Effect of ripening and in vitro digestion on the evolution and fate of bioactive peptides in Parmigiano-Reggiano cheese

Abstract The influence of ripening and in vitro digestion on the peptidomic profile of Parmigiano-Reggiano (PR) cheeses was investigated. Ripening and in vitro digestion thoroughly modified the peptidomic profile of the three cheeses. Twenty-six bioactive peptides were identified in undigested PR. Some peptides were degraded and others released during ripening. After digestion, 52 bioactive peptides were identified. Semi-quantitative data suggested that bioactive peptides released after digestion can be clustered in 5 groups according to the ripening time. VPP and IPP peptide levels in undigested samples were in the range of 4.52–11.34 and 0.66–4.24 mg kg−1, with the highest amounts found in 18-month ripened PR. YPFPGPI peptide was absent in undigested PRs but was released after digestion, especially in the 12-month-old sample (20.18 mg kg−1). The present study suggests possible differences in bioactive peptide levels after digestion as a function of the duration of ripening of PR cheese.


13
Cheese ripening is characterized by a complex chain of events that entails an intricate set of 14 biochemical reactions. Among the different biochemical events occurring during cheese ripening, 15 proteolysis is undeniably one of the most important. Several enzymatic activities originating from 16 various sources are involved in the proteolysis process during cheese ripening. Curd and 17 endogenous milk proteolytic enzymes (such as plasmin) initially hydrolyze caseins, generating large 18 or intermediate-size peptides. The released peptides are further cleaved by the action of proteinases 19 and peptidases coming from starter (S-LAB) and nonstarter (NS-LAB) lactic acid bacteria (Sforza 20 et al., 2012). 21 Bioactive peptides can be defined as short amino acid sequences, originally encrypted within the 22 sequence of the parent protein, which can be released after proteolysis and may have a positive 23 impact on human health (Rizzello et al., 2016). Numerous bioactive peptides have been identified 24 and characterized after hydrolysis of different food proteins or in fermented dairy products, 25 presenting different functional activities including antimicrobial, antioxidative, dipeptidyl 26 peptidase-IV (DPP-IV) and angiotensin-converting enzyme (ACE) inhibition, antihypertensive, 27 immunomodulatory and opioid activities (Nongonierma & FitzGerald, 2015;Rizzello et al., 2016). 28 In cheese, the presence of bioactive peptides is the result of a sensitive equilibrium between their 29 release and their degradation by the activity of lactic acid bacteria proteinases and peptidases during 30 cheese ripening (Sforza et al., 2004;Sforza et al., 2012). Numerous bioactive peptides, especially 31 ACE-inhibitory and anti-hypertensive peptides, have been identified in various cheeses (Sieber et 32 al., 2010;Lu, Govindasamy-Lucey, & Lucey, 2015;Stuknyte, Cattaneo, Masotti, & De Noni, 2015;33 intestinal digestion of Parmigiano-Reggiano. The concentration of some peptides such as VPP and 65 IPP was mostly un-affected by the in vitro digestion, whereas HLPLP and LHLPLP levels greatly 66 increased after the digestive process. Some other peptides, such as AYFYPE and AYFYPEL, were 67 not found in the Parmigiano-Reggiano samples but they were released during in vitro digestion. 68 These results suggest that in vitro digestion greatly influences the peptidomic profile of cheese. 69 The present study was designed to compare the peptidomic profile of Parmigiano-Reggiano cheese 70 at different times of ripening as well as the influence of in vitro gastro-intestinal digestion.

71
Bioactive peptides were identified, relatively quantified (by integration of the peak area of 72 individual peptides) and their fate was followed during ripening and in vitro digestion. Finally, three 73 well-known bioactive peptides, namely VPP, IPP and YPFPGPI, were quantified in the different 74 samples before and after in vitro digestion.  Water-soluble peptides extracts were obtained as described by Sforza et al. (2012) with slight 94 modifications. Five grams of cheese samples were mixed with 45 mL of 0.1 mmol L -1 HCl and 95 homogenized for 1 min (3 cycles) using an Ultra-Turrax homogenizer. The samples were then 96 centrifuged at 4000g for 40 min at 4°C. At the end of the centrifugation, the supernatants were 97 collected and filtered through Whatman filters paper 4 (Maidstone, Kent, UK). The in vitro digestion of Parmigiano-Reggiano samples was carried out by following the protocol 102 previously developed within the COST Action INFOGEST (Minekus et al., 2014). Simulated 103 salivary, gastric and intestinal fluids (SSF, SGF and SIF) were prepared exactly as described by 104 Minekus et al. (2014). Cheese samples (5 g) were mixed with 5 mL of SSF (containing 150 U mL -1 105 of salivary α-amylase), ground and incubated for 5 min at 37°C to reproduce mastication. The bolus 106 was then mixed with 10 mL of SGF (containing 4000 U mL -1 of porcine pepsin) and the pH analysis. For each digestion, aliquots were taken at the end of the gastric and intestinal phases of Low molecular weight peptides from WSPE and digested samples were extracted by ultrafiltration 119 (cut-off 3 kDa) as described in Tagliazucchi et al. (2017). The peptide content in these peptide 120 fractions was determined by measuring the amount of released amino groups using the 2,4,6-121 trinitrobenzenesulfonic acid (TNBS) assay and leucine as standard (Adler-Nissen, 1979). The 122 obtained raw data from the digested samples were corrected by the contribution of the control 123 digestion. Data are expressed as mmol leucine equivalent g -1 of cheese. Spectrometer (Thermo Scientific, San Jose, CA, USA) using a C18 column (Zorbax SB-C18 131 reversed-phase, 2.1 × 50 mm, 1.8 μm particle size, Agilent Technologies, Santa Clara, CA, USA).

132
The mobile phase consisted of (A) H2O/formic acid (99.9:0.1, v /v) and (B) acetonitrile. The sample 133 (10 μL, 100-fold diluted) was loaded into the column at a flow rate of 0.3 mL/min. The gradient 134 started at 2% B, and grew to 3% B in 2 min. The mobile phase composition was raised to 27% B in 135 The MS/MS spectra were then converted to .mgf files and the peptides were identified by using the Peptides Database (MBPDB) (Nielsen, Beverly, Qu, & Dallas, 2017). Only peptides with 100% 151 homology to acknowledged functional peptides were considered as bioactive peptides. The relative 152 amount of the bioactive peptides was estimated by integrating the area under the peak (AUP). AUP 153 was measured from the extracted ion chromatograms (EIC) obtained for each peptide (tolerance ± 5 154 ppm). Data are expressed as AUP g -1 of cheese. for 2 min, and then increased to 15% B between 2 and 6 min. The mobile phase composition was 167 then increased to 27% B in 15 min and further raised to 90% B in 4 min. The flow rate was set at 168 0.4 mL min -1 . 169 Ion source parameters was as follow: spray voltage 4 kV, capillary temperature 320 °C, sheath gas 170 50 and auxiliary gas 25. PRM parameters were as follow: resolution 17500, AGC target 5e5, max 171 IT 150 ms, MSX count 1 and isolation window 3.0 m/z.     An increase in the level of low molecular weight peptides was observed for PR cheeses at different 206 ripening time-points after gastric digestion ( Figure 1). The amount of peptides released from PR24 207 after gastric digestion was significantly higher (P < 0.001) than that released from PR12 and PR18.  (Table S1). According to the TNBS assay data, the PR24 sample contained the 219 highest amount of peptides (257 peptides), whereas the amount of peptides identified in PR12 and 220 PR18 samples was similar (84 and 72 peptides, respectively) and lower compared to the number 221 observed in the PR24 sample ( Fig. 2A). The majority of the peptides identified in PR12 and PR18 222 samples were from β-casein (63.1% and 58.3%, respectively), whereas the remaining identified 223 peptides were from αS1-casein ( Fig. 2A). PR24 sample also contained peptides released from αS2-   (Solieri et al., 2018;Hebert et al., 2008;Lozo et al., 2011;Juillard et al., 1995;Miyamoto et al., 2015). It is important to emphasize that a 239 complex and dynamic population of LAB, which thoroughly changes during ripening, is 240 characteristic of PR cheeses (Solieri et al., 2012).

241
The action of the extracellular proteinase and intracellular peptidases in Lactobacillus can explain 242 the presence of the high number of unique peptides in PR24. Several studies showed that the LAB 243 population decreases as the ripening of PR proceeds (Solieri et al., 2012;Coppola et al., 1999).

255
The Venn diagram indicated that 158 identified peptides (corresponding to the 33.7% of total 256 identified peptides) co-existed in the three PR peptide fractions after in vitro digestion (Fig. 3B).

257
There were 115, 31 and 24 peptides exclusively found in digested PR12, PR18 and PR24, 258 respectively. The highest similarity in peptide profiles was found between digested PR18 and PR24 259 samples, with 258 common peptides.

260
Only 22, 21 and 37 peptides were commonly found in undigested and digested samples from PR12,

296
Of the 11 commonly identified peptides, the relative abundance of 10 was significantly higher in 297 P24 sample respect to P12 and P18 samples (P<0.05). In contrast, the ACE-inhibitory peptide 298 FFVAPFPEVFGK displayed a decreasing trend during ripening with the highest relative abundance 299 found in P12 sample (P<0.05). The ACE-inhibitory peptide SKVLPVPQ was not detected in 300 sample P12 but showed an increasing trend during ripening with the highest relative abundance 301 found in sample P24 (P<0.05).

317
The resulting data from semi-quantitative analysis demonstrated that the majority of identified 318 bioactive peptides were not present at a constant level after digestion with respect to the ripening 319 time, but each peptide showed a characteristic trend. Bioactive peptides identified after in vitro 320 digestion can be clustered into 5 different groups as a function of the evolutive trend respect to the 321 ripening time (Table 2).

330
Peptide LHLPLP was able to resist in vitro gastro-intestinal digestion but it was hydrolyzed to 331 HLPLP by cellular peptidases prior to being transported across Caco-2 cells (Quirós et al., 2008;332 Tagliazucchi et al., 2006). The latter can actually be absorbed by intestinal cells and has been found 333 in human plasma after oral administration (Van Platerink et al., 2006). It has been suggested that the  The second group was characterized by bioactive peptides whose release after in vitro digestion 338 increased according to the ripening time reaching a plateau after 18 months of ripening (Table 2).

339
This group contained the majority of anti-microbial peptides and some ACE-inhibitory peptides 340 with demonstrated in vivo activity on spontaneously hypertensive rats (SHR) and low IC50 values.
The peptide AVPYPQR was able to decrease the blood pressure in SHR and behaved as a 342 multifunctional bioactive peptide also showing anti-microbial, anticoagulant and antioxidant 343 activities (Karaki et al., 1990;Tonolo et al., 2018;Tu et al., 2019).

344
The third group was characterized by bioactive peptides the release of which after in vitro digestion 345 increased according to the ripening time reaching a maximum value at 18 months of ripening (Table   346 2). To this group belonged peptides with different biological activities. The peptide AYFYPEL 347 presented a very low IC50 value against ACE and was able to reduce blood pressure in SHR

349
The fourth group was represented by bioactive peptides whose release after in vitro digestion 350 decreased according to the ripening time (Table 2). This group was characterized for the presence of 351 the peptide YPFPGPI (also known as β-casomorphin-7) and its precursors. Some ACE-inhibitory 352 peptides were also found in this group but, with the exception of YPFPGPIPN, they displayed 353 higher IC50 values.

354
Finally, the last group contained peptides whose amount after in vitro digestion remained constant 355 throughout ripening (Table 2).  (Table 3). The amount of VPP and IPP found in the 12-month ripened PR were 368 6.87 ± 0.68 and 1.63 ± 0.82 mg kg -1 , respectively. These data are in accordance with the range 369 reported by Basiricò et al. (2015) in PR sample at 12 months of ripening. The amount of VPP and 370 IPP increased in the sample at 18 months of ripening, reaching a concentration of 11.34 ± 0.21 and 371 4.24 ± 2.85 mg kg -1 , respectively. After that, we observed a strong decline in the concentration of 372 VPP and IPP at 24 months of ripening (4.52 ± 0.28 and 0.66 ± 0.05 mg kg -1 , respectively).

373
Peptide YPFPGPI, in contrast, was not detected in any undigested PR sample. This is consistent

Funding
This research did not receive any specific grant from funding agencies in the public, commercial, or not-for-profit sectors.