Rat Renal and Plasma Prorenin Are Activated In Vitro by Different Mechanisms
Abstract—The aim of the present study was to purify and identify a plasma protein fraction (PreR-Co) involved in renal prorenin activation and to explore its capacity to process plasma prorenin. PreR-Co was obtained from plasma as a single electrophoretic band by (NH4)2SO4 precipitation, Sephacryl S-200 HR gel filtration, anti-rat albumin immunoaffinity, and ion-exchange chromatography. The amidase, esterase, and kallikrein activities of PreR-Co were studied, as was its N-terminal amino acid sequence. Rat kidney extract or plasma (normal or previously treated with acid to pH 2.8) were incubated with PreR-Co for 15 minutes at 37°C. Renin concentration was measured by incubation with homologous angiotensinogen. The same protocol was repeated with samples activated by trypsin. The N-terminal amino acid sequence was IIGGSMDAKGSFP, which had a homology of 90% with the β-chain of haptoglobin, 69% with serine-proteases, and 65% with kallikreins. The renin concentration in rat kidney extract was 34±4 ng of angiotensin I (Ang I) · mg of tissue−1 · h−1. After PreR-Co or trypsin treatments, renin concentrations were 211±7 and 110±11 ng of Ang I · mg of tissue−1 · h−1, respectively. The plasma renin concentration in normal plasma was 67.6±13.3 ng of Ang I · mL−1 · h−1, and no significant difference was observed after PreR-Co treatment. However, a significant increase (202.8±7.8 ng of Ang I · mL−1 · h−1; P<0.01) was found after trypsin treatment. The isolated PreR-Co acts on renal prorenin but not on plasma prorenin. These results suggest that active renin is processed in the kidney by a circulating enzyme that may have a role in the regulation of circulating renin.
In 1970, De Vito et al1 reported that active renin in rat kidney slices originated from an inactive precursor. Later, acidification and limited proteolysis experiments showed that inactive renin was present in amniotic fluid and plasma.2 Inactive renin was also found in rabbit kidney extracts3 and in blood and tumor extracts from patients with Wilms’ tumors.4 The presence of inactive renin in plasma, biological fluids, and tissues has generated a large list of possible functions; however, its physiological significance remains obscure. Several blood enzymes have been implicated in the activation of plasma prorenin in vitro5 ; nevertheless, it has not yet been demonstrated that these enzymes have a similar role in vivo. In addition, the precise mechanism of renin conversion in kidneys and plasma, if it does occur, remains to be elucidated.
Recently, we isolated a protein fraction from rat plasma that activates renal prorenin.6 Despite its isolation as a single electrophoretic band, this protein fraction was not pure, and it seemed to contain significant amounts of albumin, which interfered with the precise identification of the enzyme. In addition, some preparations also had small amounts of angiotensinogen. The purposes of the present study were, therefore, to (1) change the purification method to avoid these obstacles; (2) identify the enzyme; and (3) study whether the enzyme that processes prorenin to renin in kidney homogenates can also activate prorenin in plasma.
Experiments were performed on female albino rats of a locally bred strain (originally Sprague-Dawley stock). They were housed in group cages (6 animals per cage) and kept under conditions of controlled temperature (24±2°C) and light (lights on between 6:00 am and 8:00 pm). The rats had free access to a balanced diet and water. Per the guidelines of the American Physiological Society and approved by the animal board of the Consejo Nacional de Investigaciones Científicas y Técnicas, only rats weighing 220 to 250 g were used in this study.
Rat albumin (99% agarose electrophoresis, Lot 73H9320), Freund’s complete adjuvant, EDTA disodium salt (ACS reagent), PMSF, neomycin sulfate, trypsin (from bovine pancreas type III; 7857 U of BAEE/mg), soybean trypsin inhibitor (SBTI) (type I-S; inhibitor capacity, 1.5 mg of trypsin/mg or ≈10 000 U of BAEE/mg), BAPNA, bovine serum albumin (Cohn fraction V), glycine, and CaCl2 (ACS reagent) were obtained from Sigma. Sephacryl S-200 HR, cyanogen bromide-activated Sepharose 4B, and Sephadex G-25 were purchased from Pharmacia, and BCECF acetoxymethyl ester (AM) was from Molecular Probes.
Purification of a Plasma Protein That Activates Renal Prorenin
The purification of the active fraction was performed by (NH4)2SO4 precipitation, gel filtration on Sephacryl S-200 HR, and ion-exchange chromatography, as described previously.6
Preparation of Anti-Rat Albumin Antibody
Anti-albumin antibody was raised in rabbits, and the serum antibodies were partially purified by precipitation with (NH4)2SO4. Total protein was 26.93 mg/mL7 ; it was stored at −40°C if not immediately used.
Antibodies were linked to cyanogen bromide-activated Sepharose 4B according to the method of Cuatrecasas and Anfinsen,8 and the gel was stored in 0.9% NaCl containing 0.02% sodium azide at 4°C. A total of 2 mL of active fraction from the gel filtration column (10 mg of protein) was applied to a column containing 24 mL of immunoaffinity gel, which was previously equilibrated with 0.9% NaCl solution. Elution was performed with the same solution, and proteins were monitored by absorbance at 280 nm. Activation of renal prorenin was investigated in every fraction as described below. The presence of albumin in each fraction was determined by immunoprecipitation in liquid medium (ring-test).
Ion Exchange Chromatography
The active fraction from the immunoaffinity column was concentrated 20 times in a Speed Vac evaporator, and 100 μL of the concentrate (200 μg of protein) was run through an anion-exchange column (Mono Q HR 5/5; Pharmacia), as previously described.6 The active fraction obtained from the second column was named PreR-Co.
Determination of N-terminal Amino Acid Sequence
For N-terminal amino acid sequencing, the band on the SDS-PAGE gel was blotted onto a polyvinylidene fluoride membrane and sent to 2 separate laboratories: Laboratorio Nacional de Investigación y Servicios en Péptidos y Proteínas (LANAIS-PRO, Buenos Aires, Argentina) and BIO-SYNTHESIS (Lewisville, Texas).
Determination of Amidase Activity
Spectrophotometric assays9 with BAPNA were performed to determine amidase activity. A mixture containing 400 μL of the BAPNA solution and 100 μL of the PreR-Co solution sample (3 μg of protein) was incubated at 25°C; the p-nitroaniline that was liberated was determined spectrophotometrically at 410 nm. The amount of substrate hydrolyzed was expressed as BAPNA units (amount of enzyme that will hydrolyze 1 μmol/L substrate per minute).
Determination of Esterase Activity
Fluorometric assays with a fluorogenic esterase substrate (BCECF AM) were done by incubating 1.0 μL of the sample (0.5 μg of protein) with 1.0 μL of BCECF AM. The cleavage of the ester bond was assayed fluorometrically by measuring the emission at 540 nm with excitation at 500 nm in an Amicon-Bowman Spectrophotofluorometer (American Instrument Co, Inc).
Determination of Kininogenase Activity
Kininogenase activity was measured by incubating 20 or 40 μL of PreR-Co solution (2 to 4 μg of protein) with 200 μL of partially purified dog kininogen (2 μg of kinin-releasing capability) in the presence of 1 mL of fresh, 0.1 mol/L Tris-HCl buffer (pH 8.5) containing EDTA (15 mg), 1,10-phenanthroline (1 mg), and 8 OH-quinoline (1 mg), with or without the addition of SBTI (100 μg/mL). Inactive kallikrein was also investigated by incubating the sample with 20 μg of trypsin for 30 minutes at 37°C. The reaction was stopped by adding 100 μg of SBTI. Kinins generated during incubation were measured by radioimmunoassay. Kininogenase activity was expressed as the amount of kinins generated per micron of protein per minute of incubation with kininogen.
Determination of Renin Concentration
Renin was measured by incubating the samples with a homologous angiotensinogen preparation obtained from male rats that had a nephrectomy 48 hours previously.10 The powder was dissolved immediately before use in 0.9% NaCl buffered with phosphate-citrate buffer (75 mmol/L, pH 6.8) containing 4 mmol/L EDTA, 1.4 mmol/L PMSF, and 1.1 mmol/L neomycin sulfate. The mixture was incubated for 1 hour at 37°C, and the angiotensin I (Ang I) that was generated was measured using a radioimmunoassay kit manufactured by DuPont Medical Products.
Prorenin Activation In Vitro
Prorenin Activation in Kidney Extract
Fresh kidney extract (100 mg/mL) was diluted 1:10 with 50 mmol/L phosphate buffer (pH 7.4) immediately before use. A total of 10 μL of diluted kidney extract (100 μg of tissue) was pipetted into siliconized tubes containing 100 μL of 50 mmol/L phosphate buffer (pH 7.4) or 50 μL of samples from chromatographic fractions and 50 μL of phosphate buffer; the extract was then incubated in a shaker water bath for 15 minutes at 37°C. The Ang I formed was removed by filtration on Sephadex G-25.11 Briefly, the tubes were rapidly chilled, and the incubation mixture was applied to a small column (0.4×3.5 cm) of Sephadex G-25, which had been equilibrated with 50 mmol/L phosphate buffer (pH 7.4). The effluent from the column was collected in a decapped Eppendorf tube by centrifugation at 1600g for 4 minutes at 4°C. Renin was measured by incubating 100 μL of the effluent of Sephadex G-25 with 0.9 mL of angiotensinogen solution (2 μg of Ang I). Column effluents incubated without angiotensinogen solution and incubations at 4°C were included to check for the presence of background Ang I. The results were expressed as ng of Ang I · mg of tissue−1 · h−1. Renin recovery from the Sephadex G-25 column was 97±4% (n=12); angiotensin was not detected in the assays.
Prorenin Activation in Plasma
A total of 2 aliquots of 200 μL of normal plasma was incubated for 15 minutes at 37°C with and without 50 μL of PreR-Co (10 μg of protein). Then, the plasma renin concentration (PRC) was measured in both aliquots by incubating samples with 750 μL of the angiotensinogen solution (1.2 μg of Ang I · mL−1 · h−1). The Ang I formed was measured by radioimmunoassay, and the results were expressed as ng of Ang I · mL−1 · h−1.
Renin activation was calculated in both kidney extract and plasma as the difference between the renin concentration in samples incubated with and without PreR-Co.
To remove plasma protease inhibitors, in 4 experiments, plasma was acidified to pH 2.8±0.1, incubated at 5°C overnight, and brought to pH 6.8. After that, plasma was treated with PreR-Co as previously described, but the amount of enzyme was increased to 50 μg of protein. The remaining protease inhibitors were measured by the amidase activity of papain and trypsin added to acidified and nonacidified plasmas.
Prorenin Activation by Trypsin
Plasma prorenin was activated by trypsin treatment according to the method of Johannessen et al.12 Plasma was incubated for 60 minutes at 0°C with 4 mg/mL trypsin. The reaction was stopped by adding an excess of SBTI. Then, the incubation mixture was treated with semidry Dowex 50W-X2 100 to 200 mesh resin (Bio-Rad AG). Plasma renin was measured before and after trypsin treatment by radioimmunoassay. Prorenin in the kidney extract was measured by the same method, but 10 μL of diluted kidney extract (100 μg of tissue) was incubated with trypsin (0.25 mg/mL) in the presence of bovine serum albumin (10 mg/mL) and 5 mmol/L CaCl2. The reaction was stopped by adding SBTI (0.5 mg/mL).
Data are expressed as mean±SEM. Differences between averages were evaluated by Student’s t test when 2 groups were compared or by ANOVA and the least significant difference between means test when >2 means were compared.13
Three main peaks were obtained from the Sephacryl S-200 HR, but only the peak eluting at 52 kDa could promote activation of inactive kidney renin. A significant amount of albumin was present in this fraction, and some preparations also had small amounts of angiotensinogen. After passing the 52-kDa peak through the immunoaffinity column, a peak that eluted at 10 mL (after elution started) was the only one capable of activating prorenin in kidney homogenates. According to the ring-test, albumin was completely removed by the immunoaffinity column. The fractions containing activity were concentrated and applied to the Mono Q HR 5/5 column. The peak that could activate inactive kidney renin emerged at 200 mmol/L NaCl. This peak was concentrated and applied to a second Mono Q HR 5/5. A single peak emerging at 200 mmol/L NaCl was found (Figure 1⇓). This peak was named PreR-Co. The purification achieved in each step is shown in the Table⇓. Plasma renin substrate and albumin were not detected in the peak from the second Mono Q HR 5/5 purification. The active peak was assayed by SDS-PAGE followed by silver staining. Only a single band of ≈37 kDa could be detected (results not shown), and the amino acid sequence from the N-terminal end of this band (determined by 2 independent groups) was IIGGSMDAKGSFP. A search for homologous proteins was performed with the BLAST (Best Local Alignment Search Tool) algorithm14 in the GenBank database. This sequence had a homology of 90% with the β-chain of haptoglobin, 69% with serine-proteases, and 65% with kallikreins.
The amidase activity of PreR-Co was measured in 3 experiments. We found that 47 μg of enzyme could hydrolyze 1 μmol/L BAPNA substrate per minute (specific activity, 20 U/mg).
The ability of PreR-Co to hydrolyze the fluorogenic esterase substrate (BCECF AM) is shown in Figure 2⇓. The esterase activity of PreR-Co was abolished by adding 2 mmol/L PMSF to the incubation medium.
The incubation of PreR-Co (4 μg of protein) with dog kininogen produced 25.2±0.9 pg of bradykinin · min−1 · μg of protein−1; and no difference was found when it was incubated in the presence of SBTI (28.3±0.1 pg of bradykinin · min−1 · μg of protein−1). After processing the sample with trypsin, the result was 24±2.7 pg of bradykinin · min−1 · μg of protein−1, indicating that PreR-Co is not a precursor that can be activated.
Activation of Kidney and Plasma Prorenin by PreR-Co
Renin concentration in the kidney cortex homogenate was 34±4 ng of Ang I · mg of tissue−1 · h−1. A significant increase (P<0.01) was found after 15 minutes of incubation with 50 μL of PreR-Co (10 μg of protein) (211±7 ng of Ang I · mg of tissue−1 · h−1). Moreover, after trypsin treatment of extract, the amount of renin in the kidney homogenate was 110±11 ng of Ang I · mg of tissue−1 · h−1 (Figure 3A⇓). The property of PreR-Co to activate renal prorenin was abolished by the presence of 2.0 mmol/L of DFP, 2.0 mmol/L PMSF, or 0.23 mmol/L aprotinin. However, the presence of 10 mmol/L EDTA, 2.0 mmol/L N-ethylmaleimide, 2.0 mmol/L leupeptin, or 2.0 mmol/L pepstatin did not inhibit the activation of renal prorenin.
PRC in normal rats was 67.6±13.3 ng of Ang I · mL−1 · h−1, and no significant difference (73.7±11.1 ng of Ang I · mL−1 · h−1) was found after treatment with different amounts of PreR-Co (10 to 50 μg of protein). On the contrary, a significant increase in PRC (202.8±7.8 ng of Ang I · mL−1 · h−1) was found in plasma after trypsin treatment (Figure 3B⇑).
Results similar to those found in nontreated plasma were obtained with plasma that was previously acidified (68.3±5.0 ng of Ang I · mL−1 · h−1) and treated with 50 μg of PreR-Co (70.2±0.2 ng of Ang I · mL−1 · h−1). A total of 75% of trypsin amidase activity and 95% of papain amidase activity occurred in the acidified plasmas relative to their activity in buffer solution; this indicates that most protease inhibitors had been removed. However, no activity of these enzymes was observed in nontreated plasma. Furthermore, PreR-Co could activate renal prorenin from rat kidney extract added to acidified plasma (45.4±8.7 and 159.2±6.6 ng of Ang I · mg of tissue−1 · h−1 before and after treatment with PreR-Co, respectively). In contrast, no significant difference was observed when rat kidney extract was added to normal plasma (56.1±6.4 and 59.2±7.5 ng of Ang I · mg of tissue−1 · h−1 before and after treatment with PreR-Co, respectively). Renal prorenin activation without plasma was also performed as a control (32.3±3.2 and 120.2±7.5 ng of Ang I · mg of tissue−1 · h−1 before and after treatment with PreR-Co, respectively).
Two main conclusions emerged from the present study. First, an enzyme capable of activating renal prorenin was isolated from plasma; and second, this enzyme, contrary to what was expected, did not cleave plasma prorenin. The enzyme was described in an earlier study from this laboratory.6 Earlier, we found that the presence of albumin and angiotensinogen in the purified protein fraction interfered with identification of the substance. In the present experiments, albumin was fully removed with a specific antiserum, anti-albumin, coupled to Sepharose 4B. Moreover, angiotensinogen was eliminated by a Mono Q HR 5/5 column. The final purified fraction appeared as a single band in SDS-PAGE with a molecular mass near 37 kDa. The N-terminal amino acid sequence (IIGGSMDAKGSFP) of this band had high homology with the β-chain of haptoglobin (90%) and kallikrein (65%), but it did not entirely match any known proteolytic enzyme. This finding supports the idea that PreR-Co could be a novel enzyme that processes renal prorenin. Haptoglobin is a glycoprotein without any known proteolytic activity, but it shows strong sequence homology with the serine protease family. Furthermore, PreR-Co has the capability to hydrolyze ester bonds and peptide bonds, and both are common properties of several serine proteases. However, PreR-Co cleaves kininogen and also has a 65% homology with kallikreins, which tempts us to think that it may belong to the kallikrein family. Nevertheless, the validity of this hypothesis requires further investigation.
The mechanism by which newly synthesized renin is sorted into secretory or storage pools is complex. Inactive renal renin is a heterogeneous protein mainly caused by variable glycosylation.15 Active renal renin is a glycoprotein16 and is heterogeneous with respect to isoelectric focusing,17 which may be due to differential glycosylation.18 19 The constitutive secretory pathway of active renin seems to contain relatively acidic forms of renin20 ; therefore, the cleavage of renin from an inactive form to an active one could be regulated by a differential glycosylation mechanism. With respect to the cleavage of inactive forms, many enzymes capable of activating prorenin in tissues exist,5 but the enzyme and the precise mechanism of conversion remain unknown. Our results show a significant increase in active renin after a kidney cortex extract was incubated with PreR-Co. This observation certainly suggests the existence of mechanisms of activation of an inactive Ang I–releasing enzyme. The product of activation proved to be renin when a specific antibody against rat renal renin neutralized the Ang I–generating activity.6 However, no activation occurred when plasma was incubated with PreR-Co at different pHs. By contrast, prorenin in kidney or plasma samples was activated by trypsin. The possibility that plasma protease inhibitors may act as inhibitors of PreR-Co seems unlikely for several reasons. First, PreR-Co was tested in plasma in which most of the protease inhibitors were removed by acidification. Second, the concentration of PreR-Co used was high enough (50 μg) to inhibit the remaining protease inhibitors. Third, PreR-Co was capable of activating prorenin from kidney extract added to plasma in which PreR-Co was unable to activate plasma prorenin. These results clearly show that PreR-Co is capable of interacting with renal prorenin but not with plasma prorenin. Thus, we speculate that plasma prorenin may have a different configuration than renal prorenin, which may protect the cleavage sites from attack by the enzyme.
Although several enzymes have been proposed to process renal prorenin, the precise enzyme involved in this process remains unknown. Moreover, the probability that the same enzyme may also act in other tissues or on plasma prorenin is also not known. Kim et al21 purified an enzyme from the mouse submandibular gland (named prorenin-converting enzyme) that specifically cleaves mouse submandibular gland prorenin, but this enzyme is not active in forming renal renin. In addition, no pressor or vasoconstrictor activity was detected after intravenous infusion of recombinant prorenin22 ; furthermore, transgenic rats carrying an additional renin gene (Ren-2) have elevated plasma prorenin but suppressed active renin concentration in both the blood and kidneys.23 These last 2 findings suggest that plasma prorenin is not the source of active plasma renin and, furthermore, that little or no conversion takes place in the circulating blood. These findings are also consistent with our results: we could not find prorenin activation in plasma with a preparation that was fully active in the kidney homogenate.
If it is assumed that a circulating enzyme is responsible for the activation of kidney prorenin, one important question is: how it is transported from its site of synthesis to the kidney? At this point, we are unable to define the precise mechanism; however, in light of the fact that the enzyme was associated with albumin, we speculate that albumin may play an important role in the transport of the enzyme to its site of activation. Another intriguing question raised by our present results is: why is a circulating plasma enzyme able to cleave kidney prorenin? Probably, it is because the processing of inactive to active in the storage pool of renin should reflect changes that occur in the circulating active renin.
In summary, our results indicate the presence of a protein in rat plasma that promotes a 3-fold increase in renin activity when incubated for a few minutes with a renal extract; nevertheless, it was unable to activate plasma prorenin. This protein (not yet identified) was purified 900-fold and seems to be an enzyme that may regulate the conversion of prorenin to the active form in the kidney. The possibility that this protein acts as an activator of the renin-substrate reaction may be discarded because the same concentration of substrate was used in both plasma and renal assays.
This study was funded by grants from the Consejo Nacional de Investigaciones Científicas y Técnicas de la República Argentina and Consejo de Investigaciones de la Universidad Nacional de Tucumán. The authors thank Dr Héctor L. Nolly for his help with the determination of kininogenase activity.
Reprint requests to Dr Eduardo De Vito, INSIBIO, Universidad Nacional de Tucumán, Chacabuco 461, 4000 S.M. de Tucumán, Argentina.
- Received July 21, 1998.
- Revision received September 22, 1998.
- Accepted April 26, 1999.
De Vito E, Cabrera RR, Fasciolo JC. Renin production and release by rat kidney slices. Am J Physiol. 1970;219:1042–1045.
Leckie BJ. The activation of a possible zymogen of renin in rabbit kidney. Clin Sci. 1973;44:301–304.
Robertson JIS, Nicholls MG, eds. The Renin-Angiotensin System. Kent, UK: Times Mirror International Publishing Ltd; 1993:chaps 6 and 7.
De Vito E, Martinez de Melian ER, Guardia DC. Activation of renal renin by a protein plasma fraction: a novel enzymatic mechanism. Comp Biochem Physiol. 1996;113B:433–438.
Arnon R. Papain. In: Perlmann GE, Loran L, eds. Methods in Enzymology: Individual Proteolytic Enzymes. Vol XIX. New York, NY: Academic Press; 1970:228–230.
De Vito E, Gordon SB, Cabrera RR, Fasciolo JC. Release of renin by rat kidney slices. Am J Physiol. 1970;219:1036–1041.
Maniatis T, Fritsch EF, Sambrook J. Molecular Cloning: A Laboratory Manual. Cold Spring Harbor, NY: Cold Spring Harbor Laboratory; 1982.
Snedecor GW, Cochran WG. Statistical Methods. 2nd ed. Ames, Iowa: Iowa State University Press; 1967.
Druilhet RE, Overturf ML. Separation of multiple forms of renin from human, rabbit, hog, and baboon kidney by isoelectric focusing. In: Sambhi MP, ed. Heterogeneity of Renin and Renin Substrate. New York, NY: Elsevier North Holland Inc; 1981:89–99.
Sessler FM, Jacquez JA, Malvin RL. Different production and decay rates of six renin forms isolated from rat plasma. Am J Physiol. 1986;250:E551–E557.
Abraham PA, Katz SA, Opsahl JA, Miller RP, Stanchfield WR, Andersen RC. Renal secretion and hepatic clearance of human multiple renin forms. Hypertension. 1990;16:669–676.
Kim WS, Nakayama K, Nakagawa T, Kawamura Y, Haraguchi K, Murakami K. Mouse submandibular gland prorenin-converting enzyme is a member of glandular kallicrein family. J Biol Chem. 1991;266:19283–19287.
Lenz T, Sealey JE, Maack T, James GD, Heinrikson RL, Marion D, Laragh JH. Half-life, hemodynamic, renal and hormonal effects of prorenin in cynomolgus monkeys. Am J Physiol. 1991;260:R804–R810.