Results and Discussion

Results and Discussion

Isolation and screening of ethanologenic wild yeast strains

Fruits serve as microhabitats for a variety of yeast species. Therefore bioprospecting of yeasts from fruits is advantageous [32]. Yeasts isolated from 100 samples (98 fruit sources and two fermented products) were screened in the current study. Colonies were observed in 89 samples, and colony morphology was recorded by incubating the strains at 35^oC for 24h. Colony morphology of isolated yeast presented elongated (24.72%), oval (14.61%), rounded (60.67%) shape, Large (6.7%), medium (23.6%), or small (69.66%) size, cream (39.33%) or white (57.3%) or yellow (3.37%), irregular (48.31%) or regular (51.69%) borders, bright (49.44%) or opaque (22.47%) or smooth (28.09%) texture.

Yeast exhibit variation in the ability to ferment and assimilate various sugars, which also aids in the identification of yeast than morphological and physiological characteristics. Certain genera of yeast such as Saccharomyces, Torulaspora, and Zygosaccharomyces ferment glucose readily, whereas; Lipomyces and Sterigmatomyces are strictly non-fermentative genera of the yeast. In this study, it was seen that about 74.16% and 71.91% of yeast strains readily fermented sucrose and glucose, respectively. Glucose as a carbon source allows faster growth within 24h, which is due to the presence of 20 different glucose transporters in their plasma membrane [32]. Sucrose, a disaccharide, is assimilated extracellularly by secreting enzyme invertase. Lactose was fermented by 57.3% yeast strains, galactose by 55.06% yeast strains, and lowest being maltose by 41.57% yeast. Production of biofuels from second and third-generation biomass encourages isolation of yeast strains capable of fermenting xylose. About 69.66% of yeast strains fermented xylose in the current study. In a similar study[32], 45 yeast strains were isolated from fruits and chicken litter and observed that yeast strains readily fermented glucose (22%), sucrose (12.5%), lactose (2.1%), xylose (40%), galactose (8.3%) and maltose (2.1%). Yeast strains were tested in glucose and xylose medium as these sugars are major constituents across the taxonomical groups of green macroalgae. Strains with the highest biomass in glucose media (>0.5 OD₆₀₀) and xylose media (>0.1 OD₆₀₀) were considered for further investigations (Supplementary, S1). The screening was done using the phenotypic microarray method by eliminating strains with lower redox signal intensity (RSI) in glucose medium [37]. About 40.45% of yeast strains exhibited good growth in glucose media with the biomass >0.5 OD, and 47.19% of yeast strains obtained biomass > 0.1 OD in xylose medium. Yeast strain with full gas production in Durham’s fermentation tube were selected. About 19 yeast strains (CY, TY, CHY, MY, MFY, GVY, TNY, PLY FBY2, BAY, FBJY, RJY, GWY, CKY, PWY, WTY, YKY, POY, CUIY) were screened down for further characterization and fermentation capabilities. A growth curve study was carried out for the strains, incubated at 35^oC for 24h with samples drawn at every 1h interval and biomass growth observed at 600nm. Strains exhibiting a more prolonged exponential phase (Supplementary, S1) were selected, as it is a proxy for higher ethanol production as most of the primary metabolites are formed during this phase [38, 39].

Screening of multi-stress tolerant ethanologenic yeast strain

Ethanol endurance is an important property that decides the fermentation efficiency of the yeast strain. Ethanol tolerance of yeast has been determined as the accumulation of ethanol during the fermentation process, and ethanol is toxic to yeast organisms as it inhibits the activity of crucial glycolytic enzymes involved in ethanol production and hinders amino acid and glucose transport leading to the loss of cell viability and inhibition of cell growth [34, 35]. Isolated yeast growth was recorded in a spectrophotometer (600nm) in terms of turbidity at different ethanol concentrations. CHY had the highest ethanol tolerance, up to 10%, followed by CY, TY, and MY, after which the growth decreased (Supplementary, S2). In a similar study, yeast strains CHY1011 and CHFY0901 belonging to Saccharomyces genera exhibited ethanol tolerance up to 5% [42], comparable tolerance level was recorded for baker’s yeast (positive control) in the current study whereas, an isolated yeast (Y-1) from wine (Jiuqu) had ethanol tolerance up to 14% [40]. S.cerevisiae isolated from Nuruk [43], cashew [44], and soil [45] exhibited tolerance in a medium containing 15% alcohol.

Yeast growth at different temperatures were monitored. Thermo-tolerance of yeast is evident up to 37^oC, and growth is inhibited at higher temperatures. PLY strain exhibited consistent high growth up to 40^oC, whereas TY strain exhibited higher thermotolerance up to 45^oC. In a similar study earlier [46], Pichia kudriavzevii DMKU 3-ET15 isolated from fermented pork sausage displayed thermo-tolerance up to 45 ^oC, which is comparable to TY strain (Pichia kudriavzevii) in the current study (Supplementary, S3). Candida tropicalis, Pichia kudriavzevii, Candida orthopsilosis, Candida glabrata and Kodamea ohmeri were reported as thermotolerant, and high ethanol-producing yeasts strains [47]. Pichia caribbica isolated from ripe banana was subjected to different temperatures, displayed good growth at temperatures 28, 30, 40 ^oC, which declined at 45^oC [48], contrary to which CY strain (identified as P.carribbica) exhibited good growth at 35, 40 and 45 ^oC, which declined after 45 ^oC.

Bioethanol process from seaweeds encounters a high concentration of NaCl due to its habitat [49]. Therefore, isolated yeast strains were subjected to different salt concentrations, which show a decline of cell growth with an increase in salt concentration, similar to S.cerevisiae KCTC 1126. But, S.cerevisiae KCTC 1126 adapted to NaCl and yielded an ethanol concentration of 0.48 [50]. Issatchenkia orientalis MF-121 produced 2.9% (w/v) ethanol in a medium containing Na₂SO₄ (50g/L), while tolerant to multi-stress factors such as temperature, ethanol, and salt [51]. Halotolerance up to 14% NaCl concentration was exhibited by WTY strain, however, the strain achieved a lower ethanol conversion efficiency of glucose and xylose fermentation (Supplementary, S4).

Multivariate analyses through Principal Component Analysis (PCA), given in Fig. 1a, illustrate strains clustering into three groups with the clustering of temperature tolerant strains and overlapping of ethanol and salt-tolerant strains. Principal components are accounting for 56.5% of the total variance, with PC1 contributing 38.8% and PC2 contributing 17.7%. Salt tolerant strains POY, YKY, CUIY, and PWY were located at the positive side along PC1; on the other hand, thermotolerant species CY, TY MY, MFY and CHY were closely loaded at the negative side along PC1. Ethanol-tolerant strains were loaded at the positive side of PC2.

Ethanol production by yeast strains using a synthetic medium

Fermentation of glucose is an established technology; however, fermentation of xylose has been posing challenges. During the fermentation process, 70% of the sugar is converted to ethanol, whereas 20% assimilated by the yeast cells yield glycerol, organic acids, etc. [52, 53]. Production of glycerol at a higher concentration inside the yeast cell is stimulated by factors such as higher pH, a lower flux of pyruvate (due to the utilization of glycolytic intermediates), increase in osmotic pressure, etc. with the formation of by-products (higher alcohols and organic acids at lower level), affecting the ethanol yield as the growth of yeast cells invariably directs the glycolytic intermediates to corresponding pathways. Ethanologenic yeast strains are evaluated based on the ability of strains to utilize all sugars (glucose, xylose, galactose, mannose, rhamnose, and arabinose) and convert to ethanol with minimal by-product formation [54]. Conventionally, ethanol yield at an industrial scale is calculated based on the total sugar-fed into the fermentation system, and 90-93% ethanol bioconversion is considered for an efficient ethanologenic strain.

Fig. 1a. PCA score plot of yeast strains tolerant to stress factors; temperature, ethanol and salt concentration

Fig. 1b. Multivariate cluster analysis of yeast strains

Glucose is abundantly found sugar in the feedstock and is readily fermented by yeast microorganisms. Yeast prefers glucose over xylose, and xylose uptake is regulated by glucose concentration [55, 56]. Glucose is metabolized in a series of enzyme catalysed reaction process called glycolysis, to yield two molecules of three carbon compound pyruvate, under hypoxia or anaerobic condition, pyruvate is decarboxylated and acetaldehyde is reduced to ethanol through alcohol dehydrogenase [57]. Xylose is converted to Xylulose and phosphorylated to Xylulose-5-phosphate and further metabolized to glyceraldehyde-3-phosphate, and fructose-6-phosphate, which then enters the glycolysis pathway for subsequent pyruvate and ethanol production [58]. It was seen that about 60-80% of glucose was assimilated and fermented by yeast strains within 24 h except for CY and TY strains, which consumed less glucose but achieved a higher conversion efficiency of 83% and 94%, respectively, compared to other yeast strains. The highest ethanol concentration of 5.04 g/L was recorded for MY strain with 65.3% conversion efficiency. The least ethanol concentration was recorded for PLY, CKY, and FBJY strains.

Xylose is the main component (1/3) of lignocellulosic biomass, and xylose is not fermented by Saccharomyces cerevisiae due to lack of transport system. Yeast species capable of fermenting xylose belong to the genera Brettanomyces, Candida, Claviospora, Kluyveromyces, Pachyysolen, Pichia, and Schizosacchzromyces. Among which Candida shehatea, Pachysolen tannophilus, and Pichia stipitis ferment xylose at high concentrations [58], and studies are being carried out to isolate xylose-fermenting yeast strains [59–61]. In the current study (Table 1), a higher ethanol concentration of 28.2% was recorded for GVY followed by CY strain with 15.3% conversion efficiency indicating the ability to ferment xylose sugar. GVY strain produced lower conversion efficiency in glucose medium. Co-fermentation using two different yeast species has been carried out to efficiently utilize both xylose and glucose [19, 62]. Co-fermentation of sugars present in lignocellulose biomass was carried out using P.stipitis and K.marxianus and P. stiptis, and S.cerevisiae yielded 31.87g/L and 29.45g/L of ethanol [15]. In this study, co-fermentation of glucose and xylose using yeast strain was carried out. Acid hydrolysis of biomass releases mixed sugars into the medium, and therefore mixed sugar fermentation was carried out, and CY strain produced 1.83 g/L of ethanol, achieving 27% efficiency, whereas GVY strain produced 1.8g/L of ethanol achieving 20.6% conversion efficiency. Ethanol production was affected in other yeast strains due to xylose, hence the lower ethanol yield despite the presence of glucose. Studies report that in order to improve xylose fermentation, isolated yeast strain was grown on xylose rich media, and efficacy of yeast strain in ethanol production was evaluated by supplementing xylose in concentration of 10g/L or 25g/L [63].

Ethanol production by prioritized strain using macroalgal sugars

Fermentation of macroalgal hydrolysate was carried out to validate the potential of isolated wild yeast strains for selection of ethanol production. Hydrolysate of E. intestinalis and U. lactuca obtained from acid hydrolysis were neutralized and subjected to fermentation using each of the screened yeast individually at 35^oC for 24h on an orbital shaker with 100 rpm. Reducing sugar of Ulva and Enteromorpha are illustrated in Table 1.

Fermentation progress was determined by measuring the reducing sugar (after the fermentation process) and comparing it with the theoretical yield (51% of fermented sugar) by estimating the sugar conversion efficiency of each strain, which is detailed in Table 2. A higher conversion efficiency of 49.4% was obtained for CY strain for E. intestinalis followed by TY strain 42.9%. A similar study was carried out using wild yeast strains S.cerevisiae Y12 and YPS128, derived from clean lineages with no alternations to their genome due to human interventions or domestication [37, 64], and this strain fermented the hydrolysate of U.lactuca producing 7g/L of ethanol. A multi-tolerant strain of six Saccharomyces strains was selected and utilized for fermentation of lignocellulose hydrolysate. This study indicates that the natural strains outcompeted other strains for specific traits. Yeast strains isolated from the natural environment have the potential for bioethanol production and superior to industrial strains obtained by tweaking the strain through breeding, experimental evolution, or genetic engineering [34].

Free amino acid nitrogen (FAN) content in green seaweeds is > 0.15 g/L, and this avoids the need to supply additional nitrogen sources during fermentation [37]. The FAN required for yeast growth during fermentation and metabolism is 0.15g/L. Hence, fermentation of seaweed hydrolysate was carried out without the addition of nitrogen sources. Multivariate cluster analysis was performed by selecting biochemical compositions of fruit sources (carbohydrate, protein, fat, dietary fibers, vitamins, moisture content and minerals) as independent variables and ethanol production from synthetic sugar as a dependent variable. Fig. 1(b) illustrates the clustering of strains CY and TY from the rest of the isolated strains indicating its unique properties with the higher performance capabilities. These two strains achieved higher biomass, longer exponential growth, maximum conversion efficiency concerning glucose fermentation, and exhibited temperature tolerance. Based on these criteria, strains CY and TY were prioritized for fermentation of macroalgal sugar.

Yeast Identification

The identity of the prioritized yeast strains CY was confirmed as Meyerozyma (Pichia) caribbica and TY as Pichia kudriavzevii based on 16S rRNA nucleotide sequences homology match within the NCBI GenBank (Supplementary, S5). The yeast cells were stained with methylene blue and observed under an Olympus BX-51 bright field, phase contrast microscope, live cells reduce the dye (methylene blue) [65] and remain colorless, whereas dead cells retain the color and are stained blue (Fig 2 a-d). Pichia kudriavzevii cells are oval or ellipsoidal to elongate in the study. P.kudriavzevii is a thermo-tolerant yeast strain isolated from fruits and food sources. In contrast, Meyerozyma caribbica are isolated from fermented beverages having the capabilities to ferment xylose with high efficiency [66], which was observed in this study as well.

Fig 2a. Microscopic image and Scanning electron micrographs of (CY) Meyerozyma (Pichia) caribbica	Fig 2b. Microscopic image and Scanning electron micrographs of (TY) Pichia kudriavzevii
Fig 3. (a) Effect of different concentration of salt on CY strain growth and ethanol production (RS : Reducing sugar)	Fig 3. (b) Effect of different concentration of salt on TY strain growth and ethanol production (RS : Reducing sugar)

Fermentation of macroalgal sugar using prioritized yeast strains

Separate Hydrolysis (acid) and Fermentation (SHF)

Reducing sugar from E. intestinalis and U. lactuca was obtained by using 0.7N and 0.5N H₂SO₄ concentration. About 22.4% and 19.2% sugar conversion with respect to biomass were achieved for E. intestinalis and U. lactuca, respectively (Table 4). The acid hydrolysate obtained was subjected to fermentation using prioritized yeast strains CY and TY in different combinations at 35^oC, 100 rpm for 24h. Fermentation of E. intestinalis hydrolysate using CY and TY produced 0.14g/L and 0.16 g/L of ethanol with fermentation efficiencies of 46.9% and 51.8%, respectively. Co-fermentation of E. intestinalis hydrolysate using CY and TY yielded lower fermentation efficiency of 33%. Candida sp. isolated from marine fermented red algae, Kappaphycus alvarezii acid hydrolysate achieving 50% fermentation efficiency [27]. E.intestinalis subjected to SHF produced 8.6 g/L of ethanol with 30% conversion efficiency within 48h (Cho et al., 2013). SHF of K.alvarezii [67] and G.amansii [68] yielded 0.25g and 3.33 g of ethanol achieving 55.9% and 74.7% efficiency respectively. Acid hydrolysis (1% v/v, H₂SO₄ for 90 min) of sugarcane bagasse pith were subjected to fermentation obtaining 2.58 g/L of ethanol in 30h. Fermentation time for lignocellulose biomass is longer than macroalgal biomass due to the presence of complex polysaccharide lignin. Fermentation of U. lactuca hydrolysate yielded lower ethanol of 0.04g/L and 0.05g/L for both CY and TY strain with fermentation efficiencies 24 % and 48.7% respectively. Bioethanol has been obtained from all the three types of algae, however appropriate microorganism is yet to be isolated which consumes pentose sugar and mixed sugar [69]. Lower ethanol yield in this study can be attributed to inhibitor formation during acid hydrolysis.

Simultaneous Saccharification and Fermentation (SSF)

Acid pre-treated macroalgal biomass was subjected to enzyme hydrolysis using enzyme extracted from V. parahaemolyticus [14] and subjected to the subsequent fermentation. SSF of E. intestinalis and U. lactuca using CY strain produced ethanol of 0.12g/L and 0.08g/L achieving conversion efficiencies of 45.5% and 48.7% (Table 5). Higher conversion efficiency of 80.9% was achieved for U. lactuca followed by 65.2% for E. intestinalis biomass using TY yeast strain indicating its thermo-tolerance capabilities. Similarly, Kluyveromyces marxianus was recognized as a safe (GRAS) thermo-tolerant yeast strain with tolerance range of 38-45oC and producing high ethanol concentration SSF process [66].

Higher sugar conversion efficiency by these non-domesticated (“wild”) strains Pichia kudriavzevii and Meyerozyma caribbica indicate potential to be used at industrial level, with strain improvement through experimental evolution, hybridization, or genetic engineering.

Effect of salt on ethanol production

Marine yeast is utilized in several applications as they thrive in harsh conditions hence tolerate higher process conditions (salinity and temperature) [70]. In the bioethanol process, pretreatment using dilute acid hydrolysis of marine macroalgal biomass results in salty hydrolysate, which requires a desalination process when employing terrestrial yeast strains, but in the case of halotolerant yeast strains, the hydrolysate is directly fermented to bioethanol.

Screening of cellulolytic yeast and ethanol production by CBP

A single strain of microorganisms that expresses cellulolytic activity and fermentation capabilities is of potential interest in bioethanol production as it brings down the economic burden of enzyme production and the overall bioethanol production and is regarded as the low-cost biomass processing [72, 73]. In this study, prioritized strains were isolated on plates comprising 1% CMC as the sole source of carbon (Fig 4). Hydrolytic activity was recorded for both strains CY: 2.06 and TY: 2.69. Enzyme activity of 1.15 U/ml and 1.19 U/ml was recorded for M. caribbica and P. kudriavzevii at 24h respectively (Table 6).

Fig. 3 a Efect of diferent concentrations of salt on CY strain growth and ethanol production (RS, reducing sugar). b Efect of diferent concentrations of salt on TY strain growth and ethanol production (RS, reducing sugar)

In CBP, cellulase production, cellulose hydrolysis, and fermentation of subsequent sugar occur in a single reactor by a single microbial community [74] compared to SSF. The advantage of CBP is lower or zero capital costs for enzyme production and compatibility of enzymatic and fermentation processes. Anaerobic bacteria have been tested for ethanol production via CBP. However, lower ethanol tolerance (<2%) of bacteria is a limitation for its application at an industrial scale. In this study, wild ethanologenic yeast strains M. caribbica and P. kudriavzevii were used to ferment the pre-treated macroalgal biomass E. intestinalis and U. lactuca (Table 7). Higher ethanol conversion efficiency was recorded for P. kudriavzevii fermenting E. intestinalis and U. lactuca compared to the conversion efficiency achieved through SSF process. K.marxianus PT -1 isolated from grape fermented Jerusalem artichoke tuber flour consisting of inulin at 40^oC for 48h and achieved 90% conversion efficiency through CBP [39]. CBP was carried out for brown algae Saccharina japonica using engineered E.coli (BAL1611) bacteria for 150h and obtained 4.7% ethanol [75].

SEM analysis was carried out for macroalgal biomass after CBP process, the initial dilute-acid pretreatment provided surface area for the yeast cells to attach and secret enzymes to degrade the biomass. Biomass was disintegrated after CBP process indicating the cellulolytic yeast activity (Fig 5).

CBP- compatible microorganism extensively studied is S.cerevisiae; however, it is not suitable while employing second-generation feedstock as it only yields higher ethanol from hexose and not from pentose sugar [72]. Therefore, there is a need to explore wilder ethanologenic yeast strains that exhibit higher cellulolytic activity and be employed for CBP of third-generation macroalgal feedstock and second-generation feedstock.

Fig 5. SEM micrographs of the interaction between cellulolytic yeast strains Meyerozyma caribbica on (A) E.intestinalis and (B) U.lactuca, Pichia kudriavzevii on (C) E.intestinalis and (D) U.lactuca