Four peptides: IY (Isoleucine-Tyrosine), VW (Valine-Tryptophan), RIY (Arginine-Isoleucine-Tyrosine; rapakinin) and VWIS (Valine-Tryptophan-Isoleucine-Serine) were prepared. All peptides were prepared by solid phase peptide synthesis, using standard Fmoc/tBu methodology. The properly preloaded Wang resins, common Fmoc-protected amino acid building blocks in standard Fmoc-Xxx-OH/TBTU/DIPEA protocol (2eq/2eq/4eq respectively) was used. The desired compounds were produced by deprotection with Reagent B: TFA/phenol/H₂O/TIPS (88:5:5:2; v/v/v/v). Crude peptides were purified by reverse-phase HPLC (high-performance liquid chromatography), using gradients of H₂O–ACN–0.1% TFA and the relevant fractions were lyophilized. The peptide purity, estimated by analytical HPLC was more than 95%. The molecular weights were confirmed by HR-ESI-MS (high-resolution electrospray ionization mass spectrometry).

Chymase digestion of ANG I in the presence of synthetized peptides was performed in vitro. The concentration of remaining ANG I was calculated in order to assess the ability of peptides to inhibit chymase.

Human chymase (72 units/mg protein; 259 µg protein/ml, vial volume 187 μl, Sigma) stock solution was prepared by diluting the initial sample of enzyme with Tris*HCl buffer (0.27 mM Tris*HCl, pH 7.8 and 150 mM NaCl) to obtain the final chymase concentration of 1.379 μg/ml. The ANG I (DRVYIHPFHL) stock solution was prepared by dissolving the peptide in the same Tris*HCl buffer (as above) to obtain a final peptide concentration of 1 mM. The peptide basic solutions of each of four peptides (RIY, VWIS, IY, VW) were prepared by dissolving the proper TFA salt of peptide in ANG I stock solution to obtain a final peptide concentration of 5 mM (5-fold excess to ANG I).

The samples of 100 μl of chymase stock solution were temperature-equilibrated at 37 ± 1°C for 15 min before adding 100 μl of angiotensin I stock solution or mixture of proper peptide in angiotensin I stock solution. Reactions (during which ANG I was converted to ANG II through removal of two C-terminal residues) were stopped at the desired times (0, 10, 20, 30, 40, 60, 80 min) by addition of 10 μl of 98% FA. After FA inhibition, reactions were freeze-dried, re-dissolved in 0.5 ml of methanol and analyzed using HPLC-MS system on C-12 column at linear AB gradient of 20–44% for B in 20 min where eluent A was 0.05% trifluoroacetic acid/water and eluent B was 0.05% trifluoroacetic acid/acetonitrile. Positive electrospray (ES+) was used as the ionization method for mass spectrometry.

The inhibitory effect of each peptide on chymase activity was expressed as remaining amount of substrate vs. time of digestion and presented as percentage ratio and calculated by the following formula [ANG I_t=X min/ANG I_t=0 min]x100% where ANG I_t=X min is the peak area of angiotensin I at desired time of digestion, ANG I_t=0 min is the peak area of angiotensin I at 0 min of digestion. The results were independently repeated in a second experiment and were used to calculate the half-life time of ANG I in absence or presence of inhibitory peptides.

In acute experiments rats were anesthetized with intraperitoneal (ip) sodium thiopental (100 mg/kg; Sandoz GmbH, Kundl, Austria), which provided stable anesthesia for about 4 h. All surgical preparations were already described in detail (Gawrys et al., 2014; Badzynska et al., 2020). Throughout the experiment the temperature of the animal was kept constant (37°C). A cannula was placed in the trachea to ensure a free airway. The jugular vein was cannulated for infusion of fluids and drugs; during surgical preparations 3% bovine albumin in Ringer solution was infused at 10 ml/kg/h. For mean arterial blood pressure (MBP) measurement, a Teflon catheter was placed in the femoral artery (Stoelting Blood Pressure Meter, Wood Dale, IL, United States). After surgical preparations and stabilization of all parameters, a short control period (30 min) was made to obtain baseline for measured parameters. Chymostatin or its solvent (final concentration of DMSO was 0.05%) were administered for 1 h. After cessation of chymostatin infusion, a recovery period (30 min) was made. In the end, the animals were sacrificed by sodium thiopental overdose (iv); and vital life signs (blood pressure, breathing) were monitored until death.

Urine volume was determined by gravimetric method. Sodium and potassium concentration in urine and plasma were measured by flame photometry (PFP7/C, Jenway Ltd, Stone, United Kingdom). Urine and plasma osmolality were measured by freezing point depression (Osmomat® 030 M, Gonotec, Berlin, Germany). These parameters were used to determine the excretion rates of total solutes (U_osmV), sodium (U_NaV) and potassium (U_KV). Commercially available ELISA kits were used to measure nitric oxide metabolites (NOx, SunRed, Shanghai) and albumin (Albumin, ICL, Portland) concentration in urine. Excretion rate of NOx (U_NOxV) and albumin (UAE) were calculated.

The kidneys were fixed in 4% buffered formalin, embedded in paraffin and stained with hematoxylin/eosin for morphology evaluation. Glomerulosclerosis/hyalinosis index (GSI) was defined as segmental solidification of the glomerular capillary tuft, segmental collapse, obliteration of capillary lumen and accumulation of hyaline. Fifty glomeruli per section were randomly selected and evaluated using semi-quantitative scoring method as described in detail previously (Maric et al., 2004; Gawrys et al., 2018). Blind group allocation of the investigators was used.

Values are expressed as means ± SEM. The significance of changes was evaluated by multivariate analysis of variance ANOVA with repeated measurements (followed by Duncan’s or Dunnett’s post hoc tests). For some parameters one-way ANOVA was used. A value of p < 0.05 was considered to be significant. Calculations were made in STATISTICA (version 10.0; StatSoft Inc. Krakow, Poland) or GraphPad Prism (Version 6.07).

Out of four tested peptides only RIY and VWIS exhibited blocking activity toward chymase. The conversion of ANG I to ANG II by chymase was visibly slowed down, hence these two peptides were used in subsequent in vivo experiments (Figure 1).

Analysis of preliminary acute experiments with intravenous administration of a single dose of chymostatin to anesthetized rats, showed some antihypertensive properties. The observed decrease in blood pressure was significant but delayed, observed after cessation of chymostatin infusion, which suggested some role in blood pressure regulation in adult SHR. These results prompted us to study the effects of chymostatin after chronic administration to conscious rats, to evaluate possible prolonged inhibition of chymase on blood pressure.

Two-week treatment with chymostatin did not lower blood pressure in SHR. Surprisingly, this commercially available chymase inhibitor had an opposite effect, and a significant increase in SBP was observed.

Kirimura et al. (2005), studied NK3201, another chymase inhibitor. The authors found that a single oral dose of NK3201 did not lower blood pressure in adult SHR and concluded that chymase pathway of ANG II formation does not play an important role in this rat model of hypertension. However, the study by Kirimura et al. (2005), in addition to the obvious difference in the protocol (single oral dose vs. our chronic intravenous infusion) has another limitation: the tail cuff method was used to measure blood pressure. This method is now considered far from ideal and does not allow to record subtle BP changes (Wilde et al., 2017). Moreover, spontaneously hypertensive rats are susceptible to stress, e.g., that due to prolonged immobilization during the tail cuff measurement, which might contribute to the hypertension development (Yamori, 1977; Henry et al., 1993).

Even though we obtained the results similar as Kirimura et al. (2005), we do not dismiss the importance of chymase in ANG II formation. We suggest that chronical inhibition of local/tissue RAAS results in activation of the main pathway of ANG II production, through systemic ACE. It is known that chymostatin does not block ACE (Petrie et al., 2001), hence more of endogenously generated ANG I is available for ACE. It was previously suggested that inhibition of ACE alone does not suffice to suppress RAAS (Petrie et al., 2001), which provides a rationale to employ chymase inhibitors to treat hypertension. It seems that blocking both pathways, at least partially, should be considered as a new therapeutic option.

Surprisingly, administration of DMSO in young animals resulted in a substantial increase in the excretion rate of nitric oxide metabolites. This somewhat confounding finding calls for special attention. We believe that young rats are more susceptible to toxic effects of DMSO, yet since they are still normotensive with healthy endothelium, the observed increase might be caused by NOS stimulation as a counteraction to vasoconstriction due to DMSO. Such mobilization seems more likely in young animals, because of their endothelium is not yet damaged by hypertension and in adult rats this mechanisms is probably severely impaired. Moreover, it seems that susceptibility to DMSO toxicity of adult rats is lower. Assuredly, this speculation requires more in-depth studies in the future.

Previous research by our collaborators demonstrated that rapeseed-derived peptides possess some antihypertensive potential, despite only weak affinity toward ACE (Marczak et al., 2003; Yamada et al., 2010; Yamada et al., 2011). Current in vitro research confirmed that they also inhibit chymase, therefore we tested them further in prolonged, chronic studies in conscious SHR. The most promising effects were observed after RIY administration (rapakinin). It significantly reduced blood pressure, and increased nitric oxide (NO) metabolite excretion almost 3-fold. Also the glomerulosclerosis index (GSI) was lower, suggesting that RIY alleviated renal impairment to some extent. It is worth emphasizing that the beneficial action of RIY was observed in adult rats in established phase of the disease, with already developed renal and cardiovascular damage, a condition which represents a greater challenge than prevention of the disease in young subjects.

Based on our results it seems that chymase does not play any significant role in the developmental phase of hypertension, since no blood pressure or any other change was observed after chymase inhibition (similarly after chymostatin and rapeseed-derived peptides). This is somewhat surprising, considering previous reports suggesting the important role of mast cells, the main source of chymase, in tissue fibrosis (Kotov et al., 2020). Possibly, mast cells-related tissue remodeling might be more pronounced in more advanced hypertensive heart disease. Chymase-related tissue remodeling probably occurs also in the kidneys and may be a strong contributor to hypertension in older SHR, however this needs to be confirmed in further research.

In view of the above considerations, we suggest that the reason for the significantly higher effectiveness of rapakinin (when compared to chymostatin) is its ability to block both systemic and local pathway of ANG II production. Previous study by Marczak et al. (2003) also demonstrated that RIY is more stable in the presence of ACE, i.e., is not converted to less active dipeptide IY in such quantity as VWIS to VW. The study by Yamada et al. with isolated mesenteric arteries of SHR, demonstrated that RIY relaxes the vessels via an endothelium-dependent mechanism (Yamada et al., 2010). The authors claim that the vasoactive action of RIY is not mediated through nitric oxide and bradykinin. In the present study we observed significantly higher excretion of NO metabolites in rats receiving RIY (in which significant reduction in blood pressure was recorded). This observation somehow contradicts the NO-independent pathway suggested by the aforementioned authors. However, it should be noted that experiments on isolated tissues differ significantly from functional studies performed on whole conscious animals, in which the observed effect is a result of interaction of many complex regulatory systems.

Another important element in the complex nature of RAAS function is the existence of other substrates which can be converted to angiotensin II, such as angiotensin (1–12). It is suggested that ANG (1–12) is a superior substrate for chymase compared to ANG I (Ahmad et al., 2016). Additionally, it has been shown that chymase, not ACE, is the main enzyme generating ANG II from ANG (1–12) in the heart (Ahmad et al., 2011; Ahmad et al., 2013; Nagata et al., 2015; Baghel et al., 2016; De Mello et al., 2016). Recently, the role of intracellular RAAS is increasingly recognized and the discovery of non-canonical, intracellular ANG II production might explain the limited effectiveness of RAAS inhibitors (both ACE inhibitors and ARBs i.e., angiotensin II receptor blockers), because of their inability to reach intracellular sites and to inhibit chymase-dependent production of ANG II from other substrates, like ANG (1–12) (Ferrario and Mullick, 2017). One of the new suggested strategies to improve antihypertensive therapy is to develop therapeutic agents, such as chymase by RIY, which will be able to penetrate renal and cardiac tissues and inhibit intracellular components of the RAAS (Ferrario and Mullick, 2017).

Lately, protein therapy with so called cell-penetrating peptides (CPPs) is gaining more importance (Milletti, 2012; Ramsey and Flynn, 2015; Patel et al., 2019). CPPs are small molecules which possess the ability to cross the biological membranes and deliver therapeutics into the cells, usually rich in arginine. This amino-acid is involved in the mechanism of cell internalization by CPPs, due to its ability to disrupt and cross phospholipid bi-layers (Takeuchi and Futaki, 2016). Interestingly, RIY is a small tripeptide containing arginine, which could potentially augment its ability to pass across cell membrane and act inside the cells. It is possible that the significant effectiveness of rapakinin is a result of not only blocking two main pathways of ANG II synthesis (chymase and ACE), but also of its ability to penetrate the cell membrane and inhibiting intracellular synthesis of ANG II from ANG (1–12). However, more complex membrane permeability studies are necessary to prove this hypothesis.

The second of the tested peptides, VWIS, did not exhibit any antihypertensive activity. A previous study by Marczak et al. (2003) demonstrated that of the two peptides, VWIS exhibited higher affinity to block ACE than did RIY, however, it was also showed that in the presence of ACE it transformed rapidly to dipeptide VW, which in the present experiments failed to block chymase (see Figure 1). The aforementioned authors conducted also in vivo studies in which they measured blood pressure during 8 h following a single oral administration of selected peptides (RIY, VWIS and VW) to SHR. Interestingly, they observed transient (lasting 2 h) significant blood pressure reduction after each of the peptides (Marczak et al., 2003). Therefore we expected that prolonged administration of the peptides (over two weeks) will result in maintained blood pressure reduction in SHR. This was so only in the case of RIY. Possibly, the reason for the lack of antihypertensive activity of VWIS is its transformation to less active VW, which inhibits ACE only weakly and does not block chymase.

In adult rats the highest antihypertensive activity was observed after treatment with RIY, which might be a result of a few factors. First, it is probable that rapakinin affects both the chymase and ACE pathway of ANG II synthesis. Moreover, RIY contains arginine, which could augment the cell membrane penetration, act inside the cells and inhibit the generation of ANG II from ANG (1–12). Such complex action could account for the positive effects of RIY treatment: a decrease in BP, lowering of GSI and an increase in the excretion of nitric oxide metabolites.

It should be noted that the observed decrease in blood pressure after rapakinin was significant, however not as high as after treatment with the currently used drugs, such as ACEi. More extensive studies are necessary to fully evaluate the clinical value of RIY: prolongation of the treatment and/or elevation of the dose might be useful. Additionally, the efficacy of oral administration of the selected peptides should be examined. Previous studies demonstrated a positive but transient antihypertensive activity of a single oral dose of RIY in SHR (Marczak et al., 2003), therefore tests with chronic intra-gastric administration of the peptides are needed.

In summary, it appears that the peptide RIY might be useful in the treatment of hypertension, especially in the cases resistant to other therapy. Notably, rapeseed-derived peptides are easy and cheap to obtain and therefore might find application as an ingredient of functional food.