Bioprospecting and Diversity of Yeast Producing Ethanol Isolated from Indonesia

Bioethanol is considered the most environmentally friendly as renewable fuels. Indonesia has abundant microbe diversity which is potential for bioprospecting such as fermenting agents using agriculture product as raw materials for producing bioethanol. This study aims to isolate, characterise and molecular identify of 15 isolates of bioethanol-producing yeasts from various sources. Characterisation based on ethanol production, cell morphology and various substrate utilisation has been carried out. Molecular characterisation of 15 yeast isolates using tree sets of primers had been carried out. Amplification in the internal area of transcribe spacers (ITS) was successfully carried out with an amplitude of 400 bp–900 bp. Amplifiers in the D1/D2 26s rDNA domain are 250 bp. Amplification with ScerF2 and ScerR2 specific primers was carried out successfully and proved that there were two isolates which were not Saccharomyces cerevisiae analysis of yeast genetic diversity showed 12 yeast isolates classified as S. cerevisiae and the rest belonged to the genus Clavispora, Candida and Kodamaea (Pichia).


INTRODUCTION
Fossil fuels such as oil, coal and natural gas emit greenhouse gases (GHG) emission, caused air pollution and harm to the environment. Spikes in oil prices and concerning on the effects of climate change, have lead new urgency to the search for clean, renewable fuels (Selim et al. 2020;IRENA 2019). National efforts to reduce the need for fossil fuel by utilising alternative renewable energy is stated in the Republic of Indonesia Presidential Regulation (PERPRES) Number 5 of 2006 concerning national energy policy through the development of bio-based renewable energy. Biofuel is energy derived from living matter, especially plants.

2
There are three types of biofuels bigest produced in the world: bioethanol, biodiesel and biogas. Bioethanol has potential to be used as fossil fuels substitution, as it can be mixed with gasoline called gasohol E-10, a mixture of pure gasoline with 10% pure bioethanol with an octane value of almost 92. The advantages of bioethanol include producing high energy, can reduce 12% of greenhouse gases, and are more environmentally friendly than biodiesel.
There are various ways of making biofuels, but they generally use mechanical, physical, chemical reactions, enzymatic reaction and heat to break down the starches, sugars, and other molecules in plants. The resulting products are then refined to produce a fuel that cars or other vehicles can use. Production of biofuel so far has used agricultural products as its basic material. Raw materials for bioethanol are sugar (sugar cane and sorghum juice) or carbohydrates (cassava, wheat, corn, sweet potatoes, soybeans and sorghum seeds) which is called as first generation of bioethanol (Tse et al. 2021;Salelign & Duraisamy 2021;Appiah-Nkansah et al. 2019). Second generation bioethanol utilises biomass in the form of non-food residues such as lignocellulosic waste (Fivga et al. 2019;Robak & Balcerek 2020). Fermentation is a series of reactions aided by microorganisms as biocatalysts, one of which is yeast. The type of yeast that is often used for alcohol production is Saccharomyces cereviseae. S. cereviseae is the oldest, most robust ethanol producer (Parapouli et al. 2020;Malairuang et al. 2020;Casas-Godoy et al. 2021) Report on the isolation of ethanol producing yeast have been done from diverse sources such as fruit, soil and molases and others (Johansen et al. 2019;Phong et al. 2019;Riyanti et al. 2020;Kechkar et al. 2019). Yeast identification can be determined based on morphology and biochemistry but still need further identification to determine the type to be more accurate. Yeast identification methods with molecular approaches have high accuracy and can provide information to species level (Arastehfar et al. 2019). Ribosomal DNA (rDNA) is the genome encoding region for the ribosomal RNA component. The rDNA region is separated from one another by a barrier called a spacer (Schoch et al. 2012;O'Brien et al. 2005). The method commonly used to identify types of yeast isolates is by sequencing methods in the internal area of transcribe spacers (ITS) (Chanprasartsuk et al. 2013;Pramateftaki et al. 2000). Other molecular markers to ensure yeast identification are D1/D2 domains in 26s rDNA, almost all species of yeast data sequences in this area and the ITS area are available on an international database (Hesham et al. 2014).
Some yeast isolates derived from fermented foods from various cities in Indonesia were isolated and identified in Molecular Biology Laboratory, Centre for Research and Development of Biotechnology and Agricultural Genetic Resources (BB Biogen). This study aims to select yeast that produces high levels of bioethanol using high performance liquid chromatography (HPLC), morphological and biochemical characterisation and molecular identification or phylogenetic analysis of bioethanol-producing yeast by amplifying parts of the ITS region.

Yeast Isolation, Purification and Screening for Ethanol Production
Yeast was isolated from various sources using yeast potato dextrose (YPD) agar medium containing yeast extract (10 g/L), bacto peptone (20 g/L) and glucose (20 g/L) following Cold Spring Harbour (CSH) Protocol with modification (2010). The material sample was suspended into sterile water followed by dilution to 10 -5 in sterile water, and then 100 μL aliquot was spread cultured into YPD agar medium in triplicate. The culture was then incubated in 30°C for several days. Single colony grown in the YPD agar medium was then purified by streaking on to new YPD agar medium. Several pure cultures were chosen and cultured into 10 mL of YPD broth, and incubate at 30°C for 24 h, with 125 rpm shaking. Ethanol production was measured using High Performance Liquid Chromatography (HPLC) machine (Agilent 126 Infinity, International Equipment Trading Ltd, Illinois, USA). Strain producing ethanol was chosen for evaluation using different sugar sources.

Cell Morphology
The chosen isolate was examined morphologically using light microscope with 1000× magnitude attached with image capture device. Cell measurement was done using hemocytometer with the calculation based on the comparison of grid size and volume.

Fermentation Process
Fermentation proccess was carried out using batch system in Mac Cartney bottles. A total of 1 mL of overnight seed culture were put into 9 mL of YPD media for fermentation process. Fermentation was carried out for 24 h on YPD media at 30°C, 150 rpm. Observations were made every 24 h.

Products Evaluation of Selected Isolate
Two chosen isolate with highes ethanol production was characterise based on ethanol yield. Ethanol yield was calculated based on dp/ds, when dp is ethanol production and ds is carbon used during the fermentation process as follows.
Yp s S P / T = where P (product) is ethanol concentration, s (substrate), ΔS is the total glucose consumed where obtained from initial glucose level and S is the remaining glucose concentration in the end of fermentation.

Sample Analysis Using HPLC
Sample was taken from the culture at 2 mL volume followed by centrifugation at 5,000 rpm, 4°C for 5 min. The cell biomass was used for biomass determination. Supernatant was then filtered using disposable 0.2 µL diameter filter and transferred into HPLC vial for analysis using HPLC (Agilent 126 Infinity, Agilent Technology) equipped with Refraction Index Detector (RID) detector (Agilent 126 Infinity, International Equipment Trading Ltd, Illinois, USA). Hi-plex H for carbohydrate collumn was used for the analysis using ddH 2 O as mobile phase and maintained at at 0.7 mL flow rate, and 60°C-70°C collumn temperature.
Filtrate was then analysed for ethanol production, sugars (glucose, xylose) content and side products (acetic acid, lactic acid and glycerol) content. Substrate were identified related to their peaks at various retention time which are glucose at 8.49 min, xylose for 9.01 min, ethanol for 18.96 min, acetic acid for 13.27 min, lactic acid for 11.27 min and glycerol for 11.62 min. The concentration of each compound was determined based on its peak area compared to series of standard's peak area. Data on the ethanol and side product were analysed using F-test.

Isolation of Genomic DNA
The DNA isolation process was carried out following the standard protocol of YeaStar TM Genomic DNA Kit with protocol II (Thermo Fisher Scientific, Massachusetts, USA). The amount of 1.5 mL cell cultures were centrifuged at 5,000 rpm for 2 min, and supernatant was discarded. A total of 120 µL YD digestion buffer and 5 µL R-zymolyase was then added into pellets. The pellets were resuspended with vortex and incubated at 37°C for 40 min-60 min. Then, 120 µL of YD lysis buffer was added and mixed with vortex for 10 s to 20 s. The mixture was centrifuged at 10,000 rpm for 2 min. The supernatant was then transferred to the zymo-spin III column and re-centrifuged at 10,000 rpm for 1 min. Amount of 300 µL DNA washing buffer was added to the column and centrifuged at a speed of 10,000 rpm for 1 min. Then, 300 µL DNA washing buffer was added to the column and centrifuged at 10,000 rpm for 1 min. The zymo-spin III column was 5 transferred into a sterile 1.5 mL centrifuge tube and ddH 2 O was added to the 60 µL column membrane. The column was incubated for 1 min at room temperature then centrifuged again at 10,000 rpm for 10 s. The liquid DNA was then collected in a 1.5 mL tube and added to 1 µL of RNAse A then homogenised. Several isolates need modification method during cell lysis. Qualitative DNA examination was performed with 1% agarose gel electrophoresis dyed with Etidium Bromide and visualised using UV translulimator (Sambrook & Russel 2001).

DNA Quantity Measurement
The quantity of DNA was determined using NanoDrop™ 2000/2000c Spectrophotometer (Thermo Scientific, Waltham, Massachusetts, USA) according to its manual. The data obtained in the form of concentration (ng/µL) and DNA purity (A260/280 ratio).

Gene Sequencing and Bioinformatics Analysis
Sequencing was done in First Base Company, Singapore. Sequence analysis was performed using BioEdit software. The sequence homology was anlysed using BLASTn programme (Altschul et al. 1990;Boratyn et al. 2013). Phylogenetic analysis using the MEGA 7.0 program with the neighbour-joining (NJ) method (Kumar et al. 2016).

Screening Yeast Isolates for Bioethanol Production
Fifteen pure isolates were obtained from the isolation process. Data of ethanol and side products analysis was presented in Table 1. In this screening process, the isolates produced ethanol at 0.8 up to 2.61 g/L, which most of the isolate produced side products such as acetic acid, lactic acid and glycerol. Using 24 h fermentation process, glucose was detected in the culture, which mean the fermentation process need longer than 24 h.

Growth Analysis on Various Carbon Sources
The growth of 15 isolates in four media with different carbon sources were presented in Table 2. It shows that all isolates can grow in YPG, YPX and YPGX media with different growth rate observed from the colony formation. Meanwhile, 7 all isolates could not grow in CMC media which indicated that 15 isolates did not have cellulolytic activity. Isolates grown in media containing glucose can grow faster compared to media containing xylose and mixture glucose and xylose.

8
Under light microscope, cell morphology of 15 isolates were haploid with oval to round shape cells as displayed in Fig. 1.

Yield Observation
Two isolates with highest ethanol product were compared based on their yields. Table 3 shows the yield comparison between isolates NKL 10.5.2.3 and TKH (BGR) 10.5.2.1. Two isolates produced relatively the same amount of ethanol, but it produced diff erent side product and glucose consumed. Based on the glucose consumed during the fermentation process, isolate NKL 10.5.2.3 gave higher yield compared to TKH(BGR)10.5.2.1. which is around ± 400 bp ( Fig. 2A). Amplifi cation of the D1/D2 domain shown in Fig. 2B was around ± 250 bp. Amplifi cation using specifi c primer for S. cerevisiae

DISCUSSION
Fifteen yeast isolates had been characterised for the ability to produce ethanol (Table 1), and capability using various sugars sources (Table 3). All the isolate grow on glucose medium and mixture sugars glucose and xylose, however slightly growing on sole xylose medium (Table 2). This could indicate in the fermentation process that the isolates uptaken glucose, and then xylose after the glucose depleted. However, it could also mean that the isolate could co-fermenting glucose and xylose with different rate. Morphological examination under microscope shows that all of them was haploid cell, with diverse shape from oval to circular. It was reported that the diversity of yeast species in different habitat is determined by its capability of utilising different carbon source and its nutritional selectivity as it exhibits their great specialisation (Mohd Azhar et al. 2017).
The highest ethanol concentration of 2.61 g/L produced by NKL 10.5.2.3 and TKH (BGR) 10.5.2.1, eventhough not significantly different statistically. The highest isolates were able to produce bioethanol with the same level of 2.61 g/L but the isolate TKH (BGR) 10.5.2.1 changed less sugar in the media into fermented product compounds, such as TKH isolate (BGR) 10.5.2.1 only able to convert as much as 13.47% glucose (Table 1). This value is smaller than NKL 10.5.2.3 isolate which was able to convert 13.51% glucose. The low concentration of ethanol in this result was due to low sugar concentration in the medium (20 g/L glucose) and short period of fermentation process (24 h fermentation time) as sugar remained in the medium after fermentation process (Table 2). Muin et al. (2015) stated ethanol levels will continue to rise until the 72nd hour but the ethanol and CO 2 levels produced can inhibit ethanol production in the next hour. In addition, according to Rahmana et al. (2016), at the 72nd hour yeast is still in a stationary state so that most likely the sugar will be consumed optimally. The fermentation process will slow down when the concentration of ethanol produced is around 12%-15% because yeast cells are only tolerant at certain ethanol concentrations.
Using 20 g/L glucose in the medium, isolates could give higher ethanol product in prolong time as the glucose remained 0.69 g/L and 0.63 g/L, respectively in the media. Moreover, it is predicted that all the isolates could produce higher products if the medium contain higher carbon source concentration, longer fermentation process and optimum condition after bioprocess optimisation ( Table 2).
Capability of using wide range carbon sources including hexose and pentose is very important for industrial purpoces. The capability of isolate using pentose is very important for further application using lignocellulosic hydrolysate which contained mixture hexose and pentose carbon. Research on second generation of ethanol production is still hot topic for lowering cost production from using cheap raw material such as agriculture waste containing lignocellulosic materials. However, lignocellulosic hydrolysate is recalcitrant for microbial fermentation due to sugars mixture pentose and hexose sugars and other compounds. This pentose sugars and other inhibitors act as inhibitor for industrial yeast so far (Robak & Balcerek 2020). Based on the screening on Table 2, the capability of growing in medium with pentose (xylose) is promising for this isolate to be used in lignocellulosic hydrolysate medium for second generation of ethanol production. In this result, all isolate is capable using glucose and xylose, however less growth on xylose compared to on glucose or mixture glucose and xylose observed.
In this result, all isolate tested could not grow on the cellulose medium, this condition indicated that yeast isolates did not have cellulolytic activity. Yeast isolates that have been tested do not prove the existence of cellulolytic activity so that if large-scale biethanol production will be carried out with a source of agricultural biomass material it is necessary to have pretreatment.
The low yield of ethanol obtained in this study might due to the low concentration of carbon and the length of fermentation process. Batch culture system with S. cerevisiae was used for ethanol fermentation with glucose concentrations ranging 10 g/L-260 g/L, as a result, 0.2 g/L-7.0 g/L biomass and 5.1 g/L-115.0 g/L ethanol were obtained. The maximum yield of ethanol by yeast is theoretically about 51% or 0.51 g ethanol/g glucose. High ethanol concentration and ethanol yield could be achieved by the fed-batch cultures with total glucose concentrations up to 260 g/L (Chang et al. 2018).
Generally, the maximum identity value obtained is more than 98% and there are six isolates that have a maximum identity value in the range of 86%-97% (Table 3). A high maximum identity value indicates a greater degree of homology between sequences. In addition, the e-value is also considered. The e-value is a value that states the level of statistical probability to have similarities with the tested sequence or commonly referred to as the percentage error. The smaller the e-value indicates the level of error and the higher the level of homology. The maximum identity value listed in the BLAST results is influenced by other parameters displayed in the BLAST (Claverie & Notredame 2006). E-value of all isolates is 0.0 except isolates NKL 10.5.2.3 which is 4e-176 (Table 3). This indicates that NKL 10.5.2.3 with the results of the alignment has a fairly large error rate (Nugraha et al. 2014). A total of 13 isolates were parallel to S. cerevisiae with a maximum identity value in the range of 86%-99%. Isolate TKP (NTB) 10.5.2.1 belongs to the genus Candida with a maximum identity of 94%. Isolate TKP (KBM) 10.5.2.2 belongs to the genus Kodamaea (Pichia) with a maximum identity of 99%. Isolate NKL 10.5.2.3 belongs to the genus Clavispora with a maximum identity value of 95%, but this alignment has a bad e-value. Sequences with a maximum identity value <95% are labelled as unknown organisms (Rosa et al. 2010). Meanwhile, sequences with a maximum identity value more than 97% indicate identical species. Another theory states, the maximum identity value in the range 93%-97% represents the same genus level with the possibility of different species levels. Maximum identity values in the range 82%-93% indicate novelty at the genus level (Hagstrom et al. 2000). In this regard, isolate P (KBM) 10.5.1.4 has only 86.2% homology to S. cerevisiae strain YR29. Fig. 3 shows that this isolate has far distance to other identified S. cerevisiae. Therefore, this isolate could be a candidate for new species, but need further investigation.
Phylogenetic analysis in this study uses the neighbour-joining reconstruction method with a boostrap evaluation of 1,000 repetitions. Evaluation of phylogenetic trees using boostrap is carried out to determine the level of trust in phylogenetic trees. Determination of phylogenetic reliability of reconstructed trees is based on boostrap values, if the value is more than 70% of tree branches has high stability (Osawa et al. 2004).
Relationship analysis of test sequences with BLAST result sequences must include outgroup groups for apomorphic and plesiomorphic character polarisation. This study uses Schizosaccharomycess pombe as an outgroup species.
S. cereviseae and Pichia stipitis are common yeast used for bioethanol production. S. cereviseae is unable converting pentose sugar which is 45% content of the lignocellulosichydrolysate (Nijland & Driessen 2020). Meanwhile, Pichia stipitis has great potential to ferment hexose and pentose sugars such as xylose to be used as ethanol with high levels (Ishizaki & Hasumi 2013). Kinship between species is known by phylogenetic analysis which is a combination of molecular techniques and statistics. It has been known that the results of BLAST showed four large genera that correspond to the test sequences namely Saccharomyces, Clavispora, Kodamae (Pichia) and Candida. Yeast isolated from all indigenous sub-Saharan African fermented food and beverages was also reported by Johansen et al. (2019) which being the predominant by S. cerevisiae.