Species-specific pharmacology of maximakinin, an amphibian homologue of bradykinin: putative prodrug activity at the human B₂ receptor and peptidase resistance in rats

Drugs

BK was purchased from Bachem (Torrance, CA, USA), the B₂R antagonist icatibant, from Phoenix Pharmaceuticals (Burlingame, CA, USA), enalaprilat dehydrate, from Kemprotec Ltd. (Maltby, Middlesbrough, UK) and MK from Tocris Bioscience (Minneapolis, MN, USA; sequence of BK-related peptides in Table 1). MK-des-Arg (MK sequence without the C-terminal Arg residue) has been synthesized by Peptide 2.0 Inc. (Chantilly, VA, USA) via standard solid-phase methodology and provided as a >99% pure reagent (mass spectroscopy and HPLC analyses). The activity of MK-des-Arg has recently reported at the human BK B₁ receptor (Charest-Morin & Marceau, 2016). The other drugs were purchased from Sigma-Aldrich (St. Louis, MO, USA). Compound 48/80 is the condensation product of N-methyl-p-methoxyphenethylamine with formaldehyde and is a mixture of 3- to 6-mers as supplied (Sigma-Aldrich).

Receptor constructions, radioligand binding assays

The construction of myc-tagged B₂R expression vectors containing the human and rat receptor sequence has been recently reported, as well as the derivation of HEK 293a cell lines that stably express each one of these constructions (Charest-Morin et al., 2015). The HEK 293a cell line was originally obtained from Sigma-Aldrich. Affinity of N-terminally extended kinins for the B₂R was evaluated using a radioligand binding competition assay performed at 0 °C in the presence of peptidase inhibitors that included captopril and PMSF (Houle et al., 2000). Briefly, the binding of 3 nM [³H]bradykinin (Perkin Elmer Life Sciences; 90 Ci/mmol) to adherent intact HEK 293a cells stably expressing a myc-B₂R construction was applied to construct binding competition curves for a series of unlabeled  peptides.

The vector MRGPRX2-Tango (human receptor sequence; Kroeze et al., 2015) was a gift from Dr. Bryan Roth (Addgene plasmid # 66440). This vector was transiently expressed in HEK 293a cells using the MegaTran transfection reagent as directed by the supplier (Origene). The encoded receptor notably signals via calcium and responds to the mast cell activator, compound 48/80 (Kroeze et al., 2015; McNeil et al., 2015).

Calcium mobilization

To quantify the calcium mobilization induced by kinins, Fura-2 fluorometry (Molecular Probes, Invitrogen Detection Technologies) was applied to HEK 293a cells stably expressing myc-B₂R constructions possessing the human or rat sequence or to HEK 293a cell transiently transfected with the MRGPRX2-coding vector. Cells were cultivated in 75 cm² flask and each flask allowed for three determinations. The cells were detached using enzyme-free Cell Dissociation Buffer (Life Technologies) resuspended in serum-free DMEM before being centrifuged 5 min at 1,100 rpm at room temperature. Following centrifugation, cells were resuspended in Hank’s balanced salt solution (1×, pH 7.4, prepared from 10× concentrate, Multicell Wisent, St. Bruno, Canada) with 10 mM HEPES and 1.6 mM CaCl₂. At this point, Fura-2-AM was added to cell suspensions (final concentration 2 µM) and incubated 30 min in a 37 °C bath with agitation. After the incubation, cells were centrifuged and resuspended in the appropriate volume of buffer. Calcium mobilization was read with a thermostated (37 °C) spectrofluorimeter (SLM8000C; excitation 340 nm and emission 510 nm) in 2 ml suspension of cells loaded with Fura-2. After the readings, the maximum mean fluorescence (F_max) was measured by adding 15 µL of 10% Triton X100 and the minimum mean fluorescence (F_min) with further addition of 15 µl of NaOH 1 N and 100 µl of EGTA 100 mM. Calcium mobilization concentrations were established with the following equation ([Ca⁺²] = 224((y − F_min)∕(F_max − y)), where y represents the fluorescence reading from the sample. To allow a better comparison of the results, the calcium mobilizations were expressed as intracellular Ca²⁺ fold increase vs. the baseline over time.

Human umbilical vein contractility assay

The institutional research ethics board (CHU de Québec) approved the anonymous use of human umbilical cord segments obtained after elective cesarean section deliveries (file number: 2012-323). Informed written consent was obtained from mothers. Umbilical vein rings, used as a contractile bioassay for the BK B₂R and histamine H₁ receptor (H₁R), were prepared and suspended in organ baths and submitted to equilibration in Krebs’ solution as described (Charest-Morin et al., 2014). The vascular preparation was used to assess the potency of individual peptides, the effect of B₂R and H₁R antagonists and of peptidase inhibitors (introduced 30 min before the agonist) on the apparent potency of MK and the kinetics of kinin-induced contraction, as outlined in Results.

Liquid chromatography–mass spectrometry (LC–MS)

Synthetic MK (1 µM) was submitted to digestion in 2 ml of sterile-filtered Krebs solution in the presence of a freshly prepared ring of human umbilical vein (wet weight 25–37 mg). The tubes containing these materials were incubated at 37 °C for 5 or 15 min; the tissues were then removed and five volumes of cold ethanol (10 ml) were added to each tube. The tubes were incubated for 1 h on ice, centrifuged at 10,000 g to remove precipitated proteins and the kinin-containing supernatants were collected (procedure previously validated, Raymond et al., 1995). The supernatants were then completely evaporated to dryness in a Speedvac system and reconstituted in a small volume of water. Peptides were desalted on stage tip (C18), vacuum dried and then resuspended into 0.1% formic acid. One hundred fmol of each sample (starting MK quantity) was analyzed by mass spectrometry.

Mass spectrometry: peptide samples were separated by online reversed-phase (RP) nanoscale capillary liquid chromatography (nanoLC) and analyzed by electrospray mass spectrometry (ES MS/MS). The experiments were performed with a Ekspert NanoLC425 (Eksigent) coupled to a 5600+ mass spectrometer (AB Sciex, Framingham, MA, USA) equipped with a nanoelectrospray ion source. Peptide separation took place on a picofrit column (Reprosil 3u, 120A C18, 15 cm × 0.075 mm internal diameter). Peptides were eluted with a linear gradient from 5–35% solvent B (acetonitrile, 0.1% formic acid) in 35 min, at 300 nL/min. Mass spectra were acquired using a data dependent acquisition mode using Analyst software version 1.7. Each full scan mass spectrum (400 to 1,250 m/z) was followed by collision-induced dissociation of the twenty most intense ions. Dynamic exclusion was set for a period of 3 s and a tolerance of 100 ppm

Database searching: MGF peak list files were created using Protein Pilot version 5.0 software (Sciex) utilizing the Paragon and Progroup algorithms. MGF sample files were then analyzed using Mascot (Matrix Science, London, UK; version 2.5.1).

Mascot was set up to search the MK sequence assuming that the digestion enzyme was non-specific. Mascot was searched with a fragment ion mass tolerance of 0.100 Da and a parent ion tolerance of 0.100 Da. Deamidation of asparagine and glutamine and oxidation of methionine were specified in Mascot as variable modifications. Scaffold (version Scaffold_4.7.1; Proteome Software Inc., Portland, OR, USA) was used to validate MS/MS based peptide and protein identifications.

Data processing: Peak View 1.2.0.1 (Sciex) was used to extract the specific precursor masses and calculate the area under the curve. Extracted ion chromatoram (XIC) were generated from the TOF MS using a peak width of 0.05.

Surgical preparation

All surgical and experimental procedures were reviewed and approved by the Animal Care and Handling Committee of Laval University, in accordance with the Canadian Council on Animal Care (file no. 2012-223). Experiments were performed on male Sprague-Dawley rats (300–375 g) purchased from Charles River Laboratories (St-Constant, Qc, Canada). Only animals of the male gender have been studied to be consistent with related and recently published data (Jean et al., 2016a). The rats were housed in a temperature-controlled room (22 ± 1 °C) on a 12 h/12 h light-dark cycle (lights on at 0600) and had free access to normal chow diet and tap water. They were allowed to acclimate to their environmental conditions for one week prior to being studied. At the end of the acclimation period, the rats were anesthetised with sodium pentobarbital (50 mg kg⁻¹, i.p., supplemented as required) and had one catheter implanted into the right jugular vein (for intravenous ( i.v.) injection) and one into the left femoral artery (for direct and continuous measurement of blood pressure and heart rate as previously described, Jean et al., 2016a). In some experiments, a miniaturized pulsed Doppler flow probe (Haywood et al., 1981) was implanted around the distal abdominal aorta (below the level of the ileocaecal artery) through a midline abdominal incision to monitor changes in hindquarter hemodynamic, according to the method previously developed by Gardiner & Bennett (1988) and as previously described (Bachelard et al., 1992; Jean et al., 2016a). Experiments started at least 30 min following the end of surgery in anesthetised rats.

In vivo hemodynamics in anesthetized rats

Baseline heart rate and phasic and mean arterial blood pressure were recorded over a period of 15 min in anesthetized rats. A dose response curve was then obtained by recording changes in blood pressure and heart rate elicited by i.v. injection of peptide vehicle followed by increasing doses of MK, or BK. Peptides were dissolved in isotonic saline (0.9% NaCl) containing 0.1% BSA to prevent the adsorption of peptide to the glassware and plastic surfaces. All i.v. injections were given as 100 µl boluses which were washed in with a further 100 µl of saline (the dead space of the catheter). Only one peptide was tested per group of rats and each injection started with saline-BSA 0.1% followed by the lowest dose of peptide. The next dose was administered once all recorded cardiovascular parameters had returned to baseline after the previous injection (usually 2–10 min). At the end of the experiments each animal was euthanized with an overdose of sodium pentobarbital (240 mg/kg, i.v.).

The mechanism subserving the cardiovascular responses to i.v. injections of increasing doses of MK was first investigated in animals pretreated with the ACE inhibitor, enalaprilat. In these experiments, enalaprilat was intravenously administered as bolus (0.1 mg/kg, 0.1 ml) following a 10 min period of baseline measurements of heart rate and blood pressure. Fifteen minutes later, a dose–response curve to MK was obtained, as described above. Further experiments were made in rats pretreated with a potent, long acting and selective B₂R antagonist, icatibant (Hoe 140) (D-Arg-[Hyp³, Thi⁵, D-Tic⁷, Oic⁸] bradykinin) (Hock et al., 1991; Wirth et al., 1991; Rhaleb et al., 1992; Marceau et al., 1994). In these experiments, icatibant was intravenously administered as bolus (25 µg/kg, 0.1 ml) 15 min before the i.v. injection of increasing doses of MK, as above. Further dose–response curves were also obtained from rats pretreated with pyrilamine, a histamine H₁R antagonist. In these experiments, pyrilamine was intravenously administered as bolus (5 mg/kg, 0.1 ml) 15 min before the i.v. injection of increasing doses of MK, as above. Control experiments were also performed to ensure that changes in blood pressure responses elicited by a high dose of histamine (6.4 µg/kg i.v.) could be inhibited by 5 mg/kg pyrilamine.

Acute hindquarter hemodynamic effects in anesthetized rats

Before each experiment, a 30-min stabilization period was allowed, during which continuous recording of heart rate, phasic and mean blood pressure and phasic and mean Doppler shift signals from the hindquarter probe were made in anesthetized rats, using the Biopac data acquisition and analysis system previously described (Jean et al., 2016a). Then, a first group of rats (n = 8) received successive i.v. injections of peptide vehicle (saline-BSA 0.1%), BK (6.4 µg/kg), and MK (1.6 µg/kg). All i.v. injections were given as 100 µl boluses which were washed in with a further 100 µl of saline. The next dose was administered once all recorded cardiovascular parameters had returned to baseline after the previous injection (usually 5–7 min). Further experiments were made in a second group of rats (n = 6) pretreated with the B₂R antagonist, icatibant. In these experiments, the antagonist was intravenously administered as bolus (25 µg/kg, 0.1 ml) followed 15-min later by the i.v. injections of peptide vehicle, BK (6.4 µg/kg) and MK (1.6 µg/kg), as described above. At the end of the experiments the rats were euthanized with an overdose of anesthetic (pentobarbital 240 mg/kg, i.v.).

Microscopy

HEK 293a cells stably expressing myc-B₂R constructions, either with human or rat sequence, were used for a side-by-side comparison of labeling by the EGFP-MK fusion protein. As previously reported (Charest-Morin et al., 2013), EGFP-MK was produced as the lysate of producer cells and diluted in the culture medium of receptor expressing cells for a 30 min-period of stimulation at 37 °C. The intact cells were prelabeled with an anti-myc tag monoclonal antibody conjugated with AlexaFluor 594 (clone 9E10, R&D Systems, dilution 1:100, 15-min incubation at 37 °C) to evidence myc-B₂R expression at the cell surface and eventual subcellular translocation. Cells were photographed using an Olympus BX51 microscope coupled to a CoolSnap HQ digital camera (filters for GFP: excitation 460–500 nm, emission 510–560 nm; for AlexaFluor 594: excitation 525–555 nm, emission 600–660).

Data analysis

Results are presented as means ± s.e.m. Radioligand binding data were fitted by nonlinear regression to a one-site competition equation using a least-square method (Prism 5.0; GraphPad Software Inc., San Diego, CA, USA) and IC₅₀ values calculated from this procedure. The same computer program was used to draw concentration-effect curves (least square fitting of sigmoidal dose–response equation with variable slope) and to derive contractile EC₅₀ values. Data describing baseline values of heart rate and mean arterial blood pressure and hypotensive responses to peptides in anesthetized rats were assessed by using one-way analysis of variance (ANOVA) followed by the Dunnett’s test (repeated comparison with a common control). The effects of drugs on the hypotensive responses to each dose of MK were assessed by using ANOVA followed by the Dunnett’s test. The effect of icatibant on hemodynamic parameters influenced by doses of BK or MK in rats instrumented for Doppler flow recording was tested using Student’s t test (Prism 5.0 software).

Affinity of the N-terminally extended kinin MK for recombinant myc-B₂Rs

A [³H]BK binding competition assay performed at water-ice temperature was exploited to determine the true receptor affinity of MK (Figs. 1A and 1B). This N-terminally prolonged BK sequence exhibited a very low affinity for the myc tagged B₂R construction possessing the human sequence (Fig. 1A, relative potency vs. that of BK reported in Table 1). Thus, MK was ∼1,500-fold less potent than BK. At the rat recombinant myc-B₂R, the radioligand binding competition assay indicated that MK had a lesser affinity than that of BK, but with narrower margin (6.2-fold; Fig. 1B). The relative potency of MK vs. BK was similar to the one established with an identical experiment performed with the rabbit B₂R-GFP construction (Bawolak et al., 2012), thus substantiating a significant species-dependent affinity variation for MK across mammalian species. MK-des-Arg did not significantly displace [³H]BK binding to the human myc-B₂R construction (Fig. 1A), consistent with the obligatory role of the C-terminal arginine residue of kinins for a good affinity at B₂Rs (Leeb-Lundberg et al., 2005).

Calcium signaling in HEK 293a cells stably expressing either human or rat myc-B₂Rs was conducted with a constant agonist concentration (10 nM, Fig. 2A). A sizeable and immediate signal was elicited by BK in cells expressing receptors from either species; MK also induced a strong but slightly protracted response via the rat receptor, but was virtually inactive at the human myc-B₂R. The mast cell receptor MRGPRX2 (human sequence), transiently expressed in HEK 293a cells, mediated calcium transients in response to compound 48/80 (10 µg/ml, positive control; McNeil et al., 2015), but MK was marginally active only at 10 µM at this receptor (Fig. 2B).

Human umbilical vein contractility and LC–MS

The freshly isolated human umbilical vein is a contractile bioassay for the B₂R exploited in several laboratories (Altura et al., 1972; Marceau et al., 1994; Marceau et al., 2010; Gobeil et al., 1996). BK elicits a response at subnanomolar concentrations in this preparation (Fig. 3A). Contrasting with the very low affinity of MK for the human B₂R in the binding competition assay, this prolonged peptide was only ∼20-fold less potent than BK in the contractility assay (Fig. 3A, EC₅₀ values reported in Table 1). This suggests a conversion of MK by attack at the N-terminus, BK being the shortest B₂R agonist of high affinity (Leeb-Lundberg et al., 2005). However, MK-des-Arg has negligible effects in the contractility assay, in line with the binding assay.

Equipotent concentrations of kinins (approximately EC₅₀ values, from Fig. 3A) were chosen to compare the contractile kinetics of the venous preparation (Fig. 3B). It was verified that BK and MK produced a similar contractile response expressed in gram-weight (Fig. 3B, right). However, the contraction induced by BK (10 nM) was rapidly developing and reaching a plateau, whereas MK (200 nM) induced a slowly developing response (Fig. 3B, left). These observations further support an indirect response of the N-terminally prolonged peptides on the human B₂Rs. That the effect of MK on the umbilical vein is ultimately mediated by the BK B₂Rs is shown by the profound inhibitory effect of icatibant (1 µM), an antagonist of this receptor (Fig. 3C). The H₁R antagonist pyrilamine (1 µM) did not reduce the venous effect of MK, while profoundly and competitively inhibiting the contractile effect of histamine (Fig. 3C). Thus, histamine release does not contribute significantly to the contractile effect of MK in this preparation.

Whether the venous preparation at 37 °C has peptidase/protease activities that could produce shorter peptides with high receptor affinities from MK has been tested in different groups of tissues pre-treated with various inhibitors (Fig. 3D). The combination of aminopeptidase inhibitors, bestatin and puromycin (Hitzerd et al., 2014), had no effect on the apparent contractile potency of MK. The same applied to the metallopeptidase inhibitor EDTA calcium disodium, an agent that modulated the apparent potency of kinins in a previous study of the rabbit isolated aorta (Babiuk et al., 1982), and to the wide spectrum protease inhibitor, leupeptin (Zhang et al., 2005). The cysteine protease inhibitor E-64 and the serine protease inhibitor pefabloc SC also failed to change the concentration-effect curve of MK (Fig. 3D). In principle, this observation rules out the metabolic activation of MK by many serine proteases, including kallikreins. The dual ACE and neutral peptidase inhibitor omapatrilat also failed to modify the contractile effect of MK (data not shown).

The conversion of MK into shorter, high affinity kinin(s) in the human vein is a logical necessity due to the marginal affinity of MK for the human B₂R and to the slow contraction kinetics of MK; however, individual proteases covered by the panel of inhibitors may not be kinetically limiting for this conversion. We addressed the possible generation of pharmacologically active C-terminal fragments of MK using LC-MS analysis of peptides extracted from Krebs solution containing MK (1 µM) and incubated with rings of umbilical veins (Fig. 4). Among the most intense ions, two C-terminal fragments were found. The release of Lys-Gly-Pro-BK and Gly-Pro-BK was evidenced in a time-dependent manner (Fig. 4; sequence confirmed by tandem mass spectroscopy, Fig. S1), supporting the time-dependent generation of BK-like peptides at the vicinity of tissue receptors. Surprisingly, the commercial peptide MK contained small amounts of both C-terminal fragments as contaminants, but the reaction with the tissue increased 3-4-fold the intensity of the corresponding ions. Abundant N-terminal fragments of MK were also identified, such as DLPKINRKGPRPP and DLPKINRKGPRPPGF (data not shown), supporting that multiple cleavage sites are present in the MK sequence under these conditions.

Species-dependent variation of cell labeling with EGFP-MK

MK is the B₂R-binding module in the fusion protein EGFP-MK (Charest-Morin et al., 2013). We reported that the protein is transported in endosomes by both the rabbit and human B₂Rs (Charest-Morin et al., 2013; Charest-Morin & Marceau, 2016), but wished to run a side-by-side comparison of the rat and human B₂R-mediated uptake as a function of the concentration of EFGP-MK (Fig. 5). In these experiments, myc-B₂R possessing the human or rat receptor sequence, were prelabeled at the surface of live cells with anti-myc tag monoclonal antibodies conjugated to a red fluorophore. Then the intrinsically green fluorescent protein EGFP-MK was added for incubation at 37 °C. Results show that the fusion protein is more potent at the rat myc-B₂R, which mediated the endocytosis of EGFP-MK along with extensively co-localized anti-myc antibodies, supporting the persistence of agonist-receptor complexes in endosomes at the examined time point. EGFP-MK has comparatively little effect at the human myc-B₂R construction, except for initiating a partial endocytosis of labeled receptors at 1:50 dilution without significant green signal (Fig. 5). This response is comparable to that of the 1:1,250 dilution of EGFP-MK at the rat receptor.

In vivo hemodynamic responses to MK

Baseline values for MAP and HR measured in the untreated control group or 15 min after i.v. pretreatment with enalaprilat, icatibant or pyrilamine are shown in Table 2. While no significant changes in basal values of MAP were noted between the groups, a slight but significant increase in basal HR was found between the enalaprilat pretreated groups and the control group (Table 2).

We recently described the brief hypotensive responses associated with tachycardia to the i.v. injection of increasing doses of BK in anesthetized rats and the strong potentiation of the responses following pharmacologic ACE blockade (Jean et al., 2016a). Using the same methods, MK was consistently more potent than BK to produce hypotension (Figs. 6 and 7). Further, the hypotensive episodes elicited by MK lasted longer and the tachycardia was generally more intense than what was recorded in rats injected with increasing doses of BK (Figs.7D and 7E). In sharp contrast to the effects of i.v. BK (Jean et al., 2016a), MK was not significantly potentiated by pretreatment with the ACE inhibitor enalaprilat, but the effect of the prolonged peptide was abated by the B₂R antagonist icatibant (Fig. 7A). Pyrilamine did not antagonize the effects of MK (Fig. 7A), and had no additional effect if combined with icatibant (data not shown). Pyrilamine, at the employed dose level, abated the hypotensive effect of histamine tested in separate groups of rats (Fig. 7B).

Figure 8 shows the simultaneous changes in mean arterial blood pressure, mean hindquarter Doppler shift signal and heart rate induced by i.v. injections of vehicle, BK and MK administered at doses producing a sizeable hypotensive effect of 40–55 mmHg. We found that, for both tested agonist, the observed hypotensive responses were accompanied with concomitant increases in mean Doppler shift signals and heart rate when compared with vehicle values. The increases in mean hindquarter Doppler shift might reflect a reduction in vascular resistance (vasodilation) redistributing flow to the hindlimb muscles, as previously shown in conscious rats i.v. injected with BK (Bachelard et al., 1992). Interestingly, the cardiovascular responses elicited by BK or MK were all significantly abated by systemic pretreatment of rats with icatibant, suggesting actions mediated by cardiovascular B₂Rs.