Determinants of renal afferent arteriolar actions of bradykinin: evidence that m - 《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 ... 21 下一页

返回列表

查看: 350242\|回复: 209	go Determinants of renal afferent arteriolar actions of bradykinin: evidence that m [复制链接]

轻羽

新手上路

Rank: 1

积分: 0
威望: 0
包包: 0

楼主

发表于 2009-4-21 13:42 |只看该作者 |倒序浏览 |打印

作者：Xuemei Wang, Greg Trottier,  Rodger Loutzenhiser作者单位：Smooth Muscle Research Group, Department of Pharmacology andTherapeutics, University of Calgary, Calgary, Alberta, Canada T2N 4N1 / l/ F: m1 _) l. q: \2 Z/ }


      ) T( e! U) N# v

         4 s! J7 J' Y1 M" Z

         5 l) G6 A! f& T: J0 l! S9 \# H
         / q0 E- G# w) J4 w* w

      ; {' \4 \, P/ s7 x# R


      【摘要】$ R) Z1 }6 `* D
   The determinants of bradykinin (BK)-induced afferent arteriolar vasodilation were investigated in the in vitro perfused hydronephrotic ratkidney. BK elicited a concentration-dependent vasodilation of afferentarterioles that had been preconstricted with ANG II (0.1 nmol/l), but thisdilation was transient in character. Pretreatment with the nitric oxidesynthase inhibitor N -nitro- L -arginine methyl ester (100 µmol/l) and the cyclooxygenase inhibitor ibuprofen (10µmol/l) did not prevent this dilation when tone was established by ANG IIbut fully blocked the response when tone was established by elevatedextracellular KCl, which suggests roles for both NO and endothelium-derivedhyperpolarizing factor (EDHF). We had previously shown that the EDHF-likeresponse of the afferent arteriole evoked by ACh was fully abolished by acombination of charybdotoxin (ChTX;10 nmol/l) and apamin (AP; 1 µmol/l).However, in the current study, treatment with ChTX plus AP only reduced theEDHF-like component of the BK response from 98 ± 5 to 53 ± 6%dilation. Tetraethylammonium (TEA; 1 mmol/l), which had no effect on the EDHF-induced vasodilation associated with ACh, reduced the EDHF-like responseto BK to 88 ± 3% dilation. However, the combination of TEA plus ChTXplus AP abolished the response (0.3 ± 1% dilation). Similarly,17-octadecynoic acid (17-ODYA) did not prevent the dilation when it wasadministered alone (77 ± 9% dilation) but fully abolished the EDHF-likeresponse when added in combination with ChTX plus AP (-0.5 ± 4% dilation). These findings suggest that BK acts via multiple EDHFs: one that issimilar to that evoked by ACh in that it is blocked by ChTX plus AP, and asecond that is blocked by either TEA or 17-ODYA. Our finding that a componentof the BK response is sensitive to TEA and 17-ODYA is consistent with previoussuggestions that the EDHF released by BK is an epoxyeicosatrienoic acid. ( h3 Z1 x' J2 s7 A$ O
      【关键词】 arteriole endotheliumderived hyperpolarizing factor acetylcholine octadecynoic epoxyeicosatrienoic acids tetraethylammonium charybdotoxin apamin potassium channels+ x5 ?. }. p" w. X
   THE ENDOTHELIUM PLAYS AN IMPORTANT role in modulation ofvascular reactivity and transmission of signals along blood vessels. A numberof vasodilator agents including ACh and bradykinin (BK) elicit vasodilation inan indirect manner by releasing endothelium-derived relaxing factors (EDRFs)such as nitric oxide (NO), prostacyclin (PGI 2 ), and less-definedfactors that act through hyperpolarization [endothelium-derivedhyperpolarizing factors (EDHFs)]. Although the term EDHF implies a factor that is released from the endothelium, in terminal arterioles such as the afferentarteriole, it is quite possible that the responses ascribed to EDHF aremediated by endothelial hyperpolarization and electrical coupling of theendothelial layer to the underlying smooth muscle myocytes (e.g., 3, 18, 30 ). For the purposes of thepresent report, we continue to use the conventional term EDHF or EDHF-likewhile acknowledging that the responses may not actually involve a releasedfactor. The properties of the EDHF that contribute to the component of thedilation that is insensitive to inhibition of both nitric oxide synthase (NOS)and cyclooxygenase (COX) vary among different vascular beds( 23 ). Moreover, the relativecontribution of EDHF vs. NOS and COX products varies between vascular beds andbetween different endothelium-dependent vasodilators( 23, 34 ). In regard to the renalcirculation, knowledge of the relative contributions of NO, PGI 2,and EDHF to the renal microvascular actions of different endothelium-dependentvasodilators and the nature of the renal EDHF(s) involved is limited.6 }& M2 q# I; V! ?
! ^" {; W* C/ q. ^
We recently demonstrated( 36 ) that the EDHF associatedwith the renal microvascular response to ACh is similar to that seen inseveral other vascular beds in that it is abolished by a combination ofcharybdotoxin (ChTX) and apamin (AP). This EDHF-like component accounted for95% of the initial phasic vasodilatory response to ACh but did not contributeto the sustained phase of the vasodilation nor did it contribute to theefferent arteriolar actions of ACh. Although the pharmacological properties of the renal EDHF associated with ACh were similar to those described for othervascular beds, these properties differed from the properties of the EDHFassociated with BK-induced renal vasodilation. Specifically, we found 1 mmol/ltetraethylammonium (TEA) to have no effect on the EDHF associated with ACh,whereas several other laboratories found TEA to attenuate the EDHF component of the BK response ( 10, 26, 29 ). Moreover, Imig et al.( 16 ) observed BK to releaseepoxyeicosatrienoic acids (EETs) from the renal vasculature, and thehyperpolarization elicited by EETs in isolated myocytes and the preglomerularvasodilatory actions of EETs are blocked by TEA( 38 ). When interpreted in concert with our observations, these findings suggest that BK may release anEDHF that is distinct from that associated with ACh. In support of thispremise, Fulton et al. ( 11 )observed that the cytochrome P -450 inhibitor5,8,11,14-eicosatetraynoic acid (ETYA) attenuated the response of the isolatedperfused kidney to BK but did not alter the response to ACh.; H% Z+ x. H+ }' M
) m4 M  h5 S+ N. u
Thus in the present study, we used the in vitro perfused hydronephrotic ratkidney model to investigate the determinants of the renal afferent arteriolarresponse to BK, and we compared these results to previous studies thatemployed this model to characterize the EDHF associated with ACh. Our findingssuggest the presence of at least two differing pathways that mediate the COX-and NOS-independent EDHF-like responses in the renal microcirculation. Onecomponent is abolished by ChTX plus AP. The second pathway is resistant tothis treatment but is blocked by either TEA (1 mmol/l) or the cytochrome P -450 inhibitor 17-octadecynoic acid (17-ODYA).
% p, z& o- [  U5 o2 V, M# C
METHODS

Unilateral hydronephrosis was induced to facilitate direct observations ofrenal afferent arteriolar actions of BK (see Ref. 21 for detailed description ofmodel). The use of animals complied with the Canadian Council on Animal Careregulations. The left ureters of 6- to 7-wk-old male Sprague-Dawley andLong-Evans rats were ligated under halothane-induced anesthesia to induce hydronephrosis. After 6-8 wk, the renal artery was cannulated in situ and thehydronephrotic kidney was excised and transferred to a heated chamber on thestage of an inverted microscope with continuous perfusion. The perfusate, aDMEM (GIBCO Life Technologies, Gaithersburg, MD) that contained (in mmol/l) 30 bicarbonate, 5 glucose, and 5 HEPES, was equilibrated with 95% air-5%CO 2. Temperature and pH were maintained at 37°C and 7.40,respectively. Medium was circulated through a heat exchanger to a pressurizedreservoir connected to the renal arterial cannula. Perfusion pressure, whichwas monitored within the renal artery, was maintained at 80 mmHg. The kidneyswere allowed to equilibrate from surgical manipulations for at least 1 hbefore experimental protocols were initiated. Cortical afferent arteriolesthat originated from interlobular arteries was measured over a 20- to 30-µm segment nearthe midpoint of the vessel and was determined by online image processing at asampling rate of 3 Hz. Mean diameter values were then averaged over theplateau or peak of the response. Typically, each determination was derivedfrom the mean of 50-100 individual measurements of the average diameter overthe vessel segment.
7 Y8 ~3 ]: Q1 y" k7 L: Z  x
All agents were added directly to the perfusate. The venous effluentemptied into the perfusion chamber, thus agents reached both the luminal andadluminal surfaces of the arterioles. Experiments that employed AP and ChTXrequired the use of a recirculating perfusion system as previously described( 36 ). In all otherexperiments, single-pass perfusion was employed. The data are expressed asmeans ± SE. Differences between treatment groups were assessed byBonferroni's t -test and one-way ANOVA. P values rat strains, Sprague-Dawley andLong-Evans, were used based on availability. No significant differences wereobserved between these two strains, and the data were combined.

RESULTS

BK-induced vasodilation. The concentration-dependent actions of BKon the afferent arteriole are depicted in Fig. 1. ANG II (0.1 nmol/l),which was used to establish basal tone, reduced diameter values from 16.2± 0.9 to 4.7 ± 0.5 µm( P n = 6). Figure 1 A depicts arepresentative tracing that illustrates the transient character of theBK-induced vasodilation in this setting. At concentrations of 0.1, 1, 10, and100 nmol/l, BK increased diameter values to peaks of 7.0 ± 0.7, 10.1± 0.7, 14.0 ± 1.0, and 15.9 ± 0.9 µm, respectively( Fig. 1 B ). The datawere converted to percent reversal of the ANG II-induced vasoconstriction (percent vasodilation). The corresponding values were 20 ± 3, 48± 5, 81 ± 5, and 98 ± 5% vasodilation ( Fig. 1 C ). Althoughthe afferent arteriole consistently exhibited transient BK-inducedvasodilation, sustained responses were frequently observed in the interlobularartery (ILA). Figure 2 depictsa representative example of a sustained response. Of eight ILAs studied, fourexhibited transient dilations (mean initial diameter, 67 ± 11 µm)and four exhibited sustained responses (mean diameter, 88 ± 6µm).

Fig. 1. A : original tracings illustrate the transient vasodilation evokedby increasing concentrations of bradykinin () in the afferent arteriole. B and C : concentration-response relationship is expressed asboth diameter and percent vasodilation. Vessels were preconstricted with 0.1nmol/l ANG II.: x3 c" ]' C) \  e9 m4 l, d$ e. l
+ E( ~6 x' K0 t; R8 N) K
Fig. 2. Representative tracing depicts sustained BK-induced vasodilation in aninterlobular artery (ILA).

In a separate series of kidneys ( n = 7), we assessed the effects of pretreatment with 10 µmol/l ibuprofen and 100 µmol/l N -nitro- L -arginine methyl ester( L -NAME) on the response to BK. As shown in Fig. 3 B and subsequentfigures (i.e., see Figs. 5, 6, 7 ), inhibition of COX and NOShad little effect on basal afferent arteriolar diameter. Similar findings were reported in past publications using this model( 35, 36 ). Moreover, we alsoobserved this treatment to have no effect on basal perfusate flow of theisolated perfused normal rat kidney in the absence of an added vasoconstrictor( 13 ). As further depicted in Fig. 3, inhibition of COX andNOS had very modest effects on the profile of the response to BK (100 nmol/l),which suggests a prominent role of additional factors such as EDHF. In the control kidneys, BK (100 nmol/l) elicited a peak dilation of 97 ± 2%(from 6.2 ± 1.1 to 16.4 ± 0.4 µm; basal, 16.7 ± 0.6µm; n = 7). In the kidneys pretreated with L -NAME andibuprofen, BK elicited a peak dilation of 83 ± 4% ( n = 7; P = 0.009; from 5.4 ± 0.3 to 15.5 ± 0.7 µm; basal,16.9 ± 1.7 µm). The vasodilatory responses decayed with a similar time course in these two settings. For example at 4 and 6 min, diametermeasurements of controls (10.8 ± 2.7 and 7.1 ± 1.2 µm) werenot statistically different from those of kidneys pretreated with L -NAME (5.3 ± 0.6 and 4.8 ± 0.5 µm; P =0.06 and P = 0.11, respectively).) T: j8 y1 _1 Z/ j, t( F. e9 D. _

Fig. 3. BK elicits a phasic vasodilation of the afferent arteriole regardless ofthe presence or absence of N -nitro- L -arginine methyl ester( L -NAME). A : original tracings of vessels preconstrictedwith 0.1 nmol/l ANG II illustrate control response obtained in the presence of10 µmol/l ibuprofin ( left ) and the response after the addition of100 µmol/l L -NAME ( right ). B : mean data depicttime course of the response to BK (added at t = 0 ) incontrols (ibuprofen treated) and in vessels pretreated with 100 µmol/l L -NAME (and ibuprofen).

Fig. 5. In contrast with our observations with ACh, the endothelium-derivedhyperpolarizing factor (EDHF)-like response to BK was not prevented by thecombination of 10 nmol/l charybdotoxin (ChTX) plus 1 µmol/l apamin (AP). A : original tracing. B : mean data. C : mean data arecompared with the effects of AP plus ChTX on EDHF-like response to ACh (datataken from Ref. 36 ).
) f9 w1 R8 S7 j  Z. ~  `! r
Fig. 6. A : tetraethylammonium chloride (TEA; 1 mmol/l) had no effect onthe EDHF-like response to BK when administered alone ( left ). Meandata are shown ( right; n = 5). B : however, when TEAtreatment was combined with 10 nmol/l ChTX plus 1 µmol/l AP, the responsewas totally eliminated ( n = 7). C : effects of TEA alone, APplus ChTX, and the combination of TEA plus AP plus ChTX are shown. Allexperiments were performed in the presence of L -NAME plusibuprofen. ANG II indicates application of 0.1 nmol/l ANG II.

Fig. 7. A : pretreatment with 17-octadecynoic acid (ODYA; 50 µmol/l)alone did not prevent the EDHF-like response to BK ( left ). Mean dataare shown ( right; n = 4). B : however, when combinedwith 10 nmol/l ChTX plus 1 µmol/l AP, ODYA eliminated this response( n = 6). C : summary of the effects of ODYA alone and ODYAplus AP plus ChTX. All experiments were performed in the presence of L -NAME plus ibuprofen.

To further assess the contribution of a hyperpolarizing factor, wedetermined the effects of these inhibitors on the actions of BK on arteriolesthat had been preconstricted with elevated extracellular K  . Asillustrated in Fig. 4, BKpartially reversed the vasoconstrictor actions of 25 mmol/l KCl. Treatmentwith KCl reduced diameter measurements from 18.7 ± 0.6 to 9.9 ±0.7 µm, and BK caused a peak dilation to 14.7 ± 1.4 µm( n = 5; P = 0.007) thus eliciting a 55 ± 11% dilation( Fig. 4 B, solid bar).Pretreatment with ibuprofen alone (10 µmol/l) did not significantly alterthis response (50 ± 13% vasodilation, Fig. 4 B, open bar),whereas treatment with L -NAME (100 µmol/l) abolished this action(-0.1 ± 1.5% vasodilation). We interpret these findings as indicatingthat the phasic vasodilatory response to BK seen in the presence ofKCl-induced vasoconstriction is mediated by NO production, and that the NOS-and COX-independent actions of BK, which are seen when the afferent arterioleis preconstricted by ANG II, were blocked by an elevation of extracellular K  . The latter observation is consistent with a proposed EDHF-like component of the afferent arteriolar actions of BK.1 q/ A0 x; S  a8 p& a3 o

Fig. 4. BK-induced vasodilation of afferent arterioles preconstricted with 25mmol/l KCl. A : original tracing illustrates the transient nature ofthis response. B : mean data ( n = 5) illustrate lack ofeffect of 10 µmol/l ibuprofen (Ibu) and complete abolition of response by100 µmol/l L -NAME., m. `( p* S2 V( P: m+ Z: T" J
' G, @) X1 {. k7 [# H$ H! z
Effects of K -channel blocking agents on NOS- andCOX-independent actions of BK. We have shown that like BK, ACh elicits aphasic vasodilator response that is independent of COX and NOS but is blockedby elevated K ( 14, 36 ). In the case of ACh, this EDHF-like response was abolished by a combination of 10 nmol/l ChTX plus 1µmol/l AP. We therefore examined the effects of this treatment on the BKresponse that is elicited in the presence of ibuprofen and L -NAME.These findings are summarized in Fig.5. In these experiments ( n = 6), diameter measurements inthe presence of ibuprofen (10 µmol/l) and following treatment with L -NAME (100 µmol/l) and then the combination of ChTX (10 nmol/l)and AP (1 µmol/l) were 17.8 ± 0.8, 17.7 ± 0.8, and 17.3± 0.8 µm, respectively. The subsequent administration of ANG II (0.1nmol/l) reduced diameter values to 4.8 ± 0.8 µm. In this setting, 100 nmol/l BK caused a transient vasodilation to a peak value of 12.6 ±1.2 µm ( P = 0.003), which corresponds to a 46 ± 6%vasodilation ( P = 0.003 vs. 83 ± 4% vasodilation in thepresence of L -NAME and ibuprofen alone). Thus in contrast to ourprevious findings with ACh (presented as solid bars in Fig. 5 C forillustrative purposes; data taken from 36), ChTX plus AP attenuated but didnot abolish the EDHF-like response to BK.

We next examined the effects of TEA both alone and in combination with ChTXplus AP. TEA is a nonselective K -channel blocker at highconcentrations, but at a concentration of 1 mmol/l, TEA blockslarge-conductance Ca 2 -activated K channels(BK Ca ) while having minimal effects on other K -channelspecies ( 19 ). After treatmentwith ibuprofen and then L -NAME (diameter, 15.2 ± 0.5 and14.8 ± 0.5 µm, respectively; n = 5), the addition of 1mmol/l TEA reduced the diameter to 13.6 ± 0.7 µm ( P 0.05). In this setting, ANG II reduced the diameter to 4.9 ± 0.6 µm,and 100 nmol/l BK caused a transient dilation to 14.1 ± 0.6 µm (107± 9% vasodilation vs. diameter in TEA alone; P = 0.0004). In aseparate series of seven kidneys, basal (ibuprofen) diameter values andmeasurements after treatment with L -NAME and ChTX plus AP were 15.5± 0.8, 15.1 ± 0.7, and 14.7 ± 0.8 µm, respectively( Fig. 6 B ). Theaddition of TEA (1 mmol/l) reduced diameter to 13.5 ± 0.7 µm. Inthis setting, ANG II reduced diameter to 6.6 ± 0.6 µm, and theadministration of BK (100 nmol/l) had no effect (6.6 ± 0.6 µm, 0± 1% vasodilation). Thus whereas TEA or ChTX plus AP only partiallyattenuated the EDHF-like response to BK when added alone, these agents abolished this response when added concurrently( Fig. 6 C ).
5 a* }: b' _! i+ ~( B8 \6 B
These findings suggest that two components may contribute to the EDHF-likeresponse to BK: one that (as previously shown for ACh) is blocked by ChTX plusAP and a second that is sensitive to TEA. BK has been shown to stimulate therelease of EETs from the renal vasculature( 16 ), and 11,12-EET has beenreported to elicit afferent arteriolar vasodilation by a mechanism that isblocked by 1 mmol/l TEA ( 17, 38 ). We therefore determined whether the cytochrome P -450 inhibitor 17-ODYA would also block thecomponent of the BK response that was insensitive to ChTX plus AP. The resultsof these experiments are summarized in Fig.7. Alone or in combination with TEA, 17-ODYA (50 µmol/l) didnot prevent the EDHF-like response to BK. After treatment with ibuprofen, L -NAME, and 50 µmol/l 17-ODYA, diameter measurements were 17.0± 0.9, 16.8 ± 0.9, and 16.0 ± 1.2 µm, respectively.ANG II reduced diameter to 6.0 ± 0.5 µm, and 100 nmol/l BK eliciteda peak dilation to 13.4 ± 1.0 µm, which corresponds to a 77 ±9% vasodilation ( n = 4; Fig.7 A ). Similarly, after treatment with ibuprofen and L -NAME (diameters, 15.18 ± 1.1 and 14.8 ± 1.0 µm,respectively), the addition of 1 mmol/l TEA reduced diameter to 13.8 ± 1.4 µm ( P = 0.10) and addition of 17-ODYA reduced diameter to 11.5± 1.5 µm ( P = 0.02). In this setting, ANG II reduceddiameter to 4.0 ± 0.5 µm, and 100 nmol/l BK caused a transientdilation to 11.6 ± 1.1 µm ( n = 5; P = 0.002; 106± 13% vasodilation of ANG II response vs. 17-ODYA alone, data notshown). Thus the inhibitory effects of TEA and 17-ODYA on BK responses were not additive. In a separate series of six kidneys, basal (ibuprofen) diametermeasurements and values after treatment with L -NAME and ChTX plusAP were 15.6 ± 0.5, 15.1 ± 0.5, and 14.2 ± 0.6 µm( Fig. 7 B ). Theaddition of 17-ODYA (50 µmol/l) reduced diameter to 12.5 ± 0.8 µm( P = 0.01). In this setting, ANG II reduced diameter to 5.2 ±0.2 µm, and the administration of BK (100 nmol/l) had no effect (5.2± 0.3 µm, 0 ± 6% vasodilation). Accordingly, as seen withTEA, 17-ODYA abolished that component of the EDHF-like response to BK that wasresistant to ChTX plus AP. These results are summarized in Fig. 7 C.2 ~4 K! A8 E. H) b$ B3 E7 [  s

Contribution of NO to vasodilation elicited by BK. As shown in Fig. 4, during KCl-inducedvasoconstriction, BK elicited a dilation that was NO dependent (e.g., blockedby L -NAME). Nevertheless, L -NAME had little effect onthe magnitude or profile of the BK-induced vasodilation when ANG II was used to establish basal tone (see Fig.3 ). To further assess the contribution of NO in this setting, weexamined the response elicited following pretreatment with ibuprofen, TEA, andChTX plus AP, which would block the COX- and EDHF-dependent components, andcompared these results to those in which L -NAME was also present(see Fig. 6 A ). Thesefindings are summarized in Fig.8, A and B. Diameter measurements were 16.9± 0.5 µm in the presence of ibuprofen and 16.8 ± 0.5 and 15.7 ± 0.1 µm following ChTX plus AP and TEA, respectively. The additionof ANG II reduced diameter to 7.4 ± 1.2 µm, and the subsequentadministration of BK resulted in a maximal dilation to 12.3 ± 1.0µm, which corresponds to a 62 ± 7% vasodilation ( n = 5).These findings suggest that, as seen during KCl-induced vasoconstriction, NOproduction contributes to the transient BK-induced afferent arteriolarvasodilation that is observed during ANG II-induced vasoconstriction.5 `0 H7 e. X$ j# Q* D
7 [; R. `( i* q* }
Fig. 8. A : original tracing ( left ) shows that pretreatment withibuprofen, TEA, and plus 10 nmol/l ChTX plus 1 µmol/l AP eliminated theresponse to BK in the presence of L -NAME but not in the absence ofthe nitric oxide synthase (NOS) inhibitor. Thus it appears that NO productioncontributes to the phasic vasodilation produced by BK. Mean data ( n =5) are also depicted ( right ). B : responses ( n = 7)in presence and absence of L -NAME are compared.
: I, _' D9 v  o3 X. E& n# N
DISCUSSION" }7 J+ V6 k; |6 d) l2 r* h2 }

The results of the present study indicate that BK evokes a transient vasodilatory response in the renal afferent arteriole and that this actioninvolves both NO-dependent and NO-independent mechanisms. The NO-independentcomponent of the vasodilation that remained following NOS blockade hadfeatures similar to those ascribed to an EDHF in that it was blocked by eitherelevated external K or a combination of agents known to blockCa 2 -activated K (K Ca ) channels.A major finding of this study is that the pharmacological characteristics ofthis EDHF-like component of the afferent arteriolar response to BK differfundamentally from the previously observed characteristics of the EDHF-like response of the same vessel to ACh( 36 ). In concert, these observations suggest that multiple factors or multiple mechanisms contributeto the vasodilator responses ascribed to EDHF in the renalmicrocirculation.( d+ i9 D7 {. y! q. v0 E; }' E

We suggest that the EDHF-like response to BK consists of two components,one of which was blocked by a combination of ChTX plus AP. We had previouslyshown that this combination fully abolished the EDHF-like response of theafferent arteriole to ACh( 36 ). Indeed, treatment withChTX plus AP has been shown to block the vasodilator responses attributed toEDHF in a variety of blood vessels under diverse conditions (for review, seeRefs. 23, 34 ). At the concentrationsemployed, these toxins are known to block small- and intermediate-conductance K Ca channels ( 32, 33 ). It has been suggested bya number of investigators that these K -channel blockers mayattenuate EDHF-like responses by blocking K channels that arepresent on the endothelium rather than on the vascular myocyte (e.g., 4, 5, 7 ). An elegant demonstration ofthis was provided by Doughty et al.( 5 ), who reported that ChTX andAP inhibit EDHF responses when selectively applied to the endothelial cells within the vessel lumen but not when selectively applied to the outer surfaceof a perfused mesenteric artery. These authors interpret these observations tosuggest that these agents act on the endothelium rather than the smooth musclemyocytes. Accordingly, in small vessels, in which such EDHF-like responses aremore prevalent, endothelial agonists may activate K channels inthe endothelium, and the resultant hyperpolarization may be transmitted to theunderlying smooth muscle layer via electrical communication throughmyoendothelial gap junctions( 30 ). An obligate role ofmyoendothelial gap junctions in responses attributed to EDHF is suggested byseveral studies (e.g., 3, 18, 30 ).
0 G% o6 a! V6 n3 }% S, F/ L
The present study did not specifically address this issue. However, wepreviously demonstrated that the combination of ChTX plus AP not only blocksthe transient EDHF-like component of the afferent arteriolar dilation evokedby ACh but also inhibits the sustained phase of the ACh response that isdependent on NO formation( 36 ). This latter observationis consistent with the premise that endothelial K -channelactivation and the resultant hyperpolarization contribute toCa 2 entry (see Ref. 24 for review). Thus byinterfering with this mechanism, the blockade of endothelial K channels by ChTX and AP would be anticipated to inhibit the sustainedcomponent of NO formation as we have observed. Accordingly, our previousfinding that ChTX plus AP blocks the sustained NO-dependent component of theACh response would be consistent with the postulate put forward by others( 4, 5, 7 ) that the K channels affected by these toxins are in the endothelium. ACh and BK are both known to release intracellular Ca 2 stores in theendothelium by stimulating phospholipase C, increasing levels of inositol 1,4,5-trisphosphate (IP 3 ), and activating IP 3 receptorspresent in the endoplasmic reticulum. Thus these two agents could activate thesame population of K Ca channels in the endothelium( 24 ). Such a scheme (see Fig. 9 ) could explain ourfinding that although the characteristics of the EDHF-like response to BK andACh were quite different (e.g., see Fig.3 ), a common component of the response to each of these agents wasblocked by ChTX and AP. Although the present study does not provide a directtest of this postulate, our findings would be consistent with this scheme.0 l5 g3 S6 w1 Y( {: z4 p, ?
( U' G! y3 V- u% `/ V; P
Fig. 9. A proposed model that is consistent with our findings. BK and ACh releaseCa 2 from the endoplasmic reticulum via inositol1,4,5-trisphosphate formation. Transient elevation in intracellularCa 2 activates low- and intermediate-conductanceCa 2 -activated K channels in theendothelium, and the resultant hyperpolarization is transmitted to theunderlying myocytes via myoendothelial coupling. This pathway, which is sharedby BK and ACh, is blocked by the combined treatment with ChTX plus AP. Inaddition, BK-receptor activation results in the liberation of arachidonic acid(AA), which is converted to epoxyeicosatrienoic acids (EETs) by cytochrome P -450 (Cyt P450). Released EETs activate large-conductanceCa 2 -activated K channels in the vascularmyocytes. Thus unlike ACh, the EDHF-like response to BK is not fully blockedby ChTX plus AP. However, this second component of the BK response can beeliminated by either 17-ODYA or 1 mmol/l TEA.- v4 A7 v: A8 P
! R% Z- {" u/ h$ Z8 R4 g
We previously found that treatment with ChTX plus AP fully blocked theEDHF-like response to ACh( 36 ); however, in the present study, we found that this treatment alone was not sufficient to prevent theEDHF-like response to BK (see Fig.5 ). The component of the EDHF-like response to BK that remainedafter treatment with ChTX plus AP was completely eliminated by treatment witheither 1 mmol/l TEA or 50 µmol/l 17-ODYA. In the absence of ChTX and AP,TEA or 17-ODYA administered either alone or in combination did not block theEDHF-like response to BK. We interpret these observations to indicate that asecond, ChTX- and AP-insensitive but TEA- and 17-ODYA-sensitive component contributes to the EDHF-like afferent arteriolar response to BK. Theproperties of this second component are consistent with the suggestion putforward by other investigators that the EDHF-like response to BK involves theelaboration of an EET ( 9, 12, 16, 17, 28, 38 ) that in turn activates the TEA-sensitive BK Ca channels( 1, 17, 38 ) present in the vascular myocytes (e.g., see Fig.9 )." p# d! B- |% Y6 H1 S: L5 l
! v* W$ p/ u- C+ `2 q1 l
Growing evidence implicates cytochrome P -450 products,particularly the EETs, as candidate EDHFs (reviewed in Refs. 23, 28, 34 ). This is particularly trueof EDHF responses evoked by BK. Fulton et al.( 11 ) demonstrated that thecytochrome P -450 inhibitor ETYA attenuated the response of theisolated perfused kidney to BK. Imig and co-workers( 16 ) demonstrated not onlythat cytochrome P -450 inhibitors prevented the EDHF-like response ofthe afferent arteriole to BK, but also that BK treatment augmented EETproduction by isolated renal vascular tissues. Moreover, Imig et al.( 17 ) found 11,12-EET to dilatethe rat ILA and juxtamedullary afferent arteriole via a COX-independent mechanism, whereas 8,9-EET had no effect on these vessels. Zou et al.( 38 ) reported that 11,12-EETactivated a K channel in isolated renal vascular myocytes andelicited vasodilation in the intact vessels, and both of these actions wereabolished by 1.0 mmol/l TEA, a finding consistent with our observation that17-ODYA and 1.0 mmol/l TEA abolished the same component of the EDHF-likeresponse to BK. Gebremedhin et al.( 12 ) elegantly demonstrated,using a bioassay system, that BK released a substance from a perfused bovinecoronary artery that stimulated BK Ca channels in a downstreammyocytes and that the release of this factor was blocked by endotheliumremoval or treatment with 17-ODYA. More recently, Archer et al.( 1 ) identified 11,12-EET as alikely EDHF candidate in human internal mammary arteries and demonstrated thatthis agent activates BK Ca channels in this preparation. Fisslthaleret al. ( 9 ) demonstrated that chronic treatment of native coronary artery endothelial cells with -naphthoflavone enhanced cytochrome P -450 (2C) expression levels, augmented 11,12-EET production, and enhanced both the vasodilatorresponse and the hyperpolarization induced by BK in coronary artery segments.These findings and many others (see Refs. 23, 27, 28, 34 ) support the postulate thatBK stimulates endothelial production of an epoxygenase product that in turncan evoke vasodilation by activating TEA-sensitive BK Ca channelspresent in the underlying vascular myocytes.

It is important to emphasize that although our findings are consistent withthis postulate in regard to the actions of BK, we did not see a similarprofile of the EDHF-like response of the afferent arteriole to ACh. Thus ChTXand AP treatment alone were sufficient to prevent the EDHF-like response to this agent (see Ref. 36 ).These observations suggest that there are multiple components to the EDHF-likeresponse and that the relative contributions of these two components orpathways may differ with different endothelial-dependent vasodilator agents.Very few studies have compared the determinants of the EDHF component of thevasodilator response to differing agents. However, Fulton et al.( 11 ) previously reported that ETYA prevented the response of the isolated perfused kidney to BK but did notprevent the response of this preparation to ACh and suggested differingpathways ( 27 ). Frieden et al. ( 10 ) also observed markeddifferences in the pharmacological profile of the EDHF associated with BK vs.substance P, in that the response to the latter did not involve the cytochrome P -450 pathway. Moreover, the determinants of the EDHF-like response to an agent may vary with blood vessel type. In regard to the latter, it isimportant to note that a number of studies report EDHF-like responses to BKthat are not prevented by cytochrome P -450 inhibition alone (e.g.,Refs. 2, 6, 22, 25 ). However, a furthercomplication is that a concomitant activation of the two EDHF-like pathways(see Effects of K -channel blocking agents on NOS- andCOX-independent actions of BK ) may obscure interpretations of experimentsin which only a single pathway is inhibited. The present study is an exampleof this phenomenon, in that we could only demonstrate a major effect ofblocking the cytochrome P -450 pathway or 1.0 mmol/l TEA when thevessels were concomitantly treated with ChTX plus AP.

When both of the EDHF-like or NO-independent mechanisms were blocked (bythe combined treatment with AP, ChTX, and 1 mmol/l TEA), BK elicited a phasicvasodilation that could be blocked by further treatment with L -NAME(see Fig. 8 ). Similarly, when the afferent arteriole was preconstricted by elevation of extracellular K (which would eliminate the effects of EDHF orK -channel activation), BK elicited a transient and NO-dependentincrease in diameter (see Fig.4 ). These findings illustrate that both NO and EDHF contribute tothe phasic vasodilatory response of the afferent arteriole to BK. Moreover,although K -channel blockade (e.g., 1 mmol/l TEA, ChTX, AP) orelevation of extracellular K eliminated the vasodilator actions ofEDHF, a significant component of the NO-dependent vasodilation persisted inthese settings. Thus while evidence suggests a prominent role of BK Ca channels in the vasodilatory actions of NO (e.g., Ref. 31 ), these data suggest thatadditional mechanisms contribute to the response of the afferent arteriole toNO and/or cGMP. It is widely acknowledged that multiple pathways contribute tocGMP-dependent vasodilation and that the resultant smooth muscle relaxationmay involve both K -channel-dependent and -independent mechanisms( 20 ).

Finally, we must comment on our observation that the cortical afferentarterioles of our preparation exhibited a phasic or transient vasodilatorresponse to BK. The literature suggests a high degree of variability in regardto the profile of the renal vascular responses to BK. Thus Edwards( 8 ) reported that the afferentarteriole of the rabbit does not respond to BK, whereas Yu et al.( 37 ) found the samepreparation to exhibit vasodilation at very low BK concentrations andvasoconstriction at concentrations above 0.1 nmol/l. Ihara et al.( 15 ) reported triphasicresponses of the porcine ILA to BK, characterized by an initial dilation thatwas followed by a constriction and then a more slowly developing sustaineddilation. In each of these cases, the constriction was dependent on COX andthromboxane, whereas in our studies, the transient nature of the response persisted during COX inhibition. Imig et al.( 16 ) reported sustainedvasodilator responses of the blood-perfused juxtamedullary afferent arteriole.It is possible that observed differences in the time course of BK responsesmight reflect differences in the experimental preparations used, the impact ofacute trauma associated with the surgical preparations, the imposition ofhydronephrosis, species-related differences, or perfusate-type differences.The possibility of regional differences in the profile of the response withinthe renal microcirculation is also quite likely, as we observed sustainedresponses in larger segments of the ILA in the present study (see Fig. 2 ). Studies to addressregional differences in the temporal profile of the responses and any regionaldifferences in the determinants of the vasodilation produced by BK would be ofconsiderable interest.

In summary, the present study demonstrates that a component of theEDHF-like response of cortical afferent arterioles in the in vitro perfusedhydronephrotic rat kidney preparation to BK is sensitive to inhibition byeither cytochrome P -450 inhibition by 17-ODYA or by blockade ofBK Ca channels by 1.0 mmol/l TEA. These findings are consistent withthe proposed involvement of an EET in the EDHF-like response of this vessel toBK ( 16, 17, 38 ). However, we also found anadditional component of the EDHF-like response to BK, and this component wasnot prevented by 17-ODYA or TEA but was sensitive to inhibition by acombination of ChTX plus AP. This ChTX- and/or AP-sensitive component of theEDHF-like response was similar to that previously reported for ACh-inducedvasodilation of the afferent arteriole ( 36 ). These findings cannot beexplained by a single EDHF. Thus based on these observations, we suggest thatmultiple factors or hyperpolarizing pathways contribute to the EDHF-like response of the renal afferent arteriole.3 J) j" U6 f  E' E4 ^7 K

ACKNOWLEDGMENTS) o1 z# E1 ]) m

This study was supported by a grant from the Heart and Stroke Foundation ofAlberta, the Northwest Territories and Nunavut. R. Loutzenhiser is a SeniorMedical Scholar of the Alberta Heritage Foundation for Medical Research.
      【参考文献】
Archer SL,Gragasin FS, Wu X, Wang S, McMurtry S, Kim DH, Platonov M, Koshal A, HashimotoK, Campbell WB, Falck JR, and Michelakis ED. Endothelium-derivedhyperpolarizing factor in human internal mammary artery is 11,12-epoxyeicosatrienoic acid and causes relaxation by activating smooth muscleBK Ca channels. Circulation 107: 769-776,2003.4 ]; l8 _( f' t6 K6 Q% Y- I
7 T7 b( L! z; X: J% Y- ~" F% T) y6 S

Brandes RP,Schmitz-Winnenthal FH, Feletou M, Godecke A, Huang PL, Vanhoutte PM, FlemingI, and Busse R. An endothelium-derived hyperpolarizing factor distinctfrom NO and prostacyclin is a major endothelium-dependent vasodilator inresistance vessels of wild-type and endothelial NO synthase knockout mice. Proc Natl Acad Sci USA 97:9747-9752, 2000./ r3 g% \5 t  ~+ {

Chaytor AT,Evans WH, and Griffith TM. Central role of heterocellular gap junctionalcommunication in endothelium-dependent relaxation of rabbit arteries. J Physiol 508:561-573, 1998.
0 a7 k) Q- s2 X# `( a7 f
) B) B6 k4 `1 n& ]2 x8 j/ j

Demirel E,Rusko J, Laskey RE, Adams DJ, and Van Breemen C. TEA inhibits ACh-inducedEDRF release: endothelial Ca 2 -dependent K channels contribute to vascular tone. Am J Physiol Heart CircPhysiol 267:H1135-H1141, 1994., r7 E; H- o3 x8 G8 M; G# @& x4 j

5 `! ], S- o- k  c' [6 i
Doughty JM,Plane F, and Langton PD. Charybdotoxin and apamin block EDHF in ratmesenteric artery if selectively applied to the endothelium. Am JPhysiol Heart Circ Physiol 276:H1107-H1112, 1999.- T  H) ^- u: w9 r5 j
* r8 l& @6 _0 R  {

7 r0 P' N, K  l7 r6 W* x
Drummond GR,Selemidis S, and Cocks TM. Apamin-sensitive, non-nitric oxide (NO)endothelium-dependent relaxations to bradykinin in the bovine isolatedcoronary artery: no role for cytochrome P450 and K  . BrJ Pharmacol 129:811-819, 2000.+ z: a0 k) ?( y# k' Y* e
1 l4 o- B  `% h5 P
) k' q$ ?( k* T2 U1 P4 E
1 _" B, K* F8 G; d- T( U& `  I) H+ ?
Edwards G, DoraKA, Gardener MJ, Garland CJ, and Weston AH. K is anendothelium-derived hyperpolarizing factor in rat arteries. Nature 396:269-272, 1998.$ k3 L0 _& w, a  v

4 U. n( O! l. ]

Edwards RM. Response of isolated renal arterioles to acetylcholine, dopamine, andbradykinin. Am J Physiol Renal Fluid ElectrolytePhysiol 248:F183-F189, 1985.
* W, O: X* q$ X2 f! p
+ i+ v! Y8 [6 F# T! k  W5 Z

Fisslthaler B,Popp R, Kiss L, Pontente M, Harder DR, Fleming I, and Busse R. CytochromeP450 2C is an EDHF synthase in coronary arteries. Nature 401:493-496, 1999.& ~9 P& K$ J, |; B8 R9 P

( m7 I) q3 T, V6 Z% ~1 H

Frieden M,Sollini M, and Beny J. Substance P and bradykinin activate different typesof KCa currents to hyperpolarize cultured porcine coronary artery endothelialcells. J Physiol 519:361-371, 1999./ B* Y: F3 v( Y  a4 v
) b+ d+ `3 z$ v2 B8 ]

Fulton D,McGiff JC, and Quilley J. Contribution of NO and cytochrome P450 to thevasodilator effect of bradykinin in the rat kidney. Br JPharmacol 107:722-725, 1992.

! {3 Z0 e  n, `% x
( @- t/ o2 t- O2 _7 K, h+ S/ K; x
Gebremedhin D,Harder DR, Pratt PF, and Campbell WB. Bioassay of an endothelium-derivedhyperpolarizing factor from bovine coronary arteries: role of a cytochromeP450 metabolite. J Vasc Res 35:274-284, 1998.

4 `4 J" D" f+ _% Y0 P  X  c, G- a
Gui Y,Loutzenhiser R, and Hollenberg M. Bidirectional regulation of renalhemodynamics by PAR 1 and PAR 2 in isolated perfused ratkidney. Am J Physiol Renal Physiol 285: F95-F104,2003.. x" M" s! j# n2 j8 s' N

Hayashi K,Loutzenhiser R, Epstein M, Suzuki H, and Saruta T. Multiple factorscontribute to acetylcholine-induced renal afferent arteriolar vasodilationduring myogenic, norepinephrine and KCl-induced vasoconstriction. Circ Res 75:821-828, 1994.

+ b+ c/ E/ ~0 j% q
Ihara E, HiranoK, Derkach DN, Nishimura J, Nawata H, and Kanade H. The mechanism ofbradykinin-induced endothelium-dependent contraction and relaxation in theporcine interbar renal artery. Br J Pharmacol 129: 943-952,2000.3 t4 x& a9 q* L  `& L
  {/ E2 f, I; k$ |! R. n
5 H+ }; ]# t; l; B  J7 e
7 ^* R3 A& _5 D' \
Imig JD, FalckJR, Wei S, and Capdevila JH. Epoxygenase metabolites contribute to nitricoxide-independent afferent arteriolar vasodilation in response to bradykinin. J Vasc Res 38:247-255, 2001.4 H! F$ ^+ B- X$ k- ~

+ P2 ], v7 o$ [7 m  }; [' F  s
Imig JD, NavarLG, Roman RJ, Reddy KK, and Falck JR. Actions of epoxygenase metaboliteson the preglomerular vasculature. J Am Soc Nephrol 7: 2364-2370,1996.5 X  D* a/ W4 o; V5 _% ~: u
' I# ]9 {6 d* T' z6 M# x. s
. \, O3 y9 r+ O- Z% Z/ M5 N( Z: P  N

Kuhberger E,Groschner K, Kukovetz WR, and Brunner F. The role of myoendothelial cellcontact in non-nitric oxide-, non-prostanoid-mediated endothelium-dependentrelaxation of porcine coronary artery. Br J Pharmacol 113: 1289-1294,1994.2 |6 A, H% h- W% z

3 X- W  d; K( k+ R( u3 q
Langton PD,Nelson MT, Huang Y, and Standen NB. Block of calcium-activated potassiumchannels in mammalian arterial myocytes by tetraethylammonium ions. Am J Physiol Heart Circ Physiol 260: H927-H934,1991.
2 \8 [! p9 R/ S, l
$ s* m: b) Q; L

Lincoln TM, DeyN, and Sellak H. Invited review: cGMP-dependent protein kinase signalingmechanisms in smooth muscle: from the regulation of tone to gene expression. J Appl Physiol 91:1421-1430, 2001.6 `9 O% F5 _0 v! G. F% Q6 M. X
& d5 L7 d- L% z, W

Loutzenhiser R. In situ studies of renal arteriolar functionusing the in vitro perfused hydronephrotic rat kidney. Int RevExper Pathol 36:145-160, 1996.- {0 S/ j3 l/ k/ J5 ?. k
& O* F8 w. R2 Q% f. G! ~! q  G

Matoba T,Shimokawa H, Kubota H, Morikawa K, Fujiki T, Kunihiro I, Mukai Y, Hirakawa Y,and Takeshita A. Hydrogen peroxide is an endothelium-derivedhyperpolarizing factor in human mesenteric arteries. BiochemBiophys Res Commun 290:909-913, 2002.& m2 a5 i3 l; v, g5 a# x
% T0 B" @3 g3 X

  ^2 Q  F! s- x3 {* g2 E% Z: J# ~
McGuire JJ,Ding H, and Triggle CR. Endothelium-derived relaxing factors: a focus onendothelium-derived hyperpolarizing factors. Can J PhysiolPharmacol 79:443-470, 2001., f' _. J: N# a
# l9 D1 }6 z& P: `% \3 R
* W. }9 y" o5 V8 s) E
5 p/ \3 V% N' ^" i. N; y4 @$ A+ J/ X
Nilius B andDroogmans G. Ion channels and their functional role in vascularendothelium. Physiol Rev 81:1415-1459, 2001.( n- @6 i: E" _

) a! q5 B& m" j' @/ ^, G

Ohlmann P,Martinez MC, Schneider F, Stoclet JC, and Andriantsitohaina R. Characterization of endothelium-derived relaxing factors released bybradykinin in human resistance arteries. Br JPharmacol 121:657-664, 1997.
2 `' C0 C) `" U/ B- A# v
) T3 u4 m8 m2 [  Q: ?9 s8 A- b1 Z

Pompermayer K,Assreuy J, and Vieira MA. Involvement of nitric oxide and potassiumchannels in the bradykinin-induced vasodilatation in the rat kidney perfusedex situ. Regul Pept 105:155-162, 2002.

/ H0 G- C7 i: M2 K! e7 m0 e
Quilley J,Fulton D, and McGiff JC. Hyperpolarizing factors. BiochemPharmacol 54:1059-1070, 1997.3 K2 V5 T) P% X5 n" n) s0 G

& u( x7 b6 a  K. O8 d
Quilley J andMcGiff JC. Is EDHF an epoxyeicosatrienoic acid? TrendsPharmacol Sci 21:121-124, 2000.
3 ]1 j" f2 i  \" Y

# s) F/ G' l4 X  c$ d
Rapacon M,Mieyal P, McGiff JC, Fulton D, and Quilley J. Contribution ofcalcium-activated potassium channels to the vasodilator effect of bradykininin the isolated, perfused kidney of the rat. Br JPharmacol 118:1504-1508, 1996.

" x5 S& I# H# h$ G; a
% h- G2 D5 A4 x" f
Sandow SL, TareM, Coleman HA, Hill CE, and Parkington HC. Involvement of myoendothelialgap junctions in the actions of endothelium-derived hyperpolarizing factor. Circ Res 90:1108-1113, 2002.

( ?! C$ Q( [9 `) f! n7 v( ]$ s1 F8 r
Schubert R andNelson MT. Protein kinases: tuners of the BKCa channel in smooth muscle. Trends Pharmacol Sci 22:505-512, 2001.8 S1 Z& l6 ?  X7 @) z

3 s; H+ ]; x) f( j$ A
8 h- l! R) _9 m1 b! l: v
Schweitz H,Stansfeld CE, Bidard J, Fagni L, Maes P, and Lazdunski M. Charybdotoxinblocks dendrotoxin-sensitive voltage-activated K channels. FEBS Lett 250:519-522, 1989.

, O! j1 B! u6 J1 N+ [( q

Strong PN. Potassium channel toxins. Pharmacol Ther 46: 137-162,1990.

' e! |/ e8 T% [- O; R
3 a4 x8 d, e# X! m+ X
Triggle DR,Dong H, Waldron GJ, and Cole WC. Endothelium-derived hyperpolarizingfactor(s): species and tissue heterogeneity. Clin Exp PharmacolPhysiol 26:176-179, 1999.

% I9 _- L6 L0 k" K( E* D

Trottier G,Hollenberg M, Wang X, Gui Y, Loutzenhiser K, and Loutzenhiser R. Proteinase-activated receptors elicit afferent arteriolar vasodilation byNO-dependent and NO-independent actions. Am J Physiol RenalPhysiol 282:F891-F897, 2002." d' x3 Z5 h1 j- \# T8 I5 o

Wang X andLoutzenhiser R. Determinants of the renal microvascular response toacetylcholine: afferent and efferent arteriolar actions of EDHF. AmJ Physiol Renal Physiol 282:F124-F132, 2002.
" u, f. Z9 r; r9 D0 ?
  K$ J! D4 B+ U* A. q

Yu H, CarreteroOA, Juncos LA, and Garvin JL. Biphasic effect of bradykinin on rabbitafferent arterioles. Hypertension 32: 287-292,1998.
& Y$ _# l3 {# X

Zou AP, FlemingJT, Falck JR, Jacobs ER, Gebremedhin D, Harder DR, and Roman RJ. Stereospecific effects of epoxyeicosatrienoic acids on renal vascular tone andK-channel activity. Am J Physiol Renal Fluid ElectrolytePhysiol 270:F822-F832, 1996.

收藏0 分享0 顶0 踩0