase their Ca2+ sensitivity during sustained exposure to norepinephrine and, in this way, may compensate for the substantial decrease in the Ca2+ response and minimize the effect on the contraction. Under bicarbonate-free conditions, this sensitization seems to be largely abolished, as the contractile response attenuates much faster despite more or less similar declines in the cytosolic Ca2+ concentration.
An increase in Ca2+ sensitivity after sustained exposure to norepinephrine in the presence of CO2/bicarbonate could also explain why these arteries developed augmented myogenic tone, a phenomenon that did not appear when arteries were exposed to norepinephrine in the nominal absence of CO2/bicarbonate.
More work is needed to clarify the underlying mechanism, but 1 important purpose of bicarbonate-dependent pHi regulation in VSMC may be the maintenance of contractile ability during sustained agonist exposure.
In conclusion, we show that NBCn1 (slc4a7) mediates the Na+-dependent bicarbonate transport important for pHi recovery after acid load as well as for regulation of steady-state pHi in VSMC of mouse small arteries. This transport is largely inhibited by DIDS and is of an overall magnitude similar to the amiloride-sensitive transport.
Considering the importance of pHi for protein function in general and for VSMC growth and contractility in particular, this work significantly increases our understanding of vascular physiology. This understanding may also be important for the comprehension of pathological conditions associated with pH abnormalities.
Acknowledgments
The Water and Salt Research Center at the University of Aarhus was established and is supported by the Danish National Research Foundation (Danmarks Grundforskningsfond).
This work was supported by the Danish Medical Research Council (grant no. 22-04-0044). We thank Per Larsen and Finn Skou Pedersen, The siRNA Delivery Center, University of Aarhus, for discussions about the siRNA-technique. We thank Joergen Andresen, Susie Mogensen, and Edith Bjoern Moeller for technical assistance.
References
Aalkjaer C, Mulvany MJ. Effect of changes in intracellular pH on the contractility of rat resistance vessels. Prog Biochem Pharmacol. 1988; 23: 150eC158.
Tian R, Vogel P, Lassen NA, Mulvany MJ, Andreasen F, Aalkjaer C. Role of extracellular and intracellular acidosis for hypercapnia-induced inhibition of tension of isolated rat cerebral arteries. Circ Res. 1995; 76: 269eC275.
Apkon M, Weed RA, Boron WF. Motor responses of cultured rat cerebral vascular smooth muscle cells to intra- and extracellular pH changes. Am J Physiol. 1997; 273: H434eCH445.
Aalkjaer C, Cragoe EJ. Intracellular pH regulation in resting and contracting segments of rat mesenteric resistance vessels. J Physiol. 1988; 402: 391eC410.
Naderali EK, Buttell N, Taggart MJ, Bullock AJ, Eisner DA, Wray S. The role of the sarcolemmal Ca2+-ATPase in the pH transients associated with contraction in rat smooth muscle. J Physiol. 1997; 505: 329eC336.
Berk BC, Vallega G, Muslin AJ, Gordon HM, Canessa M, Alexander RW. Spontaneously hypertensive rat vascular smooth muscle cells in culture exhibit increased growth and Na+/H+ exchange. J Clin Invest. 1989; 83: 822eC829.
Schubert R, Krien U, Gagov H. Protons inhibit the BKCa channel of rat small artery smooth muscle cells. J Vasc Res. 2001; 38: 30eC38.
Wang X, Wu J, Li L, Chen F, Wang R, Jiang C. Hypercapnic acidosis activates KATP channels in vascular smooth muscles. Circ Res. 2003; 92: 1225eC1232.
Davis JP, Chien PF, Chipperfield AR, Gordon A, Harper AA. The three mechanisms of intracellular chloride accumulation in vascular smooth muscle of human umbilical and placental arteries. Pflugers Arch. 2000; 441: 150eC154.
Kahn AM, Cragoe EJ Jr, Allen JC, Seidel CL, Shelat H. Effects of pHi on Na+-H+, Na+-dependent, and Na+-independent CleC-HCO3eC exchangers in vascular smooth muscle. Am J Physiol. 1991; 261: C837eCC844.
Kopito RR, Lodish HF. Primary structure and transmembrane orientation of the murine anion exchange protein. Nature. 1985; 316: 234eC238.
Brosius FC 3rd, Pisoni RL, Cao X, Deshmukh G, Yannoukakos D, Stuart-Tilley AK, Haller C, Alper SL. AE anion exchanger mRNA and protein expression in vascular smooth muscle cells, aorta, and renal microvessels. Am J Physiol. 1997; 273: F1039eCF1047.
Fliegel L. The Na+/H+ exchanger isoform 1. Int J Biochem Cell Biol. 2005; 37: 33eC37.
Boyarsky G, Ganz MB, Cragoe EJ Jr, Boron WF. Intracellular-pH dependence of Na-H exchange and acid loading in quiescent and arginine vasopressin-activated mesangial cells. Proc Natl Acad Sci U S A. 1990; 87: 5921eC5924.
Kalaria RN, Premkumar DR, Lin CW, Kroon SN, Bae JY, Sayre LM, LaManna JC. Identification and expression of the Na+/H+ exchanger in mammalian cerebrovascular and choroidal tissues: characterization by amiloride-sensitive [3H]MIA binding and RT-PCR analysis. Brain Res Mol Brain Res. 1998; 58: 178eC187.
Bobulescu IA, Di Sole F, Moe OW. Na+/H+ exchangers: physiology and link to hypertension and organ ischemia. Curr Opin Nephrol Hypertens. 2005; 14: 485eC494.
Aalkjaer C, Hughes A. Chloride and bicarbonate transport in rat resistance arteries. J Physiol. 1991; 436: 57eC73.
Aickin CC. Regulation of intracellular pH in smooth muscle cells of the guinea-pig femoral artery. J Physiol. 1994; 479: 331eC340.
Damkier HH, Nielsen S, Praetorius J. An Anti-N-terminal antibody localizes the NBCn1 to heart endothelia, skeletal and vascular smooth muscle cells. Am J Physiol Heart Circ Physiol. 2006; 290: H172eCH180.
Martin JB, Young JL, Benoit JN, Dean DA. Gene transfer to intact mesenteric arteries by electroporation. J Vasc Res. 2000; 37: 372eC380.
Leem CH, Lagadic-Gossmann D, Vaughan-Jones RD. Characterization of intracellular pH regulation in the guinea-pig ventricular myocyte. J Physiol. 1999; 517: 159eC180.
Romero MF, Fulton CM, Boron WF. The SLC4 family of HCO3eC transporters. Pflugers Arch. 2004; 447: 495eC509.
Smith GL, Austin C, Crichton C, Wray S. A review of the actions and control of intracellular pH in vascular smooth muscle. Cardiovasc Res. 1998; 38: 316eC331.
Kahn AM, Cragoe EJ Jr, Allen JC, Halligan RD, Shelat H. Na+-H+ and Na+-dependent CleC-HCO3eC exchange control pHi in vascular smooth muscle. Am J Physiol. 1990; 259: C134eCC143.
Neylon CB, Little PJ, Cragoe EJ Jr, Bobik A. Intracellular pH in human arterial smooth muscle. Regulation by Na+/H+ exchange and a novel 5-(N-ethyl-N-isopropyl)amiloride-sensitive Na+- and HCO3eC-dependent mechanism. Circ Res. 1990; 67: 814eC825.
Vigne P, Breittmayer JP, Frelin C, Lazdunski M. Dual control of the intracellular pH in aortic smooth muscle cells by a cAMP-sensitive HCO3eC/CleC antiporter and a protein kinase C-sensitive Na+/H+ antiporter. J Biol Chem. 1988; 263: 18023eC18029.
Korbmacher C, Helbig H, Stahl F, Wiederholt M. Evidence for Na/H exchange and Cl/HCO3 exchange in A10 vascular smooth muscle cells. Pflugers Arch. 1988; 412: 29eC36.
Putnam RW, Grubbs RD. Steady-state pHi, buffering power, and effect of CO2 in a smooth muscle-like cell line. Am J Physiol. 1990; 258: C461eCC469.
Choi I, Aalkjaer C, Boulpaep EL, Boron WF. An electroneutral sodium/bicarbonate cotransporter NBCn1 and associated sodium channel. Nature. 2000; 405: 571eC575.
Pushkin A, Abuladze N, Lee I, Newman D, Hwang J, Kurtz I. Cloning, tissue distribution, genomic organization, and functional characterization of NBC3, a new member of the sodium bicarbonate cotransporter family. J Biol Chem. 1999; 274: 16569eC16575.
Park M, Ko SB, Choi JY, Muallem G, Thomas PJ, Pushkin A, Lee MS, Kim JY, Lee MG, Muallem S, Kurtz I. The cystic fibrosis transmembrane conductance regulator interacts with and regulates the activity of the HCO3eC salvage transporter human Na+-HCO3eC cotransport isoform 3. J Biol Chem. 2002; 277: 50503eC50509.
Odgaard E, Jakobsen JK, Frische S, Praetorius J, Nielsen S, Aalkjaer C, Leipziger J. Basolateral Na+-dependent HCO3eC transporter NBCn1-mediated HCO3eC influx in rat medullary thick ascending limb. J Physiol. 2004; 555: 205eC218.
Bouzinova EV, Praetorius J, Virkki LV, Nielsen S, Boron WF, Aalkjaer C. The Na+ dependent HCO3eC uptake into the rat choroid plexus epithelium is partially DIDS-sensitive. Am J Physiol Cell Physiol. 2005; 289: C1448eCC1456.(Ebbe Boedtkjer, Jeppe Pra)