Bradykinin and lys-bradykinin generated intrarenally appear to be important renal paracrine hormones. However, the renal effects of endogenously generated bradykinin are still not clearly defined. In this study, we measured acute changes in renal excretory and hemodynamic functions and renal cortical interstitial fluid levels of bradykinin, prostaglandin E2, and cGMP in response to an acute intrarenal arterial infusion of the bradykinin B2 receptor antagonist Hoe 140 (icatibant), cyclooxygenase inhibitor indomethacin, or nitric oxide synthase inhibitor NG-monomethyl-l-arginine (L-NMMA) given individually or combined in uninephrectomized, conscious dogs (n=10) in low sodium balance. Icatibant caused a significant decrease in urine flow, urinary sodium excretion, and renal plasma flow rate (each P<.001). Glomerular filtration rate did not change during icatibant administration. Icatibant produced an unexpected large increase in renal interstitial fluid bradykinin (P<.0001) while decreasing renal interstitial fluid prostaglandin E2 and cGMP (each P<.001). Both indomethacin and L-NMMA when given individually caused significant antidiuresis and antinatriuresis and decreased renal blood flow (each P<.001). Glomerular filtration rate decreased during L-NMMA administration (P<.001) and did not change during indomethacin administration. Combined administration of icatibant and indomethacin or L-NMMA caused significant decreases in renal excretory and hemodynamic functions, which were not different from changes observed with icatibant alone. The failure of icatibant to change renal function after inhibition of cyclooxygenase and nitric oxide synthase activity suggests that the effects of kinin B2 receptor are mediated by intrarenal prostaglandin E2 and nitric oxide generation. The increase in renal interstitial fluid bradykinin during icatibant requires further study of possible alterations in kinin synthesis, degradation, or clearance as a result of B2 receptor blockade.
Considerable evidence now suggests that the renal kallikrein-kinin system is involved in the regulation of sodium and water excretion and may participate in blood pressure control. Kinins can stimulate production of both NO and eicosanoids.1 2 At least a portion of the renal effects of kinins appears to be mediated by prostaglandins,3 and inhibition of NO synthesis reduces the renal vasodilator response to bradykinin.4 In previous studies, intrarenal infusions of bradykinin in anesthetized dogs resulted in an increase in RPF and UNaV.5 6 However, the biochemical and physiological effects of exogenously administered kinins may differ from those of endogenously formed kinins.7 Administration of Fab fragments of kinin antibodies, bradykinin antagonist,8 and Fab fragments of monoclonal antibodies to rat urinary kallikrein (Fab-kallikrein)9 decreased UV, UNaV, and PGE2 excretion in rats. The purpose of the present study was to investigate the effects of kinin B2 receptor blockade on intrarenal levels of bradykinin, PGE2, and cGMP, an indirect measure of NO generation. Because previous studies showed that urinary10 and renal interstitial11 kinins increase in response to sodium depletion, we carried out our studies in conscious dogs in sodium metabolic balance at 5 mEq/d. We used a potent and specific bradykinin B2 receptor antagonist,12 13 d-Arg-Arg-Pro-Hyp-Gly-Thi-Ser-l-Tic-Oic-Arg (Hoe 140, icatibant) (Hoechst AG), to begin to evaluate the role of intrarenal bradykinin and this receptor in the regulation of renal levels of PGE2 and NO in conscious, uninephrectomized dogs.
Renal Microdialysis Technique
For the determination of RIF kinin, PGE2, and cGMP, we constructed a microdialysis probe as previously described.11 14 15 Each end of a 1-cm-long hollow fiber dialysis tube (0.3 mm ID; molecular mass cutoff, 10 000 D; Hospal) was inserted into a manually dilated end of two 30-cm-long (inflow and outflow) hollow polyethylene tubes (0.12 mm ID, 0.65 mm OD; Bioanalytical Systems). The distance between the ends of the polyethylene tubes was 3 mm (dialysis area), and the dialysis fiber was sealed in place within the polyethylene tubes with cyanoacrylic glue. The dead volume of the dialysis tubing and outflow tube was 3.6 μL. The known kinin-, eicosanoid-, and NO-generating and -degrading enzymes (molecular weights, 34 000 to 150 000) will not cross the dialysis membrane because of their size. The microdialysis probe was sterilized by a gas sterilization method.
Animal Preparation and RIF Collection
The study protocol was approved by the University of Virginia Animal Research Committee. All studies were carried out in conscious animals to avoid disturbances in renal hormones and function induced by anesthesia.16 Ten female mongrel dogs weighing 15 to 20 kg (mean±SE, 18±0.89 kg) were used in this study. Surgery with dogs under halothane anesthesia was performed as described previously for cannula insertion into the right renal artery, aorta, and inferior vena cava.17 An ultrasonic flow probe (Transonic Systems) was placed around the right renal artery18 for measurement of renal blood flow. A left nephrectomy was performed to eliminate renal compensation and avoid ureteral catheterization in conscious dogs. All cannulas and the wires for the right renal flow probe were tunneled subcutaneously through a bevel-tipped stainless steel tube and exteriorized in separate small stab wounds near the interscapular region.
For placement of the microdialysis probe, the right renal capsule was penetrated with an 18-gauge needle that was tunneled under the capsule for 2 cm before exiting again. One end of the dialysis probe then was pulled through the needle until the dialysis fiber was situated in the outer renal cortex about 2 mm from the kidney surface. The needle then was withdrawn. The two ends of the dialysis probe (inflow and outflow tubes) were tunneled under the skin and exited near the interscapular region. The exterior ends of the vascular cannulas and microdialysis probe were protected in the pocket of a close-fitting nylon jacket (Alice King Chatham). In an earlier study,15 a histological examination of the renal tissue 6 weeks after insertion of the dialysis probe did not show any significant fibrosis or scarring. Three weeks was allowed for recovery after surgery.
General Animal Care
Prevention of clot formation inside the cannulas was ensured by daily flushing with 3 mL of 10% heparin solution (Elkins Sinn). Cannulas also were flushed after each withdrawal of blood or hemodynamic measurement. As a prophylactic measure, antibiotic treatment with 250 mg ampicillin IM (Wyeth-Ayerst Laboratories) twice daily was started immediately after surgery and continued for 7 days. Body temperatures were measured twice daily during the first week after surgery and twice weekly thereafter. Blood urea nitrogen and serum creatinine levels were determined twice weekly (Serometer model 370, Mallinckrodt). Dogs were excluded from the study in the event of elevated temperature or abnormalities (blood urea nitrogen >20 mg/dL, creatinine >1.5 mg/dL) in these tests. Hematocrit levels were 41% to 46% (mean±SE, 43.6±0.87%) and 41% to 45% (42.6±0.82%) before surgery and on the day of the acute study, respectively.
Systemic arterial pressure was monitored continuously during the acute study from the aortic cannula with a Statham P23 DB pressure transducer and Gould 2400 recorder, as previously described.17 Mean arterial blood pressure was calculated as the mean of three recordings lasting 15 seconds each and taken at 2-minute intervals at the middle of the control and icatibant infusion periods.
Urinary sodium concentrations were measured by an autoanalyzer (NOVA-1, NOVA Biomedical). Venous blood samples (1 mL) for aldosterone measurements were collected at the middle of each experimental period in glass tubes coated with lithium heparin. Peripheral PAC was measured by a Coat-a-Count radioimmunoassay kit (Diagnostic Products). Peripheral venous blood (1 mL) for the determination of PRA was collected at the middle of each experimental period in tubes that contained EDTA (4°C), and plasma was separated after centrifugation and stored at −70°C. Peripheral PRA was measured by radioimmunoassay of angiotensin I generation after 3 hours of incubation19 with reagents obtained from New England Nuclear. RIF kinin, PGE2, and cGMP samples were stored at −20°C until they were measured by radioimmunoassays.20 21 22
Sodium and Fluid Balance
Sodium metabolic balance was established in all dogs by placing each dog on a palatable diet containing 5 mmol sodium/d (Hills' H/D) for 15 days. All dogs were in low sodium balance for 10 days before study. The rationale for use of the low sodium diet is provided by our previous study15 which showed that this diet increases renal interstitial kinins in the conscious dog as well as other work suggesting that the renal kallikrein-kinin system is activated by this maneuver.10 Dietary potassium intake was 40 mEq/d. Dogs had free access to water during the study. They were kept in metabolic cages; 24-hour urine collections were made, and UNaV was measured to confirm the state of sodium balance.
For collection of RIF bradykinin, PGE2, and cGMP samples, the microdialysis inflow tube was connected to a gas-tight syringe filled with lactated Ringer's solution and perfused at 3 μL/min (pump 22, Harvard Apparatus). At this perfusion rate, in vitro recovery was 78% for bradykinin15 and 70% for cGMP.11 The effluent was collected from the outflow tube. For kinin measurements, effluent was collected in nonheparinized plastic tubes containing 0.1 mL ethanol. The effluent for PGE2 and cGMP measurements was collected in plastic tubes on ice.
Effects of Acute Intrarenal Administration of Icatibant
On the day of the acute study, a Foley catheter was inserted into the bladder for urine collections. The dogs (group 1, n=5) rested quietly in a canvas sling from 8 am until the completion of each study. The dogs were infused intravenously with inulin (Sigma Chemical Co). Five percent dextrose solution (D5W, Travenol Laboratories) was infused intrarenally at 1.0 mL/min as a vehicle for the agent. A 90-minute period was allowed for equilibration, followed by a 60-minute control period of urine and interstitial fluid collection. After the control period, and with the continuation of interstitial fluid and urine collection, icatibant (99% purity, Peninsula Laboratories) was administered intrarenally (100 ng/kg per minute) for 180 minutes. This infusion rate caused no systemic effects in preliminary studies. Icatibant was prepared initially from lyophilized powder in D5W vehicle and sterilized by filtration (Millipore filters). Blood pressure was monitored as previously described. Changes in RIF kinin, PGE2, and cGMP; PAC; and PRA were determined during intrarenal administration of D5W as a vehicle for icatibant (control period) and during intrarenal administration of icatibant (experimental period). Blood samples (5 mL) for PRA, PAC, and inulin measurements were obtained in the middle of each urine collection period. GFR was calculated on the basis of inulin clearance. Renal blood flow (RBF) was measured by an ultrasonic renal flow probe, and RPF was calculated as RPF=RBF(1−Hematocrit).
Time Control Studies
After sodium metabolic balance of 5 mEq sodium/d was established, dogs (n=5) were infused intravenously with inulin. A study similar to that described above was carried out, with the exception that after the control period, D5W was administered intrarenally instead of the kinin B2 receptor blocker. Changes in renal function; RIF kinin, PGE2, and cGMP; PAC; and PRA were determined as stated above.
Effects of Intrarenal Administration of Icatibant and Indomethacin Alone or Combined
To evaluate whether PGE2 mediates the observed changes in renal function during bradykinin receptor blockade with icatibant, we repeated the above study of intrarenal administration of the kinin B2 receptor blocker icatibant (100 ng/kg per minute) in another group of dogs (group 2, n=5) on a low sodium diet at 5 mmol/d. Under the same metabolic conditions, the study was repeated on a different day except that indomethacin (Merck) was given at 0.1 μg/kg per minute for 30 minutes instead of icatibant. This was followed by another 30 minutes of intrarenal administration of indomethacin (0.1 μg/kg per minute) and icatibant (100 ng/kg per minute). This indomethacin dose has been shown to cause a 50% reduction in renal venous plasma immunoreactive prostaglandins.18
Effects of Intrarenal Administration of L-NMMA
To evaluate whether NO mediates the observed changes in renal function during icatibant administration, we repeated the above study (group 2, n=5), except that we administered the NO synthase inhibitor L-NMMA (Bachem Bioscience Inc) at 60 μg/kg per minute instead of indomethacin. This was followed by another 30 minutes of intrarenal administration of L-NMMA (60 μg/kg per minute) and icatibant (100 ng/kg per minute). Intrarenal administration of L-NMMA at 60 μg/kg per minute was shown to decrease renal excretory and hemodynamic functions and RIF cGMP.19
Statistical Analysis of Data
Data were examined by ANOVA, including a repeated measures term, using the SAS general linear model procedure (SAS Institute). Comparison between effects of infused drugs and vehicle alone were conducted with a paired t test. Data are expressed as mean±SE. Statistical significance was identified at a level of P<.05.
Renal Excretory and Hemodynamic Changes in Response to Intrarenal Icatibant
UNaV on the day before study was 4.5±0.2 mmol/24 h. UV was stable during intrarenal vehicle administration (time control). In contrast, UV was 1.04±0.06 mL/min during the control period and fell markedly to 0.58±0.08 (P<.01) in response to icatibant (experimental period) (Fig 1⇓). Similarly, UNaV did not change during the time control study. UNaV was 18.9±1.6 μmol/min before administration of intrarenal icatibant and fell 76% during its infusion to 4.6±0.8 (P<.01).
RPF was stable during the time control study. It was 111±8 mL/min during the control period and decreased to 79±10 (P<.01) during intrarenal icatibant administration (Fig 1⇑). No significant change in GFR was produced by intrarenal vehicle administration (time control) or icatibant (23.4±1.7 versus 22.8±1.6 mL/min, control versus icatibant).
RIF Kinin, PGE2, and cGMP
During the time control study, no changes in RIF kinin, PGE2, or cGMP were observed. In contrast, kinin measured by radioimmunoassay in dialysates collected during control periods was 32±4.3 pg/min and increased significantly to 122±15 (P<.01) during icatibant administration (Fig 2⇓). To rule out the possibility of the drug interfering with bradykinin measurements in the radioimmunoassay, we also added icatibant in excess (1 μg) to the bradykinin standards. It did not interfere with bradykinin binding compared with bradykinin standard curves (n=3). These data show that the elevated kinin level after drug is not a result of its recognition by the kinin antiserum. In addition, dog serum kininase II activity was measured by a commercial radioimmunoassay (Hycor Biomedical Inc) in the presence and absence of icatibant (1 μg per tube). The kininase II activity was 112±6.8 U in the absence of icatibant and 108±5.2 U in its presence, suggesting that the drug does not inhibit kininase II activity.
In contrast to the kinin levels in the same dogs, the PGE2 appearance rate decreased during icatibant administration from 11.4±0.4 pg/min during the control periods to 5.1±0.3 during drug infusion (P<.01). The renal cGMP appearance rate also decreased from 0.65±0.03 pmol/min during the control period to 0.31±0.01 during drug infusion (P<.01).
Responses to Intrarenal Infusion of Icatibant, Indomethacin, or L-NMMA Individually or Combined
Table 1⇓ shows the lack of response of PRA, PAC, and mean arterial blood pressure to intrarenal icatibant, indomethacin, or L-NMMA individually or combined.
Changes in Response to Intrarenal Administration of Icatibant or Indomethacin Individually or Combined
Similar to the first group of dogs, icatibant caused significant decreases in UV, UNaV, and RPF (P<.001) and did not affect GFR (Table 2⇓). Indomethacin caused a significant decrease in UV and UNaV from 1.1±0.04 to 0.8±0.05 mL/min and 20.0±2.1 to 8.6±1.1 μmol/min, respectively (each P<.001). RPF and GFR did not change significantly. UV decreased significantly from 1.1±0.08 to 0.7±0.04 mL/min, UNaV from 22.6±2.2 to 8.8±0.7 μmol/min, and RPF from 111±6.8 to 99±4.8 mL/min compared with control (each P<.001) during combined intrarenal administration of indomethacin and icatibant. GFR did not change significantly during combined administration of indomethacin and icatibant. Combined administration of indomethacin and icatibant did not produce significant changes in the measured renal parameters over changes observed during administration of indomethacin alone.
Changes in Response to Intrarenal L-NMMA Alone or Combined With Icatibant
L-NMMA caused a significant decrease in UV from 1.1±0.07 to 0.5±0.03 mL/min, UNaV from 21.2±2.0 to 6.2±0.6 μmol/min, RPF from 113±2.1 to 88±3.2 mL/min, and GFR from 25±2.6 to 21±0.6 mL/min (each P<.001) (Table 2⇑). These renal excretory and hemodynamic parameters continued to decrease during combined intrarenal administration of L-NMMA and icatibant but were not significantly different from changes with L-NMMA alone.
Current knowledge suggests that the renal kallikrein-kinin system participates in the control of water and electrolyte excretion and regulation of blood pressure.7 Bradykinin is known to release either NO or eicosanoids from vascular endothelial, renal interstitial, or epithelial cells.1 2 At least a portion of the renal effects of kinins is thought to be mediated by these messengers.3 However, a lack of specific tools to interrupt the kallikrein-kinin system, as well as a lack of methodology to measure its local in situ activity, has made it difficult to clearly delineate the roles of kinins in the regulation of renal hemodynamic and excretory functions.
Here, we were able to test the hypothesis that intrarenal bradykinin plays a role in the regulation of renal function either directly or through eicosanoids and NO in conscious dogs using icatibant, a specific kinin B2 receptor antagonist,23 while measuring RIF mediators concomitantly with some renal functional parameters. We reasoned that if kinins act locally within the kidney to regulate renal function and renal generation of eicosanoids and NO, then intrarenal bradykinin receptor blockade should evoke measurable changes in both.
Acute administration of icatibant during a period of chronic low sodium balance caused significant antidiuresis and antinatriuresis and decreased RPF, whereas GFR remained unchanged. This is in agreement with a previous study using a different B2 receptor blocker.18 In that study, the kinin receptor antagonist decreased UV and UNaV in dogs during low sodium but not during high sodium intake. Collectively, these results may relate to the observation that a chronic low sodium diet is associated with high renal interstitial kinin levels, whereas during a high sodium diet, such levels are barely detectable.11 It is still unclear why an increase in the renal interstitial level of kinin occurs in response to a reduction in dietary sodium intake. However, the activation of the renal kallikrein-kinin system during chronic low sodium intake suggests that the system functions to counteract antinatriuretic or vasoconstrictor mechanisms activated in response to sodium depletion. It is unlikely that the observed increase in intrarenal bradykinin levels displaced the icatibant from B2 receptors. If this happened, an opposite response to what we observed in this study should occur. Studies in rats treated with deoxycorticosterone acetate and salt8 showed a decrease in UV and UNaV in response to Fab-bradykinin and a decrease in UV in response to bradykinin antagonist. Neither altered blood pressure, renal blood flow, or GFR.8 In another study,9 Fab-kallikrein decreased kinin excretion, UV, and UNaV and slightly decreased urinary PGE2 excretion in rats on a normal sodium diet. Fab-kallikrein did not alter blood pressure, renal blood flow, or GFR.9 In contrast to previous studies, the present study is the first to report changes in RIF kinin, PGE2, and cGMP in conscious dogs on a low sodium metabolic balance. Kinins can be formed and degraded in urine, and their measurement in RIF would avoid this problem.
The antinatriuresis associated with icatibant administration is associated with a decrease in RPF and no change in GFR. Both indomethacin, a cyclooxygenase inhibitor, and L-NMMA, an NO synthase inhibitor, caused a decrease in UV, UNaV, and RPF. A decrease in GFR was observed only during L-NMMA administration. Combined administration of the B2 receptor antagonist icatibant with either indomethacin or L-NMMA caused a decrease in UV, UNaV, and RPF. There was a lack of a synergistic effect between indomethacin and icatibant, which suggests that the effects of both agents are due to inhibition of renal kinin action. Combined icatibant and L-NMMA significantly decreased GFR. The reason for the discrepancy between the effects of icatibant alone and combined icatibant and L-NMMA on GFR is that renal hormones other than bradykinin can stimulate NO.24 It is expected that icatibant administration will not completely abolish renal NO production, whereas L-NMMA, by inhibiting the NO synthase, can produce a further decrease in renal NO levels, thereby possibly explaining the observed difference in GFR response. The decrease in RPF during icatibant and L-NMMA treatments was more pronounced than its decrease during indomethacin administration. This could be mainly due to a decrease in renal NO associated with icatibant and L-NMMA. The observed effects of indomethacin are consistent with previous studies showing that indomethacin is less effective in reducing RPF in conscious dogs.25 It is important to note that the responses observed in our uninephrectomized conscious dogs may be different in other animal models (ie, anesthetized or two-kidney model). Collectively, these data suggest that renal effects of acute administration of icatibant might be mediated by both local hemodynamic and tubular mechanisms, including the partial blockade of eicosanoids or NO release.1 2 18 Since there is a correlation between NO-induced vasodilation and NO-induced cGMP accumulation in vascular smooth muscle cells, cGMP accumulation can be a useful tool for estimation of NO release. Our ability to detect changes in the levels of RIF bradykinin, PGE2, and cGMP with variations in salt intake11 suggests that these substances play a role in the regulation of renal function. The observed increase in RIF bradykinin during administration of a kinin B2 receptor blocker could result from decreased bradykinin uptake, degradation, or clearance mediated by B2 receptors. In this study, icatibant did not affect kininase II (angiotensin-converting enzyme) activity in serum; however, we do not know whether the agent affected intrarenal kinin degradation, although we are unaware of any such effect. Since the bradykinin antiserum used in the radioimmunoassay failed to recognize icatibant, it is unlikely that an assay artifact caused the unexpected increase in RIF bradykinin.
In the present study, it is not clear whether the increased levels of RIF bradykinin during icatibant administration influence B1 receptor activity. It has been reported that both B1 and B2 receptors mediate vasodilation in canine renal arteries in vitro26 and in vivo.27 B1 receptor blockade causes transient decreases in UNaV.23 Thus, it is possible that the increase in RIF bradykinin may stimulate B1 receptors; however, renal blood flow and UNaV should increase, which is opposite to our findings. This interesting possibility can be addressed in future work assessing the relative contributions of these different receptor populations to canine renal function.
The decreased levels of RIF PGE2 and cGMP during the acute B2 receptor antagonist while dogs were on a low sodium diet suggest that bradykinin partially mediates the generation of these substances in the kidney. The simultaneous decreases in UNaV and RPF during acute icatibant, indomethacin, or L-NMMA administration suggest that these renal functions are influenced by intrarenal kinin generation, PGE2, and NO. These data represent acute responses to B2 receptor blockade. Furthermore, the failure of icatibant to decrease renal excretory and hemodynamic functions after intrarenal cyclooxygenase and NO synthase inhibition suggests that the renal effects of kinins are mediated by eicosanoids and NO.
Selected Abbreviations and Acronyms
|GFR||=||glomerular filtration rate|
|PAC||=||plasma aldosterone concentration|
|PRA||=||plasma renin activity|
|RIF||=||renal interstitial fluid|
|RPF||=||renal plasma flow|
|UNaV||=||urinary sodium excretion|
|UV||=||urine flow rate|
This work was supported by grants HL-47669, HL-17705, and HL-44671 from the National Heart, Lung, and Blood Institute, National Institutes of Health, to H.M.S. or to H.S.M. Dr Helmy M. Siragy is the recipient of a Research Career Development Award (1 K04 HL-03006) from the National Institutes of Health, Bethesda, Md.
- Received February 21, 1996.
- Revision received March 26, 1996.
- Revision received October 2, 1996.
Salvemini D, Seibert K, Masferrer JL, Misko TP, Currie MG, Needleman P. Endogenous nitric oxide enhances prostaglandin production in a model of renal inflammation. J Clin Invest.. 1994;93:1940-1947.
Flamenbaum W, Gagnon J, Ramwell P. Bradykinin-induced renal hemodynamic alterations: renin and prostaglandin relationships. Am J Physiol. 1979;237:F433-F440.
Beierwalters W, Carretero OA, Scicli AG. Renal hemodynamics in response to a kinin analogue antagonist. Am J Physiol. 1988;255:F408-F414.
Carretero OA, Scicli AG. The renal kallikrein-kinin system. Am J Physiol. 1980;238:F247-F255.
Saitoh S, Scicli AG, Peterson E, Carretero OA. Effect of inhibiting renal kallikrein on prostaglandin E2, water, and sodium excretion. Hypertension. 1995;25:1008-1013.
Barabe J, Huberdeau D, Bernoussi A. Influence of sodium balance on urinary excretion of immunoreactive kinins in rats. Am J Physiol. 1988;254:F484-F491.
Siragy HM, Ibrahim MM, Jaffa AA, Mayfield R, Margolius HS. Rat renal interstitial bradykinin, prostaglandin E2, and cyclic guanosine 3′,5′-monophosphate: effects of altered sodium intake. Hypertension. 1994;23(part 2):1068-1070.
Siragy HM, Jaffa AA, Margolius HS. Stimulation of renal interstitial bradykinin by sodium depletion. Am J Hypertens. 1993;6:863-866.
Johnson MD, Malvin RL. Plasma renin activity during pentobarbital anesthesia and gradual hemorrhage in dogs. Am J Physiol. 1975;229:1098-1101.
Siragy HM. Evidence that intrarenal bradykinin plays a role in regulation of renal function. Am J Physiol. 1993;265:E648-E654.
Sealey JE, Laragh JH. How to do a plasma renin assay. Cardiovasc Med. 1977;2:1079-1092.
Harper JF, Brooker G. Femtomole sensitive radioimmunoassay for cyclic AMP and cyclic GMP after 2′0 acetylation by acetic anhydride in aqueous solution. J Cyclic Nucl Res. 1975;1:207-218.
Lortie M, Regoli D, Rhaleb N, Plante GE. The role of B1- and B2-kinin receptors in the renal tubular and hemodynamic response to bradykinin. Am J Physiol. 1992;262:R72-R76.