Novel Saccharomyces cerevisiae × Saccharomyces mikatae Hybrids for Non-alcoholic Beer Production

: The popularity of non-alcoholic beers has been increasing over the past few years. Maltose-negative strains of different genera are frequently used to obtain beers of low alcohol content. S. cerevisiae hybrids with other Saccharomyces species offer interesting inherited ﬂavour characteristics; however, their use in non-alcoholic beer production is rare. In this work, we constructed six hybrids of maltose-negative S. cerevisiae parental strains (modiﬁed to produce higher amounts of organic acids) and S. mikatae (wild-type). Growth behaviour, osmotolerance and fermentation features of the offspring were compared with parental strains. One hybrid with mitochondrial DNA inherited from both parents was used to produce non-alcoholic beer in which organic metabolites were evaluated by HPLC and HS-SPME-GC-MS. This hybrid produced non-alcoholic beer ( ≤ 0.05% ( v / v )) with an increased organic acid content, just as its parent S. cerevisiae, but without producing increased amounts of acetic acid. The beer had a neutral aromatic proﬁle with no negative off-ﬂavours, similar to the beer produced by the parent S. mikatae , which was used for the ﬁrst time to produce non-alcoholic beer. Overall, both parents and hybrid yeast produced non-alcoholic beers with increased amounts of higher alcohols compared with esters.


Introduction
In recent years, increasing awareness about healthy nutrition and well-being has significantly affected trends in the food and beverage industry.The negative effect of alcohol on human health, together with regulations and religious beliefs, are shifting consumer's preferences toward non-alcoholic beverages [1,2].Among alcoholic beverages, beer holds a prominent share of the global market, and even though non-alcoholic beers are becoming increasingly popular, they still account for only a minor proportion of total beer products [1].The term "non-alcoholic" is associated with beers containing no more than 0.5% (v/v) ethanol [3], although worldwide, the limits may differ and be as low as 0.05% (v/v) [4].Non-alcoholic beers are traditionally produced either by arresting the fermentation process or by removing ethanol from the fermented beer [5,6].However, both of these methods have detrimental effects on the quality of the non-alcoholic beer; off-flavours and lack of aroma and complexity are only a few of the reasons why customers refrain from switching to non-alcoholic beers [4].
One of the strategies that has been extensively explored over the past decades uses maltose-negative yeast strains in beer fermentation [7].Their inability to ferment maltose results in limited ethanol production, while the metabolism of simple sugars is sufficient for the production of organic molecules that contribute to the beer's aromatic profile [8].However, the products are often associated with "worty" or "cooked" flavours and undesirable turbidity and viscosity [9].The use of naturally fruit-residing yeast strains with impaired maltose transport can significantly improve the aromatic profile of non-alcoholic beers, as they produce increased amounts of flavour-active compounds to attract insects [10].Today, research focuses mainly on the use of maltose-negative non-Saccharomyces yeasts that produce notable amounts of flavour-active compounds, flocculate easily and are consumer-safe [2].
The enormous advancement in molecular biology has opened the door to the production of non-alcoholic beers with desired flavours and properties through the employment of genetically modified yeast in beer fermentation [11].Even though the use of GMO is still frowned-upon by the general public, and has to be handled carefully, an increased number of studies has focused on the production of non-alcoholic beverages by genetically modified strains [12][13][14][15][16][17][18][19].Members of the monophyletic clade of Saccharomyces, known as the Saccharomyces sensu stricto group, are able to undergo hybridisation, which can shape the hybrid to withstand harsh environments [20].Moreover, alloploid strains that tend to be capable of producing viable diploid spores may be formed during hybridisation.Thus, allotetraploid inter-specific hybrids may undergo meiosis, enabling crossovers and gene conversions [21].
An excellent example of a natural inter-specific hybrid between S. cerevisiae and S. eubayanus is yeast known as S. pastorianus and is the powerhouse of the modern brewing industry [22].Bellon et al. [23] described the first inter-specific hybrid formed between a S. cerevisiae wine strain and S. mikatae, which introduced more flavour complexity into the produced wine.Another hybrid between S. mikatae and S. cerevisiae showed increased production of compounds yielding fruity, banana and floral aromas in white wines [24].However, there is limited information available about natural or synthetic hybrids between S. mikatae and S. cerevisiae [25,26].
This work provides a clearer insight into novel constructed yeast hybrids of maltosenegative S. cerevisiae parents (GMO) and S. mikatae (wild-type) targeted for non-alcoholic beer production.Maltose is the most abundant saccharide present in the beer wort [3], and here, we describe the application of S. mikatae for the first time in non-alcoholic beer production.Construction of hybrids and their use was led by the motivation to imitate the natural hybridisation event and reveal the influence of recombination on the inherited features of the offspring.

Microorganisms
Hybrid yeasts were constructed at the Faculty of Natural Sciences of the Comenius University in Bratislava, Slovakia.Modified deletion mutants of Saccharomyces cerevisiae ∆ACO1 G418 R , Saccharomyces cerevisiae ∆KGD1 G418 R and Saccharomyces cerevisiae ∆CIT1 G418 R were constructed by the replacement of wild-type genes with the kanMX4 allele [27] and were obtained from a systematic deletion project [28].Parental Saccharomyces mikatae CBS 8839T was obtained from the Central Bureau of Fungal Cultures in the Netherlands.Commercial brewer's strains Saccharomyces cerevisiae K97 and Saccharomyces pastorianus W34/70 were obtained from a local supplier.All yeasts used in this work are listed in Table 1 and primers are listed in Table 2.

The Construction and Verification of Hybrids
Prior to breeding, yeast cells were cultivated at 28 • C on YPDA medium.Ascospores were exposed to 50 µL of 0.5 mg mL −1 Zymolyase 20T (Amsbio, USA), incubated at 30 • C for 15-30 min and vortexed.After ascus disintegration, exposed spores were mixed with an equivalent number of cells of the opposite yeast strain and poured onto YPDA.Hybridisation or formation of zygotes was controlled microscopically.Yeasts were maintained on YPDA and YPDAG + medium at 4 • C. All yeasts (Table 1) were inoculated onto a selective MING + medium, on which only inter-species hybrids were able to grow.The parental prototrophic strain S. mikatae was G S (Geneticin sensitive), and parents of S. cerevisiae were G R (Geneticin resistant) and contained auxotrophic mutations (leu-, his-, ura-).Isolation of gDNA (genomic DNA) was performed in accordance with [30], and the mtDNA (mitochondrial DNA) was purified by differential centrifugation methods described in [31].Amplification of DNA was performed in a Mastercycler thermocycler (Eppendorf, Hauppauge, NY, USA).Template DNA was amplified in 25 µL of reaction solution: 12.3 µL of miliQ; 2.5 µL 10× B PCR buffer; 2.5 µL 2 mmol L −1 of dNTP; 2.5 µL of 25 mmol L −1 MgCl 2 ; 1 µL 25 pmol.µL −1 of forward primer; 1 µL 25 pmol.µL −1 of reverse primer; 0.2 µL Taq DNA polymerase (5 U) (FIREPOL ® ); 0.5 µL gDNA.The amplification quality was checked on a 1% (w/v) agarose gel by electrophoresis.2), PCR products were digested with HaeIII enzyme and analysed by electrophoresis on an agarose gel.mtDNA evaluation was performed according to [33] as follows: gDNA was cleaved by Hinf I (Takara Bio, USA)estriction en rzyme (5 G↓ANTC 3 ;3 CTNA↑G 5 ).The content of a 40 µL of Hinf I reaction mixture was: 19 µL of gDNA; 15 µL miliQ; 4 µL 10× buffer; 1 µL of RNase A (Serva, Germany) (10 mg.mL −1 ); 1 µL Hinf I.A total of 2 drops of paraffin oil were added to the mixture, and the solution was cultivated overnight at 37 • C. The restriction mixture was extracted with a mixture of phenol-chloroform and centrifuged (12,000× g, 15 min, 4 • C).A total of 20 µL of restriction mixture was added onto a 1% (w/v) agarose gel.The DNA was separated in 1% (w/v) agarose gel in TBE solution, which contained 0.5 µg mL −1 of ethidium bromide (Sigma Aldrich, USA).DNA was visualised under the UV light at 312 nm, and the lengths of DNA fragments were compared with a standard λ/PstI 300 ng ladder.

Yeast Starters
Yeast starters used in experiments were prepared in a 24 h submersed cultivation of individual yeast strains in 20 mL of liquid YPD medium in 100 mL Erlenmeyer flasks on an orbital shaker (Biosan ES-20, Riga, Latvia) at 2 Hz, 28 • C. Cell counting for each yeast strain was performed using a Leica DM 2500 light microscope (Leica Microsystems) with a Bürker chamber.

Sugar Fermentation
The ability of yeast strains to ferment various sugars (glucose, maltose, maltotriose, sucrose) was tested in glass tubes containing inverted Durham tubes.A total of 5 mL of saccharide solution with bromothymol blue as an acid/base indicator and 1 mol L −1 K 2 CO 3 as a buffering agent were inoculated with 10 6 cells mL −1 of liquid yeast starter.Production of CO 2 , indicating saccharide fermentation, was evaluated in Durham tubes following static cultivation at 25 • C for 10 days.The cultivations were performed in triplicate.

Osmotolerance
To determine sensitivity of strains to different conditions, 10 6 cells mL −1 of liquid yeast starter were cultivated at 20 • C for one week in sterile glass tubes, each containing 10 g L −1 of yeast extract and different concentrations of glucose (0; 1; 2; 5; 10; 30 and 50) in % (w/w).Osmotolerance was determined by A 600nm measurements, where each non-inoculated solution of a specific concentration was used as a blank against a medium of a specific concentration with inoculum.Experiments were performed in triplicate.

Aerobic Growth
Growth of yeast strains was evaluated using YPD medium.The cultivations were performed in 96-well microtiter plates (Sarstedt, Germany) filled with 200 µL of medium and inoculated with 10 6 cells mL −1 of liquid yeast starter.Plates were incubated at 25 • C and shaken (4 Hz) in a Varioskan ® Flash microplate reader (ThermoFisher Scientific, Waltham, MA, USA).Growth was monitored by measuring the absorbance at 600 nm (OD 600nm ) every hour for 2 days.

Phenotype verification of Saccharomyces cerevisiae Mutants
For phenotype verification of Saccharomyces cerevisiae mutant yeast (ACO1, KGD1 and CIT1), 8 • P and 10 • P worts were used.A total of 48 mL of wort in 50 mL fermentation flasks were inoculated with yeast at a cell concentration of 10 6 cells mL −1 .Flasks were sealed, and fermentation was performed at 12 • C for 7 days.Young beer samples were analysed for organic acid concentrations by HPLC.

Beer Production
For beer production, 480 mL of 7 • P wort in 500 mL fermentation PET flasks were inoculated with yeast starters to achieve a starting biomass concentration of 10 6 cells mL −1 .Flasks were closed and fermentation proceeded at 18 • C for one day and then at 3 • C for a month.Beer samples were analysed by HPLC (sugars, organic acids, ethanol and glycerol) and by HS-SPME-GC-MS analysis for volatile organic compounds (VOCs) (esters, higher alcohols).Each beer fermentation was performed in triplicate.

Analytical Methods
Ethanol concentration, original extract, real degree of fermentation, pH and colour of beer samples was determined using a density meter DMA 4500M coupled with an Alcolyzer Beer ME, Haze QC ME Turbidity Measuring Module and pH ME Beverage Measuring Module (Anton Paar, GmbH, Graz, Austria).

Analysis of Beer by HPLC-RID
Before analysis, the beer samples were centrifuged (10 min, 2511× g) and supernatants were diluted with deionised water.An Agilent 1260 HPLC system (Santa Clara, CA, USA) coupled to an RI (refractive index) detector using an Aminex HPX-87H column (300 mm, 7.8 mm; Bio-Rad Laboratories, Hercules, CA, USA) was used.Sulfuric acid (5 mmol L −1 ) was used as the mobile phase, with a flowrate of 0.6 mL min −1 .Separation was performed at 25 • C, and the injection volume was 20 µL.Signal detection was carried out using a refractive index detector (RID).Accurate concentrations of glucose, maltose, maltotriose, ethanol and glycero, and acetic, citric, malic and succinic acid were determined using the single standard addition method.Standards with purity ≥ 99% were obtained from Merck (Darmstadt, Germany).Each beer sample was analysed in triplicate.

Analysis of Beer by HS-SPME-GC-MS
Prior to analysis, beer samples were cooled and stored at 4 • C. A total of 50 mL of each beer sample were centrifuged (10 • C, 5054× g, 10 min), and the supernatant was poured into a 50 mL flask and closed.Flasks were shaken for 3 min to remove the CO 2 .In the meantime, 2 g of NaCl with (≥99.9% purity, Pentachemicals, Czech Republic) were put into 20 mL darkened vials together with 10 mL of beer sample and 100 µL of internal standard (IS) solution, which contained ethyl heptanoate (≥ 99% purity, Sigma Aldrich, DE) and 3-octanol (≥ 99% purity, Sigma Aldrich, USA).Each vial was vortexed for 30 s to dissolve the NaCl and homogenise the sample.Headspace solid-phase microextraction (HS-SPME) of volatile organic compounds (VOCs) was performed for 30 min at 50 • C using an 85 µm Carboxen/polydimethylsiloxane (CAR/PDMS) fibre (Supelco, USA).VOCs were separated and determined using an Agilent GC 6890N system (Agilent Technologies, Santa Clara, CA, USA) coupled to an Agilent 5975B single quadrupole mass spectrometer detector (MSD) (Agilent Technologies, USA) and to a headspace solid-phase microextraction (HS-SPME) autosampler (COMBI PAL CTC Analytics, Switzerland).Analytes were separated in a DB-624 capillary column (30 m × 0.25 mm × 1.40 µm) (Agilent Technologies, USA).VOCs were desorbed in the injector of the GC in splitless mode for 10 min, and the temperature was set at 260 • C. Helium gas was used at a flow rate of 1 mL min −1 .The initial temperature program of the oven was set at 30 • C and held for 10 min, followed by three ramps in which the gradient was 2 • C/min until reaching 52 • C and held for 2 min, then 2 • C/min to 65 • C and held for 2 min, and lastly, at 5 • C/min to 250 • C and held for 3 min.The ionisation energy was 70 eV, and detection and data acquisition were performed in a scan mode from 20 to 500 Da.

Statistical Methods
All fermentation experiments in this study were performed in triplicate.The statistical analysis of the chromatographic VOC profiles and the principal component analysis (PCA) were performed using Statistica 12 software (StatSoft, USA).The PCA was used to explain the differences between beers, and results were explained by visualisation of the first two principal components on the scatter plot.

Phenotype Verification of Saccharomyces cerevisiae Mutants
The purpose of this work was to evaluate the potential of S. cerevisiae hybrids to produce non-alcoholic beers.In order to verify the phenotype of deletion mutants of S. cerevisiae (ACO1, KGD1 and CIT1 [15]), trial fermentations with 8 • P and 10 • P worts were performed.As presumed, mutant strains produced increased amounts of organic acid compared to standard brewer's strains S. cerevisiae K97 and S. pastorianus W34/70 from both 8 • P and 10 • P worts (Table 3).However, all mutant strains of S. cerevisiae also produced increased amounts of acetic acid, which drastically influenced the beer flavour through its sharp, sour and vinegar-like taste if present above the threshold level of 200 mg L −1 [34].Values are represented as: Average ± Standard Deviation, "n"-the number of replications of each beer fermentation was 3 and the number of each beer analysis was 3. (n = 3 × 3 = 9), Note: n.d.-not detected

The Preparation of Yeast Hybrids
To confirm the correct hybridisation further, we analysed the HaeIII restriction polymorphism of the D1/D2 domain of the 26S rRNA gene with PCR.From single yeast colonies, we isolated the DNA, amplified it using NL1 and NL4 primers (Table 2) and cleaved the PCR products with HaeIII.DNA fragments were separated on an agarose gel.In the case of Saccharomyces cerevisiae mutant strains (ACO1, KGD1 and CIT1), we anticipated band lengths of 133 bp, 160 bp and 322 bp.In the case of the S. mikatae genome, band lengths of 160 bp and 457 bp were expected.All the hybrids had shown the mixed restriction profile (Figure 1) and were classified as true hybrids.
To confirm the correct hybridisation further, we analysed the HaeIII restrictio morphism of the D1/D2 domain of the 26S rRNA gene with PCR.From single yea nies, we isolated the DNA, amplified it using NL1 and NL4 primers (Table 2) and the PCR products with HaeIII.DNA fragments were separated on an agarose ge case of Saccharomyces cerevisiae mutant strains (ACO1, KGD1 and CIT1), we anti band lengths of 133 bp, 160 bp and 322 bp.In the case of the S. mikatae genom lengths of 160 bp and 457 bp were expected.All the hybrids had shown the m striction profile (Figure 1) and were classified as true hybrids.Inter-species hybrids from the group of Saccharomyces are commonly presen natural environment and are increasingly implemented in industrial productio best example of a naturally created hybrid, which is now commonly used in the p tion of lager beer, is S. pastorianus (S. cerevisiae × S. eubayanus).S. pastorianus inher tochondrial genes that are responsible for ethanol-and cryo-tolerant phenotypes The mitochondrial genome (mtDNA) of S. cerevisiae contains genes important for tion, encoding proteins for cytochrome c oxidase (cox1, cox2, cox3), apocytochrome three subunits of F0 ATP synthase complex (atp6, atp8, atp9) and one ribosomal rps3 [37].Generally, in yeast, the mitochondria are inherited biparentally [38].Du crossing of two haploids of S. cerevisiae, the mtDNA recombines, leading to hetero Inter-species hybrids from the group of Saccharomyces are commonly present in the natural environment and are increasingly implemented in industrial productions.The best example of a naturally created hybrid, which is now commonly used in the production of lager beer, is S. pastorianus (S. cerevisiae × S. eubayanus).S. pastorianus inherited mitochondrial genes that are responsible for ethanol-and cryo-tolerant phenotypes [35,36].The mitochondrial genome (mtDNA) of S. cerevisiae contains genes important for respiration, encoding proteins for cytochrome c oxidase (cox1, cox2, cox3), apocytochrome b (cob), three subunits of F 0 ATP synthase complex (atp6, atp8, atp9) and one ribosomal subunit rps3 [37].Generally, in yeast, the mitochondria are inherited biparentally [38].During the crossing of two haploids of S. cerevisiae, the mtDNA recombines, leading to heteroplasmy lasting for up to 20 generations.After that, cells achieve homoplasmy and retain either parental mitochondrial genomes or the recombined ones [39].Mimicking natural hybridisation (by crossing mutant S. cerevisiae and S. mikatae) could generate offspring with beneficial features for the brewing industry (increased organic acid [15] and ester production [24]).
To reveal which forms of mtDNA the hybrids inherited, total genomic DNA was cleaved with Hinf I, and DNA fragments were separated on an agarose gel (Figure 2).Most of the hybrids inherited the mtDNA from parental S. mikatae (lines 1, 3, 4 and 6, Figure 2).However, one hybrid Hyb2ACO1 (line 2, Figure 2) received the mtDNA from S. cerevisiae.Results also showed that hybrid Hyb1CIT1 (line 5, Figure 2) possessed the recombinant mtDNA from both parental strains.Due to this interesting feature, the Hyb1CIT1 hybrid was studied in terms of the production of non-alcoholic beer from a 7 • P wort.
To reveal which forms of mtDNA the hybrids inherited, total genomic DNA was cleaved with HinfI, and DNA fragments were separated on an agarose gel (Figure 2).Most of the hybrids inherited the mtDNA from parental S. mikatae (lines 1, 3, 4 and 6, Figure 2).However, one hybrid Hyb2ACO1 (line 2, Figure 2) received the mtDNA from S. cerevisiae.Results also showed that hybrid Hyb1CIT1 (line 5, Figure 2) possessed the recombinant mtDNA from both parental strains.Due to this interesting feature, the Hyb1CIT1 hybrid was studied in terms of the production of non-alcoholic beer from a 7°P wort.

Saccharide Fermentation Tests
As maltose contributes to the majority of fermentable sugars in wort [40], yeast with maltose or maltotriose-negative phenotypes are frequently used to produce non-alcoholic beer [2].Only of few species of Saccharomyces do not ferment maltose; therefore, we chose maltose-negative strains of S. cerevisiae and S. mikatae as parental strains for the hybridisation.Three different parental S. cerevisiae strains with mutations/deletions in genes encoding enzymes involved in the Krebs cycle (ACO1, aconitase; KGD1, α-ketoglutarate dehydrogenase; and CIT1, citrate synthase) were used in this work.Screening of the hybrids on YPD and MIN agar plates containing 200 µg mL −1 Geneticin TM confirmed the successful insertion of the kanMX4 allele, as all the constructed hybrids were able to grow in the presence of the antibiotic.The fermentative abilities of the constructed hybrids were tested (Table 4), and we showed that all the strains were able to ferment glucose and sucrose but were unable to ferment maltose and maltotriose, which emphasises their potential in the production of non-alcoholic beer.

Saccharide Fermentation Tests
As maltose contributes to the majority of fermentable sugars in wort [40], yeast with maltose or maltotriose-negative phenotypes are frequently used to produce nonalcoholic beer [2].Only of few species of Saccharomyces do not ferment maltose; therefore, we chose maltose-negative strains of S. cerevisiae and S. mikatae as parental strains for the hybridisation.Three different parental S. cerevisiae strains with mutations/deletions in genes encoding enzymes involved in the Krebs cycle (ACO1, aconitase; KGD1, αketoglutarate dehydrogenase; and CIT1, citrate synthase) were used in this work.Screening of the hybrids on YPD and MIN agar plates containing 200 µg mL −1 Geneticin TM confirmed the successful insertion of the kanMX4 allele, as all the constructed hybrids were able to grow in the presence of the antibiotic.The fermentative abilities of the constructed hybrids were tested (Table 4), and we showed that all the strains were able to ferment glucose and sucrose but were unable to ferment maltose and maltotriose, which emphasises their potential in the production of non-alcoholic beer.

Osmotolerance
Another important parameter for beer production strains is osmotolerance.Overall, the yeasts (parental and hybrid strains) were able to handle up to 10% (w/w) of glucose in the medium (Figure 3), The cell density, however, rapidly decreased with increasing glucose concentration.It is known that hyperosmotic stress (caused by high concentrations of saccharides or ethanol) causes rapid shrinking of the yeast cells and activates the highosmolarity glycerol response pathway [41].Mutant S. cerevisiae parents as well as S. mikatae parents were able to withstand 30% (w/w) glucose, which is notably higher than the ability of their hybrid offspring.

Osmotolerance
Another important parameter for beer production strains is osmotolerance.Overall, the yeasts (parental and hybrid strains) were able to handle up to 10% (w/w) of glucose in the medium (Figure 3), The cell density, however, rapidly decreased with increasing glucose concentration.It is known that hyperosmotic stress (caused by high concentrations of saccharides or ethanol) causes rapid shrinking of the yeast cells and activates the highosmolarity glycerol response pathway [41].Mutant S. cerevisiae parents as well as S. mikatae parents were able to withstand 30% (w/w) glucose, which is notably higher than the ability of their hybrid offspring.

Aerobic Growth
No significant differences were detected among the hybrids in terms of aerobic growth (Figure 4).Overall, growth curves (Figure 4) of all hybrids were similar and displayed lower absorbances, representing lower growth rates than those of parental strains S. mikatae and S. cerevisiae.Hybridisation that brings alleles together in recombination may result in unpredictable results, with the outcome of reduced hybrid fertility, viability or even growth rate, but in some cases can also boost hybrid fitness [42].

Aerobic Growth
No significant differences were detected among the hybrids in terms of aerobic growth (Figure 4).Overall, growth curves (Figure 4) of all hybrids were similar and displayed lower absorbances, representing lower growth rates than those of parental strains S. mikatae and S. cerevisiae.Hybridisation that brings alleles together in recombination may result in unpredictable results, with the outcome of reduced hybrid fertility, viability or even growth rate, but in some cases can also boost hybrid fitness [42].

Beer Production
In order to investigate the fermentation performance of the production strains (Table 1) in a real setting, a 7°P wort was used for the production of beer using the parental and hybrid strains.We picked the hybrid Hyb1CIT1 with recombinant mtDNA inherited from

Beer Production
In order to investigate the fermentation performance of the production strains (Table 1) in a real setting, a 7 • P wort was used for the production of beer using the parental and hybrid strains.We picked the hybrid Hyb1CIT1 with recombinant mtDNA inherited from both parental strains (Figure 2) to test the influence of hybridisation on the production of non-alcoholic beer.Firstly, we prepared 7 • P wort, which contains a lower number of fermentable sugars than 8 • P and 10 • P worts, in order to limit the final ethanol concentration.The results shown in Table 5 prove that we successfully obtained three non-alcoholic beers with ethanol concentrations lower than 0.05% (v/v), which was very promising.Values are represented as: Average ± Standard Deviation, "n": the number of replications of each beer fermentation was 3 and the number of each beer analysis was 3.
Glucose, acting as the sole fermentable sugar in the 7 • P wort, was present at a concentration of 4.50 ± 0.07 g L −1 (Table 6) and served as the only potential carbon source for yeast metabolism.Glycerol formation is a wasteful process in terms of metabolism; however, its importance resides in osmoregulation of the cell and low-temperature tolerance, and it also plays a role as a sink for reducing equivalents [43].The glycerol concentration in alcoholic beer is typically between 1-2 g L −1 [35], and its formation can be influenced by aeration, temperature, pH, sugar concentration and sulphite content [43].In a study of non-alcoholic beer production using non-Saccharomyces yeasts [44], the concentration of glycerol in beers ranged from 0.4 to 1.0 g L −1 .In this study, glycerol concentration in three prepared non-alcoholic beers was no more than 0.65 g L −1 (Table 6).Values are represented as: Average ± Standard Deviation, "n": the number of replications of each beer fermentation was 3 and the number of each beer analysis was 3. (n = 3 × 3 = 9) Note: "n.d." = not detected.
Formation of organic acids in beers is related to the metabolic pathways (glycolysis and the Krebs cycle) [45].These compounds can impart beer body with fruitiness and sour taste and influence the beer foam and microbial stability [45,46].A comparison of organic acids in beer samples (Table 6) revealed that concentrations did not differ dramatically, which also corresponds to pH values detected in the beers (Table 5).However, mutant S. cerevisiae CIT1 produced acetic acid (Table 6), which is considered an off-flavour in beer [47] and was slightly above its threshold level of 200 mg L −1 [34].It is important to note that hybrid Hyb1CIT1 did not produce acetic acid, which eliminates the problem with negative off-flavours introduced to the final product by its parental strain.

Production of Volatile Organic Compounds (VOCs)
The esters present in beer are classified as acetate and medium-chain fatty acid ethyl esters.The key esters that implement a fruity aroma into the beverage are ethyl acetate, isoamyl acetate, isobutyl acetate, phenylethyl acetate, ethyl hexanoate and ethyl octanoate [48].These compounds are formed during the beer fermentation process by yeast metabolism and diffuse from the yeast cell to the medium [49].Concentrations of esters in beer can be affected by hydrostatic pressure, wort aeration and its composition, and lastly, by altering the fermentation conditions [50].However, their concentration in non-alcoholic beer is limited due to the maltose-negative phenotype of the strain used [51].Our results showed that ethyl acetate was the most abundant ester present in the product.Beer prepared with a hybrid yeast (Hyb1CIT1) contained 0.262 ± 0.039 mg L −1 of ethyl acetate (fruity and sweet aroma), which is similar to the concentration produced by the parental strains (Table 7).Values are represented as: Average ± Standard Deviation, "n": the number of replications of each beer fermentation was 3 and the number of each beer analysis was 3. (n = 3 × 3 = 9.) Note: "n.d." = not detected.
In the study by Catallo et al. [22], a hybrid cross between S. cerevisiae and S. eubayanus produced increased amounts of 3-methylbutyl acetate, ethyl acetate and ethyl hexanoate.The hybrids of S. jurei were also studied, and authors prepared beers with an interesting fruity profile [52], which supports the use of novel Saccharomyces hybrids in brewing.In this study, the banana flavour-inducing ester, iso-amyl acetate, with its threshold levels of 1.4-1.6 mg L −1 [4] and rose ester, 2-phenylethyl acetate, with a typical threshold level of 0.05-2.0mg L −1 [53] were detected in beers in negligible concentrations (Table 7).From the so-called "apple" esters, ethyl hexanoate and ethyl octanoate [54], whose formation is directly connected to ethanol formation [49], ethyl hexanoate was detected in all three beers at similar concentrations (Table 7) but was slightly above the threshold level of 0.2 mg L −1 [55].However, its sour apple aroma was not perceivable due to the sweet aroma caused by the residual maltose and maltotriose (Table 6).
Higher alcohols notably impact flavour and also act as the precursors for ester formation [56].They are closely related to protein synthesis and amino acid assimilation [50], and their final amount in beer can be altered by fermentation temperature [57].Increased amounts of higher alcohols compared to other VOCs were detected in all three beers (Table 7).The beer fermented with Hyb1CIT1 contained the highest amounts of 3-methyl-1-butanol, 2-methyl-1-propanol and 2-methyl-1-butanol among the other beers; however, their concentrations were far below the threshold levels [55].Comparing the results with a similar study of the inter-specific hybrid of S. mikatae and commercial wine fermenting S. cerevisiae, hybrids were able to form greater amounts of higher alcohols and esters than in this work [23].Even though the hybrids of S. mikatae are known to contribute to the aromatic profile of wine, the level of volatile compounds is strictly dependent on the number of fermentable sugars, which is limited during the production of non-alcoholic beers [23].The formation of 4-vinylguaiacol from ferulic acid during boiling or yeast fermentation is often related to an unpleasant smoky/bacon aroma [58].Its threshold concentration in beer is 0.2-0.3mg L −1 [35] and was slightly exceeded in this study in all three non-alcoholic beers (Table 7) but had no impact on the beer flavour.

Principal Component Analysis
The scatter plot (Figure 5a) with principal components 1 and 2 revealed that 47.08% of the variation in the data making up the beer VOC's profile (Table 7) is explained by first two principal components.Beers fermented with S. mikatae (Sm) are situated on the lower left side of the scatter plot; the beers CIT1 are situated on the right lower side of the scatter plot.The beers fermented with hybrid (Hyb1CIT1) are situated on the upper right and left side of the scatter plot.We can claim that the real hybrid inherited both parental features from VOCs profiles, but as we can see in Table 7, no marked differences were noted among the VOC profiles of the non-alcoholic beers.However, Hyb1CIT1 beers were characterised by higher amounts of higher alcohols, namely, 3-methyl-1-butanol, 2-methyl-1-propanol and 2-methyl-1-butanol.This analysis provided us with information about novel hybrids and fermentation similarities with parental strains of the maltose-negative mutant S. cerevisiae and wild-type S. mikatae.The scatter plot (Figure 5, a) with principal components 1 and 2 revealed that 47.08% of the variation in the data making up the beer VOC's profile (Table 7) is explained by first two principal components.Beers fermented with S. mikatae (Sm) are situated on the lower left side of the scatter plot; the beers CIT1 are situated on the right lower side of the scatter plot.The beers fermented with hybrid (Hyb1CIT1) are situated on the upper right and left side of the scatter plot.We can claim that the real hybrid inherited both parental features from VOCs profiles, but as we can see in Table 7, no marked differences were noted among the VOC profiles of the non-alcoholic beers.However, Hyb1CIT1 beers were characterised by higher amounts of higher alcohols, namely, 3-methyl-1-butanol, 2-methyl-1propanol and 2-methyl-1-butanol.This analysis provided us with information about novel hybrids and fermentation similarities with parental strains of the maltose-negative mutant S. cerevisiae and wild-type S. mikatae.   1) used in beer production.

Conclusions
This work provided a clearer insight into the breeding of hybrid Saccharomyces species crossed between mutant strains of S. cerevisiae, with higher organic acid production, and wild-type S. mikatae targeted for non-alcoholic beer production.Both parental strains as well as six hybrid strains were tested for osmotolerance and ability to propagate in 30% (w/w) glucose solution.Almost identical growth curves of maltose-negative hybrids were obtained.However, in comparison with parent strains, lower absorbance (A600nm) values were obtained for the hybrids.PCR analysis confirmed that one hybrid, Hyb1CIT1, inherited recombinant mtDNA from both parents and was highlighted for further use.Final beer production by fermenting 7°P wort with this hybrid and its parents was conducted, and the organoleptic profile of beers was determined.We successfully produced non-alcoholic beers with ≤0.05% (v/v) of ethanol using both hybrid as well as parental strains.The beer produced with hybrid Hyb1CIT1 contained a similar content of organic acids as  1) used in beer production.

Conclusions
This work provided a clearer insight into the breeding of hybrid Saccharomyces species crossed between mutant strains of S. cerevisiae, with higher organic acid production, and wild-type S. mikatae targeted for non-alcoholic beer production.Both parental strains as well as six hybrid strains were tested for osmotolerance and ability to propagate in 30% (w/w) glucose solution.Almost identical growth curves of maltose-negative hybrids were obtained.However, in comparison with parent strains, lower absorbance (A 600nm ) were obtained for the hybrids.PCR analysis confirmed that one hybrid, Hyb1CIT1, inherited recombinant mtDNA from both parents and was highlighted for further use.Final beer production by fermenting 7 • P wort with this hybrid and its parents was conducted, and the organoleptic profile of beers was determined.We successfully produced nonalcoholic beers with ≤0.05% (v/v) of ethanol using both hybrid as well as parental strains.The beer produced with hybrid Hyb1CIT1 contained a similar content of organic acids as the beer produced with the parental strain S. cerevisiae CIT1, but the production of an undesirable amount of acetic acid was successfully eliminated.The aromatic profile of beer fermented with hybrid yeast Hyb1CIT1 was neutral, with no negative off-flavours.Even though the hybrid Hyb1CIT1 was able to produce slightly higher concentrations of 3-methyl-1-butanol, 2-methyl-propanol and 2-methyl-1-butanol than the parental strains S. mikatae and S. cerevisiae, these concentrations were negligible in terms of flavour.The data for volatile organic compounds were analyzed by principal component analysis, which confirmed no significant differences between parental strains and their hybrid offspring.This work could be of help in understanding the approach of specific hybrid breeding targeted for non-alcoholic beer production.The maltose-negative parental strain S. mikatae, which was used for non-alcoholic beer production for the first time, implemented a neutral aromatic profile into the beer and could be further used in co-fermentations with other flavour-inducing yeast strains.

Fermentation 2023, 9 , 16 Figure 4 .
Figure 4. Aerobic growth of parental and hybrid strains in liquid 2% (w/w) YPD medium at 25 °C for 48 h.Each curve point represents an average value of 3 replications.

Figure 4 .
Figure 4. Aerobic growth of parental and hybrid strains in liquid 2% (w/w) YPD medium at 25 • C for 48 h.Each curve point represents an average value of 3 replications.

Figure 5 .
Figure 5. Scatter plot (a) and loading plot (b) of the principal component analysis (PCA) of VOCs profiles of beer samples.Beer sample abbreviations (Hyb1CIT1, Sm and CIT1) correspond to yeast abbreviations (Table1) used in beer production.

Figure 5 .
Figure 5. Scatter plot (a) and loading plot (b) of the principal component analysis (PCA) of VOCs profiles of beer samples.Beer sample abbreviations (Hyb1CIT1, Sm and CIT1) correspond to yeast abbreviations (Table1) used in beer production.

Table 1 .
List of hybrids, their parents and brewer s controls used in this work.

Table 4 .
The fermentation of saccharides of the constructed yeasts hybrids and standard brewer's yeasts using Durham tubes at 25 • C for 7 days.

Table 5 .
Basic parameters of final beers (Sm, CIT1 and Hyb1CIT1) produced from 7 • P wort in 500 mL.