Renin and renin blockade have no role in complement activity

Abstract

Renin, an aspartate protease, regulates the renin-angiotensin system by cleaving its only known substrate angiotensinogen to angiotensin. Recent studies have suggested that renin may also cleave complement component C3 to activate complement or contribute to its dysregulation. Typically, C3 is cleaved by C3 convertase, a serine protease that uses the hydroxyl group of a serine residue as a nucleophile. Here, we provide seven lines of evidence to show that renin does not cleave C3. First, there is no association between renin plasma levels and C3 levels in patients with C3 Glomerulopathies (C3G) and atypical Hemolytic Uremic Syndrome (aHUS), implying that serum C3 consumption is not increased in the presence of high renin. Second, in vitro tests of C3 conversion to C3b do not detect differences when sera from patients with high renin levels are compared to sera from patients with normal/low renin levels. Third, aliskiren, a renin inhibitor, does not block abnormal complement activity introduced by nephritic factors in the fluid phase. Fourth, aliskiren does not block dysregulated complement activity on cell surfaces. Fifth, recombinant renin from different sources does not cleave C3 even after 24 hours of incubation at 37 °C. Sixth, direct spiking of recombinant renin into sera samples of patients with C3G and aHUS does not enhance complement activity in either the fluid phase or on cell surfaces. And seventh, molecular modeling and docking place C3 in the active site of renin in a position that is not consistent with a productive ground state complex for catalytic hydrolysis. Thus, our study does not support a role for renin in the activation of complement.

Graphical Abstract

Introduction

Renin, an aspartate protease with a half-life of 30–90 minutes, is a key regulator of the renin-angiotensin system (RAS)¹. A decrease in either blood volume or kidney perfusion pressure triggers its release by cells of the juxtaglomerular apparatus². Renin then converts angiotensinogen (AngT) to angiotensin I (Ang1), a 10-amino acid (aa) non-active peptide, which is secondarily cleaved by angiotensin-converting enzyme (ACE) to angiotensin II, an 8-aa peptide that is biologically active and functions by binding to its two receptors^{3, 4}.

Therapeutic strategies for RAS targeting, including angiotensin-converting enzyme inhibitors (ACEis) and angiotensin II receptor blockers (ARBs), are widely used in treating cardiovascular and renal diseases^5–8. These drugs not only manage hypertension but also slow progression of mild-to-moderate chronic kidney disease (CKD)^9–14. RAS suppression with ACEis and ARBs leads to increased renin production¹⁵, which as reported by Bekassy and colleagues activates the complement alternative pathway (AP) by cleaving C3 into C3b to form a functional C3 convertase¹⁶. Aliskiren, another FDA-approved RAS inhibitor, specifically blocks renin preventing the conversion of AngT to Ang1^{11, 17}. On this basis, the authors used aliskiren to treat three DDD patients, a subtype of C3 Glomerulopathy (C3G), effectively reducing complement activation in both the circulation and the kidneys¹⁶.

Two years later, Plasse et al. described a 21-year-old male with primary atypical hemolytic uremic syndrome (aHUS), advanced renal disease, and uncontrolled malignant hypertension despite maximum doses of five antihypertensive medications. They prescribed high dose aliskiren as an adjunct to eculizumab¹⁸. Initially, aliskiren and eculizumab only achieved a partial response, but increasing aliskiren to a supratherapeutic dose allowed for reduced antihypertensive medication, increased platelets and serum C3 levels, and decreased epoetin alfa requirement¹⁸. While no other reports have linked renin to complement activity, an ongoing Phase 2 clinical trial (NCT04183101) is evaluating aliskiren’s efficacy in C3G patients.

Both C3G and aHUS are ultra-rare complement-mediated renal diseases caused by dysregulation of the AP of complement either in the fluid phase (C3G) and/or on cell surfaces (aHUS)^{19, 20}. Here, we sought to determine in a larger cohort whether renin activates complement and plays a role in exacerbating complement dysregulation in complement-mediated renal diseases. Counter to prior reports, our results show that renin does not cleave C3 and does not activate complement.

Methods

Patients and controls

The study cohort includes 54 C3G and 34 aHUS patients referred to the Molecular Otolaryngology and Renal Research Laboratories (MORL). We followed our standard procedure to collect and store serum and plasma samples at −80°C. All patients provided informed consent, and our study adhered to the Declaration of Helsinki guidelines. Our control group comprised 23 healthy individuals aged 19 to 22, who were not using ACEi and/or ARBs. The Institutional Review Board of Carver College of Medicine at the University of Iowa approved the study.

Recombinant renin, aliskiren, trypsin and complement proteins

Eight different batches of recombinant human renin were acquired from six providers (Table 1). Aliskiren was sourced from Millipore-Sigma (St. Louis, MO). Bovine and porcine trypsin were acquired from MP Biomedicals (Solon, OH) and ThermoFisher Scientific (Waltham, MA), respectively. Human complement proteins, including C3, Factor B (FB) and Factor D (FD), were obtained from Complement Technology, Inc. (Tyler, TX).

Table 1.

Recombinant renins acquired from different sources

Batch	Source	Catalog	Renin or Prorenina	Tag	Expressed as Proreninb	Trypsin Activationb	Purity by SDS-PAGEb	Trypsin by Mass Spectrometry	Renin Activity (Units/μg)c
A	Abcam	Ab183267	Renin (67–406)	C-ter HIS	Yes	Yes	> 95%	Yes	18.7
B	Abcam	Ab135012	Renin (67–406)	C-ter HIS	Yes	Yes	> 90%	No	30.9
C	Abcam	Ab155713	Prorenin (24–406)	C-ter HIS	Yes	Yes	> 95%	Yes	0.1
D	BioVision	6300–100	Renin (67–406)	C-ter HIS	Yes	Yes	> 95%d	Not Done	16.4
E	Creative Enzyme	NATE-1951	Renin (67–406)	N-ter FLAG	Direct fusion	Not needed	>90%	Not Done	24.0
F	Enzo Life	ENZ-PRT193	Renin (67–406)	N-ter FLAG	Direct fusion	Not needed	>90%	Not Done	21.4
G	AnaSpec	72041	Renin (67–406)	No tag	Yes	Yes	> 99%	No	55.5
H	Cayman	10006217	Renin (67–406)	No tag	Yes	Yes	>85%	Yes	28.8

^aAll renins were produced by HEK 293 cell line. All products were affinity purified.

^bThis information was extracted from the products’ certificates of analysis.

^c1 unit of renin activity is defined as the amount of enzyme that cleaves 1 pmol of angiotensinogen per minute at a temperature of 37°C (pH = 7.4)

^dContains 25% prorenin based on the certificate of analysis.

Renin activity by AngT tetradecapeptide cleavage

The activity of renin was evaluated through the conversion of the human AngT 14aa peptide (AnaSpec, Fremont, CA) to Ang1 10aa peptide. Renin (up to 250 ng) was added to the substrate mix containing 1 μM of AngT peptide in PBS (pH=7.4) in a total reaction volume of 40 μl. The mixture was incubated at 37°C for 1 h. Following incubation, the reaction was stopped using 1N HCl. Quantification of the generated Ang1 peptide was carried out by employing a highly specific Ang1 peptide ELISA kit (NBP2–62134, Novus Biologicals, Centennial, CO), following the manufacturer’s recommended protocol.

C3 cleavage by renin

To investigate the potential cleavage of C3 by renin, a direct C3 cleavage assay was conducted. In this assay, purified C3 at a concentration of 250 μg/ml was mixed with 40 U of renin in PBS (pH=7.4) supplemented with Mg²⁺ (final concentration 0.5 mM). [1 unit is defined as the amount of enzyme that cleaves 1 pmol of AngT 14aa peptide per minute at 37°C (pH = 7.4)]. The mixture of C3 and renin was incubated at 37°C for 1h or 24h. In parallel, the same amount of C3 was added to a mixture of FB and FD [2 μg and 20 ng, respectively]. Following the respective incubation periods, the C3 α or α’ chain was visualized using immunoblotting techniques with an anti-C3b antibody.

Plasma renin, C3, C3a and soluble C5b-9 levels

Plasma renin quantification was performed utilizing a solid-phase sandwich ELISA kit from ThermoFisher Scientific, following the manufacturer’s provided instructions. Plasma C3 was measured by a nephelometer (The Binding Site, Birmingham, UK). C3a and soluble C5b-9 (sC5b-9) were determined by using Quidel kits (A031and A020, respectively) per manufacturer instructions.

Fluid phase activity assay

Fluid phase C3 convertase activity was monitored by mixing patient serum with pooled normal human serum (1:1) in PBS (pH=7.4) supplemented with Mg²⁺ (final concentration 0.5 mM). Recombinant renin (133 nM) or aliskiren (up to 280 nM) was added to the mixture and incubated at 37°C for 30 minutes. The reaction was stopped by the addition of EDTA and resolved on a precast agarose gel (Helena Laboratories, Beaumont, TX). A diluted anti-C3 antibody (MP Biomedicals, 1:50) was applied topically to precipitate C3 and C3 activation products. The gel was then stained, de-stained, scanned, and quantified.

C3b deposition assay

The C3b deposition assay was conducted on cultured MES-13 cells (ATCC CRL-1927) with diluted patient serum (20% final concentration under EGTA-MgCl₂ [10 mM and 0.5 mM, respectively]) at 37°C for 20 minutes. Recombinant renin (1.9 and 19 nM) or aliskiren (0.4, 4 and 40 μM) was added to the diluent before incubation. Deposited C3b was visualized using Alexa-488 labeled anti-C3 antibody 7C12 (a gift from Dr. Ronald P. Taylor) under confocal microscopy.

Molecular modeling and docking

The protein-protein dock utility in the Molecular Operating Environment (MOE) software was utilized to generate docked peptide poses²¹. A coarse-grained (CG) model (residues represented as one to three “united atom” type beads was employed to reduce the computational search space, which utilized exhaustive sampling to generate a set of initial poses²². The Hopf fibration employed in grid generation²³ and a Fast Fourier Transform (FFT) was used to sample all translations for a given rotation, followed by a minimization protocol. Rapid energy minimizations were performed with unique coarse-grained rigid-bodies using a (1+r)⁻² term for electrostatics. Finally, the top peptides were minimized to account for solvation free energy using a Generalized Born / Volume Integral (GB/VI) Implicit Solvent Model²⁴.

Proteomics analysis

Proteomics analysis was conducted at the Protein Facility of Iowa State University Office of Biotechnology. For molecular weight determination, samples were run on a Waters Synapt G2-Si coupled to a Waters H-Class UHPLC. The protein of interest was isolated using chromatographic techniques (Restek Ultra C4 5um 50 × 1mm column) and introduced to the mass spectrometer. The stationary phase was 0.1% formic acid in water and the mobile phase was 0.1% formic acid in acetonitrile. The mobile phase was run in a linear gradient from 5–100% in 8 minutes at 400 μl/minute, followed by several sawtooth wash steps. The protein was introduced to the Synapt G2-Si using a dual orthogonal API source (LockSpray ESI/APCI) and subjected to CID fragmentation. Molecular weight was determined by deconvoluting the fragment masses using Waters MaxEnt software.

Statistical analysis

The statistical analyses were performed with GraphPad Prism version 9.5 including: 1) 1-way ANOVA for multiple group comparison with a Tukey post hoc test; 2) linear regression analysis; and 3) Mann-Whitney analysis for two groups. A significance level of P < 0.05 was used for all tests.

Results

1). Plasma renin in patients with complement-mediated renal diseases

In patients with complement-mediated diseases, renin levels were significantly higher compared to controls (Figure 1a). About 50% of these patients exhibited elevated plasma renin levels, peaking at 4.6 ng/ml or >0.1 nM, which was approximately five times higher than levels in normal subjects. Notably, individuals with aHUS had greater total renin levels than those with C3G. In both diseases, correlations between renin and complement biomarkers C3 and sC5b-9 did not reveal significant differences (Figure 1b and 1c). Furthermore, in C3G patients with C3Nef autoantibodies that dysregulate the AP in the fluid phase, high renin levels did not exacerbate fluid-phase C3 conversion, as no differences were observed between patients with low and high renin levels (Figure 1d).

Figure 1.

Renin concentration in patients with complement-mediated renal diseases.

a) Plasma renin levels were measured in normal individuals (n = 23) and patients with aHUS (n = 34) and C3G (n = 54). Renin levels are significantly higher in patients with C3G and aHUS; ~50% of patients have elevated renin levels (> 1 ng/ml). Patients with aHUS had higher total renin than those with C3G (P < 0.05); b) and c) C3 and sC5b-9 levels in patients with low (<1 ng/ml) and high (>1 ng/ml) renin levels (red: aHUS; blue: C3G). There are no significant differences between groups (P > 0.05), suggesting that high renin levels are not linked to increased C3 activation/consumption or higher C5 convertase activity. Patients on anti-C5 treatment are excluded in c; d) Fluid-phase activity in patients with C3G. The group with high renin (n = 30) is not enriched for patients with higher fluid phase activity as compared to the group with low renin (n =24). C3Nef-positive C3G patients with high renin levels show no greater fluid-phase C3 conversion than those with low renin levels (for all figures, horizontal dashed line, normal limit).

2). Aliskiren, a renin-specific inhibitor, does not block complement activity

To test the effect of aliskiren on fluid-phase complement activity in the presence of high renin levels, we added aliskiren to sera from four C3G patients with high renin levels and positive fluid-phase activity secondary to C3Nefs. The addition of aliskiren did not prevent complement activity at 280 nM (2800x molar excess relative to renin) (Figure 2a). Note that in patients taking aliskiren at the recommended daily dose of 300 mg, the maximum concentration (Cmax) ranges from 1 to 2 nM²⁵.

Figure 2.

The renin-specific inhibitor, aliskiren, does not block unregulated complement activity.

a) In the fluid phase. In 4 patients, each having renin levels of 848, 1006, 2027 and 1074 pg/ml, respectively, and presenting varying degrees of complement dysregulation, aliskiren was added to a 1:1 mixture of pooled normal human serum and patient serum at increasing concentrations ranging from 0 to 280 nM (representing 2800-fold molar excess to renin). The mixture was incubated in buffer containing EGTA supplemented with MgCl₂ to allow for alternative pathway (AP) activation or EDTA to stop complement activity (circle, patient #1, second lane shows concentration at the periphery of a dot blot optimal for immunoprecipitation with purified C3 (0.5 mg/ml) for the purpose of quality control for the anti-C3 antibody; C3AP, C3 activation products); b) On cell surfaces. In a patient with high renin (3036 pg/ml), factor H and aliskiren were spiked in at concentrations indicated. The addition of FH at a final concentration of 0.03 μM significantly reduced C3b deposition, and at a final concentration of 0.17 μM, C3b deposition was completely prevented. However, when aliskiren was added at low (0.4 μM), medium (4 μM) or high (40 μM) concentrations, which are 4,000x, 40,000x and 400,000x molar excess relative to renin levels, respectively, there was no reduction in C3b deposition associated with renin blockade (blue = nuclei (DAPI); green = C3b; bar = 200 μm).

We next tested whether aliskiren can prevent or modulate complement activity on cell surfaces in C3G patients with positive C3b deposition and high renin level (3.3 ng/ml). While the deposition of C3b was effectively prevented through the addition of factor H, the addition of aliskiren (up to 40 μM) had no effect (Figure 2b).

3). Recombinant renins and cleavage of C3

Except for prorenin (batch C), all recombinant renins exhibited biological activity, confirmed through both AngT tetradecapeptide cleavage assay (Table 1). To assess potential C3 cleavage activity by renin, we conducted a direct C3 cleavage assay, mixing C3 (at 250 μg/ml) with 40 U of renin (~100x higher than in patients with high renin). Immunoblot analysis followed incubations of 1 h and 24 h at 37°C. Batch A showed partial C3 cleavage after 1 h and complete cleavage after 24 h, while four other renin batches (B, D, E, F) displayed no C3 cleavage, even after 24 h (Figure 3a).

Figure 3.

Recombinant renin does not cleave purified C3.

a) Renin (40 U) and C3 (250 ug/ml) were mixed and incubated at 37°C (pH = 7.4) for 1 h (top) or 24 h (bottom). Batches A, B and D are C-terminal His-tagged renin; batches E and F are N-terminal FLAG-tagged renin (Table 1). The respective mixtures were resolved on a 4–15% polyacrylamide gel and probed with an anti-C3 antibody, which targets the Factor I cleavage site on C3 or C3b (this antibody recognizes the intact C3 α chain (116 kDa) and the C3b α’ chain (106 kDa)). Control Lane: FB and FD added to form C3 convertase. With the exception of batch A, four renins (batches B, D, E, F) did not cleave C3 even after a 24-h incubation. Batch A did show signs of C3 cleavage, which was partial after 1 h of incubation (C3/C3b mixed band, top gel) and complete after 24 h of incubation (bottom); b) Untagged renin batches G (AnaSpec) and H (Cayman) at increasing concentrations of 5 to 40 U (left to right). No cleavage activity was detected using batch G in both 1-h (top) and 24-h (bottom) incubations at 37°C. However, dose-dependent cleavage of C3 with batch H was apparent in both 1-h and 24-h incubation studies.

Further investigations involved dose studies with untagged renin batches G and H. Renin concentrations ranging from 0 to 40 U were tested. Batch G exhibited no C3 cleavage after both 1-h and 24-h incubations at 37°C, whereas batch H displayed dose-dependent C3 cleavage in the 1-h and 24-h studies (Figure 3b). In an AngT-C3 combination cleavage assay, both batches G and H exhibited dose-dependent AngT tetradecapeptide cleavage while only batch H resulted in a dose-dependent increase in C3a levels, with batch G showing no elevation in C3a (Figure 4a).

Figure 4.

Dose-Dependent AngT and C3 Cleavage in Combined AngT-C3a Assay.

a) AngT-C3 combination cleavage assay. AngT 14aa peptide and C3 were mixed in a single tube before adding renin at increasing concentrations ranging from 0.1 to 14 Units for a 1-h incubation at 37°C (pH = 7.4). The results showed a dose response in AngT cleavage for both renins (red bars). While batch G did not exhibit any noticeable increase in C3a, with batch H showed a dose-dependent increase (blue bars). FB/FD and C3 mixture served a control for C3a; b) Identification of trypsin contamination via Mass Spectrometry. Spectra: Distinct peaks in mass spectroscopy corresponding to bovine trypsin (M peaks, M+Xs peaks, marked by 3 arrows) were detected in batch H but absent in batch G.

Investigation into the C3 cleavage observed in batches A and H revealed bovine trypsin contamination, which likely originated during the renin production process. Proteomics analysis by mass spectrometry (MS) confirmed this contamination, and distinct species-specific profiles consistent with bovine trypsin were identified (Figure 4b, Supplementary Figure S1). To summarize, in batch B and G, no trypsin peaks were observed, and there was no detectable C3 cleavage activity. In contrast, both batch A and H exhibited bovine contamination along with evidence of C3 cleavage.

4). Recombinant renin (spiking) does not increase complement activity

To check whether recombinant renin increases fluid-phase activity as reflected by excessive C3 conversion, we added recombinant FLAG-tagged renin (batch F) at 133 nM (>1000x higher than the concentration observed in patients with high renin). No increase in C3 conversion was noted in any of the 15 patients tested (Figure 5a shows 3 examples). Similar results were obtained with untagged and uncontaminated renin (batch G).

Figure 5.

Recombinant renin does not induce complement activation.

a) Renin does not increase complement activity in the fluid phase. Recombinant FLAG-tagged renin (batch F) was spiked in at >1000x higher than the concentration observed in patients. Three examples are shown – patient #6 with negative pre-spiking fluid-phase activity [no C3 activation products (C3AP) shown in the second lane] and no increase with spiking (third lane); patients #7 and #8 with positive pre-spiking fluid-phase activity (second lane) and no increase with renin spiking (third lane) (total 15 patients tested; D = EDTA, complement activity stopped; G = EGTA, supplemented with MgCl₂ making AP activation possible); b) Increased renin concentration does not increase complement deposition on cell surfaces. C3b deposition does not increase as renin concentration is increased in both pooled normal human serum (top) and patient serum (bottom). As a positive control, FD (instead of renin) was spiked in at a final concentration of 10 μg/ml (~10x above normal) (right). Renin low = final concentration 1.9 nM or ~19x above the highest renin observed in patient. Renin high = final concentration 19 nM or ~190x above the highest renin observed in patients. FD added to a final concentration of 10 ug/ml (~10x higher than controls) (blue = nuclei, DAPI; green = C3b; bar = 200 μm).

We next checked whether recombinant renin increases C3b deposition on cell surfaces by directly spiking in recombinant renin (batch F) at 1.9 nM or 19 nM (~19x or ~190x above the highest renin we detected in patients) in the C3b deposition assay. No significant change in C3b deposition was observed as renin concentrations increased in the presence of pooled normal human serum and in patient serum (Figure 5b). To provide a positive control, FD was introduced (without adding renin) at a final concentration of 10 μg/ml to promote C3b deposition (Figure 5b). Similar results were obtained with untagged and uncontaminated renin (batch G).

5). Molecular modeling and docking

To investigate the possibility that C3 may bind to and undergo renin-catalyzed hydrolysis, we performed molecular docking using PDB 6i3f as a starting point for preparing the receptor. The top scoring peptide docking pose of angiotensinogen (AngT tetradecapeptide, DRVYIHPFHLVIHN) to renin resulted in a complex that very closely matched the crystal structure (RMSD ~ 0.1 Å; Figure 6a and 6b); the AngT from 6i3f (orange strand) is superposed on the docked AngT (blue colored atoms) in Figure 6a, and is catalytically competent for hydrolysis via Asp226 and Asp38, as reported²⁶. This top docking pose of AngT has a potential energy value of −63 kcal/mol.

Figure 6.

Molecular docking of AngT and C3.

a) The AngT tetradecapeptide, depicted in blue stick view, is docked into the active site of the renin enzyme (6i3f). The docking result closely resembles the co-crystal structure complex, which contains the AngT peptide, shown as the orange strand. The root-mean-square deviation (RSMD) between the docked and co-crystal structure complex is less than 0.1; b) A closeup of the enzymatic site, which cleaves the scissile bond between Leu10 and Val11 of AngT; c) The C3 tetradecapeptide is shown in green stick view, docked into the active site of renin. RSMD between the docked C3a and Renin receptor interface is 1.76 (the best fitting model among 100 docking possibility). For comparison, note the AngT tetradecapeptide again shown as the orange strand; d) A closeup of the enzymatic site in renin. This docking configuration does not result in a catalytically productive complex because there is a distance of 9.34 Å between the catalytic base (Asp38 and Asp226) and the carbonyl carbon of the scissile bond between Arg748 and Ser749, indicating that the necessary interactions for catalysis cannot be established. In figures b and d, the catalytic residues (Asp38 and Asp226) are depicted as green spheres and their oxygen atoms are colored red. The interaction surface between ligand and renin is shown with a transparent sheet view.

The same docking protocol was applied to the C3 peptide (tetradecapeptide, ARASHLGLARSNLD). The top scoring conformation (potential energy score of −45 kcal/mol) for the C3 peptide bound to the active site of renin is shown in Figure 6c and 6d (C3 peptide atoms are colored green and the catalytic residues of renin are depicted with green carbon atoms in ball and stick). The peptide’s scissile bond is distant from the closest catalytic oxygen (from Asp38 carboxylate), ~9.3 Å between the carboxylate oxygen and the carbonyl carbon of the peptide bond. There is also a poor angle of attack (~67 degrees) between the catalytic carboxylate, carbonyl carbon and carbonyl oxygen. Other lower scoring poses for C3 had even greater distances than this top pose. Although we cannot rule out catalytic hydrolysis of C3 by renin on these computational peptide docking results alone, they suggest that i) the presented approach is outstanding at capturing the known catalytic ground state between renin and AngT (nearly identical to that reported²⁶), and that ii) the best scoring pose that places C3 in the active site of renin is not consistent with a productive ground state complex for catalytic hydrolysis.

Discussion

As a result of the popular use of ACEi or ARBs, elevation in renin is common in patients with glomerular diseases who are placed on RAS blockade. Included in this category are patients with aHUS and C3G, two complement-mediated renal diseases. Indeed, ~50% of patients in our cohorts have plasma renin levels above 1 ng/ml (Figure 1a).

Renin is a key component of the RAS. It functions as a highly specific aspartate protease to cleave its only known substrate AngT to Ang1 by binding to AngT and inducing a major conformational change in its N-terminus, leading to a lock-and-key interaction between the two proteins²⁶. The active cleavage site on renin lies at the junction of two similar domains each of which contains an aspartic acid residue (Asp38 and Asp226) that form a catalytic dyad, consistent with the fact aspartic proteases use the unique properties of a pair of aspartic acid residues to activate a water molecule, which acts as a nucleophile and cleaves proteins by attacking the peptide bond at the carboxyl group²⁷.

The study by Bekassy et al. demonstrated that renin appears to cleave C3 into C3a and C3b, thereby activating the AP of the complement system¹⁶. Their findings showed that the cleavage site on C3 is the same site recognized by the C3 convertase of the AP, C3bBb. C3 convertase is a serine protease. Aspartate proteases like renin and serine proteases like C3 convertase are distinctly different and cleave highly specific peptide bonds. The catalytic center or dyad on renin hydrolyzes the peptide bond on AngT between Leu10 and Val11^{28, 29}. Its specificity is defined by the presence of a His-Pro-Phe (HPF) motif at N-terminal residue positions 6, 7 and 8 of AngT³⁰. Recognition of this motif by renin is conserved across species and is required for activity^{3, 31}. For example, changing the HPF motif to a triple alanine motif (AAA) leads to a 5-fold decrease in affinity of AngT for renin and prevents cleavage to Ang1³⁰. C3 convertase cleaves C3 into C3a and C3b, the latter of which binds to FB to generate additional C3 convertase, amplifying the complement response. Like all serine proteases, C3 convertase catalyzes the cleavage of C3 by using the hydroxyl group of a serine residue as a nucleophile; it specifically cleaves C3 between Arg748 and Ser749^{27, 32}. Importantly, with respect to renin, C3 does not carry an HPF motif and therefore would be predicted to be a very poor substrate for cleavage by renin.

If renin could cleave C3 into C3a and C3b, aliskiren, an antihypertensive drug that targets the RAS, would be preferable over ACEis or ABRs for patients with C3G or aHUS as it directly blocks the action of renin in addition to having antihypertensive and antiproteinuric benefits. However, we present seven compelling lines of evidence refuting the notion that renin cleaves C3. First, in patients with C3G and aHUS, there is no correlation between renin plasma levels and C3/sC5b-9 levels, suggesting that high renin does not lead to increased serum C3 consumption (Figure 1b and 1c). Second, in vitro tests demonstrate no differences in C3 conversion to C3b when comparing high renin patient sera to normal/low renin sera (Figure 1d). Third, despite being a renin inhibitor, aliskiren does not block fluid-phase complement activity induced by nephritic factors (Figure 2a). Fourth, aliskiren also fails to inhibit dysregulated complement activity on cell surfaces (Figure 2b). Fifth, recombinant renins from various sources do not cleave C3 in the absence of trypsin contamination, even after 24 h at 37°C (Figure 3 and 4). Sixth, spiking recombinant renin into C3G or aHUS patient sera does not enhance complement activity in the fluid phase or on cell surfaces (Figure 5). And finally, molecular modeling and docking reveal that the position of C3 relative to the active site in renin does not support productive catalytic hydrolysis (Figure 6b).

Our results conflict with the findings of Bekassy et al. who reported that all renins (plasma renin, kidney renin and recombinant renin) are equal and cleave C3 at Arg748-Ser749¹⁶. While we did observe this type of cleavage with some batches of renin, the result was consistent with our mass spectrometric findings of unexpected trypsin (serine protease) contamination (Figure 4b). Cleavage of C3 by trypsin is not unexpected as C3 can be cleaved ex vivo by several serine proteases including kallikrein, thrombin, FXIa, FXa, FIXa, and plasmin^{33, 34}. That renin cannot cleave C3 is consistent with the different mechanism of action of aspartyl proteases as compared to serine proteases, which involves the recognition of different active cleavage sites, Leu-Val versus Arg-Ser, respectively.

Bekassy et al. treated three patients with DDD with aliskiren. In patients #1 and #3, renin levels were normal before starting treatment (40 and 36 mIE/L, respectively; reference value: 5–60); in patient #2, levels were markedly elevated (459 mIE/L) due to concomitant treatment with ACEIs and ARBs. The response of serum C3 to aliskiren treatment was not consistent across the patients. Patient #1, for example, was initially seen in 2012 and showed an improvement in C3 after starting aliskiren, even to normal levels on some occasions. When aliskiren was discontinued, C3 levels dropped but upon restarting of aliskiren, C3 levels did not improve. At all times when C3a was measured, levels were markedly elevated, consistent with ongoing C3 cleavage with and without the presence of aliskiren. Patient #2 was initially seen in 2006 and started on aliskiren in 2011. There were essentially no changes in C3 levels, which always remained below normal. Patient #3 was seen in late 2012 and immediately started on eculizumab. C3 levels improved to low normal levels on this treatment for a short time before dropping, presumably prompting the start of aliskiren at the end of 2013. Eculizumab was stopped in early 2014 but C3 levels remained low thereafter even while on aliskiren treatment. During the entire disease course, all C3a levels were elevated indicating ongoing C3 cleavage with or without aliskiren. Patients #2 and #3 were also treated with mycophenolate mofetil, as recommended by KDIGO guidelines^{35, 36}. Our interpretation of these limited biomarker data is that aliskiren did not impact on-going complement activity in these patients. It is our impression that the general biochemical measures of success reported by the authors are confounded by the heterogeneity of disease and disease course, as well as the alternative benefits of renin blockade in this population making solid conclusions difficult.

Plasse et al. reported more limited experience with aHUS. They described a single 21-year-old patient diagnosed with aHUS, presenting with severe kidney injury, treatment-resistant malignant hypertension (BP 250/140 mmHg on maximum antihypertensive therapy), anemia, thrombocytopenia, low haptoglobin, and elevated lactate dehydrogenase. No schistocytes were observed, and no complement mutations or factor H autoantibodies were found. Eculizumab showed no response, and a biopsy image was unavailable, leaving uncertainty regarding complement-mediated disease. A partial response occurred with aliskiren (300 mg daily) and eculizumab, but a supratherapeutic aliskiren dose (600 mg bid) led to reduced antihypertensive medication, increased platelets, elevated C3, and decreased epoetin alfa requirement. Complement biomarker data were not reported, making it challenging to assess aliskiren’s impact on complement activity. Achieving blood pressure control without a control group complicates attributing subsequent biochemical changes to renin blockade.

The occasional cleavage of C3 by renin is due to common methods employed in its production. The conversion of preprorenin to prorenin involves removing the signaling peptide (first 24aa). Prorenin is further activated to renin by cleaving a specific peptide bond in the prosegment³⁷. During the manufacturing of recombinant renin, trypsin is commonly used for limited digestion. This practice applies to both untagged and C-terminal tagged renin. N-terminal FLAG tagged renin production may require trypsin for cell lysis. To prevent excessive cleavage by trypsin, benzamidine is added to the reaction mixture before purification. However, this process is delicate and can impact the prorenin/renin ratio, resulting in varying trypsin contamination levels from batch to batch. Most manufacturers primarily assess the purity, protein concentration, and activity of renin product, without including testing for residual trypsin in their quality control process.

In summary, our data do not support a role for renin in the cleavage of C3 and suggest that the use of aliskiren as a renin inhibitor to decrease complement activity and C3 convertase formation in patients with C3G and aHUS is misguided.

Translational Statement.

As a result of the popular use of angiotensin-converting enzyme inhibitors (ACEis) and angiotensin II receptor blockers (ARBs), elevation in renin is common in patients with glomerular diseases who are placed on RAS blockade. Included in this category of patients are those with aHUS and C3G, two complement-mediated renal diseases. Renin is a highly specific aspartate protease that cleaves AngT to Ang1. Recent studies have suggested it may also cleave C3 into C3a and C3b and enhance complement activity making aliskiren, a renin inhibitor, preferable over ACEis or ABRs for patients with C3G or aHUS. We present seven lines of evidence showing that renin does not participate in the cleavage of C3, suggesting that the use of aliskiren as a renin inhibitor to decrease C3 convertase formation in patients with C3G and aHUS is misguided.

Data Sharing Statement

All data associated with this study are presented in the article and supplementary material.