Modelling the ageing gastrointestinal system in-vitro using whey protein isolate

TL;DR

This study adapts digestion models to simulate how aging affects protein breakdown in older adults, showing reduced digestion efficiency.

Contribution

The study introduces an in-vitro model tailored to older adults' gastrointestinal conditions, revealing unique digestion patterns.

Findings

Older adult gastric conditions led to more protein aggregates and intact proteins.

Intestinal digestion showed smaller differences but distinct proteolytic profiles.

Antimicrobial peptides increased significantly in intestinal digesta of older adults.

Abstract

The aims of this study were to adapt current in-vitro digestion protocols to replicate older adults suffering from gastrointestinal infections or those on polypharmacy. The whey protein breakdown was characterised within simulated gastric and intestinal phases, by measuring degree of hydrolysis, bio-accessible peptides, molecular weight distribution by Size Exclusion Chromatography, as well as peptide sequencing using High Resolution LC-MS. Differences due to older adult digestive conditions significantly affected the digestion of whey protein, seen by the presence of some aggregates and intact proteins, and less bio accessible peptides at the end of the gastric phase. These differences were also seen in the intestinal digesta, albeit to a lesser extent wheredistinct proteolytic profiles were seen. This study shows the importance of adapting the simulated gastrointestinal conditions to better represent the ageing gastrointestinal systems of our elderly population, and to better understand the poor protein assimilation by this ageing cohort.

•Current in-vitro digestion protocols adapted to older adults suffering from gastrointestinal infections or those on polypharmacy.•Elderly gastric conditions affected whey protein breakdown with increased presence of large aggregates and intact proteins.•Smaller differences seen in intestinal digesta.•Antimicrobial peptides found to significantly increase in intestinal digesta of older adult SGID.

Current in-vitro digestion protocols adapted to older adults suffering from gastrointestinal infections or those on polypharmacy.

Elderly gastric conditions affected whey protein breakdown with increased presence of large aggregates and intact proteins.

Smaller differences seen in intestinal digesta.

Antimicrobial peptides found to significantly increase in intestinal digesta of older adult SGID.

Figures

Mean Plot of % bio-accessible peptides from whey protein isolate (WPI), digested in a simulated gastric phase for 3 h, under 4 different pH values of pH 3, pH 3.7 pH 4.5 and pH 6, and two pepsin concentration values of 2000 U/mL and 1200 U/mL. *n* = 3 with error bars determining ± standard deviation.Fig. 1

Molecular weight distribution for whey protein isolate (WPI) after Adult, Healthy Older Adult, Older Adult 2 and Older Adult 3 after simulated gastric digestive conditions calculated from SEC data. *n* = 3.Fig. 2

Molecular weight distribution for whey protein isolate (WPI) after Adult, Healthy Older Adult, Older Adult 2 and Older Adult 3 intestinal conditions calculated from size exclusion chromatography (SEC) data. n = 3.Fig. 3

Heat Map of significantly different peptides from whey protein isolate (WPI) after undergoing simulated gastrointestinal digestion (SGID), determined using liquid chromatography-mass spectrometry (LC-MS) and data analysed using Perseus, with *p*-value and Permutation-Based FDR analysis to isolate significant peptides.Fig. 4

Sequence profile of peptides on the three main whey proteins following full gastro-intestinal digestion in adults (—) heathy older adults (—), OA2 (—) and OA3 (—) SGID conditions.Fig. 5

Full text

Introduction

1

While protein is important across all life's stages, it is essential in older adults who are defined as persons over the age of 65 by the United Nations (Gerland et al., 2022). Protein assists with healthy ageing by preserving muscle mass and reducing muscle degradation or sarcopenia associated with increasing age (Murton, 2015; Olaniyan et al., 2021). Studies have shown that high-quality proteins, food-fortification and/or oral nutritional supplements (ONS) can potentially reduce issues with muscle protein turnover in older adults (Marshall et al., 2020) as well as contributing to resolving bone health issues such as osteoporosis and increased frailty (Bauer et al., 2015). However, poor digestion within the older adult upper gastrointestinal tract has been shown to reduce the ability of protein fortification to resolve these issues (Gilani & Sepehr, 2003). Specific changes within the stomach of older adults include reductions in the gastric protease pepsin (Rémond et al., 2015) and changes to gastric pH due to alterations in gastric acid secretion (Trey et al., 1997). Gastric glands that produce gastric secretions reduce in function with increasing age, contributing to the high occurrence of Helicobacter pylori infections in older adults (Huang et al., 2021). To compound the occurrence of H. pylori, this bacterium can interact with specific pathways responsible for producing gastric acid, therefore further reducing gastric acid secretion and causing an issue called hypochlorhydria which is a lack of stomach acid resulting in neutral pH of the gastric fluids (Marshall, 2001). Additionally, older adults are often prescribed proton pump inhibitors that reduce gastric acid secretion. In a cross-sectional study by Rodrigues et al. (2024) in 1200 older adults <65 years old, ∼38% of the older adults were taking proton pump inhibitors, despite 50% of them not having digestive-system related diseases (Rodrigues et al., 2024). Therefore, when examining the breakdown of protein within the older adult cohort, gastric alterations need to be thoroughly inspected and characterised.

In-vitro digestion has been studied extensively over the last number of years to develop methods that accurately replicate this human process with a very well accepted static simulated gastrointestinal digestion (SGID) model that replicates healthy adult digestion (Brodkorb et al., 2019) being widely used. When replicating digestive systems for other populations including infants and older adults, the pH of the digestive phases and the concentration of digestive enzyme used in each phase are the most common age specific alterations made to SGID protocols (Makran et al., 2022). Additionally for older adults, changes in chewing cycles, reduction in motility of the phases and increased phase duration is needed (Shani-Levi et al., 2017). Recent investigations have examined older adult digestive conditions on a range of food products including animal products of fish, meat, eggs, cheese and milk along with cereals (Hernández-Olivas et al., 2020a, Hernández-Olivas et al., 2020b, Hernández-Olivas et al., 2021a, Hernández-Olivas et al., 2021b, Hernández-Olivas et al., 2022). Trichloroacetic acid (TCA) soluble protein and free amino acids using gas chromatography/mass spectrometry methods were used to determine protein breakdown within these studies. Melchior et al. (2023) examined older adult digestive conditions against different sources of protein including whey, pea, wheat and rice proteins. Protein digestibility was determined by SDS-PAGE and soluble protein determination. Different proteolysis profile obtained between the adult and older adult in-vitro digestion models, specifically a reduction in protein digestion for the simulated older adult conditions (Melchior et al., 2023). Whey based products including whey-based desserts and skyr have also been examined in an adult versus older adult SGID method design (Dupont et al., 2023; Lavoisier et al., 2023). The o-phthalaldehyde (OPA) method to determine degree of hydrolysis (DH%), SDS-PAGE, liquid chromatography with tandem mass spectrometry (LC-MS/MS), and amino acids quantified by high-performance liquid chromatography (HPLC) were used to determine protein digestion. Results showed that proteolysis profiles differentiated in the gastric phase, with a reduction in protein digestion observed for the older adult model, but not the intestinal phase (Lavoisier et al., 2023). Other investigations have used liquid chromatography-mass spectrometry (LC-MS) to analyse peptides from milk (Aalaei et al., 2021a; Aalaei et al., 2021b) and from individual whey proteins including β-Lactoglobulin (β-Lg), α-Lactalbumin (α-Lac), lactoferrin (LF) and whey protein isolate (WPI) (Shani-Levi et al., 2017). Size-exclusion chromatography (SEC) has also been used when assessing the proteolysis of foods after in-vitro digestion (Rieder et al., 2021). While other studies have used a combination of LC-MS, DH% and SEC to characterise digestion (Zhao et al., 2024).

In previous SGID studies, changes of pH and digestive enzymes were made but each research group examined different specific values to model older adults. In particular for gastric pH, differences in values of pH include pH 4.5 (Melchior et al., 2023), pH 6 (Hernández-Olivas et al., 2021a; Hernández-Olivas et al., 2021b; Hernández-Olivas et al., 2022; Hernández-Olivas, Muñoz-Pina, Andrés, & Heredia, 2020; Hernández-Olivas, Muñoz-Pina, Sánchez-García, et al., 2020), or pH 3.7, based on the new healthy older adult method published by Menard et al. (2023) and recently used in experimental analysis by Dupont et al. (2023) and Lavoisier et al. (2023). Additionally, different concentrations of digestive enzyme have been used in these previous studies. Compared to the healthy adult model by Brodkorb et al. (2019), reductions in the gastric enzyme pepsin by 25% were used by Melchior et al. (2023) and in the studies by Hernández-Olivas et al., while a 40% reduction in pepsin has been advised in the new healthy older adult method published by Menard et al. (2023). Therefore, this highlights the need for each of these parameters to be fully characterised to determine the effect of a range of older adult conditions that may present within the ageing population, either due to health status or medication.

Therefore, the aims of this study were twofold. Firstly, to study a range of pH and enzyme concentrations kinetically within the gastric phase to determine if pH and or enzyme concentration have significant effects on protein digestion within this phase. These results helped in the determination of suitable SGID conditions to replicate a range of older adult gastrointestinal environments to add to the recently published healthy adult model by Menard et al. (2023). Then, these determined SGID conditions were used to show how whey proteins are broken down within different simulated older adult digestive conditions, compared to a healthy adult model. Both gastric and intestinal phases were examined to determine potential nutritional differences due to SGID condition as differences between digestive phases have been recently reported (Lavoisier et al., 2023).

In this study, determination of WPI digestion was examined using several techniques. The physical method of ultra-filtration (UF) was used to determine the proportion of small peptides (<10 kDa in MW) as previous research groups have characterised this fraction of soluble peptides as absorbed by the body (Le Roux et al., 2020; Rieder et al., 2021). Additionally, DH% by OPA was measured. Chromatographic techniques of SEC and LC-MS were used in a quantitative manner as they have previously been shown to analyse milk protein hydrolysis (Zhao et al., 2024). Together, these techniques did allow for a full characterisation both qualitative and quantitative of the various WPI digestas formed during in-vitro digestion.

Materials and methods

2

Materials

2.1

Whey Protein Isolate (WPI) with a 95% protein content was used, provided by Fonterra Cooperative Group Limited (Auckland, New Zealand). The digestive enzymes used was of porcine origin, pepsin (EC 3.4.23.1), trypsin (EC 3.4.21.4), and chymotrypsin (EC 3.4.21.1) obtained from Sigma-Aldrich (Ireland), supplied as a powder and stored at −20 °C. The salts (KCl, KH_2_PO_4_, NaHCO_3_, NaCl, MgCl_2_(H_2_O)6, (NH_4_)2_CO_3, CaCl_2_(H_2_O)2,) used in the simulated digestive fluids were of reagent grade and obtained from Sigma-Aldrich (Ireland), along with HCl and NaOH. Simulated salivary fluid (SSF), simulated gastric fluid (SGF) and simulated intestinal fluid (SIF) were prepared prior to SGID as stock solutions according to Brodkorb et al. (2019). 1 M HCl and 1 M NaOH for pH adjustment during SGID was also prepared prior to analysis. Enzymatic solutions were prepared on the day of each experiment according to Brodkorb et al. (2019) and Menard et al. (2023).

Reagents used in the degree of hydrolysis analysis by OPA method of Sodium tetraborate decahydrate, sodium dodecyl sulfate (SDS), OPA, ethanol, dithiothreitol 99% and l-Serine were of general reagent grade and obtained from Sigma-Aldrich. The OPA reagent was prepared as per Nielsen et al. (2001) on the day of analysis. SEC reagents of water, Acetonitrile and trifluoroacetic acid (TFA) of HPLC or LC-MS grade were obtained from Sigma-Aldrich (Ireland). Standards of bovine serum albumin (BSA), β-Lg, α-Lac, aprotinin, insulin β chain oxidised, β-Lg peptide and tryptophan were of general-purpose grade and prepared at relative concentrations in HPLC water prior to analysis, stored in the freezer at −20 °C as stock solutions and prepared in SEC mixes on the day of analysis. Mobile phase was prepared on each day of analysis.

In-vitro digestion of WPI

2.2

WPI was digested using the in-vitro SGID method based on Brodkorb et al. (2019) and Menard et al. (2023) with modifications. As the study was designed to examine the digestion of a sample containing ≥95% protein, protease enzymes were used exclusively with no lipase or bile salts (Minekus et al., 2014). WPI was prepared at 100 g/L in deionised water. SSF was added, and the 50 g/L protein mixture incubated for 2 min at 37 °C in a shaking water bath at 140 movements/min, to simulate the oral phase. For the gastric phase, SGF was added to reduce the WPI concentration to 25 g/L, and the pH reduced to the relevant pH being examined (pH 3, 3.7, 4.5 or 6) using 1 M HCl. Pepsin was added at the relevant concentration for Adult and Older Adult conditions as per Brodkorb et al. (2019) and Menard et al. (2023), and the samples incubated at 37 °C at 140 movements/min for 3 h. Aliquots were removed after 30, 60, 120 and 180 min of digestion, and the pH increased to pH 8 using 1 M NaOH to terminate the gastric digestion of these aliquots (Brodkorb et al., 2019).

Gastric phase development

2.2.1

The enzyme preparations were prepared on the day of digestion according to Brodkorb et al. (2019) and Menard et al. (2023). Pepsin at either 2000 U/mL final digesta volume as per Brodkorb et al. (2019) or 1200 U/mL final digesta volume as per Menard et al. (2023) were used to investigate the effect of enzyme concentration within each gastric digestion protocol. Based on the SGID protocols outlined by Brodkorb et al. (2019) and Menard et al. (2023) and after examining gastric pH values used to replicate older adult digestion in literature, the following gastric pH values were examined: pH 3.0 (Brodkorb et al., 2019), pH 3.7 (Menard et al., 2023), pH 4.5 (Melchior et al., 2023) and pH 6.0 (Hernández-Olivas et al., 2022), each at 2000 U/mL and 1200 U/mL pepsin. To examine the relative effect of time, pH and enzyme concentration for each of the digestion models, two methods of determining WPI digestion were determined: bio-accessible peptides using UF as well as DH%.

Intestinal phase

2.2.2

The intestinal phase was examined as per Brodkorb et al. (2019) for the Adult model, and as per Menard et al. (2023) for all Older Adult models. SIF was added and the pH adjusted to pH 7 using a recorded volume of 1 M NaOH. Trypsin at 100 U/mL and chymotrypsin at 25 U/mL were for the Adult model (Brodkorb et al., 2019) whereas these enzyme concentrations were reduced by 20% as per Menard et al. (2023) for all Older Adult conditions. The final protein concentration in these samples was 12.5 g/L. Samples were incubated at 37 °C in a shaking water bath at 140 movements/min for two hours.

All simulated digestions were completed in triplicate for each of the relevant conditions and completed on separate days. All digesta were stored at −20 °C until use.

Measuring protein digestion

2.3

Firstly, UF using 10 kDa molecular weight cut off (MWCO) filter units (Amicon® Ultra-15 Centrifugal Filters) were used to determine the % bio-accessible peptides within WPI digesta, according to research determining that the fraction of soluble peptides that may be absorbed by the body is below 10 kDa in molecular weight (Le Roux et al., 2020; Rieder et al., 2021). The filtrates from the WPI digesta were analysed using UV Spectroscopy to determine the quantity of peptides <10 kDa produced during SGID. Secondly, the OPA method as described by Nielsen et al. (2001) was used to determine DH% of WPI digesta. Briefly, digesta samples at 1 g/L were mixed with OPA reagent and stood for 2 min before spectrophotometer readings were performed at 340 nm.

Size exclusion chromatography

2.3.1

SEC was carried out on an Agilent 1200 HPLC system (Agilent Technologies, Palo Alto, CA, U.S.A.) using a Yarra 3 μm SEC-2000 (300 × 7.8 mm, Phenomenex) column. The mobile phase used was: 55:45:0.1 LC-MS grade Water:Acetonitrile:TFA % w/w in an isocratic elution. The separation was performed at 40 °C with a flow rate of 0.7 mL/min. Samples were prepared at 1 g/L in water and an injection volume of 40 μL was used. UV detection was carried out at 214 nm.

Mass modelling and MW fractionation

2.3.1.1

To create the Molecular Mass Calibration Curve analysis, the following standards were used: Bovine serum albumin (BSA, 66,460 Da) β-Lactoglobulin (β-Lg, 18,363 Da), α-Lactalbumin (α-Lac, 14,175 Da), Aprotinin (6512 Da), Insulin β chain oxidised (3496 Da), β-Lg peptide (1584 Da) and Tryptophan (204 Da). These were chosen as some proteins such as β-Lg and α-Lac are found within whey protein and therefore can be used for identifying specific peaks as well as providing a range of MW. Otherwise, the range of compounds were chosen to give a broad range of MW to allow for a good standard curve to be obtained. Firstly, the average retention time for each standard was calculated (t_Avg_, n = 8). A third-degree polynomial extension for the relation between log (MW) and elution time was then used to estimate molecular weight, which has used in other experimental analysis for a range of polymers from polystyrene (Xu et al., 2018) to peptides (Kristoffersen et al., 2020).

Molecular weight distribution

2.3.1.2

Based on McKeen (2012), number average molecular weight (Mn), weight average molecular weight (Mw) and the polydispersity index (PDI), also known as the molar-mass dispersity index, were used to determine the distribution of whole proteins and peptides from the relevant SGID digesta to gain information around the molecular weight (MW) distribution.

Number average molecular weight

2.3.1.2.1

The number average molecular weight (M_n_) is the arithmetic mean of the molecular weight of the individual peptides in the hydrolysate and can be calculated according to the equation below:

$$[eqn]$$

where i is the with peptide with weight Mi represented by a concentration proportional to its absorbance Abs_i_ recorded absorbance at UV 214 nm (McKeen, 2012).

Weight average molecular weight

2.3.1.2.2

The weight average molecular weight (Mw) is a measure that includes the mass of individual peptides within the hydrolysate which contributes to the overall weight of the protein hydrolysate and can be calculated according to the equation below:

$$[eqn]$$

Polydispersity index

2.3.1.2.3

The polydispersity index (PDI), also known as the molar-mass dispersity index, is the ratio of Weight average molecular weight to Number average molecular weight (M_w_/M_n_) (McKeen, 2012).

$$[eqn]$$

This value represents the distribution of size within the samples and gives a measure of broadness of molecular weight distribution. Samples with a low PDI will have a narrow molecular weight distribution, while polydisperse hydrolysates display an array of peptides chain lengths which broaden the molecular weight distribution and characterised with a large PDI (Gavrilov & Monteiro, 2015).

Liquid chromatography -mass spectrometry (LC-MS)

2.3.2

To establish if differences due to gastrointestinal conditions on WPI were present, intestinal digesta were analysed using mass-spectrometry. Peptides (500-1000 ng) were loaded onto Evotips as per manufacturer's instructions (EvoSep, Odense, Denmark). The samples were analysed by the Mass Spectrometry Resource (MSR) in University College Dublin on a Bruker TimsTOF Pro mass spectrometer connected to a Evosep One chromatography system. Peptides were separated on an 8 cm analytical C18 column (Evosep, 3 μm beads, 100 μm ID) using the pre-set 33 samples per day gradient on the Evosep one.

The Bruker TimsTOF Pro mass spectrometer was operated in positive ion polarity with TIMS (Trapped Ion Mobility Spectrometry) and PASEF (Parallel Accumulation Serial Fragmentation) modes enabled as previously described (Bollard et al., 2024). The accumulation and ramp times for the TIMS were both set to 100 ms., with an ion mobility (1/k0) range from 0.62 to 1.46 Vs/cm. Spectra were recorded in the mass range from 100 to 1700 m/z. The precursor (MS) Intensity Threshold was set to 2500 and the precursor Target Intensity set to 20,000. Each PASEF cycle consisted of one MS ramp for precursor detection followed by 10 PASEF MS/MS ramps, with a total cycle time of 1.16 s.

Data processing and bioinformatics

2.3.2.1

Bruker mass spectrometric data from the TIMS TOF was processed using the MaxQuant (version 2.4.2.0) incorporating the Andromeda search engine (Tyanova et al., 2016). To identify peptides and proteins, MS/MS spectra were matched against a custom database of 97 proteins from bos taurus. All searches were performed using the default setting of MaxQuant, using the unspecific enzyme setting and a false discovery rate of 1% on the peptide and protein level. The database searches were performed with acetylation (protein N terminus) and oxidation (M) as variable modifications. For the generation of label free quantitative (LFQ) ion intensities for protein profiles, signals of corresponding peptides in different nano-HPLC MS/MS runs were matched by MaxQuant in a maximum time window of 1 min. The data was log transformed, and an ANOVA comparison of intestinal digesta was carried out. For the generation of visualizations, missing values were imputed with values from a normal distribution, and the dataset was normalized by z-score. Volcano plots and Principal component analysis (PCA) were carried out for visualization of global peptide signatures between the four SGID. Heat maps were employed for visualization of peptide signatures of significantly different peptides. These were determined using ANOVA analysis with Benjamini-Hochberg FDR, with an FDR of 0.05 used as a post hoc test across the four SGID conditions groups to examine systematic trends with the data.

Statistical analysis

2.4

In-vitro digestion at each pH and enzyme level (n = 8) was carried out in triplicate on different days (n = 3). The protein breakdown analysis methods were carried out on each of the sample triplicates. Statistical analysis was carried out using IBM SPSS 27 Statistics software. Three-Way analysis of variance (ANOVA), with Repeated-Measure analysis was completed to evaluate the effect of Time, pH and Enzyme concentration. One-Way ANOVA was completed where appropriate. Tukey was used as a post hoc test for all ANOVA completed. Values of p < 0.05 were significant, and values of p < 0.001 were very significant. Results are expressed as mean ± standard deviation (SD) of n = 3 unless stated otherwise.

Results and discussion

3

Gastric phase

3.1

Gastric phase development

3.1.1

Gastric digestion of WPI within SGID was examined over a 3-h period, under 4 different pH values (pH 3.0, pH 3, pH 4.5, pH 6.0) and two different pepsin enzyme concentrations of 2000 U/mL, as shown in Fig. 1. This range of time, pH and enzyme concentrations reflects published protocols for adult digestion (pH 3, 120 min, 2000 U/mL as per Brodkorb et al., 2019) and older adult SGID (pH 3.7, 180 min, 1200 U/mL as per Menard et al., 2023). This was to determine the impact of pH as well as pepsin enzyme concentration on the hydrolysis profiles of gastric digested WPI. A Three-Way ANOVA was conducted using Time, pH and Enzyme as main effects, as well as their interactions.Fig. 1. Mean Plot of % bio-accessible peptides from whey protein isolate (WPI), digested in a simulated gastric phase for 3 h, under 4 different pH values of pH 3, pH 3.7 pH 4.5 and pH 6, and two pepsin concentration values of 2000 U/mL and 1200 U/mL. n = 3 with error bars determining ± standard deviation.Fig. 1

From Fig. 1, it is clear that time had a significant influence on % bio-accessible peptides, with increasing digestion time resulting in increased % peptides at all pH values except at pH 6 where very low bio-accessible peptides values were obtained throughout the 180 min (2.4 ± 1% for all time points regardless of Enzyme concentration). pH is also shown to have an important impact on the digestion of whey proteins into bio-accessible peptides, with large and significant (p < 0.001) differences seen, indicating that pepsin's ability to hydrolyse whey protein is highly dependent upon gastric pH.

Significant differences at p < 0.05 between pepsin concentrations at 2000 U/mL and 1200 U/mL within the same pH were observed, where the higher pepsin concentration resulted in higher % bio-accessible peptides, except at pH 6. Results also showed an interaction between Time and pH, where the longer durations and the lower the pH of gastric digestion resulted in the more % bio-accessible peptides, again with the exception of pH 6 where time was not a relevant factor.

The OPA derived DH% results of gastric digestion were examined over a 3-h period, under the same 4 different pH values and two different pepsin enzyme concentrations of 2000 U/mL and 1200 U/mL. A Three-Way ANOVA was conducted using Time, pH and Enzyme as main effects, as well as their interactions. These results showed an interaction between Time and pH. For pH 3, the results follow the trend above seen in Fig. 1 for % bio accessible peptides where the longer the duration, the higher the DH%. Although this trend was not clearly seen at any other pH value examined. No significant differences between pepsin concentrations at 2000 U/mL and 1200 U/mL within the same pH were observed as DH% remained relatively stable within each pH.

Taking the results of both bio-accessible peptides and DH% into consideration, pH was found to be the most relevant factor with large and highly statistically significant changes (p < 0.001) in whey protein digestibility seen for all four conditions studied across two different methods of determining digestion, regardless of enzyme concentrations, justifying studying a wide pH range to better represent the diversity of gastric conditions seen in the elderly population. Time was the second most significant factor, where increases from 120 to 180 min digestion showed increases in both DH% and bio-accessible peptides for all conditions at p < 0.001, except pH 6, justifying the use of 180 min duration for the gastric digestion step in older adult conditions (pH > 3). Enzyme concentration was found to result in significant different results for % bio-accessible peptides at a 5% significance level only, so that the lower enzyme concentration alone can justifiably be used to represent the gastric conditions of the elderly population.

Using the information obtained from the examination of gastric pH and enzyme conditions to simulate different older adult models, and after information from literature examination, Table 1 summarises the range of SGID conditions used within the following experimental analysis.Table 1SGID conditions for whey protein isolate (WPI) detailing the Oral, Gastric and Intestinal phases for four different digestion models to be examined, replicating different populations of Adult, Healthy Older Adult (Healthy OA), Older Adult 2 (OA2) and Older Adult 3 (OA3). SSF (simulated saliva fluid), SGF (simulated gastric fluid), SIF (simulated intestinal fluid).Table 1SGID ModelSGID PhaseParameterAdultHealthy OAOA2OA3OralSSF: Sample1:11:11:11:1Phase duration (mins)2222GastricSGF: SSF digesta1:11:11:11:1pH33.74.56Pepsin (U/mL)2000120012001200Phase duration (mins)120180180180IntestinalSIF: SGF digesta1:11:11:11:1pH7777Trypsin (U/mL)100808080Chymotrypsin (mg/mL)25202020Phase duration (mins)120120120120

pH values of pH 3 and 3.7 were used to simulate Adult and Healthy Older Adult (Healthy OA) as described by the published SGID methods of healthy Adults (Brodkorb et al., 2019) and Healthy OA (Menard et al., 2023). pH values of 4.5 and 6 were used to simulate Older Adult 2 (OA2), and Older Adult 3 (OA3) respectively. These higher pH values were used to mimic older adults who may not be in optimum health, for example who may be suffering from H. pylori infection, a common infection in older adults (Pilotto & Franceschi, 2014) known to reduce gastric acid secretion (Saha et al., 2010). They were also used to simulate those on prescription medications including proton pump inhibitors which are a common medication among the elderly (Rodrigues et al., 2024).

Protein digestion

3.2

UF and DH%

3.2.1

To determine the quantity of the bio-accessible peptides from each sample, UF methodology using 10 kDa MWCO filters was implemented. pH very significantly affected the quantity of bio-accessible peptides produced during the SGID, with an inverse relationship being observed between pH and bio-accessible peptide %, shown in Table 2 The SGID conditions for Adult, Healthy OA, OA2 and OA3, with gastric pH values of pH 3, pH 3.7, pH 4.5 and pH 6 respectively, resulted in higher quantity of absorbable peptides (p < 0.001). Values ranged from 23.52% ± 0.50 to 2.89% ± 0.72. Therefore, we can see that gastric pH had a very strong effect on the peptides generated from WPI, which may affect the extent of peptides that may be absorbed by the body.Table 2. Bio-accessible peptides and Degree of Hydrolysis (DH%) for final gastric digesta of WPI under replicating Adult, Healthy Older Adult (Healthy OA), Older Adult 2 (OA2) and Older Adult 3 (OA3) simulated gastrointestinal digestion (SGID) conditions, n = 3 ± standard deviation.Table 2Gastric SGID Condition****% Bio-accessible Peptide****% Degree of HydrolysisAdult23.52± 0.50 ^a^7.91± 0.37 ^x^Healthy Older Adult13.53± 1.36 ^b^6.11± 0.19 ^y^Older Adult 27.58± 1.27 ^c^5.40± 0.31 ^yz^Older Adult 32.89± 0.72 ^d^5.65± 1.68 ^z^Letter indices denote significant difference by ANOVA, with ^a-d^ at p < 0.001 and ^x-z^ at p < 0.05.

The magnitude of change seen in DH% results varied between 5% and 8% as shown in Table 2 Variations within experimental set up, especially in the case of OA3 DH% results, often exceeded variations between the other experimental set ups, resulting in a high variance and therefore high standard deviation within one sample treatment. DH% similarly showed a trend of a pH effect, with lower pH resulting in higher DH%, but this is not clear between OA2 and OA3 due to the large standard deviation obtained for OA3 samples.

As both % bio-accessible peptides and DH% results showed statistical differences between SGID conditions for WPI gastric digesta, further in-depth analysis was completed using SEC to determine quantitative differences between the digested WPI.

Size exclusion chromatography (SEC)

3.2.2

To quantify the digestibility of the WPI under the different SGID parameters, the size exclusion chromatograms were classified into different molecular weight zones: >30 kDa, 30<kDa>10, 10<kDa>5, 5 < kDa>1, and < 1 kDa. This classification allowed for the quantification of large undigested proteins and protein aggregates (>30 kDa), individual whey proteins (10-30 kDa), as well as large (10-5 kDa), medium (5-1 kDa) and small (<1 kDa) peptides.

Fig. 2 shows the percentage of each MW fraction within the Adult, Healthy OA, OA2 and OA3 gastric digesta samples. The MW distribution of all gastric digesta samples predominantly consisted of large aggregates and intact proteins (>10 kDa) as a minimum of 70%, indicating that undigested whey proteins are still present after gastric digestion regardless of simulated gastric condition used. This may call into question the notion of whey proteins being ‘fast proteins’ compared to caseins which are commonly denoted as ‘slower proteins’ (Boirie et al., 1997; Dangin et al., 2002). But this agrees with the fact that β-Lg is known to be gastric resistant (Sutantawong et al., 2025) and that the gastric pH of older adult conditions is close to the known isoelectric point of the whey proteins that could explain some of the aggregates (>30 kDa) seen in these samples.Fig. 2. Molecular weight distribution for whey protein isolate (WPI) after Adult, Healthy Older Adult, Older Adult 2 and Older Adult 3 after simulated gastric digestive conditions calculated from SEC data. n = 3.Fig. 2

Examining the MW distribution in more detail, each gastric digesta sample had a minimum of 15% of proteins >30 kDa, with OA2 and OA3 having the largest amount of this fraction. Adult and Healthy OA had a minimum of 20% of bio-accessible peptides (<10 kDa), with a large percentage between 5 kDa and 1 kDa being present within the Adult digesta. The OA3 conditions resulted in less than 10% of bio-accessible peptides indicating very limited hydrolysis. Therefore, these results further indicate that Adult and Healthy OA gastric conditions are more favourable to WPI breakdown as observed by the increased in bio-accessible peptides, compared to OA2 and OA3 gastric conditions. Roberts et al. (1999) showed that polypeptides from 3 to 51 amino acids in chain length were absorbed in physiologically significant quantities in an animal study, with the potency of the bioactivity of the peptides decreasing with increased chain length. Therefore, SGID conditions that result in increased quantity of small peptides, in the case of Adult and Healthy OA conditions observed for WPI, may have more chance of containing more bioactive peptides that the OA2 and OA3 digesta as these had lower amounts of small peptides.

Using the size exclusion chromatography information, M_n,_ M_w_ and PDI can be calculated according to eqs. (1), (2), (3). As shown in Table 3, the M_w_ or weight average molecular weight values for each SGID condition were relatively similar to one another and showed no statistical significance nor trend between digestion samples. In contrast, M_n_ or number average molecular weight showed some variations between SGID conditions with higher values for OA2 and OA3 gastric samples compared to Adult and Healthy OA samples, although statistical significance was not met.Table 3. Number average molecular weight (Mn), Weight average molecular weight (Mw) and Polydispersity Index (PDI) (PDI = Mn/Mw) as determined by Size Exclusion Chromatography (SEC) for gastric whey protein isolate (WPI) samples simulated gastric digestion (SGID) under Adult, Healthy Older Adult (Healthy OA), Older Adult 2 (OA2) and Older Adult 3 (OA3).Table 3Simulated Gastric Condition****MnMwPDIAdult15,948^a^26,973 ^a^1.69 ^a^Healthy Older Adult18,136 ^a^27,668 ^a^1.52 ^ab^Older Adult 220,104 ^a^30,259 ^a^1.50 ^b^Older Adult 320,048 ^a^28,809 ^a^1.43 ^b^^a-b^ Mean values with different letter indices indicate significant difference at p < 0.05.

The Polydispersity index indicates the broadness of molecular weight distribution with higher values indicating broader distributions. All gastric digesta showed a narrowly disperse distribution with significant difference only seen for the adult gastric digesta, where a significantly higher value was observed compared to OA2 and OA3 conditions. These results agree with the molecular weight distributions shown in Fig. 2 where the vast majority of peptide sizes seen in all samples consisted of large aggregates and intact proteins with little contribution from peptides of size 10 kDa and lower.

Intestinal phase

3.3

UF and DH%

3.3.1

WPI samples were further digested in a simulated intestinal phase as per the relevant SGID model described in Table 1. Results for bio-accessible peptide % and DH% post intestinal digestion are displayed in Table 4.Table 4% Bio-accessible peptides and % Degree of Hydrolysis (DH%) for final intestinal digesta of WPI under Adult, Healthy Older Adult (Healthy OA), Older Adult 2 (OA2) and Older Adult 3 (OA3) simulated gastrointestinal digestive (SGID) conditions, n = 3 ± standard deviation.Table 4**Intestinal SGID Condition****Bio-accessible Peptide %****Degree of Hydrolysis %**Adult76.86± 1.98 ^ab^18.00± 0.40 ^xy^Healthy Older Adult73.86± 2.29 ^b^16.61± 0.65 ^y^Older Adult 269.55± 1.11 ^bc^15.38± 0.06 ^yz^Older Adult 364.41± 1.03 ^d^15.70± 0.31 ^z^Letter indices denote significant difference by ANOVA, with ^a-d^ and ^x-z^ at p < 0.05.

There were large increases in the % bio-accessible peptides and DH% after the intestinal phase compared to the gastric phase, showing the action of the intestinal enzymes to breakdown and digest WPI in-vitro. Trends for both bio-accessible peptides and DH% observed in the gastric phase were still seen in the intestinal phase, with the gastric conditions (OA3) resulting in less digested WPI maintaining somewhat limited digestion after the intestinal phase. This shows that prior gastric conditions can potentially impact further gastrointestinal digestion. Therefore, further investigation into the discrete differences between these digesta samples was carried out using SEC to determine to what extent the intestinal phase could improve on the altered gastric phase preceding it.

Size exclusion chromatography (SEC)

3.3.2

The intestinal digesta were analysed using SEC and the chromatograms were classified into different molecular weight zones. This allowed for the quantification into large undigested proteins and protein aggregates (>30 kDa), individual whey proteins (10-30 kDa), large (10-5 kDa), medium (5-1 kDa) and small (<1 kDa) peptides. Fig. 3 shows the percentage of each MW fraction within the Adult, Healthy OA, OA2 and OA3 final intestinal digesta samples.Fig. 3. Molecular weight distribution for whey protein isolate (WPI) after Adult, Healthy Older Adult, Older Adult 2 and Older Adult 3 intestinal conditions calculated from size exclusion chromatography (SEC) data. n = 3.Fig. 3

All WPI intestinal digesta samples were predominantly made up of medium and small peptides (<5 kDa) for all SGID conditions with 84.04% - 91.75% recorded. This contrasts with the gastric digesta where minimal amount of medium and small peptides was obtained (Fig. 2).

Although some undigested proteins and protein aggregates (>10 kDa) remain in the older adult conditions, their proportion is small (below 10%) with only OA2 and OA3 conditions still showing any aggregates over 30 kDa. As the intestinal molecular weight distribution data have more similarities to one another than the gastric MW data, this indicates that the WPI digesta for OA2 and OA3 may ‘catch up’ after the intestinal phase, as differences between MW fractions are not as extensive.

Examining the calculated molecular weight averages shown in Table 5, the values for both M_w_ and M_n_ are much lower than those observed within the gastric phase, indicating a high degree of hydrolysis after exposure to the intestinal phase. The trend of high M_n_ value for OA2 and OA3 observed for the gastric digesta was similarly observed in the intestinal digesta. M_n_ values for OA2 and OA3 were significantly different from one another and significantly different from Adult and Healthy OA conditions at p < 0.05. M_w_ value was highest for OA2 and OA3, with these values being significantly different from one another and significantly different from Adult and Healthy OA at p < 0.05.Table 5. Number average molecular weight (M_n_), Weight average molecular weight (M_w_) and Polydispersity Index (PDI) (PDI = M_n_/M_w_) as determined by Size Exclusion Chromatography (SEC) Mass modelling for intestinal whey protein isolate (WPI) samples after simulated gastrointestinal digestion (SGID), n = 3 ± standard deviation.Table 5Intestinal SGID Condition****MnMwPDIAdult2618 ^c^10,947 ^c^4.14 ^b^Healthy Older Adult2590 ^c^10,612 ^c^4.10 ^b^Older Adult 23150 ^b^15,098 ^b^4.79 ^b^Older Adult 34087 ^a^23,207 ^a^5.68 ^a^^a-c^ Mean values with different letter indices indicate significant difference between SGID conditions at p < 0.05.

In contrast to the gastric phase, the PDI of all digestion conditions were high indicating polydisperse digesta. OA3 had the highest PDI value which was significantly different from all other SGID conditions, showing that this digesta consisted of a very broad distribution of different sized peptides, with whole proteins and possibly protein aggregates still present within this sample. All other PDI values were lower, indicating less broadness of molecular weight distribution within the Adult, Healthy OA and OA2 digestas which contain fewer whole proteins and were more dominant in small peptides.

LC-MS

3.3.3

LC-MS analysis of the intestinal digesta was completed to determine at a peptide sequence level if individual peptides in these digesta were different from one another. A typical MS chromatogram is shown in Fig. S1.

A total of 1216 peptides were identified over the twelve samples (triplicates of the four SGID conditions) with an average LFQ intensity (Label-Free Quantification intensity) of 3 × 10^9^.

It should be noted that samples were analysed for peptides >2 amino acids (AA) although only peptide with at least 5 AAs were identified. Of the 1216 peptides identified, 2.1% were 5 AA in length, 41.5% were between 6 and 10 AAs in length, and 56.4% were above 11 AAs in length. Therefore, there may be an under-representation of small peptides within the Adult digesta via LC-MS as the data obtained from the SEC analysis indicated that there are small peptides less than 5AA long present within the Adult intestinal digesta.

Automated data cleaning was performed to eliminate peptides seen in only 3 or fewer of the 12 replicates which led to 354 peptides kept for further analysis representing 96% of the initial LFQ Intensity. This drastic reduction in peptide number only affected low occurrence peptides as can be visually seen from Fig. S2 where the cumulative LFQ intensity is plotted as a function of peptide numbers. It is evident from this figure that the vast majority of the initial 1216 identified peptides are only in negligible amount in the digesta, thus justifying the clean-up operation.

Analysis of the protein origin of these peptides showed that more than 99% of Label-Free Quantification (LFQ) Intensity originates from bovine milk proteins, with between 82 and 89% of the LFQ coming from β-Lg, with α-Lac accounting for between 3.5 and 7.5% depending on SGID conditions and BSA at around 1%. The rest of the peptides originated mainly from the casein proteins. These figures are in agreement with the substrate used (whey protein isolate) and reflect the fact that while β-Lg is gastric digestion resistant, this gastric resistance may lead to a relatively high abundance of β-Lactoglobulin peptides in intestinal digesta. On the other hand, the high digestibility of α-Lac in the gastric environment may have resulted in single amino acids and very small peptides following intestinal digestion, that could explain the under-representation of this protein in the intestinal digesta samples.

Multi-scatter plots (Fig. S3) and volcano plots (Figs. S4) show good correlation within replicates and some differences between SGID samples, with marked differences between the adult and all older adult conditions and small differences between OA2 and OA3 samples.

Principal component analysis (PCA) (Fig. S5) was employed for visualization of global peptide signatures across the four SGID conditions and to examine systematic trends with the data (Jacquier et al., 2024). The two principal components identified accounted for 46.3% and 15.1% of the differences in the data. As can be seen on Fig. S5, good clustering is quite apparent for the Adult and Healthy OA SGID triplicate samples, although the OA2 and OA3 samples clustered poorly. This PCA analysis also shows a good separation of the Adult from the Healthy OA SGID triplicate samples and from both the OA2 and OA3 samples which are clustered together.

SGID significant peptides

3.3.3.1

Fig. 4 is a heatmap denoting the peptides determined within the intestinal samples of significance between the four SGID conditions, shown for the three replicates per condition. Peptides of significant difference between SGID conditions were determined using p-value and permutation-based FDR analysis at p < 0.05 (Jacquier et al., 2024). The heat map shows variation between peptides that are SGID condition specific, with red indicating higher occurrence of the peptide and green absence of the peptide. It is evident from the figure that differences between samples is greater than replication differences within samples, thus allowing for relevant differences due to SGID conditions to be discussed. Nevertheless, only small differences are seen between the Adult and Healthy OA peptides with both SGID conditions resulting in some significant peptides in common as seen in the top left-hand corner of Fig. 4, while these peptides are absent in OA2 and OA3 SGID conditions.Fig. 4. Heat Map of significantly different peptides from whey protein isolate (WPI) after undergoing simulated gastrointestinal digestion (SGID), determined using liquid chromatography-mass spectrometry (LC-MS) and data analysed using Perseus, with p-value and Permutation-Based FDR analysis to isolate significant peptides.Fig. 4

Also, the OA2 and OA3 SGID conditions appear quite similar to each other in terms of significant peptides both showing large similar regions of clustered significant peptides in Fig. 4. These results highlight the difference of SGID conditions on the digestion of WPI, particularly how Adult conditions vary from Healthy OA conditions as well as from OA2 and OA3. These results bring highly novel insights into the need to define SGID conditions that reflect various elderly populations, in particular how changes in gastric conditions have repercussions on peptide profiles of intestinal digesta, both in terms degree of hydrolysis, in terms of molecular weight and down to the peptide sequence level. In particular, the INFOGEST consensus (Menard et al., 2023) conditions lead to similar peptides profiles following full SGID digestion of whey proteins while the older adult conditions defined in this work lead to poorer digestion.

To better understand the mechanism of peptide generation, the peptide profiles coming from the three main whey proteins (β-Lg, α-Lac and BSA) were visualized as shown in Fig. 5, where the height of the graph represents the relative abundance of peptides overlapping at a given position along the protein sequence. The sharp increase and decrease in relative abundance seen at specific amino acid position shows that peptide generation during SGID conditions occur at specific cleavage sites. Remarkably, while differences are seen for the three proteins in relative occurrence, there seems to be preservation of cleavage sites for all four SGID conditions studied. For example, in the case of BSA, peptides come from three main sequence zones only, 1–21, 103–114 and 413–422. Little or no peptides are seen from the other sites along the protein sequence. For α-Lac, the peptides generated during digestion extend over most of the protein sequence with peptides covering positions 19–26, 32–50, 53–60, 79–90 and 94–104. In the case of β-Lg, the sequence coverage is near total with peptides seen in positions 1–19, 20–39, 41–59, 76–100, 123–138 and 149–156. The 41–59 and 123–138 zones of β-Lg were also seen to dominate the peptides issued from pasteurised milk digestion (Li et al., 2024).Fig. 5. Sequence profile of peptides on the three main whey proteins following full gastro-intestinal digestion in adults (—) heathy older adults (—), OA2 (—) and OA3 (—) SGID conditions.Fig. 5

26 of the most noteworthy peptides (by relative abundance) are listed in Table S1. It is clear that the P1 cleavage sites of pepsin (at Phenylalanine and Leucine residues) and Trypsin (at Lysine and Arginine residues) give rise to the most dominant peptides for all three proteins. Nevertheless, it is also notable that the relative abundance of some peptides with a Leucine or a Phenylalanine in position P1 is seen to significantly reduce (e.g. KHLVDEPQNL of BSA 377–386, GGVSLPEW of α-Lac 19–26 and TKIPAVF of β-Lg 76–82) in ageing digestive conditions OA2 and OA3 reflecting less peptic proteolytic activity in these SGID conditions. On the opposite, other peptides are seen to become more abundant in these ageing conditions such as (FKDLGEEHFK of BSA 11–20, ILDKVGINY of α-Lac 79–90 and TPEVDDEALEKFDK of β-Lg 125–138, characterised by an Arginine or a Lysine residue in P1 and all containing within their sequence a Phenylalanine and or a Leucine, once again indicating poorer peptic proteolytic activity in these SGID conditions.

Health effects and peptide bioactivity

3.3.4

Of the 354 relevant peptides identified within each SGID condition, a minimum of 87% contained at least one branched chain amino acid of leucine, valine or isoleucine. Branched chain amino acids are important in ageing as they play a role in protein synthesis and therefore influence sarcopenia (Le Couteur et al., 2020). Some studies have demonstrated that whey protein supplementation with added leucine improved the bone health of older adults with sarcopenia (Hill et al., 2019) showing the benefit of BCAAs in older adults. Our results indicate that despite the digestion profile differences for WPI due to different gastrointestinal conditions, the peptides produced are of nutritional benefit especially to older adults as all intestinal digesta contained BCAAs.

To identify if the peptides observed within the intestinal samples had any previously recorded bioactivity associated with them, the Milk Bioactive Protein Database (MBPDB) by Nielsen et al. (2024) was consulted. This MBPDB has been used by other researchers including Masotti et al. (2024) who determined specific peptides heat denatured WPI and determined bioactive peptides with ACE-inhibitory, antioxidant, DPP-IV-inhibitory. Of the 354 relevant peptides identified within WPI intestinal digesta, a total of 17 peptides were previously associated with bioactivity including antimicrobial activity, DPP-IV Inhibition, ACE-Inhibition, cholesterol management and antioxidant activity (Appendix A, Demers-Mathieu et al., 2013, Hirose, Kurimoto, Yuda and Tanaka, 2024, Jacquot, Gauthier, Drouin and Boutin, 2010, Lacroix, Chen, Kitts and Li-Chan, 2017, Lacroix, Meng, Cheung and Li-Chan, 2016, Li et al., 2022, Maes et al., 2004, Mullally, Meisel and Fitzgerald, 1997, Nagaoka et al., 2001, Pellegrini, Dettling, Thomas and Hunziker, 2001, Pihlanto-Leppälä, 2000, Power, Nongonierma, Jakeman and Fitzgerald, 2014, Sedaghati, Ezzatpanah, Mashhadiakbar Boojar, Tajabadi Ebrahimi and Aminafshar, 2015, Silveira, Martínez-Maqueda, Recio and Hernández-Ledesma, 2013, Tavares et al., 2011, Tavares and Xavier Malcata, 2012). 14 of the peptides were issued from β-Lg and the remaining 3 were from α-Lac. Knowledge around peptides with bioactivity is limited as they are usually produced using single enzymes not reflecting simulated GI conditions. Potential novel peptides produced within the simulated SGID of this research may not have been identified before and therefore not recorded to be bioactive. Therefore, there may be more than 17 peptides within the 354 identified within this SGID that may show additional bioactivity which has yet to be recorded.

Examining Appendix A, Demers-Mathieu et al., 2013, Hirose, Kurimoto, Yuda and Tanaka, 2024, Jacquot, Gauthier, Drouin and Boutin, 2010, Lacroix, Chen, Kitts and Li-Chan, 2017, Lacroix, Meng, Cheung and Li-Chan, 2016, Li et al., 2022, Maes et al., 2004, Mullally, Meisel and Fitzgerald, 1997, Nagaoka et al., 2001, Pellegrini, Dettling, Thomas and Hunziker, 2001, Pihlanto-Leppälä, 2000, Power, Nongonierma, Jakeman and Fitzgerald, 2014, Sedaghati, Ezzatpanah, Mashhadiakbar Boojar, Tajabadi Ebrahimi and Aminafshar, 2015, Silveira, Martínez-Maqueda, Recio and Hernández-Ledesma, 2013, Tavares et al., 2011, Tavares and Xavier Malcata, 2012, there are a number of peptides associated with in-vitro DPP-IV inhibition and ACE-inhibition. DPP-IV inhibitors are used to manage type-2 diabetes which is a risk factor for cardiovascular conditions including heart disease and stroke (Lacroix & Li-Chan, 2014), both of which have an increased risk with increasing age (Rodgers et al., 2019). Additionally, DPP-IV inhibitors are used to manage appetite as explored in a review by Tulipano et al. (2017) where they have been noted to reduce the degradation of appetite related hormones including glucagon-like peptide-1 (GLP-1) and glucose-dependent insulinotropic peptide (GIP), which in turn increase their circulation in the blood stream, suppressing appetite. A large proportion of the predicted DPPIV activity of the digesta comes from 3 β-Lg peptides situated near the C terminal, namely VLDTDY (94–99), VLDTDYK (94–100) and TPEVDDEALEK (125–135).

ACE-Inhibitors are used to lower blood pressure, as high blood pressure is another medical condition common with increasing age (Rodgers et al., 2019). Therefore, identification that older adult SGID conditions can produce peptides with established ACE-Inhibition and DPP-IV inhibition from WPI are of importance and could contribute to maintaining overall health with ageing. The same two peptides VLDTDY (94–99), VLDTDYK (94–100) contributed also to the predicted ACE inhibition activity of the intestinal digesta, together with another β-Lg peptide GLDIQK (9–14) situated near the N terminal of the protein.

Additionally, antimicrobial peptides were present within the intestinal samples for all SGID examined. Interestingly, the presence of these peptides increased significantly from Adult conditions to OA3 conditions with much of this activity due to a single peptide TPEVDDEALEK (125–135). This higher occurrence of antimicrobial peptides due to older adult digestive conditions is promising as older adults are more at risk of antibacterial infections including H. pylori (Pilotto & Franceschi, 2014). Bioactive peptides to manage cholesterol also followed this trend as Adult had the lowest occurrence of these peptides which increased through Healthy OA and OA2, with OA3 having the highest occurrence. Again, this bioactivity score was mainly predicted to be due to one peptide GLDIQK (25–30). Although these values were not deemed to be statistically significant, it is interesting that some bioactivities that may assist with managing the health of older adults seem to be modulated by gastrointestinal conditions related to ageing.

Conclusion

4

Examination of in-vitro digestion needs to accurately represent the relevant population age groups, as specific changes occur in the gastrointestinal tract due to increasing age. While the development of a suitable model to simulate Healthy OA has been recently published (Menard et al., 2023), this consensus established SGID conditions reflecting the general older healthy adult population and may not represent a more elderly, medicated population, or one suffering from gastro-intestinal infections. Therefore, this study used a wide range of experimental conditions representing a broad range of ageing and health conditions of the gastrointestinal tract, focusing on gastric pH, time, and digestive enzyme concentration to improve on current in-vitro digestion experimental analysis which resulted in the establishment of 2 novel older adult SGID protocols reflecting elderly population with either GI infections (OA2) or population under medications that alter their gastric pH (OA3). These protocols showed larges differences in the digestibility of whey proteins as shown by different proteolysis profiles. Specifically, pH had a very significant effect on WPI digestion within the simulated gastric phase with the higher gastric pH values of OA2 an OA3 protocols close to the isoelectric point of the whey proteins resulting in marked reduced breakdown into absorbable peptides, regardless of simulated digestion time. This finding could explain the poor in-vivo effect of protein supplements seen in elderly populations in mitigating sarcopenia and increasing muscle turn-over.

Trends in reduced digestion observed in the gastric phase were also observed within the intestinal phase, showing that gastric conditions have an impact further down the gastrointestinal tract. Degree of Hydrolysis, bio accessible peptide content and SEC data all show that the older adult conditions OA2 and OA3 result in lower protein digestion at the end of the intestinal digestion step. LC-MS analysis of intestinal digesta showed that Adult and Healthy OA WPI digesta had a statistically different peptide profile to that of OA2 and OA3 conditions Significant differences were also shown on the significant peptides between the healthy adult protocols (Adult, Healthy OA) and those newly developed protocols to simulate older populations (OA2) or medicated ones (OA3). Differences in abundance of peptides with specific bioactivity were also observed, with significantly higher occurrence of antimicrobial peptides, and higher occurrence of ACE-inhibition and cholesterol managing bioactive peptides being present within OA2 and OA3 conditions.

These results show that there is huge merit in considering a range of operating conditions in establishing SGID protocols to represent the elderly digestive systems of populations suffering from poor muscle turnover and sarcopenia, as reduced acid conditions in the gastric environment resulted in lasting effects on the protein digestibility of whey proteins.

CRediT authorship contribution statement

Laura Gunning: Writing – original draft, Methodology, Investigation, Funding acquisition, Formal analysis, Data curation, Conceptualization. Michael O'Sullivan: Writing – review & editing, Methodology, Formal analysis. Eugene Dillon: Visualization, Methodology, Investigation, Formal analysis. Raquel Cama-Moncunill: Writing – review & editing, Formal analysis, Data curation. Jean-Christophe Jacquier: Writing – review & editing, Supervision, Project administration, Methodology, Funding acquisition, Formal analysis, Conceptualization.

Declaration of competing interest

The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.