EUROPEAN M.SC. DEGREE IN FOOD SCIENCE, TECHNOLOGY AND NUTRITION THESIS AUTHOR MARIA GABRIELA ARAUJO MIÑO TITLE Influence of standardization, rennet type, curd wash level and cook temperature on the composition, microbiology, functionality, flavour and ripening of novel Swiss-type cheeses June 2012 Dublin Institute of Technology Teagasc Food Research Centre Moorepark
101
Embed
EUROPEAN M.SC. DEGREE IN FOOD SCIENCE, TECHNOLOGY ANDrepositorio.educacionsuperior.gob.ec/bitstream/28000/307/1/T... · M.Sc. degree in Food Science, Technology and Nutrition is entirely
This document is posted to help you gain knowledge. Please leave a comment to let me know what you think about it! Share it to your friends and learn new things together.
Transcript
EUROPEAN M.SC. DEGREE IN FOOD SCIENCE, TECHNOLOGY AND
NUTRITION
THESIS
AUTHOR
MARIA GABRIELA ARAUJO MIÑO
TITLE
Influence of standardization, rennet type, curd wash level and cook
temperature on the composition, microbiology, functionality, flavour
and ripening of novel Swiss-type cheeses
June 2012
Dublin Institute of Technology
Teagasc Food Research Centre Moorepark
II
DECLARATION
I hereby certify that the material which is submitted in this thesis towards award of the European
M.Sc. degree in Food Science, Technology and Nutrition is entirely my own work and has not
been submitted for any academic assessment other than part-fulfillment of the award named
above.
Signature of candidate:…………………………………..
Date:………………………
III
ACKNOWLEDGEMENTS
I would like to thank The National Secretary of Higher education, Science, Innovation and
technology in Ecuador (SENESCYT) for providing me the funding which enabled me to carry out
this project.
I would like to thank the Irish Dairy Board and Teagasc for providing me the opportunity to become
part of their team.
This dissertation would not have been possible without the guidance and the help of several
individuals who in one way or another contributed and extended their valuable assistance in the
preparation and completion of this study.
First and foremost, I would like to acknowledge the advice and guidance of my placement
supervisor Dr. Diarmuid Sheehan Research Official of the Food Chemistry and Technology
department in Teagasc Food Research Centre Moorepark (TFRCM).
I would also like to express my thanks and gratitude to my laboratory supervisor Dr. Nuria Costa,
who has supported me throughout this thesis with her patience, knowledge, encouragement
friendship, helpful advice, and valuable guidance.
I would like to extend my sincere appreciation to Mairead Stack for putting time on her own
schedule to read this thesis and for the opportunity she gave me to perform my intership in the Irish
Dairy Board and Teagasc.
I thank all Teagasc staff and especially Joanne Hayes, Paula O’Connor, Siobhán Ryan and Anne
Marie McAuliffe for their technical support during the course of this study.
To my friends Julia Adriana, Cristina, Daniel, Maria, thanks for your help, friendship and for the
many laughs shared.
It gives me immense pleasure to thank all my family members, especially my parents, Ana and
Bolivar, my brothers, Emilio and Vanessa, my aunt, Patricia, and my uncle, Fernando for supporting
and encouraging me to pursue this degree.
I am deeply grateful with Carlos for his patience, support, tolerance, love and friendship.
IV
ABBREVIATIONS
BSI British Standard Institution
Ca Calcium
CFU Colony Forming Units
FAA Free amino acids
FDM Fat in Dry Matter
FDM Fat in dry matter
g grams
GLM General linear model
HPLC High Performance Liquid Chromatography
IDF International Dairy Federation
kPa Kilopascal
L Litres
LAB Lactic Acid Bacteria
Lb Lactobacillus
LBS Lactobacillus selective medium
LM17 Lactose medium
MCA Milk clotting activity
MNFS Moisture in the non fat substance
N Newton
NSLAB Non -starter lactic acid bacteria
P Probability value
PAB Propionic acid bacteria
PCA Principal component analysis
pH 4.6 SN pH 4.6-soluble nitrogen
pH 4.6 SN% TN pH 4.6-soluble nitrogen expressed as a % of total nitrogen
S/M Salt in moisture
SLA Sodium lactate agar
TCA Trichloroacetic acid
TN Total nitrogen
YGC Yeast extract agar
V
TABLE OF CONTENTS
Declaration ........................................................................................................................................ II
Acknowledgements .......................................................................................................................... III
Abbreviations ................................................................................................................................... IV
Table of contents ............................................................................................................................... V
CHAPTER 3. Influence of cooking temperature during cheese manufacture on the composition, microbiology, proteolysis, functionality and flavour of novel Swiss-type cheese made with yeast adjunct...........................................................................................57 Abstract ........................................................................................................................................... 58
Ash content. - Determined gravimetrically by heating a sample in a furnace at or below
550 °C until completely ashed. IDB method 27, 1964.
Protein. – Determined by measuring the Nitrogen content of cheese by the macro-block
digestion. IDF method 20B, 1993.
Lactates were measured using the Megazyme kit (D-/L-Lactic Acid Kit, Megazyme
International Ireland Ltd., Bray, Ireland).
Sugars (lactose and galactose) were analyzed by HPLC as described by the method of
Zeppa et al. (2001).
2.3.6. Assessment of Proteolysis in Cheese.
2.3.6.1. Primary proteolysis
The combined effect of milk standardization, rennet type and curd wash level on primary
proteolysis, was measured by monitoring levels of pH 4.6 soluble nitrogen (SN), at 120 d, 150 d
and 180 d. The amount of nitrogen soluble in water at pH 4.6, expressed as a percentage of total
nitrogen (TN) in cheese was measured using the method of Kuchroo and Fox (1982). A sample of
grated cheese (60 g) was placed in a stomacher bag to which distilled water at 55°C (120 g) was
added and the contents blended into a Stomacher (Lab-Blender 400; Seward Medical, London) for
5 min at room temperature. The resulting homogenate was incubated in a water bath at 55°C and
allowed to stand for 1 hour. The contents of the stomacher bag were centrifuged at 2500 rfc for 20
min at 4°C. (Mistral 3000 centrifuge, Block Scientific Inc, Germany). After centrifuging the
supernatant was filtered through glass wool and the filtrate was adjusted to pH 4.6 using 10% HCl
24
and centrifuged at the same conditions as before. The supernatant was filtered though glass wood
and the pH 4.6 soluble extract was analysed by a macro-block digestion method (IDF, 1993) to
determine the content of water soluble nitrogen.
2.3.6.2. Secondary proteolysis
Free amino acids contents were determined in pH 4.6 SN extracts on cheeses of 120 d, 150 d and
180 d of ripening. Samples were deproteinised by mixing equal volumes of pH 4.6 SN and
trichloroacetic acid (240 g/L). Free amino acids were separated using ion-exchange
chromatography with post-column ninhydrin and visible colorimetric detection as described by
Fenelon et al. (2002). Samples were analysed in duplicate.
2.3.7. Functionality
2.3.7.1. Texture
Cheese samples (25 mm3 cubes) were cut from the slab of cheese (Cheese Blocker; Boos
Kaasgreedschap, Bodengraven, Netherlands) and stored at 4°C overnight before analysis. Six
cheeses cubes were analyzed by compression on a TA-HDi Texture Profile analyzer (model TA-
HDI, Stable Micro Systems, Godalming, UK) with a 5 mm compression plate and a 100 kg load cell
at room temperature. Each sample was subjected to 2 consecutive compressions at a speed of 1
mm/s, each to 30% of original sample height, as described in Rynne et al., (2004). Texture profile
analysis parameters were calculated. Hardness (N) was measured as the force at maximum
compression on the first bite. Fracture stress (kPa) was measured as the force per unit area at the
point of fracture on the first bite, fracture strain (dimensionless) was measured as the strain
corresponding to the minimum slope on the force-displacement curve. Cohesiveness
(dimensionless) was calculated as the ratio of the area of the second bite to that of the first bite.
Springiness (dimensionless) was calculated as the ratio of distance of the second bite (peak) to the
distance of the first bite (peak), and chewiness (N) was calculated as the product of harness x
cohesiveness x springiness. Adhesiveness (N * S) was calculated as the negative area after the
first bite in the texture profile curve as described in Van Vliet (1991), Bourne (1978) and expressed
in absolute values.
25
2.3.7.2. Flowability of the heated cheese
Flowability was measured by (i) the Schereiber method modified by Guinee et al. (2002), defined as
the percentage increase in the diameter of a disc of cheese (45 mm diameter, 6.5 mm thick)
melting at 280°C for 4 min, and (ii) by Olson/Price method as modified by Rynne et al. (2004); in
which a 15 g cylindrical cheese sample (diameter, 22 mm; height, 35 mm) was placed in the centre
of a 100 ml graduated glass tube, one end of which was closed and the other fitted with a rubber
bung. The tube was placed in a horizontal position, on a stainless steel tray, in an electric fan oven
at 180 °C for 7.5 min. The tray was then removed and allowed to cool to room temperature (~ 20
min), and the percentage flow was defined as the percentage increase in the length of the cylinder
of cheese.
2.3.8. Flavour
2.3.8.1. Assessment of short chain volatile fatty acids
Acetate, propionate and n-butyrate (C2:0, C3:0, C4:0) contents were determined in cheeses after
120 d, 150 d, and 180 d of ripening. Short chain volatile fatty acids (SCVFA) were obtained as
described by Kilcawley et al. (2001). Five grams of grated cheese were placed in a distillation tube
with 5 ml of 10% sulphuric acid, 0.5 ml of valeric acid at a concentration of 5 mg/ml and 10 ml of
distilled water. The solution was distilled in the distillation unit (2100 Kjeltec, Foss Tecator) and 100
ml of distillate were collected, filtered (0.2 µm filter) and injected onto the HPLC (Water Alliance
system 2695). Individual fatty acids were quantified by relating the area of each peak to the area of
the peaks of the fatty acids used in the internal standard. The final concentration of individual FFA
was expressed as mg of individual short chain fatty acid per kg of cheese. Analyses were
performed in triplicate and results averaged.
2.3.8.2. Volatiles profile
The volatile profiles of the headspace of each sample was analysed by solid phase micro-extraction
(SPME) gas chromatography mass spectrometry (GCMS).
For volatile analysis, 5 g of sample was added to a 20 ml amber screw capped SPME vial and
equilibrated at 40°C for 5 min with pulsed agitation of 4 s at 400 rpm. Sample introduction was
accomplished using a CTC Analytics CombiPal Autosampler. A single DVD/Carboxen/PDMS 1 cm
fiber was used for all analysis. The SPME fiber was exposed to the headspace above the samples
26
for 25 min at depth of 1 cm with pulsed agitation of 4s at 350 rpm. The fiber was retracted and
injected into the GC inlet at 250°C and desorbed for 2 min. Injections were made on a Varian 450
GC with a Perkin Elmer Elite DMS (60 m x 0.25 mm ID x 0.25 DF µm) column. The detector used
was a Varian 320 triple quad mass spectrometer. Individual compounds were identified using mass
spectral comparisons to the NIST 2005 mass spectral library. Individual compounds were assigned
quantification and qualifier ions to ensure that only the individual compounds were identified and
quantified, especially in the case of co-eluting or semi-co-eluting samples. Compounds were
quantified by calculating the area under the peak of each compound and are expressed in arbitrary
units. An autotune of the GCMS was carried out immediately prior to analysis to confirm that the
GCMS was operating under optimal conditions. Each sample was analysed in duplicate.
2.3.9. Statistical analysis
All statistical analysis was carried out using SAS (version 9.1.3, SAS Institute, Cary, NC). Analysis
of variance was carried out on data using the general linear model procedure of SAS (SAS
Institute). The Tukey honestly significant difference test was used to determine the significance of
difference between the means. The level of significance was determined at P < 0.05.
For variables analysed at several times during ripening, analysis of variance for the split-plot design
was carried out on data using the mixed procedure of SAS (SAS Institute).
Statistically significant differences (P < 0.05) between different treatment levels were determined by
using Tukey honestly significant difference.
Principal component analysis (PCA) of the individual amino acids, short chain fatty acids, texture
parameters, and volatiles compounds were performed by using the statistical software The
Unscrambler (v 9.7, CAMO, Norway).
27
2.4. Results
2.4.1. Raw milk composition
Milk used for the manufacture of S1 cheeses contained on average: 4.20% of fat, 3.55% of protein
and 4.63% of lactose, while milk used for manufacture of S2 cheeses contained: 4.28% of fat,
3.63% of protein and 4.627% of lactose.
2.4.2. Composition of standardized milk
The composition of the standardized milk used for this study can be observed in table 2.2
Table 2.2. Composition of standardized milk used for the manufacture of the novel Swiss-type cheeses.
Composition* Treatment
S1 SD** S2 SD**
Fat % 3.61 ± 0.17 4.02 ± 0.19
Protein % 3.56 ± 0.16 3.61 ± 0.15
Lactose % 4.67 ± 0.09 4.63 ± 0.03
Protein: Fat %
0.99 ± 0.01 0.90 ± 0.00
Casein: Fat %
0.73 ± 0.01 0.67 ± 0.02
Protein: Lactose %
0.76 ± 0.05 0.78 ± 0.03
S1 = Standardized milk for the manufacture of a novel Swiss-type cheese with recombinant rennet, with curd washing step and 0.99:1 of protein: fat ratio S2 = Standardized milk for the manufacture of a novel Swiss-type cheese with microbial rennet, without curd washing step and 0.90:1 of protein: fat ratio. *Values presented are the means of 3 replicates **Standard deviation of 3 replicates
2.4.3. Gross composition
The mean composition of the S1 and S2 cheeses are given in Table 2.3 and are typical for a Swiss-
type cheese (Fox et al., 2000; Sheehan et al., 2007).
28
Table 2.3. Gross composition of the novel Swiss-type cheeses.
a, b Values within a row not sharing a common superscript, differ significantly ( P < 0.05) S1 = cheese produced with recombinant rennet, with curd washing step and std. milk 0.99:1 of protein: fat ratio S2 = cheese produced with microbial rennet, without curd washing step and std. milk 0.90:1 of protein: fat ratio. * Values presented are the means of 3 replicates ** Standard deviation of 3 replicates ¹MNFS= Moisture in the non-fat substance ² FDM= Fat in dry matter ³ S/M= Salt in moisture
2.4.3.1. Lactates and sugars
The mean levels of lactates and sugars are shown in Table 2.4. Significant differences (P < 0.05)
were found in levels of L-lactate, total lactate, galactose, and protein: lactose ratio.
29
Table 2.4. Lactate and sugar contents of the novel Swiss-type cheeses.
Composition* Treatment
S1 SD** S2 SD**
d-lactate g/100g cheese
0.147a ± 0.078 0.215a ± 0.035
l-lactate g/100g cheese
0.549a ± 0.064 1.139b ± 0.073
Total lactate g/100g cheese
0.696a ± 0.058 1.354b ± 0.092
Protein: lactate
40.91a ± 3.272 19.693b ± 1.269
Lactose g/100g cheese
0.006a ± 0.01 0.0006a ± 0.001
Galactose g/100g cheese
0.347a ± 0.12 0.165b ± 0.131
a, b Values within a row not sharing a common superscript, differ significantly (P < 0.05) S1 = cheese produced with recombinant rennet, with curd washing step and 0.99:1 of protein: fat ratio S2 = cheese produced with microbial rennet, without curd washing step and 0.90:1 of protein: fat ratio. * Values presented are the means of 3 replicates ** Standard deviation of 3 replicates
2.4.4. Proteolysis
Primary proteolysis, as measured by levels of pH 4.6 SN and expressed as a percentage of total
nitrogen (Table 2.5), increased significantly (P < 0.05) in all cheeses during the ripening period.
The mean levels of pH 4.6 SN were significantly affected (P < 0.05) by the combined effect of
coagulant type; milk composition and curd wash level.
The levels of total free amino acids (secondary proteolysis) are shown in Table 2.5 and Figure 2.1.
30
Table 2.5. Levels of pH 4.6 soluble nitrogen and total free amino acids during ripening of novel
Swiss-type cheeses
Ripening time (d)
Treatment
S1 S2
pH 4.6 SN (% of total N)
120 16.62 a,A 25.24 b,A
150 18.99 a,B 27.86 b,B
180 21.37 a,C 30.99 b,C
Total free AA (mg/kg of cheese)
120 13395 a,A 23945 b,A
150 18126 a,B 27964 b,B
180 19772 a,B 29061 b,B
a, b Values within a row not sharing a common superscript, differ significantly (P < 0.05) A,B Values within a column not sharing a common superscript, differ significantly (P < 0.05). S1 = cheese produced with recombinant rennet, with curd washing step and std. milk 0.99:1 of protein: fat ratio S2 = cheese produced with microbial rennet, without curd washing step and std. milk 0.90:1 of protein: fat ratio
15.00
20.00
25.00
30.00
35.00
120 140 160 180 200
pH
4.6
-SN
/ %
TN
Time (days)
Primary proteolysis pH4.6-SN (% of total N)
S1
S2
Fig 2.1. The combined effects of S1 = cheese produced with recombinant rennet, with curd washing step and std. milk 0.99:1 of protein: fat ratio and S2 = cheese produced with microbial rennet, without curd washing step and std milk. 0.90:1 of protein: fat ratio, on levels of pH 4.6 soluble nitrogen, expressed as % of total N, during ripening. Values presented are the means of tree replicates.
31
0
10000
20000
30000
40000
120 140 160 180 200
FA
A (
mg
/K
g c
he
ese
)
Time (days)
Secondary proteolysis (as determined by the levels of total FAA) over time
S1
S2
Fig 2.2. The combined effect of S1 = cheese produced with recombinant rennet, with curd washing step and std. milk 0.99:1 of protein: fat ratio and S2 = cheese produced with microbial rennet, without curd washing step and std. milk 0.90:1 of protein: fat ratio, on levels of total free amino acids in pH 4.6 soluble nitrogen extract over ripening. Values presented are the means of tree replicates.
Secondary Proteolysis Individual free amino acids
0
1000
2000
3000
4000
5000
Asp
Thre Se
r
Glu
Gly
Ala
Cys Val
Met Ile Leu
Tyr
Ph
e
His
Lys
Arg
Pro
FAA
mg/
kg o
f ch
ee
se
Free AA120 d of ripenig
S1
S2
0
1000
2000
3000
4000
5000
Asp
Thre Se
r
Glu
Gly
Ala
Cys Val
Met Ile Leu
Tyr
Ph
e
His
Lys
Arg
Pro
FAA
mg/
kg o
f ch
ee
se
Free AA150 d of ripenig
S1
S2
32
0
1000
2000
3000
4000
5000
Asp
Thre Se
r
Glu
Gly
Ala
Cys Val
Met Ile Leu
Tyr
Ph
e
His
Lys
Arg
Pro
FAA
mg/
kg o
f ch
ee
se
Free AA 180 d of ripening
S1
S2
Fig 2.3. The combined effect of S1 = cheese produced with recombinant rennet, with curd washing step and std. milk 0.99:1 of protein: fat ratio and S2 = cheese produced with microbial rennet, without curd washing step and std. milk 0.90:1 of protein: fat ratio, on levels of individual free amino acids in pH 4.6 soluble nitrogen extract from cheeses at 120 d, 150 d and 180 d of ripening. Values presented are the means of tree replicates.
gumminess and cohesiveness) of the cheeses over ripening.
Hardness, defined as the high resistance to deformation by applied stress, was significantly (P <
0.001) lower in S2 cheeses. Fracture stress, defined as the force at which a cheese crumbles,
cracks, or shatters when deformed, it is the result of a high degree of hardness and a low degree of
adhesiveness (Fox et al., 2000). Fracture stress was significantly lower (P < 0.001) in S2 cheeses.
Gumminess determined as the energy required for disintegrating a piece of cheese to a state ready
for swallowing, was significantly (P < 0.001) lower in S2 cheeses. Gumminess is correlated with
hardness (Fox et al., 2000).
Chewiness, the length of time or the number of chews required to masticate a cheese to a state
ready for swallowing, was significantly lower (P < 0.05) in S2 cheeses.
Springiness (elasticity), defined as the tendency to recover from large deformation (strain) after
removal of deforming stress, was significantly (P < 0.05) higher in S2 cheese. Springiness
increases upon elevation of fat levels (Fox et al., 2000).
No significant differences were found in the parameter of cohesiveness, indicating that the extent to
which cheeses were deformed before they ruptured was the same for both treatments.
33
0
50
100
150
200
250
300
350
120 180
N
Time (days)
Hardness
S1
S2
05
101520253035
120 180
kP
a
Time (days)
Fracture Stress
S1
S2
0.0
0.5
1.0
1.5
2.0
2.5
3.0
3.5
120 180
-
Time (days)
Springiness
S1
S2
0.00
0.05
0.10
0.15
0.20
0.25
120 180
-
Time (days)
Cohesiveness
S1
S2
0
10
20
30
40
50
60
120 180
N
Time (days)
Gumminess
S1
S2
020406080
100120140
120 180
N
Time (days)
Chewiness
S1
S2
Fig 2.4. Evolution of texture parameters over ripening of novel Swiss-type cheeses S1= cheese produced with recombinant rennet, with curd washing step and std. milk 0.99:1 of protein: fat ratio S2 = cheese produced with microbial rennet, without curd washing step and std milk. 0.90:1 of protein: fat ratio.
34
2.4.5.2. Flowability of the heated cheese
The ability of the melted cheese to flow during ripening is shown in Figure 2.5. Flow levels of S2
cheese were significantly higher (P < 0.01) than that of the S1 cheese at all ripening times and
measured by the 2 methods (Schreiber and Olson). Flowability significantly (P < 0.001) increased
with ripening time as measured by the Schreiber method for both cheeses, while this increase was
not significant as measured by the Olson test
0
20
40
60
80
120 180
Flo
w %
Time (days)
Flowability Schreiber Test
S1
S2
0
100
200
300
400
120 180
% F
low
Time (days)
FlowabilityOlson Test
S1
S2
Fig. 2.5. Evolution of the flowability as measured by the Schreiber method and the Olson method, during ripening of novel Swiss-type cheeses. S1= cheese produced with recombinant rennet, with curd washing step and std. milk 0.99:1 of protein: fat ratio S2 = cheese produced with microbial rennet, without curd washing step and std. milk 0.90:1 of protein: fat ratio.
2.4.6. Flavour
2.4.6.1. Acetic, propionic and butyric acid levels during ripening
Figure 2.6 shows the mean levels of acetic, propionic and butyric acid over ripening..
35
S1 cheeses had significantly higher (P < 0.05) levels of acetic acid (3000 mg/kg) than S2 cheeses
(1500 mg/kg) (Fig. 2.6). Levels of acetate did not change significantly over ripening for either
cheese.
Levels of propionate were significantly higher (P < 0.001), at 180 d of ripening in S1 cheeses
(4200mg/kg) than S2 cheeses (800mg/kg) (Fig. 2.6) Levels of propionate increase significantly in
S2 and numerically in S1.
Levels of butyrate were not affected. Butyrate increased during ripening but the increment was
numerical rather than statistical (Fig. 2.6)
0
1000
2000
3000
4000
120 150 180
mg
/kg
ch
eese
Time (days)
Acetic Acid
S1S2
0100020003000400050006000
120 150 180
mg
/kg
ch
eese
Time (days)
Propionic Acid
S1
S2
0
200
400
600
800
120 150 180
mg
/kg
ch
eese
Time (days)
Butyric Acid
S1S2
Fig 2.6. Evolution of acetic, propionic and butyric acid during ripening, S1= cheese produced with recombinant rennet, with curd washing step and std. milk 0.99:1 of protein: fat ratio and S2 = cheese produced with microbial rennet, without curd washing step and std. milk 0.90:1 of protein: fat ratio.
2.4.6.2. Volatiles
36
Relevant chromatograms show that cheese samples S1 trial 1, 2 and 3 were highly similar to each
other (Fig. 2.7), as were cheese samples S2 Trial 1, 2 and 3 (Fig. 2.8), confirming good
repeatability. Principal component analysis (PCA) of S1 (Trial 1, 2 and 3) and S2 (Trial 1, 2 and 3)
indicated that these cheeses are quite different from one another (Fig. 2.9), with the cheeses
grouping on opposite sides of the PCA, S2 on the left and S1 on the right hand side. The PCA
explained a total of 72% percent of the total variance.
Lawlor et al, (2002) also grouped volatiles compounds in Swiss-type cheeses using PCA. The PCA
indicated that the main components strongly correlated with Swiss type cheeses (Fig 2.10) were :
* The LRI and Odour Database” at www.odour.org.uk (maintained by Dr. R. Mottram, Flavour Research Group, School of Food Biosciences, Univ. of Reading).
F10 Fig. 2.7. GC-MS chromatograms of the headspace volatiles for S1 cheeses at 120d of ripening. S1= cheese produced with recombinant rennet, with curd washing step and std. milk 0.99:1 of protein: fat ratio.
S1T1 4mth
S1T2 4mth
S1T3 4mth
Fig. 2.8. GC-MS chromatograms of the headspace volatiles for S2 cheeses at 120d of ripening. S2 = cheese produced with microbial rennet, without curd washing step and std. milk 0.90:1 of protein: fat ratio.
Fig. 2.9. Result of principal component analysis of volatile compounds of S1 and S2 cheeses at 120 d ripening. S1= cheese produced with recombinant rennet, with curd washing step and std. milk 0.99:1 of protein: fat ratio and S2 = cheese produced with microbial rennet, without curd washing step and std. milk 0.90:1 of protein: fat ratio.
Fig. 2.10. Result of principal component analysis of volatile compounds, free fatty acid, free amino acid and gross compositional constituents of eight hard-type cheeses (in bold font) showing the first two principal components. FDM=fat in the dry matter, S/M=salt in moisture, MNFS=moisture in the non-fat substance, pH 4.6-SN=pH 4.6-solible nitrogen (Lawlor et al., 2002)
2.5. Discussion
The production of novel cheese types on existing equipment involves the manipulation of
process and ripening variables, e.g., starter, rennet types, cook temperature, draining pH, etc;
with the aim of generating unique flavours and textures.
The combined effect of rennet-type, milk composition and curd wash, on the composition,
ripening parameters, functionality and flavour of novel Swiss-type cheeses was assessed in the
present study.
No significant differences were found in the compositional parameters of moisture, protein, salt,
ash, calcium, calcium: protein, P/F, MNFS and S/M. However, significant differences (P < 0.05)
were found in levels of pH, total lactates, fat content and FDM.
According to Fox and McSweeney (1997), Kosikowski and Mistry (1997) coagulant type does
not affect cheese gross composition. The marked differences in sugar metabolism and pH
values are probably due to the curd washing step which was applied to S1 cheeses. Curd
washing reduces lactose content and lactic acid concentration in the curd, affecting the pH of
the resultant cheese (Jia Hou et al., 2012). For instance, levels of total lactates in S1 cheeses
were significantly lower (P < 0.05), and pH was significantly (P < 0.05) higher due to low
lactose levels and reduced levels of lactic acid. Similar results were found by Huffman and
Kristoffersen, (1984) and Shakeel-ur-Rehman et al. (2004), curd washing did not affect gross
composition or proteolysis, but it did increase the pH of the resultant cheese
S2 cheeses contained significantly higher levels (P < 0.05) of fat and FDM due to the milk used
for those cheeses being standardized to a protein:fat ratio of 0.90:1, while milk for S1 cheeses
was standardized to a protein:fat ratio of 0.99:1.
Levels of primary proteolysis, as measured by pH 4.6 SN, increased significantly (P < 0.05) in
all cheeses during ripening, which is attributable to the continuous degradation of casein to low
molecular weight water-soluble peptides and amino acids by the action of the residual
coagulant and the proteolytic activity of the starter culture (Sallami et al., 2004; Awad et al.,
2005).
Levels of pH 4.6 SN were significantly different between treatments, with S2 cheeses having
significantly (P < 0.05) higher mean levels of primary proteolysis. This result is mainly attributed
to the thermostability of the microbial rennet, which was not inactivated during the high cook
temperatures (53°C) and according to Fox et al. (2000) the enzymes in rennet are the ones
responsible for initial proteolysis and the production of most of the water-soluble or pH 4.6 SN.
The current results show that recombinant chymosin from the Aspergillus Niger var. Awamori
(S1 cheeses) had lower thermal stability than the microbial rennet. These results are in
agreement with those obtained by Hyslop et al. (1979) and Thunell et al. (1979), who stated
that heat stabilities of coagulants decreases in the following order: Rhizomucor miehei
proteinase>Rhizomucor pusillus proteinase>veal rennet/fermentation produced chymosin.
However the thermo-stability of the coagulants (rennets) is also influenced by other parameters
including pH, temperature and time (Thunell et al., 1979; Fox, Guinee, Cogan and McSweeney,
2000).
There were significant differences (P < 0.05) in the levels of secondary proteolysis, as
measured by individual and total free amino acids; with S2 cheeses giving significantly (P <
0.05) higher mean levels of total amino acids. It has previously been reported by Yun et al.,
(1993) that coagulant type did not significantly influence the levels of secondary proteolysis.
The production of small peptides and free amino acids is due primarily to the action of enzymes
from starter bacteria (Fox et al., 2000). The marked differences in secondary proteolysis might
be attributed to the fact that more substrate (e.g. oligopeptides) was produced in S2 cheeses
during primary proteolysis; therefore more peptides were hydrolysed into amino acids.
The levels of amino acids in S2 cheeses (26.000 mg/kg-30.000 mg/kg) were much greater than
those reported, at comparable ages, in full fat Cheddar (Guinee et al., 2000) and Gouda (Fox
and Wallace, 1997), but were similar to those observed by Lawlor et al. (2002) in mature
Swiss-type cheeses. Total levels of free amino acids in S1 cheeses (13.000 mg/kg-20.000
mg/kg) were similar, at comparable ages, to full fat Cheddar (Guinee et al., 2000).The most
abundant amino acids found in S1 and S2 cheeses at 180 d of ripening where leucine, proline,
glutamine, lysine, valine, phenylalanine and threonine, typical of a Swiss type and cheddar
cheese varieties (Fox et al., 2000). Leucine, lysine, glutamine, valine and phenylalanine,
impart neutral flavours, proline and glycine impart sweet flavours and threonine is responsible
for the sweet bitter flavour (Fox et al., 2000).
Texture parameters: fracture stress, gumminess, chewiness and springiness, were significantly
affected by the combined effects of rennet type, milk composition and curd wash level, while
cohesiveness was not significant different between S1 and S2 cheeses.
Hardness, fracture stress and guminess were significantly lower (P < 0.001) in S2 cheeses,
which is mainly attributed to the higher proteolityc activity of the residual Rhizomucor miehei
proteinase, that readily hydrolyzes casein, the principal substrate of the proteinase, reducing
the content of intact casein and producing a softer texture (Fox et al., 2000). The higher fat
content of the S2 cheese probably contributed to the softer texture as well. Guminess is
generally correlated with hardness (Fox et al., 2000), the softer a piece of cheese is, the less
energy will be required to disintegrate it to a state ready for swallowing. Springiness (elasticity)
was significantly higher (P < 0.05) in S2 cheeses. Springiness increased upon elevation of fat,
salt in moisture levels and with maturity (Fox et al., 2000). Since S1 and S2 cheeses had
similar values of salt in moisture but differed in the levels of fat content, the higher elasticity of
S2 cheeses is linked to its high FDM and fat content.
Chewiness is the product of hardness, cohesiveness and springiness (Fox, et al 2000).
Chewiness was significantly higher (P < 0.05) in S1 cheese, because it is firmer than S2, and
hence it requires higher time of mastication to reach a state ready for swallowing.
The magnitude of most texture parameters decreased over ripening, an effect that is attributed
to reduction of intact casein content, owing to its hydrolysis by the proteolytic activity of the
residual chymosin and starter cultures enzymes (Guinee, 2003; Gunasekaran and Mehmet,
2004).
The ability of the melted cheese to flow (flowability) was significantly different in S1 and S2
treatments. Flowability of S2 cheeses was significantly higher (P < 0.01) than that of S1
cheeses at all ripening times and measured by the 2 methods (Schreiber and Olson).
According to Fox et al. (2002), increases in the levels of primary and secondary proteolysis
through the use of a more proteolytic coagulant than chymosin (e.g., fungal proteinase),
reduces the apparent viscosity and increases free oil and flowability.
Increases in flowability levels in S2 cheese are probably caused by various factors: the
increase in primary proteolysis in S2 cheeses (softer and more elastic texture) and the higher
level of fat. Increasing the fat content is also associated with greater flowability (Fox et al.,
2000).
The flow levels of S2 cheeses at 150 d of ripening (73%) are typical of a Swiss-type cheese
(Fox et al., 2000), while the flow levels of S1 cheeses at 150 d of ripening (62%) were similar to
those reported for a Cheddar cheese (Fox et al., 2007). Consistent with previous studies
(Sheehan et al., 2007), flowability significantly (P < 0.001) increased with ripening time (as
measured by the Schreiber method). These increases may be attributed to a number of factors,
inter alia, increases in proteolysis, fat coalescence and water binding capacity of the casein
matrix, which promote heat-induced displacement of adjoining layer of the casein matrix on
heating (Guinee, 2003).
Acetate, propionate and butyrate are important contributors to cheese flavour. However
excessive concentrations cause off-favors (rancidity) (Langsrud and Reinbold, 1973).
Levels of acetate and propionate were significantly different between the two treatments. S1
cheeses had significantly (P < 0.05) higher levels of acetate and propionate. This might be
attributed to the higher pH of S1 cheeses at 1d (5.58) in comparison to that of S2 cheeses at
1d (5.21). pH influences the growth of PAB, with the optimum pH for growth between 6 and 7,
maximum at 8.5 and minimum at 4.6 (Langsrud and Reinbold, 1973). High pH favors the
growth of PAB, which transform lactate to propionate, acetate and CO2 during the warm room
period (Fox et al., 2000). The higher pH of the S2 cheese probably allowed for a greater growth
of the PAB, thus translating into higher levels of acetate and propionate.
Levels of acetate in S1 cheese (3000 mg/kg) were similar to those reported in a Swiss-type
cheese, at comparable ages, which range from 3000-7000 mg/kg (Steffen et al, 1987; Lawlor et
al., 2002), while levels of acetate in S2 cheese (1500 mg/kg) were within the range of those
found in Cheddar varieties (100 mg/g kg to 6560 mg/kg) by Kristoffersen et al., (1959).
Propionate levels in S1 cheeses (4200 mg/kg) were similar, at comparable ages, to those
reported by Sheehan et al. (2008) in a Swiss-type cheese and to those reported for Emmental
cheese (5000 mg/kg) by FrÖhlich-Wyder and Bachman (2004). Meanwhile levels of propionate
in S2 cheeses (800 mg/kg) were similar to those reported by St Gelais et al. (1991) and
McGregor and White (1990) (120-750 mg/kg and 1000 mg/kg) respectively in Cheddar cheese.
There were no significant differences between the two treatments on the levels of butyrate.
Butyrate levels ( 600 mg/kg) at 180 d of ripening were similar to those reported for Emmental
cheese (650 mg/kg) (Ji, Alvarez and Harper, 2004).
PCA analysis allowed grouping of the experimental cheeses with their major volatile
compounds. PCA of S1 (Trial 1, 2 and 3) and S2 (Trial 1, 2 and 3) indicated that the grouping
of these cheese are quite different from one another. The major differences between the S1
and S2 samples were in relation to the concentration of specific compounds. For example, the
main volatiles found in S1 cheeses were propionic acid, 2-methyl-1-butanol, acetic acid, 1
Yun, J. J., Barbano, D. M. and Kindstedt, P. S. (1993) ‘Mozzarella cheese: impact of
coagulant type on chemical composition and proteolysis’. Journal of Dairy Science, 76:
3648–3659.
Yun, J. J., Kiely, J. L., Kindstedt, P. S. and Barbano, D. M. (1993) ‘Mozzarella cheese:
impact of coagulant type on functional properties. Journal of Dairy Science, 76: 3667–
3663.
Zeppa, G., Conterno, L. and Gerbi, V. (2001) ‘Determination of organic acids, sugars,
diacetyl, and acetoin in cheese by high-performance liquid chromatography’. Journal of
Agriculture Food Chemistry, 49” 2722-2726.
CHAPTER 3
Influence of cooking temperature during cheese manufacture on the composition, microbiology, proteolysis, functionality and flavour of novel Swiss-type cheese made with yeast adjunct
Abstract
Yeast species are frequently observed in cheeses and make a significant contribution to the
maturation process due to their ability to grow at low temperatures, assimilation/fermentation of
lactose, the assimilation of organic acids, resistance against high salt concentrations, tolerance
of low pH values and low water activities. However, the viability of yeast used as an adjunct
culture in Swiss type cheese subjected to high cook temperatures has not previously been fully
investigated.
Pilot scale novel Swiss-type cheeses were manufactured using Streptococcus thermophilus
and Lactobacillus helveticus as the main starter cultures plus yeast Kluyveromyces lactis and
propionic acid bacteria as cheese adjuncts using two different cook temperatures.
Compositional analysis were analyzed at 1d, while cheese microflora, proteolysis, levels of
short chain fatty acids and texture were monitoring during a ripening period of 120 d. Volatiles
compounds were analyzed at the end of ripening.
Increasing cook temperature from 48 °C to 53°C significantly reduced viable cell counts of
Streptococcus thermophilus and Kluyveromyces lactis, while it increased viable cell counts of
propionic acid bacteria and Lb. helveticus. Lactobacilli including NSLAB were not significantly
affected by the increased temperature. Composition parameters were affected in that cheese
produced from curd cooked to 53°C, which had significantly lower levels of moisture in non fat
substances, significantly higher pH, lower levels of proteolysis, and short chain fatty acids in
comparison to cheese cooked to 48°C. A firmer texture was observed in cheeses cooked to
53ºC in comparison to those cooked at 48°C.
Volatile compounds were affected by the different cook temperatures. However, both
treatments had aromatic compounds with fruity notes typical of yeast fermentation, suggesting
that the K. lactis was able to withstand high cook temperatures.
The study showed how the cook temperature process variable may be used to create novel
cheeses with varied composition, ripening characteristics, texture and flavour.
3.1. Introduction
Manufacture of most cheese varieties involves combining four ingredients: milk, rennet, micro
organisms and salt, which are processed through a number of common steps such as gel
formation, whey expulsion, acid production and salt addition, followed by a period of ripening.
Variations in ingredient, starter cultures and subsequent processing techniques have led to the
evolution and diversification of all cheese varieties.
The manufacture of Swiss-type cheeses generally requires thermophilic starter cultures (e.g.,
Streptococcus thermophilus and Lactobacillus helveticus). The growth of the starter cultures is
limited by the high cooking temperature (52-54°C) used for Swiss type cheeses, but growth
and acidification begins again as soon as the temperature decreases (Fox et al., 2000).
The manufacture process also involves pressing the curd under whey, overnight curd
fermentation, brine salting and ripening at elevated temperature during which the propionic acid
bacteria grow and transform the lactate to propionate, acetate, and CO2, which are responsible
for aroma and eye formation. Often, Swiss cheeses are covered by an orange smear, called
the morge, composed mainly of corynebacteria, micrococci, and yeast, whose function is to
improve aroma (Fox et al., 2000). In artisanal Swiss cheeses, propionic acid bacteria are
natural contaminants of the raw milk, but, in the industrial production of Swiss cheeses they are
normally added deliberately to the milk to give initial counts of about 103 to 104 cfu/ml (Fox et al,
2000).
St. thermophilus and Lb. helveticus are responsible for the production of lactic acid growing
optimally in the range of around 42.7 and 44.0 °C, respectively (Martley,1983). Heating
inoculated milk to 53 °C has been shown to result in a variable slowing of the acidification of
thermophilic lactobacilli due to decreased cellular viability, probably because of thermal stress
(Neviani et al., 1995). Similarly, the cooking curds to a maximum scald of 53 °C delayed but
did not arrest the growth of thermophilic lactobacilli in the core of Grana cheese (52 °C after 6
h; Giraffa et al., 1998). Sheehan et al. (2007) evaluated the effect of high cook temperatures
on starter and non-starter lactic acid bacteria viability, cheese composition and ripening indices.
Cheeses produced from curds cooked to 47°C had significantly higher levels of moisture in
non-fat substances, salt-in moisture, significantly lower pH and levels of butyrate compared to
cheeses produced from curds cooked to 50 or 53°C.
However, there are no studies relating to the survival of yeast in Swiss-type cheese
manufactured with different cooking temperatures and in turn its impact on cheese
composition, flavour, proteolysis and functionality.
3.2. Objective of study
The objectives of the study were to determine the influence of cooking temperature during the
manufacture on the composition, microbiology, proteolysis, functionality and flavour of novel
Swiss-type cheese made with yeast as adjunct culture.
3.3. Materials and Methods
3.3.1. Starters Strains
Streptococcus thermophilus and Lactobacillus helveticus, yeast Kluveromyces lactis and
Propionibacterium spp. freudenreichii were obtained as DVS and stored at -80°C until cheese
manufacture (Table 3.1).
3.3.2. Cheese Manufacture
Swiss-type cheeses (S3 and S4, Table 3.1) were manufacture at pilot scale in two vats of 500
L. Raw milk was obtained from a local dairy company, standardized to a casein to fat ratio of
0.80, held overnight at 6°C, pasteurized at 72°C for 15 s, and pumped into cylindrical,
jacketed, stainless steel vats with automated variable speed cutting and stirring equipments
equipment (APV Schweiz AG, Worb, Switzerland). Starter blend was added to the cheese milk.
Cheese milk (454 kg per vat) was inoculated with 0.003% (w/w) St. thermophilus, 0.0061%
(w/w) Lb. helveticus, 0.0061% (w/w) Propionic acid bacteria and 0.0061% of Yeast
kluyveromyces lactis (Table 3.1). After a 60 min ripening period, chymosin (Chymax plus),
diluted in 1:6 with de-ionised water, was added at a level of 18 mg/kg. A coagulation period of
50 min was allowed, prior a cut program of 5 min duration which produced curd particles of
approximately 5 mm3. After a 10 min healing period, the curd/whey mixture was stirred and
cooked by steam injection into the jacked of the vat. Curds were cooked at a rate of 1°C per
1.5 min until reach 53°C of maximum scald for treatment S3 and at a rate of 1°C per 1.5 min
until reach 48°C of maximum scald for treatment S4. The process variable applied in each trial
was that one vat was cooked to a maximum scald of 53°C (S3) and the second vat to 48 °C
(S4).
At pH 6.15 the whey was drained and the curds were cheddared, milled at pH 5.4, salted at a
level of 1.5% (w/w), mellowed for 15 min and moulded into 24 kg moulds. The moulds were
pressed in a vertical press at 3 kPa for 30 min and pressed overnight on a horizontal press at
265 kPa. Cheeses were vacuum packed next day and stored at 15°C for 35 days and at 8°C
thereafter.
Table 3.1 Details and differences between make procedures of Swiss T1 vs. Swiss T2.
Treatment Cheese code
S3 S4
Milk volume 454 kg 454 kg
Pasteurization 72°C * 15 seg 72°C * 15 seg
Standardization 0.80:1 0.80:1
Starter cultures (w/w) 0.003% St. thermophilus,
0.0061% Lb. helveticus
0.0061% Propionic acid bacteria
0.0061% Yeast kluyveromyces
lactis
0.003% St. thermophilus,
0.0061% Lb. helveticus
0.0061% Propionic Acid
Bacteria
0.0061% Yeast Kluyveromyces
lactis
Rennet Standard Standard
Curd Formation Firm Firm
Cook 1 °C per 1.5 min 1 °C per 1.5 min
Max scald 53 °C 48 ° C
Drain pH 6.15 6.15
Curd handling Cheddaring Cheddaring
Milling pH 5.4 pH 5.4
Salting method Dry salting 1.5% Dry salting 1.5%
Mellow time 15 min 15 min
Cheese size 2 blocks of 24 kg 2 blocks of 24 kg
Ripening regime 15 °C 35 d
8 °C up to 6 months.
15 °C 35 d
8 °C up to 6 months.
3.3.3. Enumeration of starter bacteria, non starter bacteria, propionic acid bacteria and
yeast
Cheese samples were aseptically removed at 30 d, 60 d, 90 d, and 120 d of ripening. The
cheese samples were placed in a stomacher bag diluted 1:10 with sterile trisodium citrate (2%
w/v) and homogenised in a stomacher (Stomacher, Lab-Blender 400, Seward, Thetford,
Norfolk, UK) for 5 min at room temperature. A serial dilution of the resultant slurry was
performed in 9 ml sterile maximum recovery diluent as required. Independent duplicate
samples were taken at each sampling point and the bacterial groups were enumerated on the
following agars: Streptococcus thermophilus on LM17 agar (Becton Dickson and Company,
Cockeysville, New Jersey, USA), incubated at 45°C for 3 d (Terzaghi and Sandine, 1975);
starter, Lactobacillus helveticus cells on MRS 5.4 agar after anaerobic incubation for 3 d at
42°C (IDF, 1998B). Non-starter lactic acid bacteria (NSLAB) on LBS agar (Becton Dickson and
Company, Cockeysville, New Jersey, USA) incubated aerobically with an overlay for 5 d at 30
°C (Rogosa, Mitchell and Wiseman, 1951); Propionic acid bacteria on sodium lactate agar after
incubation at 30°C for 7 d (Drinan and Cogan, 1992); Yeast kluyveromyces lactis enumerated
on YGC-agar incubated at 21°C for 7 d.
3.3.4. Cheese Analysis
3.3.4.1. Cheese sampling
Cheeses were sampled at various times throughout ripening at 1 d for gross compositional
analysis; at 30 d, 60 d, 90 d and 120 d for pH4.6 SN, individual free amino acids, and short
chain fatty acids; at 30 d, 90 d and 120 d for texture and at 120 d for volatiles compounds. At
each sampling time, a 7 to 6 cm slab of cheese was cut from the exterior face of the block; the
outer layer (1-2 cm) of the slab was discarded and the remainder was used for analysis.
3.3.4.2. Gross composition
Cheese samples were grated to yield particles of <1mm, using a food processor. Samples
were analyzed at one d ripening in triplicates for pH (British Standards Institution, 1975),
moisture (International Dairy Federation, 1982), fat (International Dairy Federation, 1996),
protein (International Dairy Federation, 1993), ash (International Dairy Federation, 1964),
calcium (International Dairy Federation, 1992) salt (International Dairy Federation, 1988),
lactates were measured using the Megazyme kit (D-/L-Lactic Acid Kit, Megazyme International
Ireland Ltd., Bray, Ireland) and sugars (lactose and galactose) were analysed by HPLC as
described by the method of Zeppa et al. (2001).
3.3.5. Assessment of Proteolysis in Cheese
3.3.5.1. Primary proteolysis
pH 4.6 soluble nitrogen (SN) was determined by the macro-Kjedahl method (International Dairy
Federation, 1993). The levels of pH 4.6 SN were determined in triplicate using the method of
Kuchroo and Fox (1982) and expressed as a percentage of total nitrogen (TN).
3.3.5.2. Secondary proteolysis
Individual free amino acids (FAA) were determined in duplicate on the pH 4.6 SN extracts of
cheeses after 30 d, 60 d, 90 d and 120 d of ripening prepared by a modification of the method
of Kuchroo and Fox (1982) as described by Fenelon, O’Connor and Fox (2000). Samples were
deproteinised by mixing equal volumes of 24% (w/v) trichloroacetic acid (TCA) and samples
were allowed to stand for 10 minutes before centrifuging at 14400 x g (Microcentaur, MSE, UK)
for 10 min. Supernatants were removed and diluted with 0.2 M sodium citrate buffer, pH 2.2 to
give approximately 250 nmol of each amino acid residue. Samples were then diluted 1 in 2 with
the internal standard, norleucine, to give a final concentration of 125 nm/ml. Amino acids were
quantified using a Jeol JLC-500/V amino acid analyser (Jeol (UK) Ltd., Garden city, Herts, UK)
fitted with a Jeol Na+ high performance cation exchange column.
3.3.6. Functionality
3.3.6.1. Texture
Cheese samples (25 mm3 cubes) were cut from the slab of cheese (Cheese Blocker; Boos
Kaasgreedschap, Bodengraven, Netherlands) and stored at 4°C overnight before analysis. Six
cheeses cubes were analyzed by compression on a TA-HDi Texture Profile analyzer (model
TA-HDI, Stable Micro Systems, Godalming, UK) with a 5mm compression plate and a 100 kg
load cell at room temperature. Each sample was subjected to 2 consecutive compressions at a
speed of 1 mm/s, each to 30% of original sample height, as described in Rynne et al., (2004).
Texture profile analysis parameters were calculated. Hardness (N) was measured as the force
at maximum compression on the first bite. Fracture stress (kPa) was measured as the force per
unit area at the point of fracture on the first bite. Cohesiveness (dimensionless) was calculated
as the ratio of the area of the second bite to that of the first bite. Springiness (dimensionless)
was calculated as the ratio of the area of the second bite to that of the first bite. Chewiness (N)
was calculated as the product of harness x cohesiveness x springiness. Adhesiveness (N * S)
was calculated as the negative area after the first bite in the texture profile curve as described
in Van Vliet (1991), Bourne (1978) and expressed in absolute values.
3.3.7. Flavour
3.3.7.1. Assessment of short chain volatile fatty acids
Acetate, Propionate and n-butyrate (C2:0, C3:0, C4:0) contents were determined in cheeses by
steam distillation and quantified by ligan-exchange, ion-exclusion HPLC as described by
Kilcawley et al. (2001).
3.3.7.2. Volatiles profile
The volatile profiles of the headspace of each sample was analysed by solid phase micro-
extraction (SPME) gas chromatography mass spectrometry (GCMS).
For volatile analysis, 5 g of sample was added to a 20 ml amber screw capped SPME vial and
equilibrated at 40°C for 5 min with pulsed agitation of 4 s at 400 rpm. Sample introduction was
accomplished using a CTC Analytics CombiPal Autosampler. A single DVD/Carboxen/PDMS 1
cm fiber was used for all analysis. The SPME fiber was exposed to the headspace above the
samples for 25 min at depth of 1 cm with pulsed agitation of 4 s at 350 rpm. The fiber was
retracted and injected into the GC inlet at 250°C and desorbed for 2 min. Injections were made
on a Varian 450 GC with a Perkin Elmer Elite DMS (60 m x 0.25 mm ID x 0.25 DF µm) column.
The detector used was a Varian 320 triple quad mass spectrometer. Individual compounds
were identified using mass spectral comparisons to the NIST 2005 mass spectral library.
Individual compounds were assigned quantification and qualifier ions to ensure that only the
individual compounds were identified and quantified, especially in the case of co-eluting or
semi-co-eluting samples. Compounds were quantified by calculating the area under the peak of
each compound and are expressed in arbitrary units. An autotune of the GCMS was carried
out immediately prior to analysis to confirm that the GCMS was operating under optimal
conditions. Each sample was analysed in duplicate.
3.3.8. Statistical analysis
All statistical analyses were carried out using SAS (version 9.1.3, SAS Institute, Cary, NC).
Analysis of variance was carried out on data using the general linear model procedure of SAS
(SAS Institute). The Tukey honestly significant difference test was used to determine the
significance of difference between the means. The level of significance was determined at P <
0.05.
For variables analysed at several times during ripening, analysis of variance for the split-plot
design was carried out on data using the mixed procedure of SAS (SAS Institute).
Statistically significant differences (P < 0.05) between different treatment levels were
determined by using Tukey honestly significant difference.
Principal component analysis (PCA) of the individual amino acids, short chain fatty acids,
texture parameters, and volatiles compounds were performed by using the statistical software
The Unscrambler (v 9.7, CAMO, Norway).
3.4. Results
3.4.1. Gross Composition
The mean composition of S3 and S4 cheeses are given in Table 3.2.
Table 3.2. Cheese composition of novel Swiss type cheeses made with yeast adjunct differing
a, b Values within a row not sharing a common superscript, differ significantly (P < 0.05) *S3= cheese cooked at 53ºC; S4=cheese cooked at 48ºC * Values presented are the means of 2 replicates ** Standard deviation of 2 replicates ¹MNFS= Moisture in the non-fat substance ² FDM= Fat in dry matter ³ S/M= Salt in moisture
3.4.2. Lactates and sugars
Mean levels of lactates and sugars are shown in Table 3.3. The mean levels of lactic acid
expressed as total lactates (D-lactate and L-lactate) and protein:lactate, were significantly lower
(P < 0.05) in cheese cooked at 53°C (S3), while the levels of residual lactose and galactose
were significantly (P < 0.05) higher.
Table 3.3 Lactates and sugars contents of novel Swiss type cheeses made with yeast adjunct
differing in their cooking temperature.
Composition Treatment*
S3 S4
d-lactate g/100g
0.17a
0.15b
l-lactate g/100g
0.65a
0.80b
Total lactate g/100g
0.83a
0.95b
Protein: lactate
31.45a 25.02b
Lactose g/100g
0.44a 0.35b
Galactose g/100
0.25a 0.07b
a, b Values within a row not sharing a common superscript, differ significantly ( P < 0.05) *S3= cheese cooked at 53ºC; S4=cheese cooked at 48ºC. ¹MNFS= Moisture in the non-fat substance ² FDM= Fat in dry matter ³ S/M= Salt in moisture
3.4.3. Viability of starter bacteria: Streptococcus thermophilus, Lactococcus lactis and
Lactobacillus helveticus during cheese ripening.
Viability of starter bacteria is shown in Figure 3.1. Cook temperature had significant effect on
mean viable cell number of St. thermophilus, and Lb. Helveticus. Cheeses cooked at 48°C had
significantly higher counts (P <0.001) of viable St. thermophilus than cheeses cooked at 53°C,
and they decreased significantly (P < 0.001) in both cheeses over ripening (Fig. 3.1).
Cheeses cooked at 53°C had significant (P < 0.001) higher counts of Lb. helveticus (from the
period 60d and thereafter), than those cooked at 48°C. Mean viable cell numbers of Lb.
helveticus decreased significantly (P < 0.001) during ripening for both cheeses (Fig. 3.1).
0.0
2.0
4.0
6.0
8.0
10.0
30 60 90 120
Lo
g1
0 c
fu/g
ch
ee
se
Time (days)
St. thermophilus
S3
S4
0.0
2.0
4.0
6.0
8.0
10.0
30 60 90 120
Lo
g1
0 c
fu/g
ch
ees
e
Time (days)
Lb. helveticus
S3
S4
Fig. 3.1 Effect of varying cook temperature during cheese manufacture on novel Swiss type cheeses: S3= cheese cooked at 53ºC; S4=cheese cooked at 48ºC, on viable cell count of Streptococcus thermophilus enumerated on LM17 agar @42ºC and viable cell counts of Lactobacillus helveticus enumerated on MRS 5.4 agar during cheese ripening.
3.4.4. Viability of Lactobacillus during cheese ripening
Mean counts of undefined lactobacilli (Lb. helveticus and NSLAB) are shown in Figure 3.2.
Viable lactobacilli numbers increased significantly (P <0.05), throughout the ripening period.
(Fig 3.2). No significant differences of viable counts of lactobacilli were found between
cheeses cooked at 53°C and those cooked at 48°C, throughout the ripening period in study.
0.01.02.03.04.05.06.07.08.09.0
30 60 90 120
Lo
g1
0 c
fu/g
ch
ees
e
Time (days)
Lactobacillus
S3
S4
Fig 3.2 Effect of varying cook temperature during cheese manufacture in novel Swiss type cheeses: S3= cheese cooked at 53ºC; S4=cheese cooked at 48ºC, on viable cell counts of non starter lactic acid bacteria enumerated on LBS agar during cheese ripening.
3.4.5. Viability of adjunct culture: yeast and propionic bacteria
Viability of yeast and propionic bacteria is shown in Figure 3.3. Cook temperatures significantly
(P < 0.05) affected viable cell counts of yeast Kluveromyces lactis and propionic acid bacteria.
Counts of viable yeast was significantly (P < 0.05) higher in cheeses cooked at 48°C than
those cooked at 53°C (Fig. 3.3).
Counts of viable propionic acid bacteria was significantly higher (P < 0.05) in cheeses cooked
at 48°C than those cooked at 53°C.
Viable counts of yeast K. Lactis decreased significantly (P < 0.01) over ripening in S1 and S2
cheeses, while viable PAB counts increased significantly (P < 0.05) over ripening in S1 and S2
cheeses (Fig. 3.3).
.
0.0
1.0
2.0
3.0
4.0
30 60 90 120
Lo
g10 c
fu/g
ch
eese
Time (days)
Yeast
S3
S4
0.0
2.0
4.0
6.0
8.0
10.0
30 60 90 120Lo
g1
0 c
fu/g
ch
ee
se
Time (days)
Propionic acid bacteria
S3
S4
Fig 3.3 Effect of varying cook temperature during cheese manufacture on novel Swiss type cheeses: S3= cheese cooked at 53ºC; S4=cheese cooked at 48ºC, on viable cell counts of Kluveromyces lactis, enumerated on yeast extract agar and propionic acid bacteria (PAB) enumerated on sodium lactate agar (SLA).
3.4.6. Proteolysis
Means levels of primary proteolysis expressed as pH 4.6 SN are shown in Figure 3.4. The
mean levels of pH 4.6 SN increased significantly (P < 0.05) during ripening in both cheeses..
The cheese cooked at 48ºC had significant higher (P < 0.001) levels of pH 4.6 SN than the
cheese cooked at 53 ºC.
Different cook temperatures had no significant effect on secondary proteolysis (Fig 3.5, Table
3.4). However, levels of total free amino acids increased significantly (P < 0.001) in all cheeses
during ripening (Fig 3.5, table 3.4). Individual free amino acids are shown in Figure 3.6.
Table 3.4. Levels of pH 4.6 soluble nitrogen and total free amino acids during ripening of novel
Swiss-type cheeses.
Ripening time (d)
Treatment
S3 S4
pH 4.6 SN (% of total N)
30 8.00a,A 12.02b,A
60 11.93a,B 16.13b,B
90 15.04a,C 17.38b,C
120 16.24a,D 20.05b,D
Total free AA (mg/kg of cheese)
30 8059 a, A 9606 a, A
60 12771 a, BC 12906 a, B
90 14755 a, CD 15489 a, C
120 16082 a, D 16470 a, C
a, b Values within a row not sharing a common superscript, differ significantly (P < 0.05) A,B Values within a column not sharing a common superscript, differ significantly (P < 0.05). S3= cheese cooked at 53ºC; S4=cheese cooked at 48ºC.
5
10
15
20
25
30 50 70 90 110
pH
4.6
-SN
/ %
TN
Time (days)
Primary proteolysis (as determined by pH4.6-SN levels) over time
S3
S4
Fig 3.4. The effect of different cooking temperatures on primary proteolysis in novel Swiss type cheeses: S3= cheese cooked at 53ºC; S4=cheese cooked at 48ºC, on levels of pH 4.6 soluble nitrogen, expressed as % of total nitrogen, during ripening. Values presented are the means from two replicates.
0
5000
10000
15000
20000
30 60 90 120
To
tal fr
ee a
min
o a
cid
s m
g/k
g
of
ch
ee
se
Time (days)
Secondary ProteolysisTotal free amino acids
S3
S4
Fig 3.5 The effect of different cooking temperatures on secondary proteolysis in novel Swiss type cheeses: S3= cheese cooked at 53ºC; S4=cheese cooked at 48ºC, on levels of total free amino acids in pH 4.6 soluble nitrogen extract during ripening. Values presented are the means from two replicates
Secondary Proteolysis determined by Individual free amino acids
0
200
400
600
800
1000
1200
1400
1600
1800
Asp
Thre Se
r
Glu
Gly
Ala
Cys
Val
Met Ile Leu
Tyr
Phe
His
Lys
Arg
Pro
Fre
e am
ino
aci
ds
mg
/kg
ch
eese
30 d of ripening
S3
S4
0
400
800
1200
1600
2000
2400
2800
Asp
Thre Se
r
Glu Gly Ala
Cys
Val
Met Ile Leu
Tyr
Phe
His
Lys
Arg
Pro
Fre
e a
min
o a
cid
s m
g/k
g
chee
se
60 d of ripening
S3
S4
0
500
1000
1500
2000
2500
3000
Asp
Thre Se
r
Glu Gly
Ala
Cys
Val
Met Ile Leu
Tyr
Phe
His
Lys
Arg
Pro
Fre
e a
min
o a
cid
s m
g/k
g
ch
ee
se
90 d of ripening
S3
S4
0
500
1000
1500
2000
2500
3000
3500
Asp
Thre Se
r
Glu Gly
Ala
Cys
Val
Met Ile Leu
Tyr
Phe
His
Lys
Arg
Pro
Fre
e am
ino
aci
ds
mg
/kg
ch
eese
120d of ripening
S3
S4
Fig 3.6. The effect of different cooking temperatures on secondary proteolysis in novel Swiss type cheeses: S3= cheese cooked at 53ºC; S4=cheese cooked at 48ºC on individual free amino acids in pH 4.6 soluble nitrogen extract at 30 d, 60 d, 90 d and 120 d of ripening. Values presented are the means from two replicates.
3.4.7. Functionality
3.4.7.1. Texture (TPA)
Figure 3.7 shows the texture profiles of S3 and S4 cheeses over ripening. Texture parameters
of hardness and gumminess were significantly (P < 0.001) higher in cheeses cooked at 53°C
while the texture parameter of adhesiveness was significantly (P < 0.001) higher in cheese
cooked at 48°C. No significant differences were found in the parameters of springiness,
cohesiveness and chewiness.
Texture Profiles
0
100
200
300
400
500
30 90 120
N
Time (days)
Hardness
S3
S4
0.00
1.00
2.00
3.00
30 90 120
-
Time (days)
Springiness
S3
S4
0
20
40
60
80
30 90 120
N
Time (days)
Gumminess
S3
S4
0
50
100
150
200
30 90 120
N
Time (days)
Chewiness
S3
S4
0.0
2.0
4.0
6.0
8.0
10.0
30 90 120
N m
m
Time (days)
Adhesiveness
S3
S4
0.000.050.100.150.200.250.30
30 90 120
-
Time (days)
Cohesiveness
S3
S4
Fig 3.7. Evolution of texture parameters over ripening of novel Swiss-type cheese: S3= cheese cook at 53ºC; S4=cheese cook at 48ºC
3.4.8. Flavour
3.4.8.1. Short chain fatty acids (acetate, propionate and butyrate)
Levels of short chain fatty acids (acetic, propionic and butyric acids) are shown Fig. 2.6. Cook
temperature significantly (P < 0.05) affected the levels of acetate, propionate and butyrate, and
they increased significantly (P < 0.001) during ripening in S1 and S2 cheeses (Fig. 3.8).
Short chain fatty acids
0
1000
2000
3000
4000
5000
30 60 120 150
mg
/Kg
o
f c
he
ese
Time (days)
Acetic acid
S2-T1
S2-T2
0
1000
2000
3000
4000
5000
30 60 90 120
mg
/kg
of
ch
eese
Time (days)
Propionic acid
S2-T1
S2-T2
0
500
1000
1500
30 60 90 120
mg
/kg
of
ch
ee
se
Time (days)
Butyric acid
S2-T1
S2-T2
Fig 3.8. The effect of different cooking temperatures on short chain volatile fatty acids in novel Swiss type cheeses: S3= cheese cooked at 53°C, and S4= cheese cooked at 48°C. Values presented are the means from two replicates.
3.4.9. Volatiles compounds.
Relevant chromatograms (Fig 3.9) show the volatile compounds found in S3 and S4 cheeses.
The PCA scores clustered the S3 and S4 cheeses into two different positions. S3 cheese
located on the positive dimension of PC1, and S4 cheese located on the negative dimension of
PC1, which means that volatiles compounds in S3 cheeses are quite different to those of S4
cheeses (Fig 3.10)
A second PCA was performed on a dataset comprised all treatments S1, S2, S3 and S4
cheeses in order to identify any correlation in volatiles compounds. PCA (Fig 3.11) grouped S1
and S3 cheeses together on the positive dimension of PC1. S4 cheeses were grouped on the
negative dimension of PC2 and S2 cheeses on the negative dimension of PC2. Volatiles
compounds in S1 and S3 are similar to each other but differ significantly with those of S2 and
S4 cheeses.
S3 and S4 cheeses differed significantly to each other. PCA explained 64% of total variance.
Fig 3.9. GC-MS chromatograms of the headspace volatiles for S3 and S4 cheeses at 120d of ripening. S3= cheese cook at 53ºC during manufacture; S4=cheese cook at 48ºC during manufactur
S3
S4
Fig. 3.10 Result of principal component analysis of volatile compounds of S4 and S3 cheeses at 120 d ripening. S3= cheese cook at 53ºC during manufacture; S4=cheese cook at 48ºC during manufacture
Fig. 3.11 Result of principal component analysis of volatile compounds of S1, S2, S3, and cheeses at 120 d ripening. S1= cheese produced with recombinant rennet, with curd washing step and std. milk 0.99:1 of protein: fat ratio; S2 = cheese produced with microbial rennet, without curd washing step and std. milk 0.90:1 of protein: fat ratio; S3= cheese cook at 53ºC during manufacture; S4=cheese cook at 48ºC during manufacture.
3.5. Discussion
In agreement with the results of Sheehan et al. (2007), cheeses produced from curds cooked to
48°C (S4) had significantly higher levels of moisture, moisture in nonfat substance (MNFS),
and ca: protein ratio, but significant lower levels of protein, fat, fat in dry matter and pH than
cheeses cooked to 53 °C (S3). The reduction in cheese moisture caused by the increased
cook temperature is most likely due to the increased syneresis (Turner et al., 1983). Gel
syneresis controls cheese moisture and hence regulates the growth of bacteria (Fox et al.,
2004). The high cook temperature applied to S3 cheeses reduced the viability of the starter
cultures and also its acidification rate (as consequence of thermal stress and moisture
reduction). Hence, the pH at d1 of S3 cheese was significantly (P < 0.05) higher, despite the
fact that the curds for both cheeses were drained and milled at similar pH values. Increases in
the concentration of moisture are parallel to the reduction in fat content and fat in dry matter.
The mean levels of lactic acid expressed as total lactate, were significantly lower (P < 0.05) in
cheese cooked at 53°C; the direct effect of the elevated cook temperature and reduced growth
rate of St. thermophilus, slowing down the consumption of lactose and hence the production
of lactic acid. Values of residual lactose were significantly higher in the cheese cooked at
53ºC; due to less lactose being converted into lactic acid. Galactose from lactose breakdown is
not utilised by the St. themophilus, but is metabolised by the Lb. helveticus (Fox et al., 2004)
and according to Martley (1983), the temperature at which most rapid acid production for 38
strains of lactobacilli occurred in the range of 41.8ºC-46.6°C with a mean temperature for
maximum acid production of 44°C (Giraffa et al., 1993). Residual galactose content was
significantly (P < 0.05) lower in cheese cook at 48°C, presumably Lb. Helvetius adapted better
to that temperature and thus consumed more galactose at 1 d after cheese manufacture.
Cheeses cooked at 48°C had significantly ( P < 0.001) higher counts of viable St. thermophilus
than cheeses cooked at 53°C, the results are similar to those obtained by Turner et al. (1983)
who observed that increasing cook temperature from 48 to 54ºC in Swiss type cheese
manufacture reduced St. thermophilus counts. Similar observations have been obtained in
Swiss cheeses (Thierry et al., 1998; Valence et al., 2000 and Sheehan et al., 2007). Mean
viable cell number of St. thermophilus decreased significantly (P < 0.001), from 108 cfu/g at 30d
of ripening to 105.5 cfu/g at 120d of ripening.
Contrary to the results obtained by Sheehan et al. (2007), cheeses cooked at 53°C (S3) had
significant (P < 0.001) higher counts of Lb. helveticus (from 60d of ripening and thereafter),
than those cooked at 48°C (S4). Mean viable cell numbers of Lb. helveticus decreased
significantly (P < 0.001) during ripening in S3 and S4 cheeses. This trend is in agreement with
the studies of Thierry et al. (1998), Valence et al. (2000). There was significant interaction
between the effect of treatment and ripening time (P < 0.001).
Viable lactobacilli numbers as enumerated in LBS agar (Lb. helveticus and NSLAB) increased
significantly (P < 0.05), from 107.6 cfu/g at 30d to 108.3 cfu/g at 120d of ripening, while the
numbers of Lb. helveticus on their own decreased. The increase in lactobacilli is an indication
of the increasing levels of NSLAB. Split plot analysis showed that the greatest increase in the
counts of lactobacilli in S3 and S4 cheeses occurred early in the ripening period of 30 d. Viable
lactobacilli numbers were slightly higher to that observed by Thierry et al. (1998); Sheehan et
al. (2007) is Swiss-type cheese and similar to those reported by Demarigny et al. (1996) and
Beuvier et al. (1997) in a Swiss-type cheese manufactured from raw milk. NSLAB may
originate from many sources including cheese milk, manufacturing equipment and the cheese
making environment as reviewed by Beresford and Williams (2004). Although lactobacilli
counts increased at varying rate, there were no significant differences of viable counts of
lactobacilli between cheeses cooked at 53°C and those cooked at 48°C, throughout the
ripening period in study.
Cook temperature had a significant effect on counts of the yeast Kluveromyces lactis. Cheese
cooked at 48ºC (S4) had significant (P < 0.05) higher counts of viable yeast than cheese
cooked at 53°C (S3) at 30d and 60d of ripening; thereafter the difference was no longer
significant. The differences may be attributed to sensitivity of yeast to high cook temperatures.
Viable yeast count decreased significantly (P < 0.01) from 103.5 cfu/g (S4) and 102.9
cfu/g (S3)
at 30d of ripening to 102.6 cfu/g (S4) and 102.5 cfu/g (S3), at 120d of ripening. These results are
similar to those obtained by Ferreira and Viljoen (2003), where the yeast number of D. hanseii
and Y. lipolitica decreased gradually alter 6 months of maturation to a minimun value of 102.3
cfu/g. In the study D. hanseii and Y. lipolitica were added as adjunct cultures in the
manufacture of mature Cheddar. According to Fleet (1990) and Welthagen & Viljoen (1998),
survival of yeasts in the cheese might be attributed to the utilisation of organic acids produced
by the lactic acid bacteria, and the proteolytic and lipolytic abilities of yeast. Furthermore,
yeasts grow particularly well during the initial period of ripening, due to their tolerance to low pH
values and high NaCl concentrations (Eliskases-Lechner and Ginzinger, 1995) and according
to Bartchi et al. (1994), a decrease in yeast counts towards the end of the ripening period is
observed in cheeses.
Cheese cooked at 48°C (S4) had significant (P < 0.05) higher counts of viable PAB compared
to cheese cooked at 53ºC (S3) during the period: 30d, 60d and 90d. At 120d both treatments
had similar values on viable PAB. Counts of viable PAB were 106.46 cfu/g (S4) and 105.44
cfu/g
(S3) at 30d of ripening and increased significantly (P < 0.05) until 108.1 cfu/g (S4) and 107.8 cfu/g
(S3) at 120d of ripening. The greatest increment occurred during the hot room period (35 d). A
significant interaction occurred between treatment and ripening time (P < 0.001) on viable PAB
counts. Thierry et al. (1998), Gilles et al. (1983) and Sheehan et al. (2007) reported similar
trends in numbers of PAB during the ripening of Swiss-type cheeses.
Cook temperature had a significant effect on primary proteolysis expressed as mean levels of
pH 4.6-SN. Fox et al. (2000) stated that primary proteolysis in low-cooked cheese, in which the
chymosin is not inactivated during cooking, is due mainly to chymosin.
Cheese cooked at 48ºC had significant higher (P < 0.001) levels of pH 4.6 SN and it might be
attributed to a greater activity of residual chymosin (Delacroix-Buchet & Fournier, 1992), which
presumably was less inactivated by the cook temperature applied to S3 cheeses (48°C) along
with its higher moisture content. Recombinant chymosin was used in S3 and S4 treatments,
and as discussed in Chapter 2, it is heat liable, probably it was partially inactivated by the high
cook temperature (53°C) applied during the manufacture of S3 cheeses.
Secondary protelysis, assessed as levels of individual and total free amino acids was not
affected by different cook temperatures. However, total free aminoacids increased significantly
from 8000-9000 mg/kg to 16000-16500 mg/kg during ripening. These levels were much greater
than those reported, at comparable ages, in full fat Cheddar (Guinee et al., 2000) and Gouda
(Fox and Wallace, 1997), but similar to those observed by Lawlor et al. (2000) and Sheehan et
al. (2007) in Swiss-type cheeses.
An increase in levels of total free amino acids during ripening is associated with the release of
intracellular peptidases, particularly from starter lactic acid bacteria (LAB) as a result of cell
lysis as reviewed by Khalid and Marth (1990). When the population of starter LAB declines, the