Okara (soybean residue) biotransformation by yeast Yarrowia lipolytica

    Okara (soybean residue) biotransformation by yeast Yarrowia lipolytica Weng Chan VONG, Kai Ling Corrine A.U. YANG, Shao-Quan LIU PII: DOI: Reference:

S0168-1605(16)30332-4 doi: 10.1016/j.ijfoodmicro.2016.06.039 FOOD 7286

To appear in:

International Journal of Food Microbiology

Received date: Revised date: Accepted date:

6 May 2016 23 June 2016 28 June 2016

Please cite this article as: VONG, Weng Chan, YANG, Kai Ling Corrine A.U., LIU, Shao-Quan, Okara (soybean residue) biotransformation by yeast Yarrowia lipolytica, International Journal of Food Microbiology (2016), doi: 10.1016/j.ijfoodmicro.2016.06.039

This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.

ACCEPTED MANUSCRIPT Okara (soybean residue) biotransformation by yeast Yarrowia lipolytica

Food Science and Technology Programme, Department of Chemistry, National University of

SC

a

RI

PT

Weng Chan VONGa, Kai Ling Corrine AU YANGa and Shao-Quan LIUa, b*

b

NU

Singapore, 3 Science Drive 3, Singapore 117543, Singapore

National University of Singapore (Suzhou) Research Institute, No. 377 Linquan Street,

D

MA

Suzhou Industrial Park, Suzhou, Jiangsu, China 215123

AC CE P

TE

*Corresponding author: Shao-Quan LIU: [email protected]

1

ACCEPTED MANUSCRIPT Abstract Okara, or soybean residue, is a soy food processing by-product from the manufacture of

PT

soymilk and soybean curd (tofu). In this study, solid state fermentation of okara was conducted

RI

over five days using yeast Yarrowia lipolytica, and the changes in proximate composition,

SC

antioxidant capacity, non-volatiles and volatiles were investigated. Yeast metabolism of okara significantly increased the amounts of lipid, succinate and free amino acids, and enhanced the

NU

antioxidant capacity. In particular, there was a marked increase in important umami tastants after

MA

fermentation, with 3-fold increase in succinate and a 20-fold increase in glutamate. The final fermented okara contained 3.37 g succinate and 335 mg glutamate/100 g dry matter. Aldehydes

D

and their derived acids in the fresh okara were catabolised by Y. lipolytica mainly to methyl

TE

ketones, leading to a reduced grassy off-odour and a slightly pungent, musty and cheese-like odour in the fermented okara. Amino acid-derived volatiles, such as 3-methylbutanal and 2-

AC CE P

phenylethanol, were also produced. Overall, the okara fermented by Y. lipolytica had a greater amount of umami-tasting substances, a cheese-like odour, improved digestibility and enhanced antioxidant capacity. These changes highlight the potential of Yarrowia-fermented okara as a more nutritious, savoury food product or ingredient. Y. lipolytica was thus demonstrated to be suitable for the biovalorisation of this soy food processing by-product. Keywords: Okara; soybean residue; Yarrowia lipolytica; fermentation; flavour; yeast

2

ACCEPTED MANUSCRIPT 1. Introduction Okara, or soybean residue, refers to the insoluble residue left after grinding soybeans and

PT

extracting the water-soluble components for the manufacture of soy-based food products, such as

RI

soymilk and bean curd (tofu). About 1.1 kg of fresh okara is produced from processing 1.0 kg of

SC

soybean to produce soymilk or tofu (Khare et al., 1995). Large quantities of okara are produced by the soy product manufacture industry, especially in Asian countries where soy foods are

NU

hugely popular. Okara is normally incorporated into animal feed or discarded by companies due

MA

to its high perishability and undesirable flavour and texture. However, okara is still highly nutritious; it contains about 40 – 60% carbohydrates (mostly as insoluble fibre), 20 – 30%

D

protein and 10 – 20% lipid (all dry basis) (Li et al., 2012), and so okara is a suitable substrate for

TE

biotransformation.

AC CE P

The fermentation of okara by fungi and bacteria has been well-studied, as reviewed by Vong and Liu (2016a), but there are few studies on yeast fermentation of okara. Recently, Rashad et al. (2013) showed that yeast-fermented okara had changes in proximate composition and increased antioxidant capacity. In our previous work, we observed that yeast fermentation of autoclaved okara led to interesting changes in its volatile profiles (Vong and Liu, 2016b). Aldehydes in the okara, such as hexanal and trans-2-hexenal, were biotransformed into alcohols, ketones and/or esters, and the effect varied with the yeast employed. Okara fermented by dairy yeasts (yeasts typically associated with fermented dairy products) led to increase in cheese-like volatiles. Therefore, the present study further explored other biochemical changes in okara fermented by the dairy yeast, Yarrowia lipolytica.

3

ACCEPTED MANUSCRIPT Y. lipolytica is a strict aerobe and non-conventional dimorphic yeast that is generally recognised as safe (GRAS) (Groenewald et al., 2014). Naturally found on fermented dairy and

PT

meat products, such as cheese and sausage, Y. lipolytica plays an important role in the flavour

RI

development of these foods due to its high lipolytic and proteolytic activities. This yeast secretes microbial enzymes to break down milk lipids and proteins to generate amino acids, short-chain

SC

fatty acids and methyl ketones, all of which contribute to the flavour of cheese and sausage

NU

(Patrignani et al., 2007; Sørensen et al., 2011). The ability of Y. lipolytica to assimilate a wide range of hydrophobic substrates and carbon sources, including sugars, hydrocarbons and

MA

alcohols has also driven the study of its potential biotechnological applications. Some of the products of Y. lipolytica in synthetic media include γ-decalactone (peach-like aroma) and organic

TE

D

acids from the tricarboxylic acid (TCA) cycle (Fickers et al., 2005a; Gomes et al., 2012). The application of Y. lipolytica in fermenting various agro-industrial wastes, such as olive-mill

AC CE P

wastewater, waste cooking oil and industrial fats, has also been explored. Some of the valueadded products include lipase, citrate and single-cell oil (Lanciotti et al., 2005a; Papanikolaou and Aggelis, 2003; Papanikolaou et al., 2008). The aforementioned studies highlighted the robust characteristics of Y. lipolytica in producing a variety of substances from biotransforming protein- and lipid-rich substances. As okara also contains a significant amount of protein and lipid, we hypothesised that various compounds of interest, such as organic acids and flavour compounds, might also be produced after okara fermentation. Therefore, the present study was undertaken to investigate the biochemical changes during Y. lipolytica fermentation of okara in order to determine the feasibility of okara fermentation by this yeast, and to explore the possible applications of the fermented whole okara. By exploiting the biotechnological potential of Y. lipolytica, we aim to 4

ACCEPTED MANUSCRIPT add value to okara such that it can be returned back to the food value chain as a nutritionally improved, wholesome food product or ingredient to achieve the goal of zero waste.

PT

2. Materials and methods

RI

2.1. Yeast strain and okara

SC

Freeze-dried Y. lipolytica NCYC 2904 culture was obtained from the National Collection of

NU

Yeast Culture (Norwich, UK) and propagated in sterile yeast-malt (YM) broth (2% glucose, 0.25% yeast extract, 0.25% bacteriological peptone and 0.2% malt extract, all w/v, pH 5.0). The yeast

MA

culture was incubated at 30 ◦C, 150 rpm for 48 h to obtain a cell population of about 7 log CFU/mL. Glycerol was added to the pure culture at 15% v/v and it was stored at -80 ◦C before

TE

D

use.

Fresh okara was provided by Super Bean International Pte Ltd (Singapore) from a single

AC CE P

batch. Raw, soaked and non-genetically modified soybeans were ground into fine particles at room temperature and then filtered to obtain the okara. The okara was stored at -20 ◦C before use. 2.2. Solid-state fermentation of okara A thawed pure yeast culture was added to sterile YM broth at 1% v/v and sub-cultured twice under the aforementioned conditions to obtain a cell count of about 6 log CFU/mL. It was then centrifuged (8,000 g, 15 min, 4 ◦C) and the supernatant discarded. The pellet was washed twice with 10% v/v phosphate saline buffer (pH 7.4), and the washed cells were resuspended in the same buffer to obtain the yeast pre-culture. Frozen okara was thawed, and 600 g of okara were placed in a glass container and autoclaved at 121 ◦C for 15 min. Yeast pre-culture was added to the sterilised okara at 2% v/w to and mixed

5

ACCEPTED MANUSCRIPT evenly with a sterilised metal spoon, obtaining an initial cell count of about 4 log CFU/g okara. The airtight containers were incubated at 30 ◦C for 5 days. Aerobic condition within the

PT

container was maintained by ensuring sufficient headspace (bed height: 3-3.5 cm; headspace:

RI

3.5-4 cm) and by mixing the substrate bed daily to introduce oxygen during sampling under aseptic conditions. Uninoculated, autoclaved okara incubated under the same conditions served

SC

as the control, while fresh, unheated okara served as the blank. All treatments were prepared in

NU

triplicate. Sampling was conducted daily over 5 days for the determination of viable yeast cell count, sugars, organic acids, amino acids and volatiles. Samples collected on days 0 and 5 were

MA

also subjected to proximate composition and antioxidant capacity analyses. Samples collected

TE

2.3. Yeast growth determination

D

were stored at -20 ◦C before the analyses.

AC CE P

To 90 mL of 0.1% w/v sterilised peptone water, 10 g of okara was added and the mixture was homogenised with a stomacher for 90 s. The homogenised mixture was appropriately diluted and spread plated on potato dextrose agar plates, which were then incubated at 25 ◦C for 48 h before yeast enumeration. At least two samples were taken from each container to obtain an average count.

2.4. Proximate composition Moisture content was measured with a moisture analyser (MOC-120H Shimadzu, Kyoto, Japan). Total nitrogen content was determined following AOAC 920.87 semi-micro Kjeldahl method (AOAC, 1995). The conversion factor 5.71 was used to transform nitrogen into protein. Ash content was determined by the direct ashing method (AOAC, 2000). Fat content was measured with a Soxtec apparatus (SoxtecTM 2050 FOSS, Hillerød, Denmark). Fat was extracted 6

ACCEPTED MANUSCRIPT from 3 g of freeze-dried okara with 90 mL of petroleum ether at 80 ◦C for 120 min. The defatted okara was collected and dried under nitrogen gas, and used for subsequent non-volatile and

PT

antioxidant extractions. It was stored at -20 ◦C before use. The combined carbohydrate and fibre

SC

2.5. Analysis of sugars, organic acids and amino acids

RI

content was calculated by mass differences.

NU

Sugars and amino acids were extracted from okara following the procedures of Giannoccaro et al. (2006) with slight modifications. Five grams of freeze-dried, defatted okara were extracted

MA

with 80 mL of 80% v/v ethanol in a water bath (50 ◦C, 150 rpm, 30 min). Lactose was added as an internal standard. Extraction was conducted thrice, and the combined extracts were

D

concentrated to 10 mL using a rotary evaporator. One part of concentrated extract was added to 2

TE

parts of acetonitrile, and the mixture was stored at 4 ◦C for at least 24 h to allow complete

AC CE P

precipitation of proteins. The precipitate was removed by centrifugation and the supernatant was stored at -20 ◦C before use.

Organic acids were extracted from 3 g of defatted freeze-dried okara with 50 mL of 0.1% v/v sulphuric acid in a water bath (50 ◦C, 150 rpm, 45 min). The extract was then centrifuged (10,000 g, 20 min) to remove the okara, and the supernatant was stored at 4 ◦C for at least 24 h to allow complete precipitation of proteins. The precipitate was removed by centrifugation and the supernatant was stored at -20 ◦C before use. HPLC analysis of sugars, organic acids and amino acids followed the methods of Chen et al. (2014) with slight modifications. For HPLC analysis of sugars, it was conducted at 30 ◦C with a flow rate of 1 mL/min instead of 40 ◦C at 1.4 mL/min. 2.6. Antioxidant capacity 7

ACCEPTED MANUSCRIPT Extraction of antioxidants followed the method optimised by Singh et al. (2011). One gram of freeze-dried okara was extracted with 50 mL of 25% v/v acetone in a water bath (40 ◦C, 150

PT

rpm, 15 min). The mixture was then centrifuged (5,000 g, 10 min) to remove the okara, and the

RI

supernatant was used in subsequent antioxidant assays, both measured using a microplate reader (BioTek, Winooski, Vermont, USA). Total phenolic content (TPC), expressed as gallic acid

SC

equivalent, was determined with Folin-Ciocalteu reagent as described by Isabelle et al. (2008).

MA

following the method of Huang et al. (2002).

NU

Oxygen radical absorbance capacity (ORAC), expressed as Trolox equivalent, was measured

2.7. Analysis of volatiles

D

In a 20-mL glass vial, 3 g of sample and 3 mL of saturated sodium chloride solution

TE

(Goodrich Chemical Enterprise, Singapore) were added and capped with a

AC CE P

polytetrafluoroethylene septum. Volatiles were extracted at 60 ◦C for 50 min using headspacesolid phase microextraction (HS-SPME) and then analysed by gas chromatography-mass spectrometry/flame ionisation detector (GC-MS/FID). The conditions for separation of volatiles by Agilent 5975C triple-axis MS and FID (Santa Clara, CA, USA) followed that as described by Chen et al. (2014). Identification of volatiles was based on comparison of their mass spectra with those from NIST 8.0 and Wiley 275 MS libraries. Alkane standards (C10– C40) (Fluka, Buchs, Switzerland) were run under the same conditions, and linear retention index values were calculated to verify the identities of volatiles. Semi-quantification was done using the GC-FID peak areas. The relative peak area refers to the peak area of the class of compounds expressed as a percentage of the total peak area of all volatiles present in the sample. The complete list of volatiles in the blank, control and fermented okara and their LRIs are given as supplementary data to this work. 8

ACCEPTED MANUSCRIPT 2.8. Statistical analysis

using SPSS ® 20.0 (SPSS Inc, Chicago, IL, USA) at p ≤ 0.05

RI

3. Results and discussion

PT

Data analysis was done using one-way analysis of variance (ANOVA) and Scheffe’s test

SC

3.1. Yeast population

NU

The viable yeast cell count increased from 3.94 to 7.09 log CFU/g wet matter from day 0 to 2, and remained relatively stable thereafter, with a final cell count of 7.73 log CFU/g wet matter on

MA

day 5 (Fig. 1A). Therefore, day 0 to 2 marked the exponential phase, while day 2 to 5 marked the stationary phase. Untreated (but autoclaved) okara supported the growth of Y. lipolytica well in

AC CE P

3.2. Proximate composition

TE

D

SSF set-up without any carbon or nitrogen supplementation.

The proximate composition of okara is shown in Table 1.Yeast metabolism of okara was the main reason for the observed changes in proximate composition, as the effects of yeast biomass growth and moisture loss were negligible (data not shown). No significant changes in carbohydrates and fibre, protein and ash content in okara were observed after fermentation. A lack of significant changes in protein content was likely due to the principle behind the Kjeldahl method: it measures the total nitrogen content in the sample without discriminating between protein and free amino acids. This explains the seemingly contradicting data between the lack of significant changes in protein content but a significant increase in total free amino acids (Table 2).

9

ACCEPTED MANUSCRIPT The fat content increased significantly from 7.41 to 8.73 g/100 g dry matter after fermentation. The fat content had decreased initially from day 0 to 2 before increasing

PT

subsequently (Fig. 1A). Y. lipolytica produces a few extracellular and cell wall-bound lipases

RI

(Fickers et al., 2011). These lipases become active at the growth phase, increase rapidly at the start of stationary phase and peak at late stationary phase (Papanikolaou and Aggelis, 2003). It

SC

was reported that when Y. lipolytica was cultivated in rapeseed oil without glucose

NU

supplementation, extracellular lipase activity increased immediately and remained high throughout fermentation (Kamzolova et al., 2011). Similarly, in this study, as the autoclaved

MA

okara contained a greater amount of fat but no glucose, a high initial level of extracellular lipase

D

activity was likely.

TE

The yeast lipases produced were likely to have hydrolysed the lipid fraction of okara, either partially to mono- or di-glycerides and fatty acids or completely to glycerol and fatty acids.

AC CE P

Some of the fatty acids in okara, such as C18:0 and C18:1, which makes up 4.7% and 20.4% of the total fatty acids respectively, can trigger triglyceride accumulation in Y. lipolytica (Beopoulos et al., 2008; Mateos-Aparicio, Redondo-Cuenca and Villanueva-Suárez, 2010; Saygün et al., 2014). Therefore, the lipase-hydrolysed and existing free fatty acids in okara could have been actively transported into the Y. lipolytica cells for intracellular conversion into triglycerides via the Kennedy pathway, which can then be consumed for energy or stored (Papanikolaou and Aggelis, 2011). The accumulated triglycerides were likely to have been consumed as the main carbon source during the growth phase (day 0 to 2), as there were low amounts of simple sugars (Fig. 1B). Intracellular triglycerides can be mobilised and broken down into glycerol and fatty acids, as illustrated in Fig. 2. Fatty acids are then further degraded via β-oxidation to give acyl- and 10

ACCEPTED MANUSCRIPT acetyl-CoA, and the latter can be channelled into the TCA cycle for energy production (Beopoulos et al., 2008).

PT

During the stationary phase (day 2 to 5), the concentrations of simple sugars increased. It was

RI

previously found that Y. lipolytica preferentially assimilated glucose over oleic acid as a carbon

SC

source (Kamzolova et al., 2011), so the rate of intracellular lipid degradation was likely to have decreased during the stationary phase. As lipase activity was probably highest during this period,

NU

the fatty acids produced could have accumulated and been converted into triglycerides inside the

MA

cells for storage. These intracellular lipid bodies were then released during fat extraction, leading to the increase in fat content from day 2 to 5.

TE

3.3.1. Sugars

D

3.3. Non-volatiles

AC CE P

In fresh okara, the free sugars present were glucose and fructose, at 1.42 g/100 g dry matter (Fig. 1C) and 1.37 g/100 g dry matter (data not shown) respectively. However, fructose was not consumed throughout fermentation (data not shown) as Y. lipolytica hexokinase has a low affinity for fructose (Lazar et al., 2014). Autoclave treatment slightly changed the sugar composition in okara. Glucose concentration decreased probably due to its participation in the Maillard reaction with okara amino acids. Galactose, which was not detected in fresh okara, could have been formed from the thermal degradation of okara hemicellulose (Tsubaki et al., 2009). Despite the slight increase in total free sugars following autoclave treatment, there was still a limited amount of free sugars in the autoclaved okara. Other sugar monomers in okara polysaccharides include arabinose, xylose and mannose, but these sugars were not detected in all okara samples. They likely remained bound as 11

ACCEPTED MANUSCRIPT part of the okara hemicellulose (Mateos-Aparicio, Redondo-Cuenca and Villanueva-Suárez, 2010).

PT

During the exponential phase, Y. lipolytica mainly catabolised galactose via the Leloir

RI

pathway (Lazar et al., 2015), as observed from the decrease in galactose concentration from day

SC

0 to 2 (Fig. 1B). Y. lipolytica requires glucose for synthesis of sufficient Leloir pathway enzymes (Lazar et al., 2015), but as glucose was not detected initially, Y. lipolytica was likely to have

NU

utilised fat, free fatty acids and amino acids as the alternative carbon sources.

MA

From day 2 to 3, the glucose concentration increased from 0.34 g to 1.83 g/100 g dry matter, and then decreased slightly thereafter. The increase in glucose level might be attributed to

D

gluconeogenesis via the glyoxylate cycle pathway (Fig. 2), where glucose can be generated from

TE

the acetyl-CoA of fatty acid origin, or from glucogenic amino acids (Fickers et al., 2005b;

AC CE P

Mansour et al., 2009). The rise in galactose from day 2 to 3 may be partly contributed by decreased consumption of galactose due to preferential metabolism of glucose by Y. lipolytica (Fickers et al., 2005b). Y. lipolytica may have also broken down the stachyose and raffinose in okara to galactose, as it contains genes encoding for β-glycosidases (Kanehisa et al., 2016; Mateos-Aparicio, Redondo-Cuenca, Villanueva-Suárez, et al., 2010). 3.3.2. Organic acids In general, significant changes in organic acids concentrations mainly occurred during the stationary phase (day 2 to 5), as shown in Fig. 3. Citrate concentration decreased at the start of stationary phase likely due to its metabolism via the TCA cycle and/or glyoxylate pathway (Fig. 2 and 3A). The channelling of citrate for gluconeogensis via the glyoxylate pathway coincided with the significant increase in glucose on 12

ACCEPTED MANUSCRIPT day 3. While some studies have demonstrated the use of Y. lipolytica for the production of citrate (Levinson et al., 2007; Papanikolaou et al., 2008; Saygün et al., 2014), this phenomenon was not

PT

observed in the present work. In those studies, citrate production by Y. lipolytica was induced by

RI

excess carbon and limited nitrogen in the media, while the okara substrate in this study contained low amounts of free sugars (limited carbon) and considerable amounts of free amino acids

SC

(excess nitrogen).

NU

The highest citrate concentration (0.375 g/100 g dry matter) was observed on day 1, which is

MA

equivalent to 5.06 g/100 g fat. This amount was about half of that produced by Y. lipolytica when soybean oil was the fermentation medium, which yielded 11.5 g/100 g oil after 7 days (Darvishi

D

et al., 2009). This difference was likely due to the variation in the strain used and fermentation

TE

set-up. The solid-state fermentation set-up employed in our study could have restricted the access to carbon sources by yeast cells, while the submerged-state fermentation set-up by Darvishi et al.

AC CE P

(2009) did not encounter this limitation, leading to a greater amount of citrate produced. In addition, ammonium ion, which was produced in relatively large amounts in this study (Table 2), is a negative modulator of citrate synthase and has been associated with a high level of activity of aconitase (Il'chenko et al., 2002). Hence, the okara substrate was unfavourable for citrate production, and any synthesised citrate was likely to have been converted into isocitrate and metabolised. Consequently, citrate concentration did not increase but decreased when Y. lipolytica was inoculated on okara. On the other hand, pyruvate and α-ketoglutarate were not detected during the exponential phase but increased at the onset of stationary phase (day 2) (Fig. 3B and 3C). The typical thiamine content in okara greatly exceeds the growth-limiting amount for Y. lipolytica (Morgunov et al., 2004; van der Riet et al., 1989). Under thiamine-sufficient conditions, pyruvate 13

ACCEPTED MANUSCRIPT and α-ketoglutarate would be degraded by pyruvate dehydrogenase and α-ketoglutarate dehydrogenase respectively (Morgunov et al., 2004; Zhou et al., 2010) (Fig. 2). From day 2

PT

onwards, as the yeast cells entered stationary phase, thiamine could have become a limiting

RI

factor. As a result, pyruvate and α-ketoglutarate were not degraded, and these accumulated acids were excreted by yeast cells, leading to an increase in concentrations of these two acids. Similar

SC

observations were also made by Morgunov et al. (2004), who also showed that the activity of Y.

NU

lipolytica pyruvate carboxylase significantly increased in the later part of the stationary phase when grown in a glycerol media. This might explain the observed decrease in pyruvate

MA

concentration from day 4 to 5, as pyruvate was being converted into oxaloacetate.

D

A remarkable 3-fold increase in succinate was achieved after fermentation (Fig. 3D). The

TE

production of succinate from day 3 was likely to have occurred via the glyoxylate cycle (Fig. 2). In the glyoxylate cycle, isocitrate is cleaved by isocitrate lyase to produce succinate and

AC CE P

glyoxylate, and the latter is then converted into malate by incorporating acetyl-CoA, some of which were derived from fatty acid catabolism in this study. Il'chenko et al. (2002) also showed that under thiamine-limiting conditions, isocitrate lyase accounts for the main production route of succinate in Y. lipolytica. This is in agreement with the observed increase in succinate and malate concentrations from day 3 to 5 (Fig. 3D and 3E). Although Morgunov et al. (2004) suggested that the glyoxylate cycle was not operative in Y. lipolytica when grown in glycerol media, the composition of okara is rather different from that of pure glycerol. Another likely route of succinate production is via the reduction of fumarate, which is derived from malate, since malate was produced and fumarate reductase activity has been shown to persist even during the stationary phase of Y. lipolytica (Il'chenko et al., 2002). Further increase in succinate

14

ACCEPTED MANUSCRIPT concentration after day 5 is likely to occur as its production only began towards the end of the fermentation period.

PT

Succinate and its derivatives are widely used in the food market as acidulants, antimicrobials or

RI

flavours (Beauprez et al., 2010). Succinate is also an important contributor of the umami taste in

SC

Swiss cheese, with a rather low detection threshold of 200 µg/mL in water and 400 µg/mL in a cheese model (Drake et al., 2007). Previously, Y. lipolytica was exploited for the production of

NU

succinate using engineered strains (Yuzbashev et al., 2010) or by chemical decarboxylation of its

MA

metabolite (Kamzolova et al., 2014). In the study by Kamzolova et al. (2014), when rapeseed oil was used as the sole carbon source, the final succinate concentration was 69.0 g/L. Our study

D

obtained a lower succinate concentration at 33.7 g/kg dry matter on Day 5. The difference could

TE

be due to the limited ability of Y. lipolytica to utilise the insoluble dietary fibre in okara as a carbon source, and the lack of process optimisation. Further work may be done to harness the

AC CE P

potential of utilising okara as an alternative substrate for producing food-grade succinate with a non-recombinant Y. lipolytica. 3.3.3. Amino acids

Table 2 shows the change in amino acid concentrations before and after fermentation. Autoclaving okara slightly increased the total free amino acids concentration, although it was not significant. Fermentation significantly increased the total free amino acid concentration by about 4-fold, and this increase was mainly contributed by changes in the concentrations of serine/asparagine, glutamate, histidine/glutamine, alanine, proline, and phenylalanine. The increase in amino acids after fermentation highlights the strong proteolytic capability of Y. lipolytica, which produces extracellular acid and alkaline proteases (Akpınar et al., 2011). 15

ACCEPTED MANUSCRIPT Although amino acid degradation/deamination could also have occurred, as reflected in the increase in ammonium concentration, it is more likely that the rate of protein hydrolysis

PT

exceeded that of amino acid catabolism, resulting in a net increase in free amino acids after

RI

fermentation.

SC

During fermentation, free glucogenic amino acids released (alanine, asparagine, aspartate, glutamate and glutamine) can undergo oxidative deamination to form TCA cycle intermediates

NU

or participate in gluconeogenesis (Mansour et al., 2009). These pathways could have partly

MA

accounted for the increase in TCA cycle organic acids and glucose. Notably, glutamate concentration increased by about 20-fold after fermentation and was the main amino acid present

D

in the fermented okara. This is consistent with the observation that glutamate is the major amino

TE

acid in okara protein isolates (Chan and Ma, 1999). Complete chemical hydrolysis of okara protein isolates produced 195 mg glutamate/g protein (Chan and Ma, 1999), which is about 13

AC CE P

times greater than the present concentration of 14.5 mg glutamate/g protein in the fermented okara. Prolonged fermentation or increasing the inoculum size may further boost the amount of glutamate produced.

The production of free amino acids by Y. lipolytica can affect the nutritional and sensorial quality of fermented okara in various ways. The breakdown of okara proteins into free amino acids and, most likely, shorter peptides improves the digestibility of okara. Some amino acids and peptides could also have contributed to the increased antioxidant capacity in the fermented okara (Section 3.4). Besides, the pronounced spike in glutamate can impart an umami taste, while the catabolism of branched-chain and aromatic amino acids influenced the volatile profile (Section 3.5). Clearly, protein and amino acids play important roles in improving the nutrition and flavour of the fermented okara. 16

ACCEPTED MANUSCRIPT 3.4. Antioxidant capacity The trends in the TPC and ORAC assays were in agreement with each other (Fig. 4).

PT

Comparing between the blank and control at day 0, heating slightly increased the antioxidant

RI

capacity, which could be due to the enhanced solubility of polyphenolic compounds and the

SC

thermal degradation of isoflavones glycosides (with their aglycone counterparts showing greater

NU

antioxidant capacity) (Tsubaki et al., 2009).

Fermentation significantly improved the antioxidant capacity of okara due to the degradation

MA

products of okara by the yeast, as yeast biomass did not affect antioxidant capacity (data not shown). The improvement in antioxidant capacity could be contributed by various factors. Firstly,

D

the increase in the amounts of certain amino acids could have enhanced antioxidant activity.

TE

Hydrophobic and aromatic amino acids, such as histidine, tryptophan and phenylalanine, have

AC CE P

antioxidant properties (Pownall et al., 2010). The increase in free amino acid contents may also indicate the production of shorter chain peptides, and the proteolysis of okara protein has been shown to release potentially bioactive or antioxidant peptides (Jiménez-Escrig et al., 2010). Secondly, the increase in TPC may be due to the tyrosine catabolism by Y. lipolytica to produce various polyphenols, such as p-hydroxyphenylethanol, p-hydroxyphenylacetic acid and homogenistic acid, although the mechanism is not well-understood to date (Williams and Withers, 2007). Thirdly, it is plausible that okara aglycone isoflavones had increased after fermentation. Majority of the okara isoflavones exist as glycoside conjugates, which can be cleaved by β-glucosidase to release the more bioactive aglycone counterparts (Anderson, 1995). In this study, the Y. lipolytica strain used can assimilate cellobiose (NCYC, 1999) and various cellobiose-utilising Y. lipolytica strains have been shown to possess putative genes encoding for

17

ACCEPTED MANUSCRIPT extracellular β-glucosidase (Guo et al., 2015; Ryu et al., 2016). Moreover, Ryu et al. (2016) noted that there was constitutive expression of some β-glucosidase genes in wild-type Y.

PT

lipolytica regardless of the sugars present in the media. It is, therefore, highly plausible that some

RI

degree of β-glucosidase activity was exhibited by the strain used in the present study during fermentation. This could have led to the bioconversion of okara isoflavones from glycosides to

SC

aglycones, resulting in the observed increase in antioxidant capacity. Further investigation on

NU

this hypothesis is certainly warranted.

MA

3.5. Volatiles

In the fresh okara (blank), the major volatiles were aldehydes (relative peak area of 97.5%).

D

After autoclaving, okara contained mainly aldehydes and alcohols. The final fermented okara

TE

contained mostly alcohols and ketones, with relative peak areas of 65.1% and 20.2% respectively

AC CE P

(supplementary data, Table S1). Biotransformation decreased the total amount of aldehydes in the control okara by about 8 times, while increasing the total amount of ketones by 4.5 times. Some volatiles unique to fermented okara include 2-phenylethanol, 2-methyl-2-propenal, and 2pentanone. The time-course changes of some key volatiles in the blank, control and fermented samples are shown in Fig. 5 and 6. Hexanal, (E,E)-2,4-nonadienal and (E,E)-2,4-decadienal are typically formed by enzymatic degradation of fatty acids (Fig. 5A, 5B and 5C). When raw soybeans are ground, soybean lipoxygenases catalyse the hydroperoxidation of unsaturated fatty acids, such as linoleic and linolenic acids, and the intermediates are then lysed by soy hydroperoxide lyases to form aldehydes (Stanojevic et al., 2014). Therefore, the fresh okara contained the greatest amount of hexanal, although it decreased after autoclaving. Autoclave treatment may have increased the

18

ACCEPTED MANUSCRIPT amounts of (E,E)-2,4-nonadienal and (E,E)-2,4-decadienal by thermal degradation of the linolenic fatty acids, and increased the amount of 3-methylbutanal (Fig. 6C) by deamination of

PT

leucine via Strecker degradation (Lozano et al., 2007).

RI

During incubation, hexanal had spontaneously oxidised to hexanoic acid, as observed from

SC

the decrease in hexanal (Fig. 5A) and the corresponding increase in hexanoic acid in the control (Fig. 5F). However, in the fermented sample, the decrease in hexanal content did not correspond

NU

to an increase in hexanol nor hexanoic acid content; instead, both of these compounds decreased

MA

during fermentation (Fig. 5E and 5F). Similarly, the amount of (E,E)-2,4-nonadienal and (E,E)2,4-decadienal decreased to undetectable levels by day 2, suggesting that these aldehydes were

D

catabolised by Y. lipolytica.

TE

These aldehydes were likely to have been oxidised to fatty acids spontaneously or

AC CE P

enzymatically, and then catabolised via β-oxidation by the yeast (Fickers et al., 2005a). Partial βoxidation of fatty acids produces methyl ketones with one less carbon than the initial fatty acid (Lanciotti et al., 2005b), such as 2-propanone and 2-pentanone (Fig. 6A and 6B). For instance, the partial β-oxidation of hexanoic acid forms 2-pentanone; in the fermented okara, the decrease in the former compound (Fig. 5F) corresponded to the increase in the latter compound (Fig. 6B). This tendency of Y. lipolytica to produce short-chain methyl ketones was also observed in a cheese-surface model (Sørensen et al., 2011). In addition, Y. lipolytica catabolism of branched-chain and aromatic amino acids via the Ehrlich pathway was likely to have produced higher aldehydes and alcohols (Hazelwood et al., 2008). Leucine and phenylalanine concentrations had increased by approximately 2- and 6-fold

19

ACCEPTED MANUSCRIPT respectively following fermentation (Table 2), and their catabolism can lead to the formation of 3-methylbutanal and 2-phenylethanol respectively (Fig. 6C and 6D).

PT

Overall, the fermented okara had no grassy and beany off-odour due to the elimination of

RI

hexanal and unsaturated aldehydes. Instead, short-chain methyl ketones, which are common

SC

aroma compounds in surface-ripened cheeses such as Camembert and Brie, gave the fermented okara a musty, cheese-like odour (Zinjarde, 2014). On the other hand, the malty aroma of 3-

NU

methylbutanal and the floral, rose-like aroma of 2-phenylethanol can give a slight floral nuance

MA

in the fermented okara, as they typically do in modulating the aroma of some cheeses (Abilleira et al., 2010). The final fermented okara did not smell grassy or beany, but slightly pungent and

TE

4. Conclusion

D

musty, similar to surface-ripened cheeses.

AC CE P

Growth of Y. lipolytica was well-supported in the okara matrix, in part due to its proteolytic and lipolytic capabilities to assimilate the lipid and protein fractions in okara. The catabolism of these components in the okara consequently led to an significant increase in succinate, free amino acids (especially glutamate), and short-chain methyl ketones, all of which could contribute to an overall umami and cheese-like flavour in the fermented okara. Moreover, yeast biotransformation of okara also enhanced its antioxidant capacity. This study therefore demonstrated that the exploitation of Y. lipolytica could go beyond the production of a single or a few value-added compounds; instead, the whole Yarrowia-fermented okara could be used as a nutritious and more palatable food product, achieving zero waste altogether. As evident, Y. lipolytica is suitable for valorising soy food processing by-products and expanding their potential applications.

20

ACCEPTED MANUSCRIPT Acknowledgements The authors thank Super Bean International Pte Ltd (Singapore) for kindly providing the

PT

okara used in this study.

SC

RI

Appendix A. Supplementary data

AC CE P

TE

D

MA

NU

Supplementary data to this article can be found online.

21

ACCEPTED MANUSCRIPT

PT

References Abilleira, E., Schlichtherle-Cerny, H., Virto, M., de Renobales, M., Barron, L.J.R., 2010.

SC

in winter and spring. Int. Dairy J. 20, 537-544.

RI

Volatile composition and aroma-active compounds of farmhouse Idiazabal cheese made

NU

Akpınar, O., Uçar, F., Yalçın, H.T., 2011. Screening and regulation of alkaline extracellular protease and ribonuclease production of Yarrowia lipolytica strains isolated and

MA

identified from different cheeses in Turkey. Ann. Microbiol. 61, 907-915. Anderson, R.L., 1995. Compositional changes in trypsin inhibitors, phytic acid, saponins and

D

isoflavones related to soybean processing. J. Nutr. 125, 581S-588S.

TE

AOAC, 1995. Method 920.87. Protein (total) in flour, final action. In: Official methods of

AC CE P

analysis of AOAC International. AOAC International, Maryland, USA, pp. 32.1.22. AOAC, 2000. Method 923.03. Ash of flour. In: Official methods of analysis of AOAC International. AOAC International, Maryland, USA, pp. 32.1.05. Beauprez, J.J., De Mey, M., Soetaert, W.K., 2010. Microbial succinic acid production: natural versus metabolic engineered producers. Process Biochem. 45, 1103-1114. Beopoulos, A., Mrozova, Z., Thevenieau, F., Le Dall, M.-T., Hapala, I., Papanikolaou, S., Chardot, T., Nicaud, J.-M., 2008. Control of lipid accumulation in the yeast Yarrowia lipolytica. Appl. Environ. Microbiol.74, 7779-7789. Chan, W.-M., Ma, C.-Y., 1999. Acid modification of proteins from soymilk residue (okara). Food Res. Int. 32, 119-127.

22

ACCEPTED MANUSCRIPT Chen, D., Chia, J.Y., Liu, S.-Q., 2014. Impact of addition of aromatic amino acids on nonvolatile and volatile compounds in lychee wine fermented with Saccharomyces cerevisiae

PT

MERIT.ferm. Int. J. Food Microbiol. 170, 12-20.

RI

Darvishi, F., Nahvi, I., Zarkesh-Esfahani, H., Momenbeik, F., 2009. Effect of plant oils upon lipase and citric acid production in Yarrowia lipolytica yeast. J. Biomed. Biotechnol.

SC

2009.

NU

Drake, S.L., Carunchia Whetstine, M.E., Drake, M.A., Courtney, P., Fligner, K., Jenkins, J., Pruitt, C., 2007. Sources of umami taste in Cheddar and Swiss cheeses. J. Food Sci. 72,

MA

S360-S366.

Fickers, P., Benetti, P.-H., Wache, Y., Marty, A., Mauersberger, S., Smit, M., Nicaud, J.-M.,

TE

D

2005a. Hydrophobic substrate utilisation by the yeast Yarrowia lipolytica, and its potential applications. FEMS Yeast Res. 5, 527-543.

AC CE P

Fickers, P., Marty, A., Nicaud, J.M., 2011. The lipases from Yarrowia lipolytica: genetics, production, regulation, biochemical characterization and biotechnological applications. Biotechnol. Adv. 29, 632-644. Fickers, P., Nicaud, J.M., Destain, J., Thonart, P., 2005b. Involvement of hexokinase Hxk1 in glucose catabolite repression of LIP2 encoding extracellular lipase in the yeast Yarrowia lipolytica. Curr. Microbiol. 50, 133-137. Giannoccaro, E., Wang, Y.-J., Chen, P., 2006. Effects of solvent, temperature, time, solvent-tosample ratio, sample size, and defatting on the extraction of soluble sugars in soybean. J. Food Sci. 71, C59-C64. Gomes, N., Teixeira, J.A., Belo, I., 2012. Fed-batch versus batch cultures of Yarrowia lipolytica for γ-decalactone production from methyl ricinoleate. Biotechnol. Lett. 34, 649-654.

23

ACCEPTED MANUSCRIPT Groenewald, M., Boekhout, T., Neuvéglise, C., Gaillardin, C., Van Dijck, P.W., Wyss, M., 2014. Yarrowia lipolytica: safety assessment of an oleaginous yeast with a great industrial

PT

potential. Crit. Rev. Microbiol. 40, 187-206.

RI

Guo, Z., Duquesne, S., Bozonnet, S., Cioci, G., Nicaud, J.-M., Marty, A., O’Donohue, M.J., 2015. Development of cellobiose-degrading ability in Yarrowia lipolytica strain by

SC

overexpression of endogenous genes. Biotechnol. Biofuels 8, 1.

NU

Hazelwood, L.A., Daran, J.-M., van Maris, A.J., Pronk, J.T., Dickinson, J.R., 2008. The Ehrlich pathway for fusel alcohol production: a century of research on Saccharomyces cerevisiae

MA

metabolism. Appl. Environ. Microbiol.74, 2259-2266. Huang, D., Ou, B., Hampsch-Woodill, M., Flanagan, J.A., Prior, R.L., 2002. High-throughput

TE

D

assay of oxygen radical absorbance capacity (ORAC) using a multichannel liquid handling system coupled with a microplate fluorescence reader in 96-well format. J.

AC CE P

Agric. Food Chem. 50, 4437-4444. Il'chenko, A.P., Chernyavskaya, O.G., Shishkanova, N.V., Finogenova, T.V., 2002. Metabolism of Yarrowia lipolytica grown on ethanol under conditions promoting the production of αketoglutaric and citric acids: a comparative study of the central metabolism enzymes. Microbiol. 71, 269-274.

Isabelle, M., Lee, B.L., Ong, C.N., Liu, X., Huang, D., 2008. Peroxyl radical scavenging capacity, polyphenolics, and lipophilic antioxidant profiles of mulberry fruits cultivated in southern China. J. Agric. Food Chem. 56, 9410-9416. Jiménez-Escrig, A., Alaiz, M., Vioque, J., Rupérez, P., 2010. Health-promoting activities of ultra-filtered okara protein hydrolysates released by in vitro gastrointestinal digestion:

24

ACCEPTED MANUSCRIPT identification of active peptide from soybean lipoxygenase. Eur. Food Res. Technol. 230, 655-663.

PT

Kamzolova, S.V., Lunina, J.N., Morgunov, I.G., 2011. Biochemistry of citric acid production

RI

from rapeseed oil by Yarrowia lipolytica yeast. J. Am. Oil Chem. Soc. 88, 1965-1976. Kamzolova, S.V., Vinokurova, N.G., Dedyukhina, E.G., Samoilenko, V.A., Lunina, J.N.,

SC

Mironov, A.A., Allayarov, R.K., Morgunov, I.G., 2014. The peculiarities of succinic acid

NU

production from rapeseed oil by Yarrowia lipolytica yeast. Appl. Microbiol. Biotechnol.

MA

98, 4149–4157.

Kanehisa, M., Sato, Y., Kawashima, M., Furumichi, M., Tanabe, M., 2016. KEGG as a reference

D

resource for gene and protein annotation. Nucleic Acids Res. 44, D457-D462.

TE

Khare, S., Jha, K., Gandhi, A., 1995. Citric acid production from okara (soy-residue) by solid-

AC CE P

state fermentation. Biores. Technol. 54, 323-325. Lanciotti, R., Gianotti, A., Baldi, D., Angrisani, R., Suzzi, G., Mastrocola, D., Guerzoni, M., 2005a. Use of Yarrowia lipolytica strains for the treatment of olive mill wastewater. Bioresour. Technol. 96, 317-322. Lanciotti, R., Vannini, L., Lopez, C.C., Gobbetti, M., Guerzoni, M.E., 2005b. Evaluation of the ability of Yarrowia lipolytica to impart strain-dependent characteristics to cheese when used as a ripening adjunct. Int. J. Dairy Technol. 58, 89-99. Lazar, Z., Dulermo, T., Neuvéglise, C., Crutz-Le Coq, A.-M., Nicaud, J.-M., 2014. Hexokinase – a limiting factor in lipid production from fructose in Yarrowia lipolytica. Metab. Eng. 26, 89-99.

25

ACCEPTED MANUSCRIPT Lazar, Z., Gamboa-Meléndez, H., Crutz-Coq, A.-M., Neuvéglise, C., Nicaud, J.-M., 2015. Awakening the endogenous Leloir pathway for efficient galactose utilization by Yarrowia

PT

lipolytica. Biotechnol. Biofuels 8, 1.

RI

Levinson, W.E., Kurtzman, C.P., Kuo, T.M., 2007. Characterization of Yarrowia lipolytica and related species for citric acid production from glycerol. Enz. Microb. Technol. 41, 292-

SC

295.

NU

Li, B., Qiao, M., Lu, F., 2012. Composition, nutrition, and utilization of okara (soybean residue). Food Rev. Int. 28, 231-252.

MA

Lozano, P.R., Drake, M., Benitez, D., Cadwallader, K.R., 2007. Instrumental and sensory

TE

Food Chem. 55, 3018-3026.

D

characterization of heat-induced odorants in aseptically packaged soy milk. J. Agric.

Mansour, S., Bailly, J., Delettre, J., Bonnarme, P., 2009. A proteomic and transcriptomic view of

AC CE P

amino acids catabolism in the yeast Yarrowia lipolytica. Proteomics 9, 4714-4725. Mateos-Aparicio, I., Redondo-Cuenca, A., Villanueva-Suárez, M.-J., Zapata-Revilla, M.-A., Tenorio-Sanz, M.-D., 2010. Pea pod, broad bean pod and okara, potential sources of functional compounds. LWT - Food Sci. Technol. 43, 1467-1470. Mateos-Aparicio, I., Redondo-Cuenca, A., Villanueva-Suárez, M., 2010. Isolation and characterisation of cell wall polysaccharides from legume by-products: okara (soymilk residue), pea pod and broad bean pod. Food Chem. 122, 339-345. Morgunov, I.G., Kamzolova, S.V., Perevoznikova, O.A., Shishkanova, N.V., Finogenova, T.V., 2004. Pyruvic acid production by a thiamine auxotroph of Yarrowia lipolytica. Process Biochem. 39, 1469-1474.

26

ACCEPTED MANUSCRIPT NCYC, 1999. National collection of yeast cultures: Yarrowia lipolytica NCYC 2904. Retrieved May 4, 2016, from https://catalogue.ncyc.co.uk/yarrowia-lipolytica-2904

PT

Papanikolaou, S., Aggelis, G., 2003. Modeling lipid accumulation and degradation in Yarrowia

RI

lipolytica cultivated on industrial fats. Curr. Microbiol. 46, 0398-0402. Papanikolaou, S., Aggelis, G., 2011. Lipids of oleaginous yeasts. Part I: Biochemistry of single

SC

cell oil production. Eur. J. Lipid Sci. Technol. 113, 1031-1051.

NU

Papanikolaou, S., Galiotou-Panayotou, M., Fakas, S., Komaitis, M., Aggelis, G., 2008. Citric acid production by Yarrowia lipolytica cultivated on olive-mill wastewater-based media.

MA

Bioresour. Technol. 99, 2419-2428.

Patrignani, F., Iucci, L., Vallicelli, M., Guerzoni, M.E., Gardini, F., Lanciotti, R., 2007. Role of

TE

D

surface-inoculated Debaryomyces hansenii and Yarrowia lipolytica strains in dried fermented sausage manufacture. Part 1: Evaluation of their effects on microbial evolution,

AC CE P

lipolytic and proteolytic patterns. Meat Sci. 75, 676-686. Pownall, T.L., Udenigwe, C.C., Aluko, R.E., 2010. Amino acid composition and antioxidant properties of pea seed (Pisum sativum L.) enzymatic protein hydrolysate fractions. J. Agric. Food Chem. 58, 4712-4718. Rashad, M.M., Mahmoud, A.E., Abdou, H.M., Nooman, M.U., 2013. Improvement of nutritional quality and antioxidant activities of yeast fermented soybean curd residue. Afr. J. Biotechnol. 10, 5504-5513. Ryu, S., Hipp, J., Trinh, C.T., 2016. Activating and elucidating metabolism of complex sugars in Yarrowia lipolytica. Appl. Environ. Microbiol. 82, 1334-1345.

27

ACCEPTED MANUSCRIPT Saygün, A., Şahin-Yeşilçubuk, N., Aran, N., 2014. Effects of different oil sources and residues on biomass and metabolite production by Yarrowia lipolytica YB 423-12. J. Am. Oil

PT

Chem. Soc. 91, 1521-1530.

RI

Singh, A., Kuila, A., Yadav, G., Banerjee, R., 2011. Process optimization for the extraction of polyphenols from okara. Food Technol. Biotechnol. 49, 322.

SC

Sørensen, L.M., Gori, K., Petersen, M.A., Jespersen, L., Arneborg, N., 2011. Flavour compound

NU

production by Yarrowia lipolytica, Saccharomyces cerevisiae and Debaryomyces hansenii in a cheese-surface model. Int. Dairy J. 21, 970-978.

MA

Stanojevic, S.P., Barac, M.B., Pesic, M.B., Zilic, S.M., Kresovic, M.M., Vucelic-Radovic, B.V., 2014. Mineral elements, lipoxygenase activity, and antioxidant capacity of okara as a

TE

D

byproduct in hydrothermal processing of soy milk. J. Agric. Food Chem. 62, 9017-9023. Tsubaki, S., Nakauchi, M., Ozaki, Y., Azuma, J.-i., 2009. Microwave heating for solubilization

AC CE P

of polysaccharide and polyphenol from soybean residue (okara). Food Sci. Technol. Res. 15, 307-314.

van der Riet, W.B., Wight, A.W., Cilliers, J.J.L., Datel, J.M., 1989. Food chemical investigation of tofu and its byproduct okara. Food Chem. 34, 193-202. Vong, W.C., Liu, S.Q., 2016a. Biovalorisation of okara (soybean residue) for food and nutrition. Trends Food Sci. Technol. 52, 139 – 147. Vong, W.C., Liu, S.Q., 2016b. Changes in volatile profile of soybean residue (okara) upon solid‐ state fermentation by yeasts. J. Sci. Food Agric. DOI: 10.1002/jsfa.7700 Williams, A.G., Withers, S.E., 2007. Tyrosine metabolism in pigment‐forming Yarrowia lipolytica strains isolated from English and European speciality mould‐ripened cheese exhibiting a brown discolouration defect. Int. J. Dairy Technol. 60, 165-174.

28

ACCEPTED MANUSCRIPT Yuzbashev, T.V., Yuzbasheva, E.Y., Sobolevskaya, T.I., Laptev, I.A., Vybornaya, T.V., Larina, A.S., Matsui, K., Fukui, K., Sineoky, S.P., 2010. Production of succinic acid at low pH

PT

by a recombinant strain of the aerobic yeast Yarrowia lipolytica. Biotechnol. Bioeng. 107,

RI

673-682.

Zhou, J., Zhou, H., Du, G., Liu, L., Chen, J., 2010. Screening of a thiamine-auxotrophic yeast for

SC

α-ketoglutaric acid overproduction. Lett. Appl. Microbiol. 51, 264-271.

AC CE P

TE

D

MA

NU

Zinjarde, S.S., 2014. Food-related applications of Yarrowia lipolytica. Food Chem. 152, 1-10.

29

ACCEPTED MANUSCRIPT Figure captions Fig. 1. Time-course changes in (A) yeast cell count (dashed line) and fat (solid line) and in (B)

PT

glucose (solid line) and galactose (dashed line). Blank (■); okara fermented by Yarrowia

RI

lipolytica (▲). For yeast cell count, error bars are standard deviations from three independent

SC

replicates, each tested in duplicate (n = 6). For fat, glucose and galactose concentrations, error

NU

bars are standard deviations from three independent replicates (n = 3). Fig. 2. Metabolism of okara sugar and triglyceride fractions by Yarrowia lipolytica for the

MA

production of organic acids via the tricarboxylic acid (TCA) cycle (solid black arrows) and glyoxylate cycle (broken orange arrows). DH = dehydrogenase.

TE

D

Fig. 3. Time-course changes in organic acids in blank (■) and okara fermented by Yarrowia

AC CE P

lipolytica (▲). Error bars are standard deviations from three independent replicates (n = 3). Fig. 4. Changes in total phenolic content (TPC) and oxygen radical absorbance capacity (ORAC) in blank, control (day 0) and okara fermented by Yarrowia lipolytica (day 5). Different letters indicate significant differences among the TPC or ORAC values at p ≤ 0.05. Error bars are standard deviations from three independent replicates (n = 3). Fig. 5. Time-course changes in selected aldehydes, alcohols and acid in blank (■), control (♦) and okara fermented by Yarrowia lipolytica (▲). Error bars are standard deviations from three independent replicates (n = 3). Fig. 6. Time-course changes in selected methyl ketones, aldehyde and alcohol in blank (■), control (♦) and okara fermented by Yarrowia lipolytica (▲). Error bars are standard deviations from three independent replicates (n = 3).

30

ACCEPTED MANUSCRIPT Yeast cell count

A

Fat

9

9.5

Glucose (Fermented) Galactose (Fermented)

2.5

8.0 7.5

6

7.0

5

6.5 6.0

4

1

2

3

4

1.0

SC

5.0 0

1.5

0.5

5.5

3

PT

7

2.0

RI

8.5

g/ 100g dry matter

9.0

8

g/100 g dry matter

log CFU/g okara (wet basis)

Glucose (Blank)

B

0.0

5

0

NU

Day

2

3

4

5

Day

AC CE P

TE

D

MA

Fig. 1.

1

31

ACCEPTED MANUSCRIPT

Galactose

Glycolysis

Lipase

Leloir pathway

Fatty acid β-Oxidation

Pyruvate DH

Acetyl-CoA

SC

Citrate synthase

NU

Oxaloacetate Acetyl-CoA Malate synthase

Fumarate reductase

MA

Glyoxylate

D

Fumarate

Citrate Aconitase

Isocitrate lyase

Isocitrate Isocitrate DH

α-Ketoglutarate α-Ketoglutarate DH

AC CE P

TE

Succinate

Fig. 2.

RI

Pyruvate

Pyruvate carboxylase

Malate

Triglyceride

PT

Glucose

32

ACCEPTED MANUSCRIPT A

B

Citrate 0.9

0.12

0.5 0.4 0.3 0.2

0.06 0.04

SC

0.02

0.1

0.00

0.0 0

2

0

4

C

1

NU

Day

D

α-Ketoglutarate

2

3

4

5

3

4

5

Day

Succinate

4.5

0.09 0.08 0.07 0.06

D

0.05

g / 100 g dry matter

MA

0.10

0.04

TE

g / 100 g dry matter

0.08

RI

0.6

PT

0.10

0.7

g / 100 g dry matter

g / 100 g dry matter

0.8

0.03 0.02 0.00 0

AC CE P

0.01 1

2

3

E

4.0 3.5 3.0 2.5 2.0 1.5 1.0 0.5 0.0

4

5

Day

0

1

2 Day

Malate

0.70 0.60 g / 100 g dry matter

Pyruvate

0.50 0.40 0.30 0.20 0.10 0.00 0

1

2

3

4

5

Day

Fig. 3.

33

ACCEPTED MANUSCRIPT

250

1000

200

a

a

800

150

600

100

400 B A

50

200

0

0 Control (Day 0)

Fermented (Day 5)

MA

Blank

RI

1200

SC

C

300

PT

1400

b

ORAC (mg Trolox equivalent/100 g dry matter)

TPC ORAC

NU

TPC (mg gallic acid equivalent/100 g dry matter)

350

D

Samples

AC CE P

TE

Fig. 4.

34

ACCEPTED MANUSCRIPT A

B

Hexanal

(E,E)-2,4-Nonadienal

14000

200

GC-FID peak area (104)

6000 4000 2000

100

50

0

0

0

1

2

3

4

5

C

0

1

NU

Day

D

(E,E)-2,4-Decadienal 800

2

4

5

Ethanol

1600

GC-FID peak area (104)

MA

1200

600

1000

500

D

400

200

0

AC CE P

100 1

2

3

4

800 600 400

TE

300

0

200 0 0

5

1

2

Day

E 7000 6000 5000

4

5

Hexanoic acid

F

1-Hexanol

8000

3 Day

1200 1000 GC-FID peak area (104)

GC-FID peak area (104)

3 Day

1400

700 GC-FID peak area (104)

PT

8000

150

RI

10000

SC

GC-FID peak area (104)

12000

4000 3000 2000 1000

800 600 400 200 0

0 0

1

2

3 Day

4

5

0

1

2

3

4

5

Day

Fig. 5.

35

ACCEPTED MANUSCRIPT B GC-FID peak area (104)

GC-FID peak area (104)

900 800 700 600 500 400 300 200

2-Pentanone 300 250 200

PT

2-Propanone

150

RI

A

100

SC

50

100

0

0 0

1

2

3

4

0

5

C

1

NU

Day

D

3-Methylbutanal

2

3

4

5

4

5

Day

2-Phenylethanol

7000

MA

250

TE

100 50 0 0

1

2

3

Day

Fig. 6.

GC-FID peak area (104)

D

150

AC CE P

GC-FID peak area (104)

6000

200

4

5000 4000 3000 2000 1000 0

5

0

1

2

3 Day

36

ACCEPTED MANUSCRIPT Table 1

PT

Proximate composition of okara before and after fermentation by Y. lipolytica.

Control (Day 0)

RI

Average (g/100 g dry matter) Fermented (Day 5)

76.30 ± 0.63a

Carbohydrate + fibre

62.53 ± 1.81a

Protein

26.20 ± 1.00a

23.48 ± 0.70a

Fat

7.41 ± 0.73a

8.73 ± 0.15b

MA

NU

SC

Moisture (g/100 g wet matter)

3.86 ± 0.31a

Ash

62.57 ± 0.61a

3.77 ± 0.09a

Different letters in the same row indicate significant differences at p ≤ 0.05. Values are the

D

a,b

75.24 ± 0.62a

AC CE P

TE

mean ± standard deviation of three independent replicates (n = 3).

37

ACCEPTED MANUSCRIPT Table 2

PT

Amino acid concentrations before and after fermentation by Y. lipolytica. Average (mg/100 g dry matter) Blank Control (Day 0) Fermented (Day 5) a b 7.30 ± 0.76 16.46 ± 1.37 24.91 ± 3.39c 16.65 ± 0.69a 20.12 ± 1.56a 144.41 ± 57.93b 14.19 ± 0.97a 17.28 ± 1.69a 339.49 ± 81.32b 6.92 ± 0.40a 8.71 ± 0.81a 32.50 ± 13.26b 12.70 ± 0.35a Trace 108.83 ± 37.44b 12.43 ± 0.63a 19.87 ± 0.29b 19.97 ± 4.76b 17.28 ± 0.38a 22.00 ± 1.19ab 33.66 ± 11.08a 18.19 ± 0.30a 23.90 ± 1.46a 131.57 ± 16.0b 49.15 ± 11.67a 79.98 ± 30.11a 160.42 ± 43.53b 15.04 ± 0.79 Trace Trace a a 20.65 ± 0.42 23.23 ± 1.87 55.76 ± 8.73b 14.75 ± 0.34a 19.39 ± 1.37a 33.26 ± 8.36b Trace Trace 12.08 ± 2.48 ND Trace Trace a b 8.07 ± 0.09 10.54 ± 0.83 34.82 ± 0.54c a b 11.96 ± 0.28 19.55 ± 1.25 41.09 ± 12.65c 15.69 ± 0.44a 18.12 ± 1.49a 119.39 ± 43.56b 12.79 ± 0.60a 11.32 ± 1.03a 35.21 ± 7.21b 4.53 ± 3.04a 6.98 ± 1.87a 127.01 ± 28.38b

AC CE P

TE

D

MA

NU

SC

RI

Amino acid Fermented : Control Asp 1.51 Ser & Asn 7.18 Glu 19.64 Gly 3.73 His & Gln 15.26 Arg 1.00 Thr 1.53 Ala 5.50 Pro 2.01 Cys Tyr 2.40 Val 1.72 Met Lys Ile 3.30 Leu 2.47 Phe 5.20 Trp 3.11 NH3 18.18 Total amino acids 253.75 ± 19.11a 310.48 ± 46.33a 1309.57 ± 301.41b 4.22 a,b,c Different letters in the same row indicate significant differences at p ≤ 0.05. Values are the mean ± standard deviation of three independent replicates (n = 3). ND = not detected.

38

ACCEPTED MANUSCRIPT Highlights Okara (soybean residue) was fermented with yeast Yarrowia lipolytica



Fermentation increased the amounts of free amino acids and succinate



Antioxidant capacity was enhanced after fermentation



Fermentation eliminated the grassy off-odour and resulted in a cheese-like odour



Yarrowia-fermented okara can be a more nutritious, savoury food product.

AC CE P

TE

D

MA

NU

SC

RI

PT



39