Topic hub — 2026-10-22
Does breaking down protein with enzymes make it more bitter or less?
Proteins from milk, soy, pulses, cereals, fish or insects, including those used in protein powders, become hydrolysates — mixtures of short peptides and amino acids — when broken down by enzymes. Protein products and peptides are often said to taste bitter; we checked, in studies where people rated bitterness, whether hydrolysis always makes protein more bitter or whether the enzyme or the way it is hydrolysed can soften it. This page collects comparisons in which the enzyme or enzyme amount, the degree or time of hydrolysis, or the after-treatment of the resulting hydrolysate was changed for the same protein raw material. Bitter peptides from cheese ripening are covered on the cheese ripening and starter page, and suppressing bitterness with sugar or salt on the sugar, salt and bitterness page. In the included studies, depending on the enzyme and the degree of hydrolysis, bitterness rose in some cases and, at the end points shown, did not differ significantly from the unhydrolysed material in others. Some studies lowered bitterness with after-treatments such as adding an enzyme that trims peptide ends, while in others bitterness rose after membrane processing or after ultrasound applied to an encapsulated hydrolysate. Raw materials, scales, raters and sample concentrations differ by study, so it cannot be said across studies how many times more bitter one sample is. In the list, rows compared with the unhydrolysed raw material come first (for clams and crickets, the label states that equivalent processing is unconfirmed), and rows of a different nature are marked with a tag at the start of the label (degree of hydrolysis or enzyme, compared with a commercial hydrolysate, after-treatment of the hydrolysate) and grouped after them. Bitterness is read separately from umami, aftertaste and acceptability. This page lines up the included studies together with their conditions: the raw material, the enzyme and hydrolysis conditions, the after-treatment, and who rated them on what scale. It then separates what was established within one experiment, what differences might explain, and what this comparison alone cannot separate. It is about perceived taste, not advice on choosing commercial protein products or on health. Fermentation and bitter peptides as a whole are covered on the fermentation and bitter peptides page.
On this page (10 sections)
- What this page establishes
- What the included studies allow us to say
- Bitterness ratings at a glance
- Conditions and results, study by study
- What was established within an experiment
- Candidate explanations for the differences
- What is not known — what this comparison alone cannot separate
- How studies and rows were counted
- What we looked at
- Read in the encyclopedia
What this page establishes
- That in studies compared with the unhydrolysed raw material, Alcalase and some other enzymes made it significantly more bitter, while at the end points shown in the table (the last time point rated for soy, the end of hydrolysis for lupin) Flavourzyme and papain did not differ significantly from unhydrolysed; that in the same soy study they were significantly higher at intermediate times; and that no significant difference does not mean the same.
- That in comparisons of the degree or time of hydrolysis, some studies found bitterness rising with further hydrolysis and others found it falling, so the direction is not one way.
- That some studies lowered bitterness by adding enzymes that trim peptide ends, or adsorbents, to an existing hydrolysate.
- That in some comparisons bitterness rose after membrane processing or after ultrasound applied to a hydrolysate encapsulated in liposomes, so after-treatment does not always lower it.
- That many comparisons have no rating of the unhydrolysed raw material, and scales and concentrations differ by study, so values cannot be compared across studies.
- That several reports come from the same institute and are not counted as independent replications.
What the included studies allow us to say
- In studies that compared hydrolysates with the unhydrolysed raw material by the same procedure, results split by enzyme. In a study of soy protein isolate (Meinlschmidt et al. 2016), hydrolysis with Alcalase for 120 min significantly raised bitterness from 2.8 to 9.2, and Neutrase also to 7.1, while at 120 min papain 0.2% (3.1) and Flavourzyme (2.1) did not differ significantly from unhydrolysed (both were significantly higher at an intermediate time). In a lupin study (Schlegel et al. 2019), too, Alcalase was the only one of 9 enzymes significantly more bitter than unhydrolysed (medians 1.1 and 7.2). Flavourzyme's 2.1 is numerically below the unhydrolysed 2.8, but this is not a significant decrease. [rows: Meinlschmidt 2016 Soy: unhydrolysed (0 min) vs Alcalase for 120 min; Meinlschmidt 2016 Soy: unhydrolysed (0 min) vs Neutrase for 120 min; Meinlschmidt 2016 Soy: unhydrolysed (0 min) vs papain 0.2% for 120 min; Meinlschmidt 2016 Soy: unhydrolysed (0 min) vs Flavourzyme for 120 min; Schlegel 2019 Lupin: unhydrolysed vs Alcalase for 2 h]
- Studies comparing the degree or time of hydrolysis also do not point one way. In a wheat gluten study (Sun et al. 2022), taking the degree of hydrolysis from 4% to 20% significantly raised bitterness with Alcalase from 0.65 to 4.38. In a study hydrolysing casein with 2 enzyme sets (Daher et al. 2021), one set fell from about 5.20 to 3.00 between 3 and 5 h, while the other rose from 2.23 at 5 h to 3.59 at 9 h (which pairs of times differ is in a figure and unconfirmed). In a cricket study (Fetriyuna et al. 2026), longer hydrolysis gave lower bitterness (marginal means 14.18 to 7.32, significant). [rows: Sun 2022 Degree of hydrolysis or enzyme: wheat gluten, Alcalase at 4% vs 20% degree of hydrolysis; Daher 2021 Degree of hydrolysis or enzyme: casein, Flavourzyme with Protamex (set 109), 3 h vs 5 h; Daher 2021 Degree of hydrolysis or enzyme: casein, Promod with Flavorpro (set 125), 5 h vs 9 h; Fetriyuna 2026 Degree of hydrolysis or enzyme: cricket, ProteAX for 2 h vs 5 h (mean over enzyme amounts)]
- Some studies lowered bitterness by treating an existing hydrolysate afterwards. In a study applying an aminopeptidase to pepsin hydrolysates of casein and cod (Nakamura et al. 2023), casein fell from 3.5 to 1.7 and cod from 2.9 to 0.54 (both p<0.01). In a study further hydrolysing an Alcalase hydrolysate of rice with Flavourzyme (Wu et al. 2026), in an evaluation of fractions below 2500 Da taken from every sample at equal nitrogen concentration, all 4 second-stage products were judged significantly less bitter than the fraction of the original hydrolysate (A6). On the other hand, in a study membrane-processing a poultry hydrolysate (Aspevik et al. 2026) bitterness rose from 4.4 to 5.1, and in a study ultrasonicating liposomes encapsulating a salmon hydrolysate (Sharma et al. 2023) it rose from 4.5 to 6.3. After-treatment does not always lower bitterness either. Scales and raw materials differ, so values are not compared across studies. [rows: Nakamura 2023 After-treatment of the hydrolysate: pepsin hydrolysate of casein, aminopeptidase LapA at 0 h vs 4 h; Nakamura 2023 After-treatment of the hydrolysate: pepsin hydrolysate of cod protein, aminopeptidase LapA at 0 h vs 21 h; Wu 2026 After-treatment of the hydrolysate: rice, Alcalase hydrolysate (A6) vs Flavourzyme with pH held for 3 h (A6-s3); Wu 2026 After-treatment of the hydrolysate: rice, Alcalase hydrolysate (A6) vs Flavourzyme without pH control for 4 h (A6-f4); Aspevik 2026 After-treatment of the hydrolysate: poultry hydrolysate, before membrane processing (crude) vs after microfiltration, nanofiltration and diafiltration; Sharma 2023 After-treatment of the hydrolysate: encapsulated salmon hydrolysate (L-PH1), ultrasound for 0 s vs 120 s]
Bitterness ratings at a glance
Human bitterness ratings only. Raw materials, scales and concentrations differ by study, so values are not compared across studies. The two samples compared are set out as the sample before and the sample after "vs" in the label. Before "vs" we put the unhydrolysed, less hydrolysed or untreated sample; for comparisons between enzymes, the order is as written in the label. The "Direction" column shows which sample was more bitter ("... more bitter"). The direction shows which published value is larger, or what the text states, and is not necessarily the result of an individual test between the two conditions shown. Whether the difference is significant is shown by the test in the same "Direction / test" column; even in rows with no significant difference, the direction shows which value is larger. Rows with no test shown and rows whose values are only in figures we did not obtain are kept as they are. In rows compared with a commercial hydrolysate the values are sums of classes, positive meaning more bitter and negative less bitter than the commercial hydrolysate. In the rice rows the value is a score for the difference in bitterness between two samples, not an intensity. These two kinds of rows are also marked as such in the column for what was measured. Labels begin with a tag for rows of a different nature; rows without a tag compare hydrolysates with the unhydrolysed raw material within one study and come first in the list. "Degree of hydrolysis or enzyme" marks rows comparing hydrolysis conditions with no rating of unhydrolysed material; "Compared with a commercial hydrolysate" marks rows whose reference is a commercial hydrolysate; and "After-treatment of the hydrolysate" marks rows in which enzymes, adsorbents, membranes, encapsulation or ultrasound were applied to a hydrolysate afterwards. Among the rows without a tag, for the cricket and clam rows the label states that it cannot be confirmed whether the unhydrolysed material and the hydrolysates went through the same fractionation step. Select a study name to jump to its detailed table.
Results in which people rated bitterness only. Where each value comes from (a table or a reading from a figure) is shown in its row; scales and methods differ between studies, so compare within a row, not across rows.
| Study | Comparison and condition | Measured as | Sample before "vs" | Sample after "vs" | Direction / test |
|---|---|---|---|---|---|
| Meinlschmidt 2016 | Soy: unhydrolysed (0 min) vs Alcalase for 120 min A 5% dispersion of soy protein isolate hydrolysed for 120 min at enzyme-specific temperature and pH (enzyme 0.5%, papain 0.2%); the 0 min value in the table is the 2.8 that the text gives as the bitterness of unhydrolysed SPI; rated as a 2.5% solution (pH 7.0); Alcalase at 50 °C and pH 8.0, degree of hydrolysis 13.0% at 120 min | 10 cm line scale from 0 (not detectable) to 10 (intense) (10 trained panellists, mean of 2 replicates) | 2.8 | 9.2 | Sample after "vs" more bitter Tested, significant |
| Meinlschmidt 2016 | Soy: unhydrolysed (0 min) vs Neutrase for 120 min A 5% dispersion of soy protein isolate hydrolysed for 120 min at enzyme-specific temperature and pH (enzyme 0.5%, papain 0.2%); the 0 min value in the table is the 2.8 that the text gives as the bitterness of unhydrolysed SPI; rated as a 2.5% solution (pH 7.0); Neutrase at 50 °C and pH 6.5, degree of hydrolysis 6.3% at 120 min | 10 cm line scale from 0 (not detectable) to 10 (intense) (10 trained panellists, mean of 2 replicates) | 2.8 | 7.1 | Sample after "vs" more bitter Tested, significant |
| Meinlschmidt 2016 | Soy: unhydrolysed (0 min) vs papain 0.2% for 120 min A 5% dispersion of soy protein isolate hydrolysed for 120 min at enzyme-specific temperature and pH (enzyme 0.5%, papain 0.2%); the 0 min value in the table is the 2.8 that the text gives as the bitterness of unhydrolysed SPI; rated as a 2.5% solution (pH 7.0); papain 0.2% at 80 °C and pH 7.0, degree of hydrolysis 4.6% at 120 min | 10 cm line scale from 0 (not detectable) to 10 (intense) (10 trained panellists, mean of 2 replicates) | 2.8 | 3.1 | Sample after "vs" more bitter Tested, not significant |
| Meinlschmidt 2016 | Soy: unhydrolysed (0 min) vs Flavourzyme for 120 min A 5% dispersion of soy protein isolate hydrolysed for 120 min at enzyme-specific temperature and pH (enzyme 0.5%, papain 0.2%); the 0 min value in the table is the 2.8 that the text gives as the bitterness of unhydrolysed SPI; rated as a 2.5% solution (pH 7.0); Flavourzyme at 50 °C and pH 6.0, degree of hydrolysis 8.5% at 120 min | 10 cm line scale from 0 (not detectable) to 10 (intense) (10 trained panellists, mean of 2 replicates) | 2.8 | 2.1 | Sample before "vs" more bitter Tested, not significant |
| Schlegel 2019 | Lupin: unhydrolysed vs Alcalase for 2 h Lupin protein isolate (50 g/kg protein) hydrolysed for 2 h at enzyme-specific temperature and pH (enzyme-to-solution ratio 0.5%); unhydrolysed controls made under the same conditions; rated as a 1% solution with a nose clip; Alcalase at 50 °C and pH 8.0, degree of hydrolysis 9.05% | median on an unstructured 10 cm line (10 trained panellists) | 1.1 | 7.2 | Sample after "vs" more bitter Tested, significant |
| Schlegel 2019 | Lupin: unhydrolysed vs Flavourzyme for 2 h Lupin protein isolate (50 g/kg protein) hydrolysed for 2 h at enzyme-specific temperature and pH (enzyme-to-solution ratio 0.5%); unhydrolysed controls made under the same conditions; rated as a 1% solution with a nose clip; Flavourzyme at 50 °C and pH 6.0, degree of hydrolysis 6.90% | median on an unstructured 10 cm line (10 trained panellists) | 1.1 | 0.7 | Sample before "vs" more bitter Tested, not significant |
| Schlegel 2020 | Lupin: unhydrolysed vs Alcalase and papain together for 4 h (OS 5) Lupin protein isolate at 5%, 50 °C and pH 8.0; Alcalase 0.5% and papain 0.2% (enzyme-to-solution ratio) added together for 4 h, degree of hydrolysis 13.3%; serving concentration not in the text | 0 (no perception) to 10 (strong) scale; the Fig. 4 legend gives means on a 10 cm line (10 panellists) | 1.9 | 5.7 | Sample after "vs" more bitter Tested, significant |
| Zhao 2026 | Clam: unhydrolysed control (equivalence of processing unconfirmed) vs alkaline protease for 4 h Clam meat powder in water at 1:50; alkaline protease (5000 U/g protein, 50 °C, pH 9.5) for 4 h; the control was freeze-dried clam protein, and the hydrolysates were centrifuged supernatants; serving concentration not stated | 0–10 scale (1–2 no bitterness, 9–10 unbearably bitter; 12 trained assessors, mean of 3 replicates) | Figure not obtained, so values unconfirmed | Figure not obtained, so values unconfirmed | Sample after "vs" more bitter No test reported |
| Fetriyuna 2026 | Cricket: raw material dried without enzyme (CP; equivalence of processing unconfirmed) vs ProteAX 1.5% for 5 h Crickets blended and pasteurised at 90 °C for 15 min, then ProteAX 1.5% (w/v) at 60 °C and pH 7.0 for 5 h; the hydrolysate was the freeze-dried centrifuged supernatant (degree of hydrolysis 67.71%, TCA-soluble nitrogen method); CP is described only as freeze-dried without enzyme treatment, and whether it went through the same centrifugation and supernatant step is not stated (degree of hydrolysis 6.85% for CP); serving concentration and solvent not stated | gLMS (150 mm, in mm; 25 semi-trained panellists) | 32.88 | 5.92 | Sample before "vs" more bitter No test reported |
| Sun 2022 | Degree of hydrolysis or enzyme: wheat gluten, Alcalase at 4% vs 20% degree of hydrolysis 15 g wheat gluten in 300 mL water; enzyme-to-substrate ratio 1:20, stopped when the pH-stat degree of hydrolysis reached 4% and 20%; the centrifuged supernatant freeze-dried and rated as a 1% solution; unhydrolysed gluten was not rated; Alcalase | 10-point line scale (anchored with caffeine; 8 experienced assessors) | 0.65 | 4.38 | Sample after "vs" more bitter Tested, significant |
| Sun 2022 | Degree of hydrolysis or enzyme: wheat gluten, trypsin at 4% vs 20% degree of hydrolysis 15 g wheat gluten in 300 mL water; enzyme-to-substrate ratio 1:20, stopped when the pH-stat degree of hydrolysis reached 4% and 20%; the centrifuged supernatant freeze-dried and rated as a 1% solution; unhydrolysed gluten was not rated; trypsin | 10-point line scale (anchored with caffeine; 8 experienced assessors) | 0.59 | 3.4 | Sample after "vs" more bitter Tested, significant |
| Fetriyuna 2026 | Degree of hydrolysis or enzyme: cricket, ProteAX for 2 h vs 5 h (mean over enzyme amounts) Pasteurised crickets with ProteAX at 0.5, 1.0, 1.5 or 2% (w/v) at 60 °C and pH 7.0 for 2, 3.5 or 5 h; the centrifuged supernatant freeze-dried (degree of hydrolysis 50.34–67.71%, TCA-soluble nitrogen method); serving concentration and solvent not stated | gLMS (150 mm, in mm; 25 semi-trained panellists) | 14.18 (2 h, marginal mean over 4 enzyme amounts) | 7.32 (5 h, likewise) | Sample before "vs" more bitter Tested, significant |
| Daher 2021 | Degree of hydrolysis or enzyme: casein, Flavourzyme with Protamex (set 109), 3 h vs 5 h Milk protein rich in micellar casein diluted to 10% of total nitrogenous matter, hydrolysed for 9 h and sampled every 1 h (recipe confidential); rated at 10% dry matter in mineral water; unhydrolysed casein was not rated; set 109 is Flavourzyme with Protamex; degree of hydrolysis 13.9% at 4 h, 44.1% at 5 h and 50.8% at 9 h | 1 (low) to 7 (high) (19 trained assessors, 2 replicates) | about 5.20 | 3.00 | Sample before "vs" more bitter Tested; significance reported, but which pair differs is unclear |
| Daher 2021 | Degree of hydrolysis or enzyme: casein, Promod with Flavorpro (set 125), 5 h vs 9 h Milk protein rich in micellar casein diluted to 10% of total nitrogenous matter, hydrolysed for 9 h and sampled every 1 h (recipe confidential); rated at 10% dry matter in mineral water; unhydrolysed casein was not rated; set 125 is Promod 523MDP with Flavorpro 937MDP; degree of hydrolysis 10.9% at 3 h, 22.6% at 5 h and 28.9% at 9 h | 1 (low) to 7 (high) (19 trained assessors, 2 replicates) | 2.23 | 3.59 | Sample after "vs" more bitter Tested; significance reported, but which pair differs is unclear |
| Jiang 2024 | Degree of hydrolysis or enzyme: skim milk, Flavourzyme vs Neutrase 12% (w/w) skim milk hydrolysed for 4.5 h and inactivated at 105 °C for 5 min; Neutrase at 50 °C and 10,000 U/g (degree of hydrolysis 24.89%), Flavourzyme at 50 °C and 6000 U/g (23.20%), Protamex at 55 °C and 10,000 U/g (28.13%), all at pH 7.0; serving concentration not stated; unhydrolysed skim milk was not rated | score against quinine standards set from 0 to 10 points (10 food-science graduate students) | 6.60 | 7.60 | Sample after "vs" more bitter Tested, significant |
| Jiang 2024 | Degree of hydrolysis or enzyme: skim milk, Protamex vs Neutrase 12% (w/w) skim milk hydrolysed for 4.5 h and inactivated at 105 °C for 5 min; Neutrase at 50 °C and 10,000 U/g (degree of hydrolysis 24.89%), Flavourzyme at 50 °C and 6000 U/g (23.20%), Protamex at 55 °C and 10,000 U/g (28.13%), all at pH 7.0; serving concentration not stated; unhydrolysed skim milk was not rated | score against quinine standards set from 0 to 10 points (10 food-science graduate students) | 6.80 | 7.60 | Sample after "vs" more bitter Tested, significant |
| Sugiyama 1991 | Degree of hydrolysis or enzyme: sardine, Alcalase vs Proleather (both alkaline) Defatted sardine meal at 8% protein with enzyme at 8% of protein for 6 h; Alcalase at 55 °C and pH 8.0, Proleather at 50 °C and pH 8.0; rated as a 5% protein solution at 30 °C; unhydrolysed meal was not rated | 6 grades from 0 (none) to 5 (very strongly), mean of a 10-member panel | 1.7 | 4.0 | Sample after "vs" more bitter No test reported |
| Knežević-Jugović 2023 | Compared with a commercial hydrolysate: soy, Neutrase alone vs Neutrase then Flavourzyme Soy protein concentrate (8% protein); 2-step samples (first enzyme for 75 min, then adjusted to 50 °C and pH 7 with Flavourzyme 4% for another 120 min) set beside the one-step samples; insolubles removed and spray-dried; each rated as a 1% solution paired with a commercial soy protein hydrolysate (R); unhydrolysed concentrate was not rated; Neutrase at pH 7 and 45 °C, enzyme 4% | sum of classes on a 5-point bipolar scale against R (positive means more bitter than R, negative less bitter than R; 20 assessors) | +14 | −25 | Sample before "vs" more bitter Tested; significance reported, but which pair differs is unclear |
| Knežević-Jugović 2023 | Compared with a commercial hydrolysate: soy, Alcalase alone vs Alcalase then Flavourzyme Soy protein concentrate (8% protein); 2-step samples (first enzyme for 75 min, then adjusted to 50 °C and pH 7 with Flavourzyme 4% for another 120 min) set beside the one-step samples; insolubles removed and spray-dried; each rated as a 1% solution paired with a commercial soy protein hydrolysate (R); unhydrolysed concentrate was not rated; Alcalase at pH 8 and 55 °C, enzyme 4% | sum of classes on a 5-point bipolar scale against R (positive means more bitter than R, negative less bitter than R; 20 assessors) | −15 | −7 | Sample after "vs" more bitter Tested; significance reported, but which pair differs is unclear |
| Nakamura 2023 | After-treatment of the hydrolysate: pepsin hydrolysate of casein, aminopeptidase LapA at 0 h vs 4 h Supernatant of 10% casein hydrolysed with pepsin (1/300) at 37 °C for 20 h in total, adjusted to pH 8.0, with purified LapA added at 1/24,500 at 50 °C; rated diluted 6-fold | 0 (no bitterness) to 5 (maximum), with Gly-Phe at 0.5, 1.0 and 1.5% as 1, 2 and 3 (9 panellists) | 3.5 | 1.7 | Sample before "vs" more bitter Tested, significant |
| Nakamura 2023 | After-treatment of the hydrolysate: pepsin hydrolysate of cod protein, aminopeptidase LapA at 0 h vs 21 h Supernatant of 10% cod protein hydrolysed with pepsin (1/300) at 37 °C for 7 h in total, adjusted to pH 8.0, with purified LapA added at 1/80,000 at 50 °C | 0 (no bitterness) to 5 (maximum), with Gly-Phe at 0.5, 1.0 and 1.5% as 1, 2 and 3 (9 panellists) | 2.9 | 0.54 | Sample before "vs" more bitter Tested, significant |
| Wu 2026 | After-treatment of the hydrolysate: rice, Alcalase hydrolysate (A6) vs Flavourzyme with pH held for 3 h (A6-s3) The 6 h Alcalase hydrolysate of rice protein (A6, DH-OPA 24.69%) given a second stage with Flavourzyme (5%, pH 6, 60 °C); every sample rated as the 2500 Da membrane permeate at equal nitrogen concentration; unhydrolysed rice protein was not rated; A6-s3 held at pH 6.0 for 3 h (DH-OPA 37.14%) | difference-in-bitterness score per pair from 0 (no difference) to 7 (extremely different), with a choice of the more bitter sample (15 panellists, 2 sessions) | Reference (one side of a difference-score pair; no intensity value) | difference score from A6, 3.64 (A6 more bitter) | Sample before "vs" more bitter Tested, significant |
| Wu 2026 | After-treatment of the hydrolysate: rice, Alcalase hydrolysate (A6) vs Flavourzyme without pH control for 4 h (A6-f4) The 6 h Alcalase hydrolysate of rice protein (A6, DH-OPA 24.69%) given a second stage with Flavourzyme (5%, pH 6, 60 °C); every sample rated as the 2500 Da membrane permeate at equal nitrogen concentration; unhydrolysed rice protein was not rated; A6-f4 without pH control (falling from 6.0 to 5.1) for 4 h (DH-OPA 40.35%) | difference-in-bitterness score per pair from 0 (no difference) to 7 (extremely different), with a choice of the more bitter sample (15 panellists, 2 sessions) | Reference (one side of a difference-score pair; no intensity value) | difference score from A6, 2.78 (A6 more bitter) | Sample before "vs" more bitter Tested, significant |
| Zhao 2026 | After-treatment of the hydrolysate: clam, alkaline protease for 4 h vs then AOAP for 3 h Clam meat powder in water at 1:50 with alkaline protease (5000 U/g protein, 50 °C, pH 9.5) for 4 h; AOAP, an aminopeptidase complex (5000 U/g, 50 °C, pH 7.0), then added to that hydrolysate for 3 h more (degree of hydrolysis given in the text as rising from 23.9% to 49.8%); serving concentration not stated | 0–10 scale (1–2 no bitterness, 9–10 unbearably bitter; 12 trained assessors, mean of 3 replicates) | Figure not obtained, so values unconfirmed | Figure not obtained, so values unconfirmed | Sample before "vs" more bitter Tested, significant |
| Sugiyama 1991 | After-treatment of the hydrolysate: sardine Alcalase hydrolysate, untreated vs aminopeptidase M 3 mg Alcalase hydrolysate of defatted sardine meal (5% protein, pH 7.0), 100 mL, with aminopeptidase M added at 3 mg as protein, 37 °C for 3 h; inactivated at 90 °C for 5 min and filtered | 6 grades from 0 (none) to 5 (very strongly), mean of a 10-member panel | 1.7 | 0.9 | Sample before "vs" more bitter No test reported |
| Sugiyama 1991 | After-treatment of the hydrolysate: sardine Alcalase hydrolysate, untreated vs synthetic adsorbent VP-10 Alcalase hydrolysate of defatted sardine meal (5% protein, pH 7.0), 150 mL, passed through a column of 15 mL of the vinylpyridine resin VP-10 (SV 2), and the eluate freeze-dried and rated at 5% protein | 6 grades from 0 (none) to 5 (very strongly), mean of a 10-member panel | 1.7 | 0.5 | Sample before "vs" more bitter No test reported |
| Aspevik 2026 | After-treatment of the hydrolysate: poultry hydrolysate, before membrane processing (crude) vs after microfiltration, nanofiltration and diafiltration A company-supplied enzymatic hydrolysate of poultry side streams processed by 0.1 μm microfiltration, 200 Da nanofiltration and diafiltration in turn; all rated at 2% (w/w) dry matter (not at equal protein concentration) | 15 cm line with end points 1 (no intensity) and 9 (high intensity) (10 trained assessors, 2 replicates) | 4.4 | 5.1 | Sample after "vs" more bitter Tested, significant |
| Sharma 2023 | After-treatment of the hydrolysate: salmon hydrolysate at 1%, not encapsulated vs encapsulated in liposomes A 1% solution of the papain hydrolysate of salmon frame protein (SFPH; papain 1% at 40 °C for 5 min), and liposomes made by hydrating a soy lecithin and glycerol film with the same solution (L-PH1) | 15 cm line scale from 0 (none) to 15 (intense) (trained panel, number not stated) | 5.8 | 4.2 | Sample before "vs" more bitter Tested, significant |
| Sharma 2023 | After-treatment of the hydrolysate: encapsulated salmon hydrolysate (L-PH1), ultrasound for 0 s vs 120 s Liposomes encapsulating 1% hydrolysate (L-PH1) ultrasonicated (20 kHz, 750 W, 20% amplitude, 2 s on and 8 s off) for 120 s, kept at 25 ± 5 °C in iced water | 15 cm line scale from 0 (none) to 15 (intense) (trained panel, number not stated) | 4.5 | 6.3 | Sample after "vs" more bitter Tested, significant |
Conditions and results, study by study
For each study, the conditions for the raw material, the enzyme and hydrolysis, the after-treatment, how samples were served and who rated them, and the results. Each row is one study × one measured result, so different comparisons in the same study are separate rows. Under each row we give where it is in the paper and whether it can be used to judge whether changing the enzyme or enzyme amount, the degree or time of hydrolysis, or the after-treatment of the resulting hydrolysate, for the same protein raw material, changes the intensity of bitterness people perceive, with the reason. The reason also says whether the row can be used for comparison with the unhydrolysed material, or only for comparing hydrolysis conditions or after-treatments. Nothing is left blank: "not given", "not read" and "not applicable" are kept apart, and values from figures we did not obtain are marked as such. The degree of hydrolysis is measured differently by study (OPA, pH-stat, TCA-soluble nitrogen), and enzyme amounts mix ratios to protein with ratios to solution, so read them with the method.
Meinlschmidt 2016 sensory only
Enzymatic treatment of soy protein isolates: effects on the potential allergenicity, technofunctionality, and sensory properties. Food Science & Nutrition 4(1):11-23 — doi:10.1002/fsn3.253
- How deeply checked
- Checked in full text — Full text via Europe PMC (methods "Enzymatic hydrolysis of SPI", "Sensory analysis" and "Statistical analysis"; results "Effects of the enzymatic treatment on the bitterness of SPI"; Tables 1 and 2; the Fig. 3 legend); Fig. 3 is a separate experiment and is not used in this table Checked by: LLM (one extracted, another checked against the source; no human check), 2026-10-22.
- Samples
- Soy protein isolate (SPI, 94.6% protein)
- Processing and cooking
- SPI dispersed in deionised water at 5% (w/w); 10 proteases (papain at 2 amounts, 0.2% and 0.05%, for 11 conditions in all) at enzyme-specific temperature and pH, sampled at 10, 30, 60 and 120 min; enzyme amount 0.5% (papain as above); inactivated at 90 °C for 20 min and freeze-dried; controls without enzyme made under the same conditions; degree of hydrolysis (OPA method) 2.1% at 0 min and, at 120 min, 13.0% for Alcalase, 8.5% for Flavourzyme, 6.3% for Neutrase and 4.6% for papain 0.2%
- How it was served or eaten
- 2.5% (w/w) solution adjusted to pH 7.0, 10 mL at room temperature, coded with random 3-digit numbers
- Who rated, on what scale
- 10 panellists trained for 2 months, including caffeine threshold tests; 10 cm line scale from 0 (not detectable) to 10 (intense); reference was Alcalase hydrolysate (the calibration section gives 1.0 g/L = 5 and 2.5 g/L = 10, while the session standards are given as 1.0% and 2.5%, inconsistent units); 2 replicates; the methods give ANOVA with Tukey, while the Table 2 footnote gives ANOVA with Bonferroni per protease, inconsistent
| Measured | Comparison and condition | Sample before "vs" | Sample after "vs" | Unit | Direction / test |
|---|---|---|---|---|---|
| Bitterness intensity (sensory) | Soy: unhydrolysed (0 min) vs Alcalase for 120 min A 5% dispersion of soy protein isolate hydrolysed for 120 min at enzyme-specific temperature and pH (enzyme 0.5%, papain 0.2%); the 0 min value in the table is the 2.8 that the text gives as the bitterness of unhydrolysed SPI; rated as a 2.5% solution (pH 7.0); Alcalase at 50 °C and pH 8.0, degree of hydrolysis 13.0% at 120 min | 2.8 | 9.2 | 10 cm line scale from 0 (not detectable) to 10 (intense) (10 trained panellists, mean of 2 replicates) | Sample after "vs" more bitter Tested, significant |
| In Table 2, 0 min (a) and 120 min (b) carry different letters, significant (p<0.05). Bitterness was already high at 10 min (8.7). The letters in Table 2 compare 0 min with each time within one enzyme; enzymes were not tested against each other. The methods give ANOVA with Tukey and the Table 2 footnote gives Bonferroni, inconsistent Where: Methods, "Enzymatic hydrolysis of SPI", "Sensory analysis" and "Statistical analysis"; Results, "Effects of the enzymatic treatment on the bitterness of SPI"; Tables 1 and 2 For whether changing the enzyme or enzyme amount, the degree or time of hydrolysis, or the after-treatment of the resulting hydrolysate, for the same protein raw material, changes the intensity of bitterness people perceive: Usable — The same soy protein isolate at 0 and 120 min, rated by 10 trained panellists with the same procedure and tested over time within each enzyme; however enzymes were not tested against each other, and the same % of enzyme differs in activity | |||||
| Bitterness intensity (sensory) | Soy: unhydrolysed (0 min) vs Neutrase for 120 min A 5% dispersion of soy protein isolate hydrolysed for 120 min at enzyme-specific temperature and pH (enzyme 0.5%, papain 0.2%); the 0 min value in the table is the 2.8 that the text gives as the bitterness of unhydrolysed SPI; rated as a 2.5% solution (pH 7.0); Neutrase at 50 °C and pH 6.5, degree of hydrolysis 6.3% at 120 min | 2.8 | 7.1 | 10 cm line scale from 0 (not detectable) to 10 (intense) (10 trained panellists, mean of 2 replicates) | Sample after "vs" more bitter Tested, significant |
| In Table 2, 0 min (a) and 120 min (c) carry different letters, significant (p<0.05). The letters in Table 2 compare 0 min with each time within one enzyme; enzymes were not tested against each other. The methods give ANOVA with Tukey and the Table 2 footnote gives Bonferroni, inconsistent Where: Methods, "Enzymatic hydrolysis of SPI", "Sensory analysis" and "Statistical analysis"; Results, "Effects of the enzymatic treatment on the bitterness of SPI"; Tables 1 and 2 For whether changing the enzyme or enzyme amount, the degree or time of hydrolysis, or the after-treatment of the resulting hydrolysate, for the same protein raw material, changes the intensity of bitterness people perceive: Usable — The same soy protein isolate at 0 and 120 min, rated by 10 trained panellists with the same procedure and tested over time within each enzyme; however enzymes were not tested against each other, and the same % of enzyme differs in activity | |||||
| Bitterness intensity (sensory) | Soy: unhydrolysed (0 min) vs papain 0.2% for 120 min A 5% dispersion of soy protein isolate hydrolysed for 120 min at enzyme-specific temperature and pH (enzyme 0.5%, papain 0.2%); the 0 min value in the table is the 2.8 that the text gives as the bitterness of unhydrolysed SPI; rated as a 2.5% solution (pH 7.0); papain 0.2% at 80 °C and pH 7.0, degree of hydrolysis 4.6% at 120 min | 2.8 | 3.1 | 10 cm line scale from 0 (not detectable) to 10 (intense) (10 trained panellists, mean of 2 replicates) | Sample after "vs" more bitter Tested, not significant |
| Direction shows which published mean is larger. In Table 2, 0 min (a) and 120 min (ab) share a letter: not significant (which does not mean the same). At 10 min (4.3) and 30 min (4.8) bitterness was significantly higher than at 0 min. The letters in Table 2 compare 0 min with each time within one enzyme; enzymes were not tested against each other. The methods give ANOVA with Tukey and the Table 2 footnote gives Bonferroni, inconsistent Where: Methods, "Enzymatic hydrolysis of SPI", "Sensory analysis" and "Statistical analysis"; Results, "Effects of the enzymatic treatment on the bitterness of SPI"; Tables 1 and 2 For whether changing the enzyme or enzyme amount, the degree or time of hydrolysis, or the after-treatment of the resulting hydrolysate, for the same protein raw material, changes the intensity of bitterness people perceive: Usable — The same soy protein isolate at 0 and 120 min, rated by 10 trained panellists with the same procedure and tested over time within each enzyme; however enzymes were not tested against each other, and the same % of enzyme differs in activity | |||||
| Bitterness intensity (sensory) | Soy: unhydrolysed (0 min) vs Flavourzyme for 120 min A 5% dispersion of soy protein isolate hydrolysed for 120 min at enzyme-specific temperature and pH (enzyme 0.5%, papain 0.2%); the 0 min value in the table is the 2.8 that the text gives as the bitterness of unhydrolysed SPI; rated as a 2.5% solution (pH 7.0); Flavourzyme at 50 °C and pH 6.0, degree of hydrolysis 8.5% at 120 min | 2.8 | 2.1 | 10 cm line scale from 0 (not detectable) to 10 (intense) (10 trained panellists, mean of 2 replicates) | Sample before "vs" more bitter Tested, not significant |
| Direction shows which published mean is larger. In Table 2, 0 min (a) and 120 min (ab) share a letter, so it cannot be said to be significantly lower than unhydrolysed. At 30 min (4.5) bitterness was significantly higher than at 0 min. The text says bitterness rose to 4.3 within 1 h and fell to 2.1 at 2 h, below unhydrolysed SPI, but this is not read as a significant decrease. The letters in Table 2 compare 0 min with each time within one enzyme; enzymes were not tested against each other. The methods give ANOVA with Tukey and the Table 2 footnote gives Bonferroni, inconsistent Where: Methods, "Enzymatic hydrolysis of SPI", "Sensory analysis" and "Statistical analysis"; Results, "Effects of the enzymatic treatment on the bitterness of SPI"; Tables 1 and 2 For whether changing the enzyme or enzyme amount, the degree or time of hydrolysis, or the after-treatment of the resulting hydrolysate, for the same protein raw material, changes the intensity of bitterness people perceive: Usable — The same soy protein isolate at 0 and 120 min, rated by 10 trained panellists with the same procedure and tested over time within each enzyme; however enzymes were not tested against each other, and the same % of enzyme differs in activity | |||||
Schlegel 2019 sensory only
Enzymatic hydrolysis of lupin protein isolates—Changes in the molecular weight distribution, technofunctional characteristics, and sensory attributes. Food Science & Nutrition 7(8):2747-2759 — doi:10.1002/fsn3.1139
- How deeply checked
- Checked in full text — Full text via Europe PMC (sections 2.3, 2.7 and 2.8; results "Sensory analysis of the protein hydrolysates"; Tables 2, 3 and 6) Checked by: LLM (one extracted, another checked against the source; no human check), 2026-10-22.
- Samples
- Lupin (Lupinus angustifolius) protein isolate (LPI, 90% dry matter, 92% protein)
- Processing and cooking
- 9 proteases at enzyme-specific temperature and pH for 2 h in a dispersion of 50 g/kg protein; enzyme amount 0.5% (papain 0.2%), given in the footnote as an enzyme-to-solution ratio; inactivated at 90 °C for 20 min, neutralised and freeze-dried; controls without enzyme made under the same conditions; degree of hydrolysis (OPA method) 9.05% for Alcalase and 6.90% for Flavourzyme
- How it was served or eaten
- 1% (w/w) solution, 20 mL; bitterness rated with a nose clip
- Who rated, on what scale
- 10 panellists trained with caffeine threshold tests and other material; unstructured 10 cm line; reference was a 1% Alcalase hydrolysate of LPI; Table 6 gives medians (the methods say mean and SD); ANOVA with Tukey (p<0.05)
| Measured | Comparison and condition | Sample before "vs" | Sample after "vs" | Unit | Direction / test |
|---|---|---|---|---|---|
| Bitterness intensity (sensory) | Lupin: unhydrolysed vs Alcalase for 2 h Lupin protein isolate (50 g/kg protein) hydrolysed for 2 h at enzyme-specific temperature and pH (enzyme-to-solution ratio 0.5%); unhydrolysed controls made under the same conditions; rated as a 1% solution with a nose clip; Alcalase at 50 °C and pH 8.0, degree of hydrolysis 9.05% | 1.1 | 7.2 | median on an unstructured 10 cm line (10 trained panellists) | Sample after "vs" more bitter Tested, significant |
| In Table 6, unhydrolysed (a) and Alcalase (b) carry different letters, significant (p<0.05). Of the 9 enzymes, only Alcalase differed in letter from unhydrolysed (Protamex 2.4 is ab) Where: Sections 2.3, 2.7 and 2.8; Results, "Sensory analysis of the protein hydrolysates"; Tables 2, 3 and 6 For whether changing the enzyme or enzyme amount, the degree or time of hydrolysis, or the after-treatment of the resulting hydrolysate, for the same protein raw material, changes the intensity of bitterness people perceive: Usable — The same lupin protein isolate with a no-enzyme control made under the same conditions, rated by 10 trained panellists and tested; values are medians, and the enzyme amount is a ratio to the solution | |||||
| Bitterness intensity (sensory) | Lupin: unhydrolysed vs Flavourzyme for 2 h Lupin protein isolate (50 g/kg protein) hydrolysed for 2 h at enzyme-specific temperature and pH (enzyme-to-solution ratio 0.5%); unhydrolysed controls made under the same conditions; rated as a 1% solution with a nose clip; Flavourzyme at 50 °C and pH 6.0, degree of hydrolysis 6.90% | 1.1 | 0.7 | median on an unstructured 10 cm line (10 trained panellists) | Sample before "vs" more bitter Tested, not significant |
| Direction shows which published median is larger. In Table 6, unhydrolysed and Flavourzyme are both a: not significant (which does not mean the same). The text says the bitterness of all hydrolysates other than Alcalase remained similar to unhydrolysed LPI. Flavourzyme had the second-highest degree of hydrolysis after Alcalase, but its bitterness was on the low side Where: Sections 2.3, 2.7 and 2.8; Results, "Sensory analysis of the protein hydrolysates"; Tables 2, 3 and 6 For whether changing the enzyme or enzyme amount, the degree or time of hydrolysis, or the after-treatment of the resulting hydrolysate, for the same protein raw material, changes the intensity of bitterness people perceive: Usable — The same lupin protein isolate with a no-enzyme control made under the same conditions, rated by 10 trained panellists and tested; values are medians, and the enzyme amount is a ratio to the solution | |||||
Schlegel 2020 sensory only
Effect of enzyme‐assisted hydrolysis on protein pattern, technofunctional, and sensory properties of lupin protein isolates using enzyme combinations. Food Science & Nutrition 8(7):3041-3051 — doi:10.1002/fsn3.1286
- How deeply checked
- Checked in full text — Full text via Europe PMC (sections 2.3, 2.7 and 2.8; the results paragraph on bitterness; Tables 2 and 3; the Fig. 4 legend); Fig. 4 not obtained Checked by: LLM (one extracted, another checked against the source; no human check), 2026-10-22.
- Samples
- Lupin (Lupinus angustifolius, cultivar Boregine) protein isolate (LPI)
- Processing and cooking
- 5% (w/w) protein at 50 °C and pH 8.0; 12 combinations of Alcalase, papain, Corolase 7089 and Neutrase: 8 one-step (added together, 4 h) and 4 two-step (second enzyme added after 1 h, then 4 h more); inactivated at 90 °C for 20 min and neutralised; degree of hydrolysis (OPA method) 13.3% for OS 5 at 4 h
- How it was served or eaten
- Serving concentration not stated in the text (said to follow the earlier paper)
- Who rated, on what scale
- 10 panellists; the methods describe a 0 (no perception) to 10 (strong) scale with a nose clamp, and the Fig. 4 legend gives means and SD on a 10 cm line; ANOVA with Tukey
| Measured | Comparison and condition | Sample before "vs" | Sample after "vs" | Unit | Direction / test |
|---|---|---|---|---|---|
| Bitterness intensity (sensory) | Lupin: unhydrolysed vs Alcalase and papain together for 4 h (OS 5) Lupin protein isolate at 5%, 50 °C and pH 8.0; Alcalase 0.5% and papain 0.2% (enzyme-to-solution ratio) added together for 4 h, degree of hydrolysis 13.3%; serving concentration not in the text | 1.9 | 5.7 | 0 (no perception) to 10 (strong) scale; the Fig. 4 legend gives means on a 10 cm line (10 panellists) | Sample after "vs" more bitter Tested, significant |
| From the text: of the 12 combinations, only OS 5 was significantly higher than unhydrolysed. The combinations without Alcalase (papain with Corolase 7089) were 1.2 and 2.1. For the other combinations and the marks in Fig. 4: Figure not obtained, so values unconfirmed. TS 3, with the same enzymes added in 2 steps, is a different sample Where: Sections 2.3, 2.7 and 2.8; the results paragraph on bitterness; Tables 2 and 3; Fig. 4 (not obtained) For whether changing the enzyme or enzyme amount, the degree or time of hydrolysis, or the after-treatment of the resulting hydrolysate, for the same protein raw material, changes the intensity of bitterness people perceive: Usable — Unhydrolysed and hydrolysed samples of the same lupin protein isolate, with a test; however the serving concentration is not in the text and the values of the other combinations are only in a figure and unconfirmed | |||||
Sun 2022 sensory only
Characteristics of the enzyme-induced release of bitter peptides from wheat gluten hydrolysates. Frontiers in Nutrition 9:1022257 — doi:10.3389/fnut.2022.1022257
- How deeply checked
- Checked in full text — Full text via Europe PMC (methods "Preparation of wheat gluten hydrolysates", "Determination of degree of hydrolysis" and the sensory paragraph; results "Bitterness intensity of wheat gluten hydrolysates"; the Fig. 1 legend); Fig. 1 not obtained Checked by: LLM (one extracted, another checked against the source; no human check), 2026-10-22.
- Samples
- Wheat gluten
- Processing and cooking
- 15 g gluten in 300 mL deionised water; Alcalase or trypsin (both at an enzyme-to-substrate ratio of 1:20), stopped when the pH-stat degree of hydrolysis reached 4, 12 or 20% and heated in boiling water for 15 min; the centrifuged supernatant freeze-dried; hydrolysis time not given
- How it was served or eaten
- 1% solution
- Who rated, on what scale
- 8 experienced assessors (4 men and 4 women, aged 20–35); 10-point line scale anchored with caffeine at 0.3, 0.8 and 1.2 g/L as 2, 5 and 10; for testing, only p<0.05 in the Fig. 1 legend
| Measured | Comparison and condition | Sample before "vs" | Sample after "vs" | Unit | Direction / test |
|---|---|---|---|---|---|
| Bitterness intensity (sensory) | Degree of hydrolysis or enzyme: wheat gluten, Alcalase at 4% vs 20% degree of hydrolysis 15 g wheat gluten in 300 mL water; enzyme-to-substrate ratio 1:20, stopped when the pH-stat degree of hydrolysis reached 4% and 20%; the centrifuged supernatant freeze-dried and rated as a 1% solution; unhydrolysed gluten was not rated; Alcalase | 0.65 | 4.38 | 10-point line scale (anchored with caffeine; 8 experienced assessors) | Sample after "vs" more bitter Tested, significant |
| From the text: bitterness rose as the degree of hydrolysis increased and rose significantly at 20% (p<0.05 in the Fig. 1 legend). For the 12% values and the marks in Fig. 1: Figure not obtained, so values unconfirmed. At the same 20%, Alcalase gave 4.38 and trypsin 3.4, and the text calls Alcalase more bitter, but an individual test between these 2 was not confirmed Where: Methods, "Preparation of wheat gluten hydrolysates" and the sensory paragraph; Results, "Bitterness intensity of wheat gluten hydrolysates"; Fig. 1 (not obtained) For whether changing the enzyme or enzyme amount, the degree or time of hydrolysis, or the after-treatment of the resulting hydrolysate, for the same protein raw material, changes the intensity of bitterness people perceive: Usable — Usable for comparing degrees of hydrolysis (same gluten and enzyme, with a test). With no rating of unhydrolysed gluten, it cannot be used for whether hydrolysis makes protein more bitter than the unhydrolysed material; hydrolysis time not given | |||||
| Bitterness intensity (sensory) | Degree of hydrolysis or enzyme: wheat gluten, trypsin at 4% vs 20% degree of hydrolysis 15 g wheat gluten in 300 mL water; enzyme-to-substrate ratio 1:20, stopped when the pH-stat degree of hydrolysis reached 4% and 20%; the centrifuged supernatant freeze-dried and rated as a 1% solution; unhydrolysed gluten was not rated; trypsin | 0.59 | 3.4 | 10-point line scale (anchored with caffeine; 8 experienced assessors) | Sample after "vs" more bitter Tested, significant |
| From the text: bitterness rose as the degree of hydrolysis increased and rose significantly at 20% (p<0.05 in the Fig. 1 legend). For the 12% values and the marks in Fig. 1: Figure not obtained, so values unconfirmed. At the same 20%, Alcalase gave 4.38 and trypsin 3.4, and the text calls Alcalase more bitter, but an individual test between these 2 was not confirmed Where: Methods, "Preparation of wheat gluten hydrolysates" and the sensory paragraph; Results, "Bitterness intensity of wheat gluten hydrolysates"; Fig. 1 (not obtained) For whether changing the enzyme or enzyme amount, the degree or time of hydrolysis, or the after-treatment of the resulting hydrolysate, for the same protein raw material, changes the intensity of bitterness people perceive: Usable — Usable for comparing degrees of hydrolysis (same gluten and enzyme, with a test). With no rating of unhydrolysed gluten, it cannot be used for whether hydrolysis makes protein more bitter than the unhydrolysed material; hydrolysis time not given | |||||
Daher 2021 sensory only
Sensopeptidomic Kinetic Approach Combined with Decision Trees and Random Forests to Study the Bitterness during Enzymatic Hydrolysis Kinetics of Micellar Caseins. Foods 10(6):1312 — doi:10.3390/foods10061312
- How deeply checked
- Checked in full text — Full text via Europe PMC (sections 2.1, 2.2.1 and 3.1; the Fig. 1 legend; discussion); Fig. 1 not obtained Checked by: LLM (one extracted, another checked against the source; no human check), 2026-10-22.
- Samples
- Milk protein rich in micellar casein (92% casein)
- Processing and cooking
- Diluted to 10% of total nitrogenous matter and hydrolysed for 9 h with 2 enzyme sets: set 109, Flavourzyme with Protamex; set 125, Promod 523MDP with Flavorpro 937MDP; recipe (enzyme amounts, pH, temperature) confidential; sampled every hour, inactivated at 98 °C for 3 min and spray-dried; degree of hydrolysis by the OPA method
- How it was served or eaten
- Dissolved in mineral water at 10% dry matter, 20 mL at room temperature
- Who rated, on what scale
- 19 assessors trained for 20 months (12 women and 7 men, aged 45–65); 1 (low) to 7 (high); 2 replicates; two-way ANOVA with Duncan test (p≤0.05)
| Measured | Comparison and condition | Sample before "vs" | Sample after "vs" | Unit | Direction / test |
|---|---|---|---|---|---|
| Bitterness intensity (sensory) | Degree of hydrolysis or enzyme: casein, Flavourzyme with Protamex (set 109), 3 h vs 5 h Milk protein rich in micellar casein diluted to 10% of total nitrogenous matter, hydrolysed for 9 h and sampled every 1 h (recipe confidential); rated at 10% dry matter in mineral water; unhydrolysed casein was not rated; set 109 is Flavourzyme with Protamex; degree of hydrolysis 13.9% at 4 h, 44.1% at 5 h and 50.8% at 9 h | about 5.20 | 3.00 | 1 (low) to 7 (high) (19 trained assessors, 2 replicates) | Sample before "vs" more bitter Tested; significance reported, but which pair differs is unclear |
| From the text: about 5.20 for the first 3 h, falling to 3.00 between 3 and 5 h and staying about 3 afterwards; correlation between degree of hydrolysis and bitterness r = −0.933 (p≤0.001). The 9 samples were significantly discriminated by ANOVA, but which pairs of times differ is in the letters of Fig. 1. Figure not obtained, so values unconfirmed Where: Sections 2.1 and 2.2.1; section 3.1; Fig. 1 (not obtained); discussion For whether changing the enzyme or enzyme amount, the degree or time of hydrolysis, or the after-treatment of the resulting hydrolysate, for the same protein raw material, changes the intensity of bitterness people perceive: Limited — Usable for comparing hydrolysis times, but the recipe is confidential and which pairs of times differ is in a figure and unconfirmed; with no rating of unhydrolysed casein, it cannot be used for comparison with unhydrolysed material | |||||
| Bitterness intensity (sensory) | Degree of hydrolysis or enzyme: casein, Promod with Flavorpro (set 125), 5 h vs 9 h Milk protein rich in micellar casein diluted to 10% of total nitrogenous matter, hydrolysed for 9 h and sampled every 1 h (recipe confidential); rated at 10% dry matter in mineral water; unhydrolysed casein was not rated; set 125 is Promod 523MDP with Flavorpro 937MDP; degree of hydrolysis 10.9% at 3 h, 22.6% at 5 h and 28.9% at 9 h | 2.23 | 3.59 | 1 (low) to 7 (high) (19 trained assessors, 2 replicates) | Sample after "vs" more bitter Tested; significance reported, but which pair differs is unclear |
| From the text: the minimum was 2.18 at 2 h, and from 5 h bitterness rose from 2.23 to 3.59 as the degree of hydrolysis increased (the text also says bitterness appears stable overall). The 9 samples were significantly discriminated by ANOVA, but which pairs of times differ is in the letters of Fig. 1. Figure not obtained, so values unconfirmed. Sets 109 and 125 differ in both enzymes and recipe, so the difference cannot be attributed to the enzymes alone Where: Sections 2.1 and 2.2.1; section 3.1; Fig. 1 (not obtained); discussion For whether changing the enzyme or enzyme amount, the degree or time of hydrolysis, or the after-treatment of the resulting hydrolysate, for the same protein raw material, changes the intensity of bitterness people perceive: Limited — Usable for comparing hydrolysis times, but the recipe is confidential and which pairs of times differ is in a figure and unconfirmed; with no rating of unhydrolysed casein, it cannot be used for comparison with unhydrolysed material | |||||
Fetriyuna 2026 sensory only
Enhancing Consumer Acceptance of Cricket (Acheta domesticus) Protein Hydrolysate Through ProteAX‐Mediated Enzymatic Hydrolysis to Improve Techno‐Functional and Organoleptic Properties. Food Science & Nutrition 14(6):e72033 — doi:10.1002/fsn3.72033
- How deeply checked
- Checked in full text — Full text via Europe PMC (sections 2.2.1–2.2.3, 3.1 and 3.3; Tables 1, 5, 7 and 9) Checked by: LLM (one extracted, another checked against the source; no human check), 2026-10-22.
- Samples
- House cricket (Acheta domesticus), head and antennae removed, blended with water at 1:2
- Processing and cooking
- Pasteurised at 90 °C for 15 min, then ProteAX at 0.5, 1.0, 1.5 or 2% (w/v) for 2, 3.5 or 5 h at 60 °C and pH 7.0 (12 conditions); inactivated at 90 °C for 20 min and the centrifuged supernatant freeze-dried (CPH); the raw CP is described only as freeze-dried without enzyme treatment, and whether it went through the same centrifugation and supernatant step is not stated; degree of hydrolysis (TCA-soluble nitrogen method) 6.85% for CP and 50.34–67.71% for CPH
- How it was served or eaten
- Sample held on the tongue for 10 s and spat out; serving concentration and solvent not stated
- Who rated, on what scale
- 25 semi-trained panellists given prior instructions (excluded if under 16 or over 25); gLMS (150 mm); two-way ANOVA (enzyme amount and time) with Duncan test
| Measured | Comparison and condition | Sample before "vs" | Sample after "vs" | Unit | Direction / test |
|---|---|---|---|---|---|
| Bitterness intensity (sensory) | Cricket: raw material dried without enzyme (CP; equivalence of processing unconfirmed) vs ProteAX 1.5% for 5 h Crickets blended and pasteurised at 90 °C for 15 min, then ProteAX 1.5% (w/v) at 60 °C and pH 7.0 for 5 h; the hydrolysate was the freeze-dried centrifuged supernatant (degree of hydrolysis 67.71%, TCA-soluble nitrogen method); CP is described only as freeze-dried without enzyme treatment, and whether it went through the same centrifugation and supernatant step is not stated (degree of hydrolysis 6.85% for CP); serving concentration and solvent not stated | 32.88 | 5.92 | gLMS (150 mm, in mm; 25 semi-trained panellists) | Sample before "vs" more bitter No test reported |
| Direction shows which published mean is larger. The two-way ANOVA covers only the 12 conditions of enzyme amount (4 levels, 0.5–2%) and time, and no test of CP against each condition is shown. All 12 conditions were 5.92–16.16 mm, lower than CP. Since CP and the hydrolysates cannot be confirmed to share the same fractionation step, this cannot be called the effect of hydrolysis alone. Aftertaste (after 5 min) was also 20.00 for CP against 2.80 for this condition Where: Sections 2.2.1 and 2.2.2; section 3.3; Tables 5 and 6 For whether changing the enzyme or enzyme amount, the degree or time of hydrolysis, or the after-treatment of the resulting hydrolysate, for the same protein raw material, changes the intensity of bitterness people perceive: Limited — Raw material and hydrolysate of the same crickets, but whether the raw material (CP) went through the same centrifugation and supernatant step as the hydrolysates is not stated, and there is no test of CP against the hydrolysates | |||||
| Bitterness intensity (sensory) | Degree of hydrolysis or enzyme: cricket, ProteAX for 2 h vs 5 h (mean over enzyme amounts) Pasteurised crickets with ProteAX at 0.5, 1.0, 1.5 or 2% (w/v) at 60 °C and pH 7.0 for 2, 3.5 or 5 h; the centrifuged supernatant freeze-dried (degree of hydrolysis 50.34–67.71%, TCA-soluble nitrogen method); serving concentration and solvent not stated | 14.18 (2 h, marginal mean over 4 enzyme amounts) | 7.32 (5 h, likewise) | gLMS (150 mm, in mm; 25 semi-trained panellists) | Sample before "vs" more bitter Tested, significant |
| In the Duncan test of Table 9, 2 h (c), 3.5 h (b, 10.60) and 5 h (a) carry different letters, significant (p<0.05). In the two-way ANOVA, time and enzyme amount were significant and the interaction was not (p = 0.086). This compares marginal means, not individual pairs of conditions (for example 16.16 and 5.92). The marginal means for enzyme amount were 12.24 at 0.5% and 11.30 at 1% (b), and 9.39 at 1.5% and 9.87 at 2% (a) Where: Sections 2.2.1 and 2.2.2; section 3.3; Tables 5, 7 and 9 For whether changing the enzyme or enzyme amount, the degree or time of hydrolysis, or the after-treatment of the resulting hydrolysate, for the same protein raw material, changes the intensity of bitterness people perceive: Limited — Usable for comparing hydrolysis times, but the test is of marginal means over enzyme amounts, not individual pairs of conditions, and the serving concentration is not stated; not a comparison with unhydrolysed material | |||||
Zhao 2026 sensory only
Impact of Aspergillus oryzae-Derived Aminopeptidase Complex in Developing the Flavor Profile of Clam Hydrolysate. Foods 15(10):1753 — doi:10.3390/foods15101753
- How deeply checked
- Checked in full text — Full text via Europe PMC (sections 2.1, 2.6, 2.7 and 2.13; sections 3.2 and 3.3; Tables 1 and 2); Fig. 2 with the bitterness values not obtained Checked by: LLM (one extracted, another checked against the source; no human check), 2026-10-22.
- Samples
- Boiled and frozen Manila clam (Ruditapes philippinarum) meat, viscera removed, homogenised with water at 1:1 and freeze-dried to powder
- Processing and cooking
- Powder and water at 1:50; alkaline protease (5000 U/g protein, 50 °C, pH 9.5) for 4 h (AP group); AOAP, an aminopeptidase complex (5000 U/g, 50 °C, pH 7.0), added to the AP hydrolysate for 3 h more (two-stage group; degree of hydrolysis given in the text as rising from 23.9% to 49.8%); an AOAP-only group (4 h) and freeze-dried clam protein (control) were also compared; hydrolysates were centrifuged supernatants
- How it was served or eaten
- Served in 20 mL cups; serving concentration not stated
- Who rated, on what scale
- 12 trained assessors (6 men and 6 women, aged 20–25); 0–10 scale (1–2 no bitterness, 9–10 unbearably bitter); mean of 3 replicates; ANOVA with Tukey (p<0.05)
| Measured | Comparison and condition | Sample before "vs" | Sample after "vs" | Unit | Direction / test |
|---|---|---|---|---|---|
| Bitterness intensity (sensory) | Clam: unhydrolysed control (equivalence of processing unconfirmed) vs alkaline protease for 4 h Clam meat powder in water at 1:50; alkaline protease (5000 U/g protein, 50 °C, pH 9.5) for 4 h; the control was freeze-dried clam protein, and the hydrolysates were centrifuged supernatants; serving concentration not stated | Figure not obtained, so values unconfirmed | Figure not obtained, so values unconfirmed | 0–10 scale (1–2 no bitterness, 9–10 unbearably bitter; 12 trained assessors, mean of 3 replicates) | Sample after "vs" more bitter No test reported |
| From the text: bitterness is said to have increased substantially with alkaline protease, but the text gives no individual significance statement for this comparison. The statistics section gives ANOVA with Tukey in general, but the marks in Fig. 2 were not obtained, so whether this comparison is significant is unconfirmed. The values are also in Fig. 2. Figure not obtained, so values unconfirmed. The control was freeze-dried protein while the hydrolysates were centrifuged supernatants, so the fractionation step may also differ. The source does not call the alkaline protease Alcalase Where: Sections 2.6, 2.7 and 2.13; sections 3.2 and 3.3; Fig. 2 (not obtained) For whether changing the enzyme or enzyme amount, the degree or time of hydrolysis, or the after-treatment of the resulting hydrolysate, for the same protein raw material, changes the intensity of bitterness people perceive: Limited — Unhydrolysed and hydrolysed samples of the same clam, but the control was freeze-dried protein while the hydrolysates were centrifuged supernatants, values are only in a figure and unconfirmed, and the text gives no test result for this comparison | |||||
| Bitterness intensity (sensory) | After-treatment of the hydrolysate: clam, alkaline protease for 4 h vs then AOAP for 3 h Clam meat powder in water at 1:50 with alkaline protease (5000 U/g protein, 50 °C, pH 9.5) for 4 h; AOAP, an aminopeptidase complex (5000 U/g, 50 °C, pH 7.0), then added to that hydrolysate for 3 h more (degree of hydrolysis given in the text as rising from 23.9% to 49.8%); serving concentration not stated | Figure not obtained, so values unconfirmed | Figure not obtained, so values unconfirmed | 0–10 scale (1–2 no bitterness, 9–10 unbearably bitter; 12 trained assessors, mean of 3 replicates) | Sample before "vs" more bitter Tested, significant |
| From the text: adding AOAP significantly reduced bitterness (ANOVA with Tukey, p<0.05), and umami is said to have increased significantly. Values and marks are in Fig. 2. Figure not obtained, so values unconfirmed. The acceptability order (two-stage > AOAP only > control > alkaline protease) is not an order of bitterness alone Where: Sections 2.6, 2.7 and 2.13; sections 3.2 and 3.3; Fig. 2 (not obtained) For whether changing the enzyme or enzyme amount, the degree or time of hydrolysis, or the after-treatment of the resulting hydrolysate, for the same protein raw material, changes the intensity of bitterness people perceive: Limited — Before and after adding AOAP to an existing hydrolysate; the text reports significance, but values and marks are in a figure and unconfirmed | |||||
Jiang 2024 sensory only
Lactiplantibacillus plantarum for the Preparation of Fermented Low-Bitter Enzymatic Skim Milk with Antioxidant Ability. Foods 13(23):3828 — doi:10.3390/foods13233828
- How deeply checked
- Checked in full text — Full text via Europe PMC (sections 2.2–2.4 and 2.11; sections 3.1 and 3.2.1; Table 1; the Fig. 2 legend) Checked by: LLM (one extracted, another checked against the source; no human check), 2026-10-22.
- Samples
- Skim milk powder made up to 12% (w/w) with water
- Processing and cooking
- Neutrase (50 °C, pH 7.0, 10,000 U/g), Flavourzyme (50 °C, pH 7.0, 6000 U/g) or Protamex (55 °C, pH 7.0, 10,000 U/g) for 4.5 h; inactivated at 105 °C for 5 min; degree of hydrolysis (pH-stat method) 28.13% for Protamex, 24.89% for Neutrase and 23.20% for Flavourzyme
- How it was served or eaten
- At room temperature; serving concentration not stated
- Who rated, on what scale
- 10 food-science graduate students (5 men and 5 women) scoring against quinine standards set at 0, 2, 4, 6, 8 and 10 points; no training described; ANOVA with Tukey (p<0.05)
| Measured | Comparison and condition | Sample before "vs" | Sample after "vs" | Unit | Direction / test |
|---|---|---|---|---|---|
| Bitterness intensity (sensory) | Degree of hydrolysis or enzyme: skim milk, Flavourzyme vs Neutrase 12% (w/w) skim milk hydrolysed for 4.5 h and inactivated at 105 °C for 5 min; Neutrase at 50 °C and 10,000 U/g (degree of hydrolysis 24.89%), Flavourzyme at 50 °C and 6000 U/g (23.20%), Protamex at 55 °C and 10,000 U/g (28.13%), all at pH 7.0; serving concentration not stated; unhydrolysed skim milk was not rated | 6.60 | 7.60 | score against quinine standards set from 0 to 10 points (10 food-science graduate students) | Sample after "vs" more bitter Tested, significant |
| From the text: Neutrase was significantly higher than Protamex (6.80) and Flavourzyme (p<0.05). The degree of hydrolysis differed among the 3, and the activity units (U/g) are defined differently for each enzyme Where: Sections 2.2–2.4 and 2.11; section 3.2.1; Fig. 2 For whether changing the enzyme or enzyme amount, the degree or time of hydrolysis, or the after-treatment of the resulting hydrolysate, for the same protein raw material, changes the intensity of bitterness people perceive: Limited — Usable for comparing enzymes, but degree of hydrolysis and enzyme units differ by enzyme, so the enzyme alone cannot be isolated; serving concentration not stated; with no rating of unhydrolysed milk, it cannot be used for comparison with unhydrolysed material | |||||
| Bitterness intensity (sensory) | Degree of hydrolysis or enzyme: skim milk, Protamex vs Neutrase 12% (w/w) skim milk hydrolysed for 4.5 h and inactivated at 105 °C for 5 min; Neutrase at 50 °C and 10,000 U/g (degree of hydrolysis 24.89%), Flavourzyme at 50 °C and 6000 U/g (23.20%), Protamex at 55 °C and 10,000 U/g (28.13%), all at pH 7.0; serving concentration not stated; unhydrolysed skim milk was not rated | 6.80 | 7.60 | score against quinine standards set from 0 to 10 points (10 food-science graduate students) | Sample after "vs" more bitter Tested, significant |
| From the text: Neutrase was significantly higher than Protamex and Flavourzyme (6.60) (p<0.05). A test between Protamex and Flavourzyme cannot be confirmed from the text Where: Sections 2.2–2.4 and 2.11; section 3.2.1; Fig. 2 For whether changing the enzyme or enzyme amount, the degree or time of hydrolysis, or the after-treatment of the resulting hydrolysate, for the same protein raw material, changes the intensity of bitterness people perceive: Limited — Usable for comparing enzymes, but degree of hydrolysis and enzyme units differ by enzyme, so the enzyme alone cannot be isolated; serving concentration not stated; with no rating of unhydrolysed milk, it cannot be used for comparison with unhydrolysed material | |||||
Sugiyama 1991 sensory only
各種プロテアーゼ処理によって得られたマイワシタンパク加水分解物の特性. 日本水産学会誌 57(3):475-479 — doi:10.2331/suisan.57.475
- How deeply checked
- Checked in full text — J-STAGE PDF (pp. 475–478; methods; results on sensory evaluation of bitterness and on removal of bitterness; Tables 1–4); the tables were read from page images Checked by: LLM (one extracted, another checked against the source; no human check), 2026-10-22.
- Samples
- Defatted sardine meal
- Processing and cooking
- Protein concentration 8% in the reaction, enzyme at 8% of protein; 8 proteases for 6 h with pH-stat: alkaline (Alcalase 55 °C, pH 8.0; Actinase and Proleather 50 °C, pH 8.0), neutral (Neutrase 45 °C, pH 7.0; papain 55 °C, pH 7.0) and acidic (Molsin and Pepsin 50 °C, pH 3.0; Neulase 45 °C, pH 3.0); inactivated by shifting pH, neutralised, and the centrifuged supernatant freeze-dried; no degree-of-hydrolysis values
- How it was served or eaten
- 50 mL of a 5% protein solution in a 100 mL bottle, held at 30 °C for 30 min before rating
- Who rated, on what scale
- Mean of a 10-member panel; 0 (none, distilled water), 1 (hardly), 2 (slightly), 3 (somewhat), 4 (strongly), 5 (very strongly); no statistical test described
| Measured | Comparison and condition | Sample before "vs" | Sample after "vs" | Unit | Direction / test |
|---|---|---|---|---|---|
| Bitterness intensity (sensory) | Degree of hydrolysis or enzyme: sardine, Alcalase vs Proleather (both alkaline) Defatted sardine meal at 8% protein with enzyme at 8% of protein for 6 h; Alcalase at 55 °C and pH 8.0, Proleather at 50 °C and pH 8.0; rated as a 5% protein solution at 30 °C; unhydrolysed meal was not rated | 1.7 | 4.0 | 6 grades from 0 (none) to 5 (very strongly), mean of a 10-member panel | Sample after "vs" more bitter No test reported |
| Direction shows which published mean is larger. No test is described. Other values in Table 2: Actinase 1.8, Neutrase 4.0, papain 3.5, Molsin 2.7, Pepsin 3.2, Neulase 2.7. Even among the alkaline enzymes the values split into 1.7 and 4.0, and the text says bitterness did not follow protein solubilisation or tryptophan content Where: pp. 475–477; methods, sensory evaluation of bitterness; results, sensory evaluation of bitterness; Tables 1 and 2 For whether changing the enzyme or enzyme amount, the degree or time of hydrolysis, or the after-treatment of the resulting hydrolysate, for the same protein raw material, changes the intensity of bitterness people perceive: Limited — Usable for comparing enzymes, but with no test; with no rating of unhydrolysed meal, it cannot be used for comparison with unhydrolysed material | |||||
| Bitterness intensity (sensory) | After-treatment of the hydrolysate: sardine Alcalase hydrolysate, untreated vs aminopeptidase M 3 mg Alcalase hydrolysate of defatted sardine meal (5% protein, pH 7.0), 100 mL, with aminopeptidase M added at 3 mg as protein, 37 °C for 3 h; inactivated at 90 °C for 5 min and filtered | 1.7 | 0.9 | 6 grades from 0 (none) to 5 (very strongly), mean of a 10-member panel | Sample before "vs" more bitter No test reported |
| Direction shows which published mean is larger. No test is described. Other values in Table 3: leucine aminopeptidase 1.4 at 1 mg and 1.0 at 3 mg, aminopeptidase M 1.0 at 1 mg, carboxypeptidase A 1.3 at 1 mg and 1.0 at 3 mg. The text also calls the original Alcalase hydrolysate (1.7) not a practical problem Where: p. 476, methods on peptidase treatment; p. 477, results on removal of bitterness; Table 3 For whether changing the enzyme or enzyme amount, the degree or time of hydrolysis, or the after-treatment of the resulting hydrolysate, for the same protein raw material, changes the intensity of bitterness people perceive: Limited — Before and after treating an existing hydrolysate, but with no test | |||||
| Bitterness intensity (sensory) | After-treatment of the hydrolysate: sardine Alcalase hydrolysate, untreated vs synthetic adsorbent VP-10 Alcalase hydrolysate of defatted sardine meal (5% protein, pH 7.0), 150 mL, passed through a column of 15 mL of the vinylpyridine resin VP-10 (SV 2), and the eluate freeze-dried and rated at 5% protein | 1.7 | 0.5 | 6 grades from 0 (none) to 5 (very strongly), mean of a 10-member panel | Sample before "vs" more bitter No test reported |
| Direction shows which published mean is larger. No test is described. Other resins in Table 4: XFS-4022 1.0, HP-2MG 0.7. Tryptophan also fell from 1.07 to 0.68 g/16 gN Where: p. 476, methods on synthetic adsorbent treatment; pp. 477–478, results on removal of bitterness; Table 4 For whether changing the enzyme or enzyme amount, the degree or time of hydrolysis, or the after-treatment of the resulting hydrolysate, for the same protein raw material, changes the intensity of bitterness people perceive: Limited — Before and after treating an existing hydrolysate, but with no test | |||||
Knežević-Jugović 2023 sensory only
Impact of Different Enzymatic Processes on Antioxidant, Nutritional and Functional Properties of Soy Protein Hydrolysates Incorporated into Novel Cookies. Foods 12(1):24 — doi:10.3390/foods12010024
- How deeply checked
- Checked in full text — Full text via Europe PMC (sections 2.3, 2.4, 2.10.2 and 2.13; sections 3.1 and 3.4; Table 2) Checked by: LLM (one extracted, another checked against the source; no human check), 2026-10-22.
- Samples
- Soy protein concentrate (SPC)
- Processing and cooking
- 8% (w/v) protein; Alcalase (pH 8, 55 °C), papain (pH 8, 50 °C), Neutrase (pH 7, 45 °C), Umamizyme (pH 8, 50 °C) and Everlase (pH 9, 60 °C) at 4% enzyme (by weight of protein); two-step: Alcalase or Neutrase for 75 min, then adjusted to 50 °C and pH 7 and Flavourzyme 4% added for another 120 min; inactivated at 90 °C for 15 min, insolubles removed by centrifugation and spray-dried; the hydrolysis time of the one-step samples used for sensory testing was not confirmed in the text
- How it was served or eaten
- 1% solutions served at room temperature, each paired with a commercial soy protein hydrolysate (R)
- Who rated, on what scale
- 20 assessors selected under ISO 8586; paired comparison (ISO 5495) on a 5-point bipolar scale, less bitter than R negative, similar 0, more bitter positive; results as sums of classes; Friedman test (5%) only among the 5 samples below R; the statistics section gives one-way ANOVA with Tukey in general
| Measured | Comparison and condition | Sample before "vs" | Sample after "vs" | Unit | Direction / test |
|---|---|---|---|---|---|
| Sum of bitterness classes against a reference sample (sensory) | Compared with a commercial hydrolysate: soy, Neutrase alone vs Neutrase then Flavourzyme Soy protein concentrate (8% protein); 2-step samples (first enzyme for 75 min, then adjusted to 50 °C and pH 7 with Flavourzyme 4% for another 120 min) set beside the one-step samples; insolubles removed and spray-dried; each rated as a 1% solution paired with a commercial soy protein hydrolysate (R); unhydrolysed concentrate was not rated; Neutrase at pH 7 and 45 °C, enzyme 4% | +14 | −25 | sum of classes on a 5-point bipolar scale against R (positive means more bitter than R, negative less bitter than R; 20 assessors) | Sample before "vs" more bitter Tested; significance reported, but which pair differs is unclear |
| The authors write that adding Flavourzyme significantly reduced the bitterness of the Neutrase hydrolysate. However, the published Friedman test covers only the 5 samples less bitter than R and does not include the Neutrase-only sample (+14), so the details of an individual test for this pair cannot be confirmed (not treated as a confirmed significant difference). Direction shows which published sum of classes is larger; these are not bitterness intensities or sizes of difference. The statistics section gives one-way ANOVA with Tukey in general Where: Sections 2.3, 2.10.2 and 2.13; section 3.4; Table 2 For whether changing the enzyme or enzyme amount, the degree or time of hydrolysis, or the after-treatment of the resulting hydrolysate, for the same protein raw material, changes the intensity of bitterness people perceive: Limited — Usable for comparing the addition of a second enzyme, but the values are sums of classes against a commercial hydrolysate and an individual test for this pair cannot be confirmed; since the reference is not the unhydrolysed material, it cannot be used for comparison with unhydrolysed material | |||||
| Sum of bitterness classes against a reference sample (sensory) | Compared with a commercial hydrolysate: soy, Alcalase alone vs Alcalase then Flavourzyme Soy protein concentrate (8% protein); 2-step samples (first enzyme for 75 min, then adjusted to 50 °C and pH 7 with Flavourzyme 4% for another 120 min) set beside the one-step samples; insolubles removed and spray-dried; each rated as a 1% solution paired with a commercial soy protein hydrolysate (R); unhydrolysed concentrate was not rated; Alcalase at pH 8 and 55 °C, enzyme 4% | −15 | −7 | sum of classes on a 5-point bipolar scale against R (positive means more bitter than R, negative less bitter than R; 20 assessors) | Sample after "vs" more bitter Tested; significance reported, but which pair differs is unclear |
| The authors write that for the Alcalase hydrolysate, adding Flavourzyme even increased bitterness. Both are on the less-bitter side of R, and the Friedman test over the 5 samples including these 2 was significant overall (5%), but whether this pair differs significantly cannot be confirmed. Direction shows which published sum of classes is larger; these are not bitterness intensities or sizes of difference. The statistics section gives one-way ANOVA with Tukey in general Where: Sections 2.3, 2.10.2 and 2.13; section 3.4; Table 2 For whether changing the enzyme or enzyme amount, the degree or time of hydrolysis, or the after-treatment of the resulting hydrolysate, for the same protein raw material, changes the intensity of bitterness people perceive: Limited — Usable for comparing the addition of a second enzyme, but the values are sums of classes against a commercial hydrolysate and an individual test for this pair cannot be confirmed; since the reference is not the unhydrolysed material, it cannot be used for comparison with unhydrolysed material | |||||
Nakamura 2023 sensory only
Reduction in the bitterness of protein hydrolysates by an aminopeptidase from Aspergillus oryzae. Food Science and Technology Research 29(1):71-77 — doi:10.3136/fstr.FSTR-D-22-00110
- How deeply checked
- Checked in full text — J-STAGE PDF (pp. 71–74; methods "Debittering activity assay of the purified aminopeptidase" and the sensory paragraph; results; Figs. 1 and 2) Checked by: LLM (one extracted, another checked against the source; no human check), 2026-10-22.
- Samples
- Milk casein and cod protein (both 10%, w/v)
- Processing and cooking
- Pepsin at pH 2.0 (enzyme-to-substrate ratio 1/300); casein at 37 °C for 4 h and 16 h, cod for 2 h and 5 h; neutralised and inactivated at 100 °C for 15 min, and the centrifuged supernatant used as the hydrolysate; hydrolysates adjusted to pH 8.0 and the aminopeptidase LapA purified from Proteax added at 1/24,500 for casein and 1/80,000 for cod at 50 °C; sampled at 1, 2, 4, 8 and 21 h; no degree-of-hydrolysis values
- How it was served or eaten
- Casein hydrolysate served diluted 6-fold because of its strong bitterness
- Who rated, on what scale
- 9 panellists (2 men and 6 women in their 20s, 1 man in his 60s); 0 (no bitterness) to 5 (maximum), with Gly-Phe at 0.5, 1.0 and 1.5% as 1, 2 and 3; t-test against the 0 h LapA sample
| Measured | Comparison and condition | Sample before "vs" | Sample after "vs" | Unit | Direction / test |
|---|---|---|---|---|---|
| Bitterness intensity (sensory) | After-treatment of the hydrolysate: pepsin hydrolysate of casein, aminopeptidase LapA at 0 h vs 4 h Supernatant of 10% casein hydrolysed with pepsin (1/300) at 37 °C for 20 h in total, adjusted to pH 8.0, with purified LapA added at 1/24,500 at 50 °C; rated diluted 6-fold | 3.5 | 1.7 | 0 (no bitterness) to 5 (maximum), with Gly-Phe at 0.5, 1.0 and 1.5% as 1, 2 and 3 (9 panellists) | Sample before "vs" more bitter Tested, significant |
| From the text and Fig. 1: t-test against 0 h of LapA, p<0.01. The text says bitterness hardly changed further up to 21 h, while free amino acids kept increasing. The reference is the pepsin-hydrolysed sample, not unhydrolysed casein Where: pp. 72–74; methods, "Debittering activity assay of the purified aminopeptidase" and the sensory paragraph; results; Figs. 1 and 2 For whether changing the enzyme or enzyme amount, the degree or time of hydrolysis, or the after-treatment of the resulting hydrolysate, for the same protein raw material, changes the intensity of bitterness people perceive: Usable — Usable for comparing after-treatment of an existing hydrolysate (before and after adding an aminopeptidase to the same hydrolysate, with a test); since the reference is the pepsin hydrolysate, it cannot be used for comparison with the unhydrolysed raw material | |||||
| Bitterness intensity (sensory) | After-treatment of the hydrolysate: pepsin hydrolysate of cod protein, aminopeptidase LapA at 0 h vs 21 h Supernatant of 10% cod protein hydrolysed with pepsin (1/300) at 37 °C for 7 h in total, adjusted to pH 8.0, with purified LapA added at 1/80,000 at 50 °C | 2.9 | 0.54 | 0 (no bitterness) to 5 (maximum), with Gly-Phe at 0.5, 1.0 and 1.5% as 1, 2 and 3 (9 panellists) | Sample before "vs" more bitter Tested, significant |
| From the text and Fig. 2: t-test against 0 h of LapA, p<0.01. The text says most panellists did not report bitterness at this level. The reference is the pepsin-hydrolysed sample, not unhydrolysed cod Where: pp. 72–74; methods, "Debittering activity assay of the purified aminopeptidase" and the sensory paragraph; results; Figs. 1 and 2 For whether changing the enzyme or enzyme amount, the degree or time of hydrolysis, or the after-treatment of the resulting hydrolysate, for the same protein raw material, changes the intensity of bitterness people perceive: Usable — Usable for comparing after-treatment of an existing hydrolysate (before and after adding an aminopeptidase to the same hydrolysate, with a test); since the reference is the pepsin hydrolysate, it cannot be used for comparison with the unhydrolysed raw material | |||||
Wu 2026 sensory only
Peptidomic evidence for pH-stat flavourzyme-mediated debittering of Alcalase-derived rice protein hydrolysates. Current Research in Food Science 13:101529 — doi:10.1016/j.crfs.2026.101529
- How deeply checked
- Checked in full text — Full text via Europe PMC (sections 2.2, 2.3 and 2.6; sections 3.2 and 3.3; Tables 1 and 2; the Fig. 3 legend); supplementary Tables S7 and S8 not obtained Checked by: LLM (one extracted, another checked against the source; no human check), 2026-10-22.
- Samples
- Rice protein concentrate
- Processing and cooking
- First stage: rice protein in water at 1:19 (w/w) with Alcalase 5% (enzyme-to-substrate ratio, w/w), pH 9, 55 °C, pH-stat, for 6 h (A6; DH-OPA 24.69%); its supernatant then treated with Flavourzyme (enzyme-to-substrate ratio 5%, pH 6, 60 °C) with pH held at 6.0 (pH-stat, s) or left to fall (f, from pH 6.0 to 5.1) for 3 or 4 h (A6-f3, A6-f4, A6-s3, A6-s4; DH-OPA 37.14–40.35%); inactivated at 95 °C for 15 min, and the permeate of the centrifuged supernatant through a 2500 Da membrane freeze-dried
- How it was served or eaten
- Fractions below 2500 Da reconstituted to equal nitrogen concentration by the Kjeldahl method
- Who rated, on what scale
- Formal evaluation by 15 panellists comparing all 10 pairs of 5 samples over 2 sessions, rating the difference in bitterness from 0 (no difference) to 7 (extremely different) and choosing the more bitter one; Wilcoxon signed-rank and sign tests with Bonferroni correction for 10 comparisons; a preliminary evaluation before filtration (5 panellists) was used only to choose times
| Measured | Comparison and condition | Sample before "vs" | Sample after "vs" | Unit | Direction / test |
|---|---|---|---|---|---|
| Difference-in-bitterness score for a pair of samples (sensory) | After-treatment of the hydrolysate: rice, Alcalase hydrolysate (A6) vs Flavourzyme with pH held for 3 h (A6-s3) The 6 h Alcalase hydrolysate of rice protein (A6, DH-OPA 24.69%) given a second stage with Flavourzyme (5%, pH 6, 60 °C); every sample rated as the 2500 Da membrane permeate at equal nitrogen concentration; unhydrolysed rice protein was not rated; A6-s3 held at pH 6.0 for 3 h (DH-OPA 37.14%) | Reference (one side of a difference-score pair; no intensity value) | difference score from A6, 3.64 (A6 more bitter) | difference-in-bitterness score per pair from 0 (no difference) to 7 (extremely different), with a choice of the more bitter sample (15 panellists, 2 sessions) | Sample before "vs" more bitter Tested, significant |
| Difference scores in Table 2: significant (p<0.05) by Wilcoxon signed-rank and sign tests with Bonferroni correction for 10 comparisons. Table 2 gives 3.64 in the A6 row and 3.63 in the A6-s3 row, inconsistent. A6-s3 is at the least-bitter end, but its difference from A6-s4 was not significant. This compares fractions below 2500 Da, not whole hydrolysates. In the preliminary evaluation before filtration (5 panellists), A6-s3 was the most bitter, the reverse order Where: Sections 2.3 and 2.6; section 3.3; Table 2 For whether changing the enzyme or enzyme amount, the degree or time of hydrolysis, or the after-treatment of the resulting hydrolysate, for the same protein raw material, changes the intensity of bitterness people perceive: Limited — Usable for comparing after-treatment (a second-stage hydrolysis), paired at equal nitrogen concentration and tested, but the samples compared are fractions below 2500 Da, not whole hydrolysates; values are difference scores; it cannot be used for comparison with unhydrolysed material | |||||
| Difference-in-bitterness score for a pair of samples (sensory) | After-treatment of the hydrolysate: rice, Alcalase hydrolysate (A6) vs Flavourzyme without pH control for 4 h (A6-f4) The 6 h Alcalase hydrolysate of rice protein (A6, DH-OPA 24.69%) given a second stage with Flavourzyme (5%, pH 6, 60 °C); every sample rated as the 2500 Da membrane permeate at equal nitrogen concentration; unhydrolysed rice protein was not rated; A6-f4 without pH control (falling from 6.0 to 5.1) for 4 h (DH-OPA 40.35%) | Reference (one side of a difference-score pair; no intensity value) | difference score from A6, 2.78 (A6 more bitter) | difference-in-bitterness score per pair from 0 (no difference) to 7 (extremely different), with a choice of the more bitter sample (15 panellists, 2 sessions) | Sample before "vs" more bitter Tested, significant |
| Difference scores in Table 2: significant after correction (p<0.05). All 4 second-stage samples were judged less bitter than A6 (A6-f3 3.75, A6-s4 4.03). The text says that without pH control 3 h was less bitter than 4 h, and that at the same time holding pH gave a less bitter sample. This compares fractions below 2500 Da Where: Sections 2.3 and 2.6; section 3.3; Table 2 For whether changing the enzyme or enzyme amount, the degree or time of hydrolysis, or the after-treatment of the resulting hydrolysate, for the same protein raw material, changes the intensity of bitterness people perceive: Limited — Usable for comparing after-treatment (a second-stage hydrolysis), paired at equal nitrogen concentration and tested, but the samples compared are fractions below 2500 Da, not whole hydrolysates; values are difference scores; it cannot be used for comparison with unhydrolysed material | |||||
Aspevik 2026 sensory only
Upcycling of poultry protein hydrolysates using membrane filtration technology – Effects on sensory properties and chemical composition. Food Chemistry: X 38:104181 — doi:10.1016/j.fochx.2026.104181
- How deeply checked
- Checked in full text — Full text via Europe PMC (sections 2.1, 2.2.2, 2.2.5 and 2.2.6; Table 3 and the paragraphs after it) Checked by: LLM (one extracted, another checked against the source; no human check), 2026-10-22.
- Samples
- Enzymatic protein hydrolysate concentrate from poultry side streams (60% dry matter, supplied by a company); the original enzyme, degree of hydrolysis and time are not given
- Processing and cooking
- Diluted to about 7% dry matter and microfiltered at 0.1 μm (MF, about 60 °C); the MF permeate nanofiltered at 200 Da (NF), and the retentate concentrated to half volume before adding water for diafiltration (NF-dia); operating conditions were not systematically controlled, the authors say
- How it was served or eaten
- 2% (w/w) dry matter in tap water; 20 mL at room temperature (20 ± 2 °C) under red light
- Who rated, on what scale
- 10 assessors trained and tested under ISO 8586; unstructured 15 cm line with end points 1 (no intensity) and 9 (high intensity); 2 replicates; two-way ANOVA with Tukey (Table 3 caption)
| Measured | Comparison and condition | Sample before "vs" | Sample after "vs" | Unit | Direction / test |
|---|---|---|---|---|---|
| Bitterness intensity (sensory) | After-treatment of the hydrolysate: poultry hydrolysate, before membrane processing (crude) vs after microfiltration, nanofiltration and diafiltration A company-supplied enzymatic hydrolysate of poultry side streams processed by 0.1 μm microfiltration, 200 Da nanofiltration and diafiltration in turn; all rated at 2% (w/w) dry matter (not at equal protein concentration) | 4.4 | 5.1 | 15 cm line with end points 1 (no intensity) and 9 (high intensity) (10 trained assessors, 2 replicates) | Sample after "vs" more bitter Tested, significant |
| In Table 3, crude (b) and NF-dia (a) carry different letters, significant (two-way ANOVA with Tukey; overall p = 0.048 for bitterness). Microfiltration (4.7) and nanofiltration (4.8) in between are both ab. The text says the difference between microfiltration and NF-dia was significant, which does not match the letters in the table. The authors suggest the removal of salt by nanofiltration and the higher protein concentration at equal dry matter as possible reasons Where: Sections 2.1, 2.2.2 and 2.2.5; Table 3 and the paragraphs after it For whether changing the enzyme or enzyme amount, the degree or time of hydrolysis, or the after-treatment of the resulting hydrolysate, for the same protein raw material, changes the intensity of bitterness people perceive: Limited — Before and after membrane processing of an existing hydrolysate, with a test, but the original hydrolysis conditions are unknown, and since dry matter was equalised, protein concentration differs too | |||||
Sharma 2023 sensory only
Effect of Liposomal Encapsulation and Ultrasonication on Debittering of Protein Hydrolysate and Plastein from Salmon Frame. Foods 12(4):761 — doi:10.3390/foods12040761
- How deeply checked
- Checked in full text — Full text via Europe PMC (sections 2.2–2.5; sections 3.1 and 3.2; Tables 2 and 4) Checked by: LLM (one extracted, another checked against the source; no human check), 2026-10-22.
- Samples
- Papain hydrolysate of salmon frame protein (SFPH; papain 1% at 40 °C for 5 min) and its plastein (SFPT)
- Processing and cooking
- Soy lecithin and glycerol films hydrated with 1, 2 or 3% SFPH or SFPT solution to encapsulate them in liposomes; the least bitter 1% preparations (L-PH1, L-PT1) were then ultrasonicated (20 kHz, 750 W, 20% amplitude) for 30, 60 or 120 s
- How it was served or eaten
- Serving volume and temperature not stated
- Who rated, on what scale
- Panel trained for 1 month with caffeine as the reference (number not stated); 15 cm line scale from 0 (none) to 15 (intense); ANOVA with Tukey
| Measured | Comparison and condition | Sample before "vs" | Sample after "vs" | Unit | Direction / test |
|---|---|---|---|---|---|
| Bitterness intensity (sensory) | After-treatment of the hydrolysate: salmon hydrolysate at 1%, not encapsulated vs encapsulated in liposomes A 1% solution of the papain hydrolysate of salmon frame protein (SFPH; papain 1% at 40 °C for 5 min), and liposomes made by hydrating a soy lecithin and glycerol film with the same solution (L-PH1) | 5.8 | 4.2 | 15 cm line scale from 0 (none) to 15 (intense) (trained panel, number not stated) | Sample before "vs" more bitter Tested, significant |
| In Table 2, before (e) and after (f) encapsulation carry different letters, significant (p<0.05). At 2% and 3% the letters before and after overlap, so the text's "significant at every concentration" is not taken as is. Encapsulation wraps the hydrolysate rather than changing it, and does not break down peptides Where: Sections 2.2–2.4; section 3.1; Table 2 For whether changing the enzyme or enzyme amount, the degree or time of hydrolysis, or the after-treatment of the resulting hydrolysate, for the same protein raw material, changes the intensity of bitterness people perceive: Limited — Before and after encapsulating an existing hydrolysate, with a test, but encapsulation wraps the bitter material without changing the hydrolysate itself; panel size not stated | |||||
| Bitterness intensity (sensory) | After-treatment of the hydrolysate: encapsulated salmon hydrolysate (L-PH1), ultrasound for 0 s vs 120 s Liposomes encapsulating 1% hydrolysate (L-PH1) ultrasonicated (20 kHz, 750 W, 20% amplitude, 2 s on and 8 s off) for 120 s, kept at 25 ± 5 °C in iced water | 4.5 | 6.3 | 15 cm line scale from 0 (none) to 15 (intense) (trained panel, number not stated) | Sample after "vs" more bitter Tested, significant |
| In Table 4, 0 s (c) and 120 s (a) carry different letters, significant (p<0.05). The plastein preparation (L-PT1) also rose significantly from 3.5 to 5.5. Together with the lower encapsulation efficiency after ultrasound, the authors attribute this to disrupted liposomes. L-PH1 is 4.2 in Table 2 and 4.5 at 0 s in Table 4; the text does not state why they differ, so comparisons are made within each table Where: Section 2.5.1; section 3.2; Tables 3 and 4 For whether changing the enzyme or enzyme amount, the degree or time of hydrolysis, or the after-treatment of the resulting hydrolysate, for the same protein raw material, changes the intensity of bitterness people perceive: Limited — Before and after ultrasonicating encapsulated hydrolysate, with a test, but the treatment acts on the encapsulated product rather than the hydrolysate, and the panel size is not stated | |||||
What was established within an experiment
Only results observed in comparisons within one study. Results with no test shown are stated as such.
- In Meinlschmidt et al., the letters in Table 2 compare 0 min with each time within one enzyme: Alcalase (significantly higher from 10 min) and Neutrase (significantly higher at 120 min) were more bitter than unhydrolysed. Papain 0.2% was significantly higher at 10 min (4.3) and 30 min (4.8), and at 120 min (3.1) no longer differed significantly from 0 min. Enzymes were not tested against each other. [rows: Meinlschmidt 2016 Soy: unhydrolysed (0 min) vs Alcalase for 120 min; Meinlschmidt 2016 Soy: unhydrolysed (0 min) vs Neutrase for 120 min; Meinlschmidt 2016 Soy: unhydrolysed (0 min) vs papain 0.2% for 120 min]
- In the 2 lupin studies, Alcalase alone (Schlegel et al. 2019, 7.2) and Alcalase with papain together for 4 h (Schlegel et al. 2020, 5.7) were significantly more bitter than unhydrolysed. Flavourzyme in the 2019 study (0.7) did not differ significantly from unhydrolysed. The 2020 study says the other combinations were not significantly higher than unhydrolysed. [rows: Schlegel 2019 Lupin: unhydrolysed vs Alcalase for 2 h; Schlegel 2019 Lupin: unhydrolysed vs Flavourzyme for 2 h; Schlegel 2020 Lupin: unhydrolysed vs Alcalase and papain together for 4 h (OS 5)]
- In wheat gluten (Sun et al. 2022), with both Alcalase and trypsin, 20% degree of hydrolysis was significantly more bitter than 4% (0.65 and 4.38; 0.59 and 3.4). Unhydrolysed gluten was not rated. [rows: Sun 2022 Degree of hydrolysis or enzyme: wheat gluten, Alcalase at 4% vs 20% degree of hydrolysis; Sun 2022 Degree of hydrolysis or enzyme: wheat gluten, trypsin at 4% vs 20% degree of hydrolysis]
- In crickets (Fetriyuna et al. 2026), the time means over enzyme amounts were 14.18 at 2 h, 10.60 at 3.5 h and 7.32 at 5 h, each significantly different. The raw material dried without enzyme was 32.88, higher than the 12 conditions (5.92–16.16), but there is no test between raw material and hydrolysates, and whether the fractionation step was the same is not stated. [rows: Fetriyuna 2026 Degree of hydrolysis or enzyme: cricket, ProteAX for 2 h vs 5 h (mean over enzyme amounts); Fetriyuna 2026 Cricket: raw material dried without enzyme (CP; equivalence of processing unconfirmed) vs ProteAX 1.5% for 5 h]
- In skim milk (Jiang et al. 2024), with the same 4.5 h of hydrolysis, Neutrase (7.60) was significantly more bitter than Protamex (6.80) and Flavourzyme (6.60). In sardine (Sugiyama et al. 1991), even among alkaline enzymes Alcalase gave 1.7 and Proleather 4.0, with no test. [rows: Jiang 2024 Degree of hydrolysis or enzyme: skim milk, Flavourzyme vs Neutrase; Jiang 2024 Degree of hydrolysis or enzyme: skim milk, Protamex vs Neutrase; Sugiyama 1991 Degree of hydrolysis or enzyme: sardine, Alcalase vs Proleather (both alkaline)]
- There are comparisons in which adding an enzyme that trims peptide ends to an existing hydrolysate lowered bitterness. In Nakamura et al., casein fell from 3.5 to 1.7 and cod from 2.9 to 0.54 (p<0.01). In Wu et al., comparing fractions below 2500 Da at equal nitrogen concentration, all 4 second-stage products were significantly less bitter than the fraction of the original hydrolysate (A6), but the difference between A6-s3, at the least-bitter end, and A6-s4 was not significant. Zhao et al. report a significant decrease when AOAP was added to a clam hydrolysate (values in a figure, unconfirmed). In Sugiyama et al., aminopeptidase M at 3 mg gave 1.7 to 0.9 and the synthetic adsorbent VP-10 gave 0.5, with no test. [rows: Nakamura 2023 After-treatment of the hydrolysate: pepsin hydrolysate of casein, aminopeptidase LapA at 0 h vs 4 h; Nakamura 2023 After-treatment of the hydrolysate: pepsin hydrolysate of cod protein, aminopeptidase LapA at 0 h vs 21 h; Wu 2026 After-treatment of the hydrolysate: rice, Alcalase hydrolysate (A6) vs Flavourzyme with pH held for 3 h (A6-s3); Zhao 2026 After-treatment of the hydrolysate: clam, alkaline protease for 4 h vs then AOAP for 3 h; Sugiyama 1991 After-treatment of the hydrolysate: sardine Alcalase hydrolysate, untreated vs aminopeptidase M 3 mg; Sugiyama 1991 After-treatment of the hydrolysate: sardine Alcalase hydrolysate, untreated vs synthetic adsorbent VP-10]
- There are also comparisons in which after-treatment raised bitterness. The poultry hydrolysate (Aspevik et al. 2026) rose significantly from 4.4 before to 5.1 after membrane processing (by the table letters, between the crude and the last stage). The salmon hydrolysate (Sharma et al. 2023) fell significantly from 5.8 to 4.2 when encapsulated in liposomes, but rose significantly from 4.5 to 6.3 when the encapsulated product was ultrasonicated for 120 s. [rows: Aspevik 2026 After-treatment of the hydrolysate: poultry hydrolysate, before membrane processing (crude) vs after microfiltration, nanofiltration and diafiltration; Sharma 2023 After-treatment of the hydrolysate: salmon hydrolysate at 1%, not encapsulated vs encapsulated in liposomes; Sharma 2023 After-treatment of the hydrolysate: encapsulated salmon hydrolysate (L-PH1), ultrasound for 0 s vs 120 s]
- In soy concentrate (Knežević-Jugović et al. 2023), sums of classes against a commercial hydrolysate were +14 for Neutrase alone, −25 for Neutrase then Flavourzyme, −15 for Alcalase alone and −7 for Alcalase then Flavourzyme. The authors write that bitterness fell significantly for Neutrase and rose for Alcalase, but the published Friedman test concerns the 5 less-bitter samples as a whole, and individual significance for these pairs cannot be confirmed. [rows: Knežević-Jugović 2023 Compared with a commercial hydrolysate: soy, Neutrase alone vs Neutrase then Flavourzyme; Knežević-Jugović 2023 Compared with a commercial hydrolysate: soy, Alcalase alone vs Alcalase then Flavourzyme]
Candidate explanations for the differences
Conditions that differ between studies and could explain the differences in bitterness results, and the authors' interpretations. Several conditions differ at once, so which one mattered is unknown. These are candidates, not causes.
- Enzymes with exopeptidase activity, which removes amino acids from peptide ends, such as Flavourzyme and the purified aminopeptidase, gave low or lowered bitterness in some results (Flavourzyme in Meinlschmidt et al., Nakamura et al., Wu et al., Zhao et al.). The authors point to the release of hydrophobic amino acids at the ends. However, in Knežević-Jugović et al. adding Flavourzyme after Alcalase moved the sum of classes up from −15 to −7, so this activity alone cannot explain the results. [rows: Meinlschmidt 2016 Soy: unhydrolysed (0 min) vs Flavourzyme for 120 min; Nakamura 2023 After-treatment of the hydrolysate: pepsin hydrolysate of casein, aminopeptidase LapA at 0 h vs 4 h; Wu 2026 After-treatment of the hydrolysate: rice, Alcalase hydrolysate (A6) vs Flavourzyme with pH held for 3 h (A6-s3); Zhao 2026 After-treatment of the hydrolysate: clam, alkaline protease for 4 h vs then AOAP for 3 h; Knežević-Jugović 2023 Compared with a commercial hydrolysate: soy, Alcalase alone vs Alcalase then Flavourzyme]
- As a reason Alcalase hydrolysates tend to be bitter, the authors point to its tendency to cleave at hydrophobic amino acids, producing bitter peptides (Meinlschmidt et al., Schlegel et al., Sun et al.). However, Alcalase in sardine (Sugiyama et al. 1991) was on the low side at 1.7, and across studies with different raw materials, amounts and scales the direction cannot be set by the enzyme name alone. [rows: Meinlschmidt 2016 Soy: unhydrolysed (0 min) vs Alcalase for 120 min; Schlegel 2019 Lupin: unhydrolysed vs Alcalase for 2 h; Sun 2022 Degree of hydrolysis or enzyme: wheat gluten, Alcalase at 4% vs 20% degree of hydrolysis; Sugiyama 1991 Degree of hydrolysis or enzyme: sardine, Alcalase vs Proleather (both alkaline)]
- The relation between degree of hydrolysis and bitterness differs by study. In Sun et al. bitterness rose with degree of hydrolysis; in set 109 of Daher et al. bitterness fell as degree of hydrolysis rose (r = −0.933); and in Schlegel et al. 2019 Flavourzyme had low bitterness despite a high degree of hydrolysis (6.90%). The methods for degree of hydrolysis (OPA, pH-stat, TCA-soluble nitrogen) also differ by study, so the values cannot be lined up and compared. [rows: Sun 2022 Degree of hydrolysis or enzyme: wheat gluten, Alcalase at 4% vs 20% degree of hydrolysis; Daher 2021 Degree of hydrolysis or enzyme: casein, Flavourzyme with Protamex (set 109), 3 h vs 5 h; Schlegel 2019 Lupin: unhydrolysed vs Flavourzyme for 2 h]
- For the rise in bitterness after membrane processing, Aspevik et al. point to the removal of salt by nanofiltration and to the higher protein concentration at equal dry matter. Sharma et al. point to liposomes disrupted by ultrasound. Both are the authors' interpretations, not confirmed by these designs. [rows: Aspevik 2026 After-treatment of the hydrolysate: poultry hydrolysate, before membrane processing (crude) vs after microfiltration, nanofiltration and diafiltration; Sharma 2023 After-treatment of the hydrolysate: encapsulated salmon hydrolysate (L-PH1), ultrasound for 0 s vs 120 s]
What is not known — what this comparison alone cannot separate
- Many comparisons have no rating of the unhydrolysed raw material (Sun et al., Daher et al., Jiang et al., Sugiyama et al., Wu et al., Nakamura et al.). They compare hydrolysing further, changing enzymes or treating afterwards, and do not directly answer whether hydrolysis makes protein bitter. Even among studies with an unhydrolysed control, for crickets (Fetriyuna et al.) and clams (Zhao et al.) it cannot be confirmed that control and hydrolysates went through the same fractionation step. The reference in Knežević-Jugović et al. is a commercial hydrolysate. [rows: Sun 2022 Degree of hydrolysis or enzyme: wheat gluten, Alcalase at 4% vs 20% degree of hydrolysis; Daher 2021 Degree of hydrolysis or enzyme: casein, Flavourzyme with Protamex (set 109), 3 h vs 5 h; Jiang 2024 Degree of hydrolysis or enzyme: skim milk, Flavourzyme vs Neutrase; Sugiyama 1991 Degree of hydrolysis or enzyme: sardine, Alcalase vs Proleather (both alkaline); Wu 2026 After-treatment of the hydrolysate: rice, Alcalase hydrolysate (A6) vs Flavourzyme with pH held for 3 h (A6-s3); Nakamura 2023 After-treatment of the hydrolysate: pepsin hydrolysate of casein, aminopeptidase LapA at 0 h vs 4 h; Fetriyuna 2026 Cricket: raw material dried without enzyme (CP; equivalence of processing unconfirmed) vs ProteAX 1.5% for 5 h; Zhao 2026 Clam: unhydrolysed control (equivalence of processing unconfirmed) vs alkaline protease for 4 h; Knežević-Jugović 2023 Compared with a commercial hydrolysate: soy, Neutrase alone vs Neutrase then Flavourzyme]
- Scales (0–10, 0–5, 1–7, 1–9, 0–15, gLMS, difference scores, sums of classes), sample concentrations and the denominator of enzyme amount (relative to protein or to solution) all differ by study. Values cannot be lined up in one table to say how many times more bitter. The degree of hydrolysis also varies together with the enzyme, so which one mattered cannot be separated. [rows: Meinlschmidt 2016 Soy: unhydrolysed (0 min) vs Alcalase for 120 min; Sugiyama 1991 Degree of hydrolysis or enzyme: sardine, Alcalase vs Proleather (both alkaline); Daher 2021 Degree of hydrolysis or enzyme: casein, Flavourzyme with Protamex (set 109), 3 h vs 5 h; Aspevik 2026 After-treatment of the hydrolysate: poultry hydrolysate, before membrane processing (crude) vs after microfiltration, nanofiltration and diafiltration; Sharma 2023 After-treatment of the hydrolysate: salmon hydrolysate at 1%, not encapsulated vs encapsulated in liposomes; Fetriyuna 2026 Degree of hydrolysis or enzyme: cricket, ProteAX for 2 h vs 5 h (mean over enzyme amounts); Wu 2026 After-treatment of the hydrolysate: rice, Alcalase hydrolysate (A6) vs Flavourzyme with pH held for 3 h (A6-s3); Knežević-Jugović 2023 Compared with a commercial hydrolysate: soy, Neutrase alone vs Neutrase then Flavourzyme; Schlegel 2019 Lupin: unhydrolysed vs Alcalase for 2 h]
- In some comparisons values or pairwise marks are only in figures we did not obtain, so they are unconfirmed (the other combinations in Schlegel et al. 2020, the 12% degree of hydrolysis in Sun et al., the pairs of times in Daher et al., and all groups in Zhao et al.). The text and whether a test was reported were checked, but the size of decreases or increases cannot be compared on this page. [rows: Schlegel 2020 Lupin: unhydrolysed vs Alcalase and papain together for 4 h (OS 5); Sun 2022 Degree of hydrolysis or enzyme: wheat gluten, Alcalase at 4% vs 20% degree of hydrolysis; Daher 2021 Degree of hydrolysis or enzyme: casein, Flavourzyme with Protamex (set 109), 3 h vs 5 h; Zhao 2026 Clam: unhydrolysed control (equivalence of processing unconfirmed) vs alkaline protease for 4 h; Zhao 2026 After-treatment of the hydrolysate: clam, alkaline protease for 4 h vs then AOAP for 3 h]
- Meinlschmidt et al. 2016, Schlegel et al. 2019 and Schlegel et al. 2020 are from the same institute (Fraunhofer IVV), with procedures following the earlier reports; the pea study placed in the scope (García Arteaga et al.) is from the same institute too. Whether the same people rated the samples is unconfirmed. They are not counted as independent replications. Daher et al. 2021 and Daher et al. 2020, placed in the scope, are also from the same group. [rows: Meinlschmidt 2016 Soy: unhydrolysed (0 min) vs Alcalase for 120 min; Schlegel 2019 Lupin: unhydrolysed vs Alcalase for 2 h; Schlegel 2020 Lupin: unhydrolysed vs Alcalase and papain together for 4 h (OS 5); Daher 2021 Degree of hydrolysis or enzyme: casein, Flavourzyme with Protamex (set 109), 3 h vs 5 h]
- Bitterness is read separately from umami, aftertaste and acceptability. The acceptability order in Zhao et al. and the aftertaste values in Fetriyuna et al. are not bitterness intensity itself. This page lines up studies of perceived taste and is not advice on choosing commercial protein products or on health. [rows: Zhao 2026 After-treatment of the hydrolysate: clam, alkaline protease for 4 h vs then AOAP for 3 h; Fetriyuna 2026 Cricket: raw material dried without enzyme (CP; equivalence of processing unconfirmed) vs ProteAX 1.5% for 5 h]
How studies and rows were counted
The number of evidence rows is not the number of studies. Table rows, ledger evidence rows, papers and research groups are counted separately. The soy and lupin reports from the same institute (Fraunhofer IVV) are counted as one research group.
- Sources in the table: 14, of which 14 have results on people's bitterness ratings in this table; for the others, only results on something else (compounds, smell) are shown, or they are used only for context.
- Rows in the table (study × measured result): 29. Rows on perceived bitterness: 29.
- Evidence rows in the lab's ledger these rows rest on: 0. One paper can supply several rows. Rows without a ledger evidence row come from other tables of the same papers, or from sources newly read for this question.
- Research groups among the studies with bitterness results in this table, as grouped in the ledger: 12 (reports from the same lab count as 1; co-authors and affiliations can overlap, so this is not a count of independent replications).
What we looked at
The sources examined for this question and why each was included or excluded. We searched Europe PMC for studies combining protein hydrolysates, degree of hydrolysis or debittering with human ratings of bitterness, together with enzyme names and raw materials (milk, casein, soy, pea, rice, wheat, fish, insects and others), searched J-STAGE for Japanese- and English-language studies, and added the sources already in the lab's ledger. Treatments combined with fermentation, physical treatments during hydrolysis, ratings the authors call preliminary, and studies whose bitterness values are only in supplementary material were kept as references outside the main table. Studies with no human bitterness rating (such as electronic tongue only), studies that cannot be confirmed as treatments of the same starting material, and studies of enzymes that do not break down protein were excluded. Exclusions taken from the previous reader's record, whose originals were not checked in this round, are marked as such. The search described here is a record, not a claim that nothing was missed.
| Source | Decision | Why |
|---|---|---|
| Enzymatic treatment of soy protein isolates: effects on the potential allergenicity, technofunctionality, and sensory properties (2016) doi | Included in the table | Soy protein isolate hydrolysed with 10 proteases and rated against unhydrolysed material by 10 trained panellists with the same procedure; read in full |
| Enzymatic hydrolysis of lupin protein isolates—Changes in the molecular weight distribution, technofunctional characteristics, and sensory attributes (2019) doi | Included in the table | Lupin protein isolate hydrolysed with 9 proteases and compared with a no-enzyme control; same institute as Meinlschmidt et al.; read in full |
| Effect of enzyme‐assisted hydrolysis on protein pattern, technofunctional, and sensory properties of lupin protein isolates using enzyme combinations (2020) doi | Included in the table | Lupin protein isolate hydrolysed with enzyme combinations and compared with unhydrolysed material; same institute; Fig. 4 not obtained; read in full |
| Impact of Aspergillus oryzae-Derived Aminopeptidase Complex in Developing the Flavor Profile of Clam Hydrolysate (2026) doi | Included in the table | Unhydrolysed, one-stage and two-stage clam samples compared; values in Fig. 2, not obtained; the heading states that it cannot be confirmed the control was processed like the hydrolysates; the second-stage row is labelled "After-treatment of the hydrolysate"; read in full |
| Enhancing Consumer Acceptance of Cricket (Acheta domesticus) Protein Hydrolysate Through ProteAX‐Mediated Enzymatic Hydrolysis to Improve Techno‐Functional and Organoleptic Properties (2026) doi | Included in the table | Crickets hydrolysed at different enzyme amounts and times; the heading states that it cannot be confirmed the raw material was processed like the hydrolysates; the time-comparison row is labelled "Degree of hydrolysis or enzyme"; read in full |
| Characteristics of the enzyme-induced release of bitter peptides from wheat gluten hydrolysates (2022) doi | Included in the table | Degrees of hydrolysis of wheat gluten compared; with no rating of unhydrolysed gluten, its label begins with "Degree of hydrolysis or enzyme"; read in full |
| Sensopeptidomic Kinetic Approach Combined with Decision Trees and Random Forests to Study the Bitterness during Enzymatic Hydrolysis Kinetics of Micellar Caseins (2021) doi | Included in the table | Casein followed over 9 h with 2 enzyme sets; recipe confidential; labelled "Degree of hydrolysis or enzyme"; read in full |
| Lactiplantibacillus plantarum for the Preparation of Fermented Low-Bitter Enzymatic Skim Milk with Antioxidant Ability (2024) doi | Included in the table | Skim milk hydrolysed with 3 enzymes; the lactic fermentation part is not used; labelled "Degree of hydrolysis or enzyme"; read in full |
| 各種プロテアーゼ処理によって得られたマイワシタンパク加水分解物の特性 (1991) doi | Included in the table | Sardine meal hydrolysed with 8 enzymes, and the Alcalase hydrolysate treated with peptidases or adsorbents; no test described; in the previous reader's search record, the one Japanese-language J-STAGE study matching these conditions (a search record, not a confirmation that it is the only one); read in full |
| Impact of Different Enzymatic Processes on Antioxidant, Nutritional and Functional Properties of Soy Protein Hydrolysates Incorporated into Novel Cookies (2023) doi | Included in the table | One- and two-step soy concentrate hydrolysates compared with a commercial hydrolysate; labelled "Compared with a commercial hydrolysate"; read in full |
| Reduction in the bitterness of protein hydrolysates by an aminopeptidase from Aspergillus oryzae (2023) doi | Included in the table | A purified aminopeptidase applied to pepsin hydrolysates; labelled "After-treatment of the hydrolysate"; read in full |
| Peptidomic evidence for pH-stat flavourzyme-mediated debittering of Alcalase-derived rice protein hydrolysates (2026) doi | Included in the table | An Alcalase hydrolysate of rice given a second stage with Flavourzyme, with fractions below 2500 Da compared in pairs; labelled "After-treatment of the hydrolysate"; read in full |
| Upcycling of poultry protein hydrolysates using membrane filtration technology – Effects on sensory properties and chemical composition (2026) doi | Included in the table | A poultry hydrolysate processed with membranes; the original hydrolysis conditions are unknown, but it is included as an after-treatment comparison; read in full |
| Effect of Liposomal Encapsulation and Ultrasonication on Debittering of Protein Hydrolysate and Plastein from Salmon Frame (2023) doi | Included in the table | A salmon hydrolysate encapsulated in liposomes and the encapsulated product ultrasonicated; encapsulation does not break down peptides, but it is included as an after-treatment comparison; read in full |
| Enzymatic Hydrolysis and Fermentation of Pea Protein Isolate and Its Effects on Antigenic Proteins, Functional Properties, and Sensory Profile (2022) doi | Read as context (not in the table) | Pea protein isolate treated with combinations of enzyme and lactic fermentation and rated by ranking (10 panellists in the text, 11 in the figure legend); only the fermentation combinations and the untreated isolate were rated, not enzyme-only samples; the significance level was p<0.10, and which combination differed is in Fig. 4. Figure not obtained, so values unconfirmed; not counted as an example of significantly increased bitterness; from Fraunhofer IVV; read in full |
| Ultrasound viscous reduction effects on the proteolysis of soy protein isolate at a limited degree of hydrolysis: Changes in the functional characteristics and protein structure (2024) doi | Read as context (not in the table) | Soy protein isolate hydrolysed with bromelain to 12% degree of hydrolysis, with or without ultrasound pretreatment; panel size, concentration and scale are referred to an earlier report; the sentence calling the ordinary hydrolysate the most bitter carries p>0.05, so it is not counted as significantly more bitter; Fig. 2D: Figure not obtained, so values unconfirmed; read separately from the electronic-tongue direction; read in full |
| Principal Component Analysis from Mass Spectrometry Data Combined to a Sensory Evaluation as a Suitable Method for Assessing Bitterness of Enzymatic Hydrolysates Produced from Micellar Casein Proteins (2020) doi | Read as context (not in the table) | 16 hydrolysates made from the same raw material with different recipes and rated by 19 assessors (3.31–5.68 in Table 1); not made into a row because the enzymes and conditions are confidential, so what was changed cannot be stated; same group as Daher et al. 2021; read in full |
| Debittering of Protein Hydrolysates by Lactobacillus LBL-4 Aminopeptidase (2011) doi | Read as context (not in the table) | Applying a lactic-bacterium aminopeptidase to subtilisin hydrolysates of casein and soy lowered bitterness from the equivalent of 0.003–0.004% quinine to 0.0005–0.001% (algae, separately, from weak to almost none); rated by 5 people, called preliminary by the authors and with no test described, so kept out of the main table; read in full |
| Effects of Moderate Enzymatic Hydrolysis on Structure and Functional Properties of Pea Protein (2022) doi | Read as context (not in the table) | Pea hydrolysed with 4 enzymes to 2–8% degree of hydrolysis (Flavourzyme up to 7%); the text says bitterness rose with degree of hydrolysis, but the sensory values and the order of enzymes are in supplementary Fig. S1, not obtained; the order of enzymes in the text (trypsin > Alcalase > Neutrase > Flavourzyme) concerns improved solubility and function, not bitterness; read in full |
| Mechanistic insights into high-intensity ultrasound-induced debittering and umami-enhancing of Antarctic krill peptides during ultrasound-assisted flavourzyme hydrolysis: Integrating E-tongue, multi-spectroscopy, and label-free peptidomics☆ (2026) doi | Read as context (not in the table) | Krill hydrolysed with Flavourzyme under simultaneous ultrasound; the text says 400–700 W gave bitterness 1.8–4.3 lower than 0–200 W (p<0.05); a physical treatment during hydrolysis rather than an enzyme condition or after-treatment, so kept out of the main table; individual differences between 400 W and stronger settings unconfirmed; read in full |
| Deciphering the Molecular Bittercode in Potato Protein Hydrolysates through a Sensoproteomics Approach (2026) doi | Excluded | Compares a potato protein isolate with commercial hydrolysates, but the hydrolysates are products of other suppliers and cannot be confirmed as treatments of the same starting material |
| Effects of Enzymatic Hydrolysis on Physicochemical Properties and Solubility and Bitterness of Milk Protein Hydrolysates (2021) doi | Excluded | Bitterness measured only by electronic tongue; no human rating |
| Thermolysin Versus Four Commercial Proteases in the Modification of Soy Protein Isolate: Structural, Functional, and Taste Characterization (2026) doi | Excluded | Bitterness measured only by electronic tongue, and the text states no human confirmation was done |
| Optimized Nitrogen Recovery and Non-Bitter Hydrolysates from Porcine Hemoglobin (2008) | Excluded | Gives only the quinine-equivalent bitterness of one optimised condition, with no bitterness comparison between conditions or with untreated material (checked in the J-STAGE PDF; DOI not confirmed in the PDF) |
| Reducing the Bitterness of Rapeseed Protein: Integrating Enzymatic Treatment, Metabolomics, and Sensory Analysis to Elucidate Underlying Mechanisms (2025) doi | Excluded | The enzymes that lowered bitterness were laccase and β-glucosidase, not a comparison of protein hydrolysis |
| Debittering of protein hydrolyzates (2001) doi | Excluded | A review in the lab's ledger, not primary data from human ratings |
| Enzymatic Debittering of Cheese Flavoring and Bitterness Characterization of Peptide Mixture Using Sensory and Peptidomics Approach (2024) doi | Full text not reachable | A study of enzymatic debittering of cheese flavouring, close to the scope, but the full text is not open and the lab's ledger has only the abstract |
| Impact of Ultrasound Pretreatment and Enzyme Concentration on Taste and Biological Activities of Porcine Lung Hydrolyzates (2025) doi | Excluded | Bitterness by electronic tongue only; reason for exclusion from the previous reader's record; not checked against the original in this round |
| Consumer Rejection Threshold of Mung Bean Protein Hydrolysate: Unsweetened and Sweetened Brewed Teas as Test Models (2026) doi | Excluded | A rejection-threshold study of one hydrolysate mixed into tea, not a comparison of hydrolysis conditions; reason for exclusion from the previous reader's record; not checked against the original in this round |
| Sensoproteomic Characterization of Lactobacillus Johnsonii-Fermented Pea Protein-Based Beverage: A Promising Strategy for Enhancing Umami and Kokumi Sensations while Mitigating Bitterness (2024) doi | Excluded | A study of lactic fermentation, not a comparison of enzymatic hydrolysis conditions; reason for exclusion from the previous reader's record; not checked against the original in this round |
| The Impact of Varying Enzymatic Pretreatment Durations of Wheat Gluten on the Flavour Characteristics of High-Moisture Plant-Based Extrudates (2026) doi | Excluded | Sensory evaluation of extrudates; no bitterness item could be confirmed; reason for exclusion from the previous reader's record; not checked against the original in this round |
| Preparation, Separation, and Identification of Low-Bitter ACE-Inhibitory Peptides from Sesame (Sesamum indicum L.) Protein (2026) doi | Excluded | No description of human sensory testing could be confirmed; reason for exclusion from the previous reader's record; not checked against the original in this round |
Read in the encyclopedia
- Protein hydrolysate debittering — separation, masking and enzyme-based methods to reduce the bitterness of protein hydrolysates
- Bitter peptides — short hydrophobic peptides released when proteins are broken down
- Ney's Q-rule — a rule of thumb predicting bitterness from the average hydrophobicity of peptides, and its limits
- Fermentation and bitter peptides — the topic hub on how fermentation makes and breaks bitter peptides
- Does cheese bitterness change with ripening and the choice of starter bacteria? — a comparison of cheese ripening and bitter peptides
- Does adding sugar or salt suppress bitterness? — a comparison of bitterness suppression
Source ledger
Each study's source can be reached from its section above. All of the lab's sources are listed in the source ledger.
How to cite this page
Aoki, M. (NIGAMI LAB) (2026). Does enzyme hydrolysis make protein more or less bitter?. https://nigamilab.com/en/topics/protein-hydrolysate-bitterness/
Each claim card has a stable anchor (#clm-…). The dataset is CC BY 4.0; see the data page. Dataset DOI: 10.5281/zenodo.22739101