foods

Article
Flavor Characterization of Animal Hydrolysates and Potential
of Glucosamine in Flavor Modulation
Kathrine H. Bak 1,2, * , Sandra S. Waehrens 1 , Yu Fu 1,3, * , Ching Yue Chow 1 , Mikael A. Petersen 1 ,
Jorge Ruiz-Carrascal 1 , Wender L. P. Bredie 1 and René Lametsch 1

1 Department of Food Science, Faculty of Science, University of Copenhagen, Rolighedsvej 26,

1958 Frederiksberg, Denmark; [email protected] (S.S.W.); [email protected] (C.Y.C.);
[email protected] (M.A.P.); [email protected] (J.R.-C.); [email protected] (W.L.P.B.); [email protected] (R.L.)
2 Institute of Food Safety, Food Technology and Veterinary Public Health, University of Veterinary Medicine
Vienna, Veterinärplatz 1, 1210 Vienna, Austria
3 College of Food Science, Southwest University, No.2 Tiansheng Road, Beibei District, Chongqing 400715, China
* Correspondence: [email protected] (K.H.B.); [email protected] (Y.F.)

Abstract: Bovine (meat and heart) and porcine (hemoglobin and plasma) raw materials were hy-
drolyzed by Protease A (both endo- and exopeptidase activity), with or without glucosamine added
during the enzyme inactivation step. Hydrolysates were characterized via peptide analysis (yield, UV-
and fluorescence scanning spectroscopy, and peptide size distribution via size exclusion chromatogra-
phy), sensory evaluation, and volatile compound analysis via gas chromatography mass-spectrometry
(GC-MS) to determine if glucosamine-induced Maillard reaction improved taste and flavor. Porcine
hemoglobin produced the most flavor-neutral hydrolysate, and could expectedly have the broadest



application in food products. Both bovine meat and -heart hydrolysates were high in umami, and


thereby good candidates for savory applications. Porcine plasma hydrolysate was high in liver flavor
Citation: Bak, K.H.; Waehrens, S.S.;
and would be suitable for addition to certain meat products where liver flavor is desirable. All
Fu, Y.; Chow, C.Y.; Petersen, M.A.;
hydrolysates had low perceived bitterness. Glucosamine-induced Maillard reaction had just a minor
Ruiz-Carrascal, J.; Bredie, W.L.P.;
influence on the sensory profile via an increased perception of sweet taste (p = 0.038), umami taste
Lametsch, R. Flavor Characterization
of Animal Hydrolysates and Potential
(p = 0.042), and yolk flavor (p = 0.038) in the hydrolysates, irrespective of raw material. Glucosamine
of Glucosamine in Flavor Modulation. addition had a statistically significant effect on 13 of 69 volatiles detected in the hydrolysates, but the
Foods 2021, 10, 3008. https:// effect was minor and raw material-specific.
doi.org/10.3390/foods10123008
Keywords: hydrolysates; sensory; volatile compounds; glucosamine; Maillard reaction; GC-MS
Academic Editor: Baohua Kong

Received: 20 October 2021

Accepted: 22 November 2021 1. Introduction
Published: 4 December 2021
By-products from the meat industry are currently underutilized, and are used mostly
in the production of low-value products such as animal feed, pet food, and fertilizer [1], or
Publisher’s Note: MDPI stays neutral
may even be disposed of as waste [2]. However, these products are rich in high-quality
with regard to jurisdictional claims in
protein [3], meaning that there is potential for the production of value-added ingredients,
published maps and institutional affil-
e.g., in the functional foods market [4]. At the same time, the global demand for high
iations.
protein foods has increased in recent years [5], signifying the potential from both an
economic and a sustainability point of view of improving the utilization of the by-products
that are a consequence of the meat industry.
Examples of by-product valorization include blood and organs [2], however low-value
Copyright: © 2021 by the authors.
meat cuts also have valorization potential. Enzymatic hydrolysis is an efficient way of
Licensee MDPI, Basel, Switzerland.
extracting amino acids and peptides from a protein-containing raw material [4]. Problems
This article is an open access article
with the bitterness of protein hydrolysates have previously been reported [6], though, in a
distributed under the terms and
previous study [7], Fu et al. found that enzymatic hydrolysates from meat and plasma had
conditions of the Creative Commons
only minor issues with regard to bitterness compared to other types of hydrolysates, e.g.,
Attribution (CC BY) license (https://
creativecommons.org/licenses/by/
casein hydrolysates [6]. Furthermore, enzymatic hydrolysis leads to the development of
4.0/).
a number of volatile compounds, including Maillard and lipid oxidation products [8]. A

Foods 2021, 10, 3008. https://doi.org/10.3390/foods10123008 https://www.mdpi.com/journal/foods

Foods 2021, 10, 3008 2 of 12

number of these volatile compounds may lead to off-flavors, while others may have an
advantageous effect, depending on the application of the hydrolysate.
Purposely inducing a Maillard reaction is a way of potentially improving hydrolysate
characteristics. Glucosamine is an amino sugar possessing salt-sour-bitter taste characteris-
tics [9]. Nonetheless, glucosamine is an appropriate reducing sugar to utilze in order to
induce a Maillard reaction in meat-related products as this increases neither the bitterness
nor the sweetness of the glycated hydrolysate. The successful application of glucosamine
in the glycation of poultry protein isolate provides evidence to support this suggestion [10].
The induced Maillard reaction between hydrolysates and sugar has also proved successful
in regard to soy protein hydrolysates with xylose, resulting in both reduced bitterness and
an increase in desirable odors and umami [11], or flavor enhancing properties in consommé
soup [12]. Similarly, Maillard reaction products formed from the glycation of sunflower
protein hydrolysate with xylose and cysteine had a meat-like flavor and flavor enhancing
properties [13]. The positive effects are largely credited to a high content of so-called Mail-
lard peptides, which are peptides in the region of 1 to 5 kDa [12–14], and to the presence of
cysteine [13], which has long been known to serve as an important precursor of meat flavor
compounds in reaction with carbohydrates [15]. A carefully controlled induced Maillard
reaction can also serve as a way of enhancing the functional properties of a protein [15].
Our previous study deemed the enzyme Protease A, which contains a combination of
endopeptidase and exopeptidase activity, to be the best choice overall out of ten enzymes
that were investigated with regard to the production of high yield hydrolysates with a
relatively low bitterness, from both bovine meat and porcine plasma [7]. For this reason,
Protease A was the enzyme of choice in this study. Furthermore, simultaneous studies
performed in our lab with an internal sensory panel have indicated that the induction of a
Maillard reaction by the addition of a small amount of glucosamine to porcine hydrolysates
may lead to an improvement in taste and flavor without having a detrimental effect on the
color of the hydrolysate [16].
Consequently, the aim of this work was to characterize hydrolysates from bovine
(minced meat and heart) and porcine (hemoglobin and plasma) sources with or without
added glucosamine based on peptide size distribution and spectral characteristics, sensory
evaluation by a trained panel, and volatile compound analysis via gas chromatography-
mass spectrometry (GC-MS). The four raw materials were chosen for their value-adding
potential, based on Danish market conditions and on their availability in food grade
form. The hypothesis is that the results will provide a reliable description of the animal
hydrolysates, which can be used to suggest suitable, characteristic-dependent applications.
Additionally, this study should serve as confirmation of our previous results [16], which
showed that the addition of glucosamine during the enzyme inactivation step of hydrolysis
presents a novel way of promoting a Maillard reaction in hydrolysates, whereby the overall
production time is reduced.

2. Materials and Methods

2.1. Hydrolysate Overview and Preparation
Selection of the raw materials and the protease for the present study was based on
previous studies including a sensory screening of the raw materials and ten different
enzymes [7,17]. Accordingly, bovine meat, bovine heart, porcine hemoglobin, and porcine
plasma, all collected in a hygienic way (i.e., food grade), were subjected to enzymatic
hydrolysis using the food grade enzyme Protease A for 5 h, followed by glucosamine-
induced Maillard reaction as described in detail below. An overview of the produced
hydrolysates is given in Supplementary Table S1.
Minced bovine meat (21.4% protein; low price, i.e., prepared from lower value cuts)
and bovine heart (18.2% protein) were purchased at a local supermarket (Inco, Copenhagen,
Denmark). Porcine hemoglobin (25.2% protein) and porcine plasma (6.2% protein) were
kindly supplied by DAT-Schaub A/S (Copenhagen, Denmark). Food grade glucosamine
(>99.8% purity) was purchased from Blackburn Distributions Ltd. (Nelson, UK). Protease
Foods 2021, 10, 3008 3 of 12

A 2SD (100,000 U/g), with both endopeptidase and exopeptidase activities, was donated
by Amano Enzyme Inc. (Nagoya, Japan). All analytical reagents were purchased from
Sigma-Aldrich (Steinheim, Germany). The raw materials were stored at −20 ◦ C until
further processing. The protein content of the raw materials (noted above) was determined
by the AOAC method using Kjeldahl analysis [18].
The enzymatic hydrolysates were prepared in a pilot plant reserved for food grade
materials by hydrolysis with Protease A at a concentration of 0.5% w/w for 5 h at pH 7.0
and 50 ◦ C. The procedure was conducted according to that described by Fu et al. [7,17].
Bovine minced meat and heart were hydrolyzed according to the method described for
minced meat [7], with the heart first being minced by passing it through a grinder once
(Scharfen Meat Mincer model X 70, Witten, Germany) with a 3 mm plate.
For half the hydrolysate preparations, food grade glucosamine was added in a ratio of
1:10 relative to the protein content of the raw material. This addition occurred at the begin-
ning of the heating step for enzyme inactivation (85 ◦ C for porcine plasma hydrolysates
and 90 ◦ C for the remaining three hydrolysates). After cooling, the supernatant of each of
the eight hydrolysates (four raw materials, with or without glucosamine) was kept frozen
at −20 ◦ C until further analysis.

2.2. Sensory Evaluation

2.2.1. Sensory Panel
A descriptive sensory analysis of the hydrolysates was conducted by ten trained
panelists (nine women and one man, with an age range of 22–47) selected from the trained
external sensory panel of the Department of Food Science at the University of Copenhagen,
Denmark. Subjects were paid for their participation. When initially becoming panel
members, the subjects underwent a selection process with subsequent training on the
basic tastes as well as other sensory acuities. Prior to this project, the panel members had
participated in the descriptive analysis of other food types.

2.2.2. Sample Preparation

Hydrolysates were normalized with water to a protein content of 1% for sensory
profiling. Preparation of hydrolysate samples for sensory analysis was done by pouring
25 mL portions into 30 mL black polypropylene cups with lids. Each sample was blinded
with three-digit random numbers. The samples were stored for two hours at 14 ◦ C to
obtain a serving temperature of 14 ◦ C.

2.2.3. Sensory Vocabulary Development and Profiling

The descriptive sensory analysis consisted of two training sessions with sensory
vocabulary development and training on scale use, followed by two sensory profiling
sessions. Each session was scheduled to two hours, which was sufficient due to the
experience of the panel members and the nature of this project. During the first training
session, the sensory panel generated the sensory vocabulary to describe the variation
between the hydrolysates based on panel consensus (see Table 1). During the second
training session, reference materials were provided to increase the panelists’ cognitive
clarity towards the developed attributes. The sensory panel was also trained in assessing
intensity on a 15 cm-line scale anchored from ‘not at all’ to ‘very much’. The reference
materials were used to support training on sensory intensity scores on the scale. The final
sensory vocabulary including definition with reference materials is shown in Table 1 and
included 17 sensory descriptors.
For the sensory profiling, the panel simultaneously evaluated the hydrolysates in a
complete block design with three blocks over the two profiling days, resulting in triplicate
evaluations of each sample. Within each block, the samples were served in a random
monadic sequence to reduce any bias related to the presentation order and carry-over effect.
The intensities of the attributes were rated on the 15 cm-line scale going from “not at all”
to “very much”. The panel was provided with sparkling water, and cold and lukewarm
Foods 2021, 10, 3008 4 of 12

water to rinse their palates. Data were collected using the Fizz Acquisition (Version2.50B,
Biosystèmes, Couternon, France). The conditions for the sensory descriptive analysis were
in accordance with the guidelines in ISO 13299:2003 [19].

Table 1. List of sensory attributes with definitions developed for the profiling.

Sensory Attribute 1 Scale Definition with Reference Materials 2

Odor Odor associated with
Animal-O ‘not at all’ → ‘very much’ Animal note reminiscent of Parma ham/Prosciutto di Parma
Raw meat-O ‘not at all’ → ‘very much’ Odor characteristic reminiscent of fresh minced beef (4–7% fat)
Taste
Umami-T ‘not at all’ → ‘very much’ Taste sensation of monosodium glutamate (0.7 g/L water)
Sweet-T ‘not at all’ → ‘very much’ Taste sensation of sucrose (12 g/L water)
Salt-T ‘not at all’ → ‘very much’ Taste sensation of sodium chloride (2.0 g/L water)
Bitter-T ‘not at all’ → ‘very much’ Taste sensation of caffeine (0.58 g/L water)
Flavor Aromatic taste sensation associated with
Metallic-F ‘not at all’ → ‘very much’ Metallic sensation reminiscent of ferrous sulfate (0.016 g/L water)
Liver-F ‘not at all’ → ‘very much’ Flavor sensation reminiscent of liver pâté
Yolk-F ‘not at all’ → ‘very much’ Flavor sensation reminiscent of pasteurized egg yolk
Fish sauce-F ‘not at all’ → ‘very much’ Flavor sensation reminiscent of Asian fish sauce
Sulfur-F ‘not at all’ → ‘very much’ Sulfury flavor note reminiscent of boiled egg yolk
Toasted Burnt-F ‘not at all’ → ‘very much’ Toast
Aftertaste (AT) Aftertaste associated with
Overall Intensity-AT ‘not at all’ → ‘very much’ Overall intensity
Umami-AT ‘not at all’ → ‘very much’ Taste sensation of monosodium glutamate (0.7 g/L water)
Bitter-AT ‘not at all’ → ‘very much’ Taste sensation of caffeine (0.58 g/L water)
Metallic-AT ‘not at all’ → ‘very much’ Metallic taste sensation reminiscent of ferrous sulfate (0.016 g/L water)
Liver-AT ‘not at all’ → ‘very much’ Flavor sensation reminiscent of liver pâté
1 Suffix to sensory attributes indicates method of assessment by panelists: -O, odor, -T, taste, -F, flavor and -AT, aftertaste. 2 Definitions of
sensory attributes as agreed on between the panelists during vocabulary development.

2.3. Peptide Characterization

All determinations were performed in triplicate and the results reported as the average
value.

2.3.1. Yield
The yield of the hydrolysates was determined as described by [7] using oven heating
at 105 ◦ C until the constant weight and subsequent calculation of the yield as the percent-
age of dry matter content in the supernatants relative to the dry matter content prior to
centrifugation.

2.3.2. Spectral Characteristics of Glycated Hydrolysates

UV and fluorescence scanning were employed to monitor the formation of intermedi-
ate Maillard reaction compounds and melanoidins from browning.
The extent of the glycation of protein hydrolysates was determined by scanning
the absorbance spectra of the samples (approximately 5 mg/mL) from each treatment
from 200 nm to 500 nm using a Helios Omega UV-Visible spectrophotometer (Thermo
Scientific, Loughborough, UK). The formation of intermediate Maillard reaction products
and melanoidins during the Maillard reaction was monitored by UV scanning at 320 and
420 nm [16,20,21].
The fluorescence spectra of native and glycated hydrolysates (approximately 5 mg/mL)
were determined using a SpectraMax multi-mode microplate reader (Molecular Devices,
Sunnyvale, CA, USA). The excitation wavelength was fixed at 347 nm, while emission
wavelengths were fixed at 400–600 nm for the detection of fluorescent products generated
by the Maillard reaction [10,16].
Foods 2021, 10, 3008 5 of 12

2.3.3. Peptide Size Distribution of Native and Glycated Hydrolysates

The peptide size distribution of native and glycated hydrolysates was measured
according to our previously reported method [7]. The peptide size distribution of sam-
ples was analyzed by size exclusion chromatography under isocratic conditions using a
Phenomenex BioSep™ SEC-S2000 column (300 mm × 4.6 mm) (Torrance, CA, USA) on
an ultra-high performance liquid chromatography (UHPLC) system (Thermo Scientific
Dionex Ultimate 3000, Denmark). From each sample (1 mg/mL), 10 µL were injected,
eluted isocratically (the buffer being 30% acetonitrile containing 0.1% trifluoroacetic acid)
at a flow rate of 0.5 mL/min, and monitored at 214 nm.
Molecular weight (MW) distribution ranges were divided into fractions, including
>10 kDa, 5–10 kDa, 1–5 kDa, 0.5–1 kDa, and <0.5 kDa. As was the case in our previous
study [7], this was achieved by plotting a calibration curve using Trp (204 Da), GLV (287 Da),
SGNIGFPGPK (1114 Da), insulin (5700 Da) and myoglobin (17,600 Da) as standards, and
calculating MW as logMW = −0.6086t + 6.9381, where MW is the molecular weight in Da
and t denotes the elution time in minutes.

2.4. Volatile Compound Analysis by GC-MS

Hydrolysates were used undiluted for GC-MS. Dynamic headspace sampling and
the subsequent GC-MS were completed in duplicate for each sample using 20 mL of
hydrolysate with 1 mL of 5 mg/L 4-methyl-1-pentanol added as internal standard. The pro-
cedure was as described in our previous studies [8,17]. In brief, the collection of the volatiles
onto 200 mg Tenax-TA traps with mesh size 60/80 (Markes International, Llantrisant, UK)
was made at 37 ◦ C for 60 min. An automatic thermal desorption unit (TurboMatrix 350,
Perkin Elmer, Shelton, CT, USA) was used to desorb the volatiles, followed by separation
on a ZB-Wax capillary column of 30 m length, 0.25 mm internal diameter, and 0.50 µm film
thickness on a gas chromatograph-mass spectrometer (7890A GC-system interfaced with a
5975C VL MSD with Triple-Axis detector (Agilent Technologies, Palo Alto, CA, USA)). The
column temperature program and other gas chromatograph and mass spectrometer details
were as described in [8].

2.5. Data Analysis

Whether or not the addition of glucosamine and the type of raw material had an
overall influence on the sensory profile of the hydrolysates required exploration and, as
such, we conducted a linear analysis of the variance (ANOVA) including the addition of
glucosamine (+/−) and raw material (bovine meat, bovine heart, porcine hemoglobin,
porcine plasma) as main effects, including their interaction. The significance level was
set to 5%. The next priority was to identify sensory sample differences. Thus, a sample
effect was tested on each sensory attribute by mixed linear ANOVA model with a sample
as fixed factor and panelist, replicate, sample-panelist interactions, and sample-replicate
interactions as random factors using the lmerTest package [20] in R v3.5.2 (R Development
Core Team). The significance level was set to 5%. Post hoc pairwise comparisons were
made by Tukey’s HSD test (p < 0.05).
For the analysis of the GC-MS data, the peak areas relative to the internal standard
and mass spectra were extracted from the chromatograms using the PARAFAC2 based
software PARADISe [21] and mass spectra were identified using the NIST05 database.
Volatile compound identifications were verified by comparison with retention indices (RI)
of either authentic reference compounds or retention indices from the available literature
(tentative identification).
ANOVA on the results of the GC-MS analysis was performed using JMP 13.0.0 (SAS
Institute Inc., Cary, NC, USA) and least square means were compared by Tukey’s test
(p ≤ 0.05).
Foods 2021, 10, 3008 6 of 12

3. Results and Discussion

3.1. Sensory Characterization
Unsurprisingly, the linear ANOVA showed that the raw material type significantly
(p < 0.05) changed the sensory profile of the hydrolysates as significant effects were found
for all the sensory attributes investigated. However, the addition of glucosamine had just a
minor influence on the sensory profile, with significant effects found only for umami taste
(p = 0.042), sweet taste (p = 0.038), and yolk flavor (p = 0.038).
For a better understanding of sample differences, the sample effect was analyzed via
mixed linear ANOVA. A significant sample effect (p < 0.05) was found for all sensory at-
tributes. After an inspection of the pairwise comparisons (Tukey’s HSD test), the significant
effects could be explained by the dominant impact of raw material on the sensory profile of
the hydrolysates. Pairwise comparison of samples with and without glucosamine within
each raw material did not find any significant difference between these with regard to
umami taste, sweet taste, and yolk flavor. This can probably be explained by the fact that
the significance levels for the attributes were very close to 0.05 and nearly non-significant
in the model testing overall, hence the minimal effect of the addition of glucosamine. How-
ever, a systematic increase in intensity levels for these attributes is visible for hydrolysates
with added glucosamine irrespective of the raw material (Table 2).

Table 2. Panel mean intensity scores of the sensory attributes for the hydrolysates rated on a 15-cm unstructured line scale,
with p-values and suffixes (a, b, c, d, e) annotating sample differences. G indicates glucosamine addition.

Sensory Attributes 1 Meat Meat + G Heart Heart + G Hemo Hemo + G Plasma Plasma + G p-Value 2
Odor (O)
Animal-O 6.9 b 7.4 b 8.4 b 8.0 b 1.8 a 3.8 a 11.4 c 12.2 c <0.001
Raw meat-O 9.2 bc 9.7 c 9.8 c 9.4 bc 2.4 a 5.2 ab 5.4 ab 5.3 ab <0.001
Taste (T)
Umami-T 10.1 ab 11.0 ab 11.6 b 12.1 b 7.7 a 8.9 ab 7.9 a 8.9 ab <0.001
Sweet-T 4.6 b 4.9 b 4.5 b 5.1 b 2.1 a 3.5 ab 3.8 ab 4.8 b <0.001
Salt-T 3.1 ab 4.0 ac 3.8 ac 3.8 ac 2.5 a 3.2 ab 6.1 bc 6.6 c 0.001
Bitter-T 5.1 a 5.9 a 5.8 a 4.5 a 3.3 a 4.3 a 6.1 a 6.1 a 0.031
Flavor (F)
Metallic-F 7.8 ac 8.6 bc 9.3 c 8.1 ac 5.9 a 6.4 ab 8.7 bc 9.8 c <0.001
Liver-F 4.4 a 4.7 a 6.1 ab 4.9 a 4.1 a 4.3 a 10.4 bc 11.2 c <0.001
Yolk-F 8.9 bc 9.7 bc 10.2 c 10.3 c 4.4 a 7.6 ac 5.9 ab 6.2 ab <0.001
Fish sauce-F 2.6 ab 4.1 b 3.7 ab 3.4 ab 1.7 a 2.5 ab 3.8 ab 4.1 b 0.011
Sulfur-F 4.9 ab 5.0 b 5.4 b 5.3 b 2.3 a 3.9 ab 4.1 ab 4.2 ab 0.014
Toasted Burnt-F 1.2 a 1.5 ab 1.2 a 1.4 ab 0.8 a 1.3 ab 3.9 bc 4.1 cb <0.001
After taste (AT)
Overall Intensity-AT 8.5 bc 9.1 cd 9.6 ce 9.3 ce 5.7 a 6.4 ab 11.0 de 11.4 e <0.001
Umami-AT 8.8 ac 9.4 bc 10.2 c 10.0 bc 5.7 a 7.1 ac 6.7 ab 6.7 ab <0.001
Bitter-AT 5.3 ab 6.6 b 6.3 ab 5.5 ab 3.7 a 3.7 a 6.3 ab 6.2 ab 0.002
Metallic-AT 5.8 ac 6.8 bcd 7.2 cd 6.5 bcd 4.1 a 5.1 ab 7.6 cd 8.3 d <0.001
Liver-AT 3.0 a 3.6 a 4.4 a 3.4 a 2.3 a 2.4 a 9.8 b 10.7 b <0.001
1Suffix to sensory terms indicates method of assessment by panelists: -O, odor, -T, taste, -F, flavor, -AT, aftertaste. 2 Significance level:
Different letters (a, b, c, d, e) within the same rows indicate significant (p < 0.05) differences between samples according to Tukey’s HSD test.

As mentioned, raw material type had a significant influence on the sensory profile
of the hydrolysates. Overall, the porcine plasma hydrolysates were characterized by high
levels of liver flavor and aftertaste, while bovine minced meat and heart hydrolysates
were high in umami taste. The porcine hemoglobin hydrolysates were low in all sensory
attributes, and thus very bland. All hydrolysates had a low perceived bitterness.
Considering the potential applications for each type of hydrolysate, and since both
bovine meat and bovine heart hydrolysates are relatively high in umami, they would
make good candidates for savory applications such as in soups or meatballs. Porcine
plasma is high in liver flavor compared to the other hydrolysates, and thus, might be a
Foods 2021, 10, 3008 7 of 12

good candidate for certain meat products where liver flavor is desirable, e.g., liver pâté.
Since porcine hemoglobin hydrolysates proved to be the most neutral in flavor, they are a
potential candidate for broader food product applications.

3.2. Peptide Characteristics

3.2.1. Yield
The yields of the hydrolysates are of a similar magnitude to those described by Fu
et al. [7]. Yields were not determined for the glycated hydrolysates, as drying at a high
temperature can lead to an excessive loss of weight due to the volatile Maillard reaction
products (results not shown). In general, the yields of hydrolysates derived from bovine
meat (80.0%) and porcine hemoglobin (81.4%) were relatively high, in part due to the
higher protein content in these raw materials. The yields for bovine heart and porcine
plasma were slightly lower at 72.7% and 73.0%, respectively.

3.2.2. UV-Visible and Fluorescence Spectra of Native and Glycated Hydrolysates

UV-visible spectra of native and glycated hydrolysates derived from bovine meat,
bovine heart, porcine hemoglobin and porcine plasma are shown in Figure 1A–D, respec-
tively. UV-visible analysis is a fast and easy way to monitor and characterize the progress
of the Maillard reaction [22]. The formation of intermediate Maillard reaction products and
late-stage melanoidin compounds during the Maillard reaction can be monitored at 320
and 420 nm, respectively [23,24]. As is especially evident for the bovine heart hydrolysates
Foods 2021, 10, x FOR PEER REVIEW
(Figure 1B), the glycated hydrolysates exhibited a slightly higher absorbance than those of
8 of 13
native hydrolysates in the range of 300–350 nm. This fact is mainly due to the formation of
intermediate Maillard reaction products.

Figure 1. UV-scanning spectra of hydrolysates with (red) or without (blue) glucosamine (G) of (A)
bovine
bovine meat, (B) bovine
meat, (B) bovine heart,
heart, (C)
(C) porcine
porcine hemoglobin,
hemoglobin, and
and (D)
(D) porcine
porcine plasma.
plasma.

Notably, the major peak was observed in each glycation treatment in the region of
270–280 nm (Figure 1). This fact could be ascribed
ascribed to
to the
the presence
presence of
of aromatic
aromatic amino
amino acids,
as well as the autocondensation products of glucosamine
glucosamine [10,24]. There was no significant
change found as a result of glucosamine addition at the wavelength of 420 nm, suggesting
that no advanced glycation end-products were formed in the glycated hydrolysates.
Fluorescence spectroscopy can serve as a method of detecting the fluorescent
products generated by the Maillard reaction [25]. The fluorescent products can be detected
at the excitation wavelength of 347 nm, as well as the emission wavelength of 400–600 nm
Foods 2021, 10, 3008 8 of 12

change found as a result of glucosamine addition at the wavelength of 420 nm, suggesting
that no advanced glycation end-products were formed in the glycated hydrolysates.
Fluorescence spectroscopy can serve as a method of detecting the fluorescent products
generated by the Maillard reaction [25]. The fluorescent products can be detected at the
excitation wavelength of 347 nm, as well as the emission wavelength of 400–600 nm [10,26].
Fluorescence spectra of native hydrolysates, glycated hydrolysates derived from bovine
meat, bovine heart, porcine hemoglobin, and porcine plasma are illustrated in Figure 2A–D,
respectively. Apart from bovine heart, the remaining samples tended to display a pro-
gressive downtrend of fluorescence intensity with the emission wavelength ranging from
400 nm to 600 nm (Figure 2A,C,D). By contrast, bovine heart samples showed a downward
trend in the region of 400–480 nm, followed by a major peak around the region of 500–
600 nm (Figure 2B). It has previously been demonstrated that several fluorescent products
contribute to the increased fluorescence intensity by glycation using glucosamine, such as
Foods 2021, 10, x FOR PEER REVIEW 9 of 13
pentosidine, argpyrimidine, pentodilysine, pirropyridine, as well as the autocondensation
products of glucosamine, even at a moderate reaction temperature [10].

Figure 2. Fluorescence
Fluorescence spectra
spectra of hydrolysates
hydrolysates with (red) or without (blue) glucosamine (G) of (A) bovine meat, (B) bovine
heart, (C) porcine hemoglobin, and (D) porcine
heart, (C) porcine hemoglobin, and (D) porcine plasma.
plasma.

3.2.3.
3.2.3. Peptide
Peptide Size
Size Distribution
Distribution of
of Native
Native and
and Glycated
Glycated Hydrolysates
Hydrolysates
The
The peptide size distributions of native hydrolysates, glycated
peptide size distributions of native hydrolysates, glycated hydrolysates
hydrolysates derived
derived
from bovine meat, bovine heart, porcine hemoglobin, and
from bovine meat, bovine heart, porcine hemoglobin, and porcine plasma were porcine plasma were
determined
determined
and displayed andin displayed
Figure 3.inPlasma
Figure 3. Plasmatended
samples samplestotended to showpercentages
show higher higher percentages
of high
of high MW peptide fraction (>5 kDa) than those derived from other raw
MW peptide fraction (>5 kDa) than those derived from other raw materials (Figure 3), materials (Figure
3), as also
as also seen
seen in our
in our previous
previous study
study [7].[7]. This
This phenomenon
phenomenon might
might be be
duedue to the
to the high
high re-
resistance of plasma proteins towards Protease A, or to its tendency to
sistance of plasma proteins towards Protease A, or to its tendency to aggregate duringaggregate during
high-temperature
high-temperature heat heat treatment
treatment [7].
[7]. By
By contrast,
contrast, heart
heart hydrolysates
hydrolysates contained
contained the
the highest
highest
proportion
proportion of low lowMW MWpeptide
peptidefraction
fraction(<1(<1 kDa)
kDa) (Figure
(Figure 3) among
3) among the hydrolysates,
the hydrolysates, sug-
suggesting
gesting thatthat heart
heart muscle
muscle proteins
proteins are easily
are easily degraded
degraded into small
into small peptides
peptides or freeor free
amino
amino acids. Generally,
acids. Generally, there
there was no was no significant
significant change inchange in the size
the peptide peptide size distribution
distribution between
between native and glycated hydrolysates, which is not surprising, considering the
molecular weight of glucosamine is 161 Da. Furthermore, this is in line with the results of
both the sensory and the volatile analysis (discussed below)—indicating that the extent of
glycation was relatively low.
 Foods 2021, 10, 3008 9 of 12

native and glycated hydrolysates, which is not surprising, considering the molecular
Foods 2021, 10, x FOR PEER REVIEW weight of glucosamine is 161 Da. Furthermore, this is in line with the results of both
10 of the
13
sensory and the volatile analysis (discussed below)—indicating that the extent of glycation
was relatively low.

Peptidesize
Figure3.3.Peptide
Figure sizedistribution
distributionofofnative
nativeand
andglycated
glycatedhydrolysates.
hydrolysates.

3.3. Volatile Profiles Determined by GC-MS

3.3. Volatile Profiles Determined by GC-MS
In total, 69 volatiles were identified including aldehydes (20), alcohols (12), and
In total, 69 volatiles were identified including aldehydes (20), alcohols (12), and
ketones (10), in addition to typical Maillard reaction compounds such as furans and
ketones (10), in addition to typical Maillard reaction compounds such as furans and
pyrazines (see Supplementary Table S2). The chromatograms are shown in Supplementary
pyrazines (see Supplementary Table S2). The chromatograms are shown in
Figure S1.
Supplementary
For 13 of theFigure S1.
69 detected volatiles (Table S2), there was a statistically significant effect
For 13 of the 69 detected
of glucosamine addition. The volatiles
relative(Table S2), there was
concentrations a statistically
of the 13 compounds significant
are showneffectin
ofTable
glucosamine
3. addition. The relative concentrations of the 13 compounds are shown in
Table 3. The different effects of glucosamine addition were observed for bovine meat, bovine
heart, porcine hemoglobin, and porcine plasma hydrolysates, indicating that the effect of
Table 3. Volatile compounds where glucosamine (G) addition had an effect on quantities present in the hydrolysates.
the addition of glucosamine on volatile compound development in animal hydrolysates is
Relative quantity in hydrolysate without glucosamine is set to 1, and hydrolysate + G is indicated as fold increase.
rather raw material specific. The lipid oxidation product 3-methyl-2-butenal was the only
Volatile compound in which
Odor 1 an Meat
effect+of
G glucosamine
Heart + G Hemo was+observed
G Plasmafor + Gmore than one
Retention indexraw3 material

(bovine meat and porcine hemoglobin). Exp. Auth. std. Literature

2-Ethyl-1-hexanol YolkRose,
flavor, the only flavor attribute
green 1.8 that increased with the1500 addition
1503 of glucosamine
2-Methylbutanal (Table 2), Cocoa,
is often associated with the presence of sulfur
almond 0.5 compounds 911 such as sulfides
880–963and
3-Methylbutanal thiazoles Malt
[28], however, many aldehydes also act as0.6odor-active 915compounds
917 within an
(E)-2-Butenal egg yolk [29].
Flower These compounds 1.9originate from either the autoxidation
1030 of unsaturated
1002–1084
(E)-2-Methyl-2-butenal fatty acids, thermal
Green, fruit oxidation
4.5 of saturated triacylglycerols, or Strecker
1088 degradation [30].
1012–1113
Both 2-methylbutanal
Almond, and 3-methylbutanal (Table 3) have been detected in egg yolk,
3-Methyl-2-butenal 6.4 4.6 1197 1189–1236
the latterroasted
being 2notable for significant odor activity [28]. In the present experiment, a
difference---in 3-methylbutanal and0.72-methylbutanal content between
5-Ethylcyclopent-1-enecarboxaldehyde 1423hydrolysates with and
1399–1428
4-Ethylbenzaldehyde without glucosamine
Sweet was
1.9 only found for porcine plasma, yet in1722 those cases, this actually
1711–1753
Estragole represented a decrease
Licorice, with the addition of glucosamine (Table
anise 2.3 3). To our knowledge,
1681 1624–1701
out of the remaining volatiles identified via GC-MS where a difference was detected
Balsamic,
Myrcene 4.4 1172 1170
between must,
hydrolysates
spice with and without glucosamine, only D-limonene has previously
D-Limonene been identified
Citrus in egg yolk, however 6.3 it does not appear to be one of the
1194 1197main odor-active
1-Methyl-4-(1-methylethylidene)- compounds [29]. The presence of the terpenes D-limonene and myrcene in the hydrolysates
Pine, plastic 10.8 1291 1233–1323
cyclohexene is likely due to the animal feed.
Benzonitrile Apparently,
Rancid there was 2.3 no correlation between the results of1614 the sensory analysis and
1570–1637
1 Odor descriptors are from the volatile compound
www.flavornet.org analysis,
(accessed indicating
on June 16 2020).that the volatile
2 Odor compounds
descriptors from
are from [27] . 3 Table
Exp. =3 were
not
experimental value, Auth. std. necessarily
= authentic very odor-active. These findings are in line with the findings of the peptide
standard.

The different effects of glucosamine addition were observed for bovine meat, bovine
heart, porcine hemoglobin, and porcine plasma hydrolysates, indicating that the effect of
the addition of glucosamine on volatile compound development in animal hydrolysates
is rather raw material specific. The lipid oxidation product 3-methyl-2-butenal was the
Foods 2021, 10, 3008 10 of 12

characteristics analyses—all in all, showing that the extent of the Maillard reaction was
relatively low under the current experimental conditions.

Table 3. Volatile compounds where glucosamine (G) addition had an effect on quantities present in the hydrolysates.
Relative quantity in hydrolysate without glucosamine is set to 1, and hydrolysate + G is indicated as fold increase.

Volatile Odor 1 Meat + G Heart + G Hemo + G Plasma + G Retention Index 3

Exp. Auth. std. Literature
2-Ethyl-1-hexanol Rose, green 1.8 1500 1503
2-Methylbutanal Cocoa, almond 0.5 911 880–963
3-Methylbutanal Malt 0.6 915 917
(E)-2-Butenal Flower 1.9 1030 1002–1084
(E)-2-Methyl-2-butenal Green, fruit 4.5 1088 1012–1113
Almond,
3-Methyl-2-butenal 6.4 4.6 1197 1189–1236
roasted 2
5-Ethylcyclopent-1-
— 0.7 1423 1399–1428
enecarboxaldehyde
4-Ethylbenzaldehyde Sweet 1.9 1722 1711–1753
Estragole Licorice, anise 2.3 1681 1624–1701
Balsamic, must,
Myrcene 4.4 1172 1170
spice
D-Limonene Citrus 6.3 1194 1197
1-Methyl-4-(1-
methylethylidene)- Pine, plastic 10.8 1291 1233–1323
cyclohexene
Benzonitrile Rancid 2.3 1614 1570–1637
1Odor descriptors are from www.flavornet.org (accessed on 16 June 2020). 2 Odor descriptors are from [27]. 3 Exp. = experimental value,
Auth. std. = authentic standard.

Further investigation into the effect of heated systems on the flavor-forming potential
of hydrolysates both with and without added glucosamine is needed, as application of the
hydrolysates at higher temperatures, e.g., during cooking, would be expected to lead to
more extensive Maillard reactions than witnessed here.

4. Conclusions
In summary, glucosamine addition at a ratio of 1:10 relative to the protein content
of the raw material (bovine minced meat and heart, porcine hemoglobin and plasma)
provided only minor changes to sensory, volatile, and peptide-characteristics of animal
hydrolysates under the conditions investigated, including a slightly increased perception
of sweet taste, umami taste, and yolk flavor.
Due to the low perceived bitterness of all native and glycated hydrolysates there is
definitely potential for application in food products. As raw material heavily influences
the sensory attributes of the hydrolysate, this will influence the specific application of each
hydrolysate.
Glucosamine-induced Maillard reaction can have an effect on the flavor and peptide
characteristics of hydrolysates, however the effect observed here was minor, indicating
the need for a more extensive Maillard reaction in order to fully take advantage of the
glycation process.

Supplementary Materials: The following are available online at https://www.mdpi.com/article/10

.3390/foods10123008/s1, Table S1: Hydrolysates produced by 5 h enzymatic hydrolysis with Protease
A based on different types of raw material and glucosamine (G) addition at a ratio of 1:10 relative to
the protein content of the raw materials. Table S2: Peak areas (×10−5 ) of the volatile compounds in
the hydrolysates with or without glucosamine (G) relative to the peak area of the internal standard.
Different letters within the same rows indicate significant (p < 0.05) differences according to Tukey’s
Foods 2021, 10, 3008 11 of 12

test. Figure S1: Chromatograms showing the volatile composition of hydrolysates of A minced
bovine meat (BEEF), B bovine heart (HEART), C porcine hemoglobin (HEMO), and D porcine plasma
(PLASMA) in dublicates (A and B) with or without glucosamine (G).
Author Contributions: Conceptualization, K.H.B., S.S.W., Y.F., J.R.-C. and R.L.; methodology, K.H.B.,
S.S.W., Y.F. and M.A.P.; formal analysis, K.H.B., S.S.W., Y.F., C.Y.C. and M.A.P.; investigation, K.H.B.,
S.S.W. and Y.F.; writing—original draft preparation, K.H.B., S.S.W., Y.F. and M.A.P.; writing—review
and editing, K.H.B., S.S.W., Y.F., C.Y.C., M.A.P., J.R.-C., W.L.P.B. and R.L.; visualization, K.H.B., S.S.W.
and Y.F.; supervision, M.A.P., J.R.-C., W.L.P.B. and R.L.; project administration, J.R.-C. and R.L.;
funding acquisition, J.R.-C. and R.L. All authors have read and agreed to the published version of the
manuscript.
Funding: This research was funded by the Norma and Frode S. Jacobsen’s Foundation project
number 274 and by the Danish Ministry of Food, Agriculture and Fisheries under GUDP grant
No. 34009-16-1085. The authors further acknowledge the support from the Science and Technology
Research Program of Chongqing Municipal Education Commission grant No. KJQN202100225. The
APC was funded by the University of Veterinary Medicine Vienna.
Institutional Review Board Statement: Not applicable.
Informed Consent Statement: Not applicable.
Data Availability Statement: The data presented in the study are available from the corresponding
author upon request.
Acknowledgments: Open Access Funding by the University of Veterinary Medicine Vienna.
Conflicts of Interest: The authors declare no conflict of interest.

References
1. Alao, O.B.; Falowo, B.A.; Chulayo, A.; Muchenje, V. The Potential of Animal By-Products in Food Systems: Production, Prospects
and Challenges. Sustainability 2017, 9, 1089. [CrossRef]
2. Jayathilakan, K.; Sultana, K.; Radhakrishna, K.; Bawa, A. Utilization of byproducts and waste materials from meat, poultry and
fish processing industries: A review. J. Food Sci. Technol. 2012, 49, 278–293. [CrossRef] [PubMed]
3. Toldrá, F.; Mora, L.; Reig, M. New insights into meat by-product utilization. Meat Sci. 2016, 120, 54–59. [CrossRef]
4. Lafarga, T.; Hayes, M. Bioactive peptides from meat muscle and by-products: Generation, functionality and application as
functional ingredients. Meat Sci. 2014, 98, 227–239. [CrossRef] [PubMed]
5. Mullen, A.M.; Álvarez, C.; Zeugolis, D.I.; Henchion, M.; O’Neill, E.; Drummond, L. Alternative uses for co-products: Harnessing
the potential of valuable compounds from meat processing chains. Meat Sci. 2017, 132, 90–98. [CrossRef] [PubMed]
6. Fitzgerald, R.; O’Cuinn, G. Enzymatic debittering of food protein hydrolysates. Biotechnol. Adv. 2006, 24, 234–237. [CrossRef]
[PubMed]
7. Fu, Y.; Liu, J.; Hansen, E.T.; Bredie, W.L.P.; Lametsch, R. Structural characteristics of low bitter and high umami protein
hydrolysates prepared from bovine muscle and porcine plasma. Food Chem. 2018, 257, 163–171. [CrossRef]
8. Bak, K.H.; Petersen, M.; Lametsch, R.; Hansen, E.; Ruiz-Carrascal, J. Development of Volatile Compounds during Hydrolysis of
Porcine Hemoglobin with Papain. Molecules 2018, 23, 357. [CrossRef]
9. Parke, S.A.; Birch, G.G.; Dijk, R. Some Taste Molecules and their Solution Properties. Chem. Senses 1999, 24, 271–279. [CrossRef]
10. Hong, P.K.; Ndagijimana, M.; Betti, M. Glucosamine-induced glycation of hydrolysed meat proteins in the presence or absence of
transglutaminase: Chemical modifications and taste-enhancing activity. Food Chem. 2016, 197, 1143–1152. [CrossRef]
11. Liu, P.; Huang, M.; Song, S.; Hayat, K.; Zhang, X.; Xia, S.; Jia, C. Sensory Characteristics and Antioxidant Activities of Maillard
Reaction Products from Soy Protein Hydrolysates with Different Molecular Weight Distribution. Food Bioprocess. Technol. 2012, 5,
1775–1789. [CrossRef]
12. Ogasawara, M.; Katsumata, T.; Egi, M. Taste properties of Maillard-reaction products prepared from 1000 to 5000 Da peptide.
Food Chem. 2006, 99, 600–604. [CrossRef]
13. Eric, K.; Raymond, L.V.; Huang, M.; Cheserek, M.J.; Hayat, K.; Savio, N.D.; Amédée, M.; Zhang, X. Sensory attributes and
antioxidant capacity of Maillard reaction products derived from xylose, cysteine and sunflower protein hydrolysate model
system. Food Res. Int. 2013, 54, 1437–1447. [CrossRef]
14. Song, N.; Tan, C.; Huang, M.; Liu, P.; Eric, K.; Zhang, X.; Xia, S.; Jia, C. Transglutaminase cross-linking effect on sensory
characteristics and antioxidant activities of Maillard reaction products from soybean protein hydrolysates. Food Chem. 2013, 136,
144–151. [CrossRef]
15. Sanmartín, E.; Arboleya, J.C.; Villamiel, M.; Moreno, F.J. Recent Advances in the Recovery and Improvement of Functional
Proteins from Fish Processing By-Products: Use of Protein Glycation as an Alternative Method. Compr. Rev. Food Sci. Food Saf.
2009, 8, 332–344. [CrossRef]
Foods 2021, 10, 3008 12 of 12

16. Fu, Y.; Liu, J.; Zhang, W.; Wæhrens, S.S.; Tøstesen, M.; Hansen, E.T.; Bredie, W.L.P.; Lametsch, R. Exopeptidase treatment
combined with Maillard reaction modification of protein hydrolysates derived from porcine muscle and plasma: Structure–taste
relationship. Food Chem. 2020, 306, 125613. [CrossRef]
17. Fu, Y.; Bak, K.H.; Liu, J.; De Gobba, C.; Tøstesen, M.; Hansen, E.T.; Petersen, M.A.; Ruiz-Carrascal, J.; Bredie, W.L.P.; Lametsch, R.
Protein hydrolysates of porcine hemoglobin and blood: Peptide characteristics in relation to taste attributes and formation of
volatile compounds. Food Res. Int. 2019, 121, 28–38. [CrossRef]
18. Horwitz, W. Method 955.04. In Official Methods of Analysis of AOAC International, 17th ed.; AOAC International: Gaithersburg,
MD, USA, 2000.
19. ISO. ISO 13299:2003 Sensory Analysis—Methodology—General Guidance for Establishing a Sensory Profile; ISO: Geneva, Switzerland,
2003.
20. Kuznetsova, A.; Brockhoff, P.; Christensen, R. LmerTest: Tests in linear mixed effects models. R Package Version 2017, 82, 1–26.
[CrossRef]
21. Johnsen, L.G.; Skou, P.B.; Khakimov, B.; Bro, R. Gas chromatography—Mass spectrometry data processing made easy. J.
Chromatogr. A 2017, 1503, 57–64. [CrossRef] [PubMed]
22. Ogutu, B.; Kim, Y.J.; Kim, D.W.; Oh, S.C.; Hong, D.L.; Lee, Y.B. Optimization of Maillard Reaction between Glucosamine and
Other Precursors by Measuring Browning with a Spectrophotometer. Prev. Nutr. Food Sci. 2017, 22, 211–215. [CrossRef]
23. Delgado-Andrade, C.; Seiquer, I.; Haro, A.; Castellano, R.; Navarro, M.P. Development of the Maillard reaction in foods cooked
by different techniques. Intake of Maillard-derived compounds. Food Chem. 2010, 122, 145–153. [CrossRef]
24. Hong, P.K.; Betti, M. Non-enzymatic browning reaction of glucosamine at mild conditions: Relationship between colour formation,
radical scavenging activity and α-dicarbonyl compounds production. Food Chem. 2016, 212, 234–243. [CrossRef]
25. Ashraf, J.M.; Ahmad, S.; Choi, I.; Ahmad, N.; Farhan, M.; Tatyana, G.; Shahab, U. Recent advances in detection of AGEs:
Immunochemical, bioanalytical and biochemical approaches. IUBMB Life 2015, 67, 897–913. [CrossRef]
26. Matiacevich, S.B.; Santagapita, P.R.; Buera, M.P. Fluorescence from the maillard reaction and its potential applications in food
science. Crit. Rev. Food Sci. Nutr. 2005, 45, 483–495. [CrossRef] [PubMed]
27. Kishimoto, T.; Wanikawa, A.; Kono, K.; Shibata, K. Comparison of the Odor-Active Compounds in Unhopped Beer and Beers
Hopped with Different Hop Varieties. J. Agric. Food Chem. 2006, 54, 8855–8861. [CrossRef]
28. Plagemann, I.; Zelena, K.; Krings, U.; Berger, R. Volatile flavours in raw egg yolk of hens fed on different diets. J. Sci. Food Agric.
2011, 91, 2061–2065. [CrossRef]
29. Cerny, C.; Guntz, R. Evaluation of potent odorants in heated egg yolk by aroma extract dilution analysis. Eur. Food Res. Technol.
2004, 219, 452–454. [CrossRef]
30. Zhang, Y.; Liu, Y.; Yang, W.; Huang, J.; Liu, Y.; Huang, M.; Sun, B.; Li, C. Characterization of the Potent Aroma Compounds in
Preserved Egg Yolk by Gas Chromatography-Olfactometry, Quantitative Measurements, and Odor Activity Value. J. Agric. Food
Chem. 2018, 66, 6132–6141. [CrossRef] [PubMed]