Optimized acid hydrolysis of the polysaccharides from the seaweed Solieria filiformis ( Kützing ) P . W . Gabrielson for bioethanol production

The seaweeds are bio-resource rich in sulfated and neutral polysaccharides. The tropical seaweed species used in this study (Solieria filiformis), after dried, shows 65.8% (w/w) carbohydrate, 9.6% (w/w) protein, 1.7% (w/w) lipid, 7.0% (w/w) moisture and 15.9% (w/w) ash. The dried seaweed was easily hydrolyzed under mild conditions (0.5 M sulfuric acid, 20 min.), generating fermentable monosaccharides with a maximum hydrolysis efficiency of 63.21%. Galactose and glucose present in the hydrolyzed were simultaneously fermented by Saccharomyces cerevisiae when the yeast was acclimated to galactose and cultivated in broth containing only galactose. The kinetic parameters of the fermentation of the seaweed hydrolyzed were Y(P⁄S) = 0.48 ± 0.02 g.g , PP = 0.27 ± 0.04 g.L .h, η = 94.1%, representing a 41% increase in bioethanol productivity. Therefore, S. filiformis was a promising renewable resource of polysaccharides easily hydrolyzed, generating a broth rich in fermentable monosaccharides for ethanol production.


Introduction
The search for renewable sources for energy production is encouraged for various reasons, including improvement in air quality, achievement of energetic independence, security and sustainability (Park, Kim, Park, Kim, & Yoon, 2014;Wei, Quarterman, & Jin, 2013).Bioethanol has received great attention as a renewable alternative fuel and, in contrast to fossil fuels, bioethanol is produced via the fermentation of sugars.
Acta Scientiarum.Biological Sciences Maringá, v. 39, n. 4, p. 423-430, Oct.-Dec, 2017 Seaweeds are photosynthetic organisms, able to absorb CO 2 and required nutrients from the surrounding water through their thalli.Carbohydrates such as cellulose and floridean starch are present in cells; however, other complex polysaccharides accumulate in brown seaweed (alginate, laminarin and fucoidan), red seaweed (galactans) and green seaweed (ulvan).Chemically, more than 50% of seaweed composition is ligninfree carbohydrate, a feature that leads to mild hydrolysis conditions (Jiang et al., 2016).This fact, associated with the low cost of production, makes biomass a feasible source for biofuel generation (Jang, Cho, Jeong, & Kim, 2012).
The Brazilian coast has a great diversity of red seaweed species (Saito & de Oliveira, 1990), among them Solieria filiformis (Gigartinales, Solieriaceae), abundant in the northeast.The cultivation of seaweeds exhibits several advantages such as high area productivity, no competition with conventional agriculture for land and no need for rain, irrigation, fertilizers or pesticides.Other advantages include: control of harmful algal blooms, maintaining healthy mariculture systems, atmospheric CO 2 sequestration and high photosynthetic efficiency (Kumar, Gupta, Kumar, Sahoo, & Kuhad, 2013;Park et al., 2012;Yanagisawa, Nakamura, Ariga, & Nakasaki, 2011).
Therefore, to avoid the diauxic process, yeast strains may be improved by: acclimation to a high concentration of sugar for a short time and manipulation of the genes involved in the galactose metabolism to enhance ethanol production (Kim et al., 2014) or application of evolutionary engineering to construct mutants with enhanced ability for bioethanol production from galactose (Kim et al., 2014;Lee et al., 2015;Ostergaard et al., 2000).
In this context, the current study aims to optimize the acid hydrolysis of the biomass from seaweed Solieria filiformis and the bioethanol productivity by simultaneous fermentation of the monosaccharides by S. cerevisiae galactose-acclimated.

Seaweed biomass
Specimens of the seaweed were harvested from farming structures located in Flecheiras beach, Trairi, Ceará, Northwest Brazil (03° 13' 06" S 39° 16' 47" W).The algal biomass was washed with water, dried at room temperature (26°C), cutted and milled by an electric mill.The milled powder was then sieved through an 80-mesh sieve (< 0.18 mm) and stored at room temperature.

Biochemical composition
The protein content was measured by the semimicro Kjeldahl method, with a factor of 6.25 for the conversion of nitrogen to protein (Fawcett, 1954).Assessment of the moisture and ash content was performed according to the Association of Official Analytical Chemists (Association of Official Analytical Chemists [AOAC], 2000).Crude lipids were extracted from the powdered dried seaweed using Soxhlet apparatus and ethanol as solvent.Lipid extraction was conducted at 80°C for 4 hours.After determination of the moisture, ash, protein and lipid contents, the total carbohydrate content was determined by the difference of the total seaweed dried biomass.

Acid hydrolysis of the seaweed
Based on the hydrolysis conditions utilized for Kappaphycus alvarezii seaweed (Hargreaves et al., 2013;Khambhaty et al., 2012;Meinita et al., 2012a), a species of the same family as S. filiformis, the levels of the variables time and acid concentration were defined.The acid hydrolysis of the dried seaweed (7.0% moisture) was conducted in an autoclave at 121°C and 50.0 g L-1 solid loading.The seaweed powder was mixed with sulfuric acid (0.2, 0.5 and 1.0 M) in a 250 mL Erlenmeyer flask and incubated at 121°C for 10, 20 or 30 min.Then, the hydrolysates were filtered and the pH of the liquid phase was adjusted to 5.0 with calcium hydroxide, producing calcium sulfate, which was further separated by filtration.Samples of the liquid phase were used for analysis of glucose, galactose, cellobiose, 5-hydroxymethylfurfural (5-HMF) and furfural by high performance liquid chromatograph (HPLC).Assays were performed in triplicate, and results represent the mean ± standard deviations of two independent experiments.The hydrolysis efficiency (HE, %) was based on the composition of S. filiformis and calculated according to Equation 1.
where S 1 and S 2 are the glucose and galactose concentrations determined by HPLC and S Sf is the carbohydrate content of the dry seaweed.Total monosaccharide (glucose + galactose), cellobiose and 5-HMF concentrations and HE obtained after acid hydrolysis were treated statistically by one-way analysis of variance (ANOVA) and pairwise comparisons were made just for total monosaccharides and HE using Tukey's test utilizing OriginPro 9.0 software (OriginLab Corporation, USA).The hydrolysate with maximum monosaccharide concentration was selected as the fermentation medium, denoted as Sfmedium.

Acclimatization of the yeast to galactose
Commercial Saccharomyces cerevisiae (Fleishmann) was used for acclimatization to galactose.The yeast was cultured on a Sabouraud (HiMedia) agar plate, a single colony was selected and the seed culture was incubated in 9 mL enrichment broth composed of 10.0 g L −1 glucose, 5.0 g L −1 peptone, 3.0 g L −1 yeast extract and 3.0 g L −1 malt extract cultured at 30°C and 150 rpm for 48 hours.Then, 2.5 mL of the culture (OD 600 = 1.9) was transferred to YPGGC (3.0 g L −1 yeast extract, 5.0 g L −1 peptone, 20.0 g L −1 galactose, 20.0 g L −1 glucose and 10.0 g L −1 cellobiose) and cultured under the same conditions.After 48 hours of incubation, 2.5 mL of the yeast culture was transferred to fresh medium and five successive batch cultures were carried out with YPGGC medium, supplemented with Sf-medium in ratios of 25/75, 50/50, 60/40, 70/30, 80/20 and 90/10 (v/v) for each batch.The incubation procedure was done at 30°C and 150 rpm for 24 hour to obtain acclimated S. cerevisiae.The acclimated yeasts were denominated Sc Gal and were maintained on plates containing Sf-supplemented YPGGC agar medium, and utilized in subsequent experiments.

Inoculum preparation
For fermentation assays, single colonies of the Sc Gal strain were inoculated in 100 mL of two different growth broths: i) B Glu : 40 g L −1 glucose, 10 g L −1 peptone, pH 3.5 and ii) B Gal : 40 g L −1 galactose, 10 g L −1 peptone, pH 3.5.In both media, the cultures were incubated for 30h at 30°C and 150 rpm in a shaker (Tecnal TE-420, São Paulo, Brazil).The inoculums were denominated Sc Gal -B Glu and Sc Gal -B Gal , respectively.

Fermentation medium
The pH of the hydrolysate selected and denoted Sf-medium was adjusted to 5.5 with calcium hydroxide under constant agitation.Then, it was filtered by vacuum filtration in a sintered funnel plate to remove the CaSO 4 formed, and the liquid fraction was utilized for fermentation assays.

Fermentation assays
S. cerevisiae acclimated to galactose was used for ethanol production.In this step, the influence of inoculum medium (B Glu and B Gal medium) on the process was evaluated.Ethanol fermentation was carried out in a working volume of 50 mL in a 125 mL Erlenmeyer flask, and 10% v/v of yeast (Sc Gal -B Glu or Sc Gal -B Gal , with OD 600 adjusted to 1.9 ± 0.1) was inoculated to the hydrolysate from seaweed (Sfmedium).The bioprocesses were conducted at 30°C and 150 rpm for 30 hours.The experiments were done in duplicate.Samples of 1 mL were taken every 2 hours and filtered through a 0.22 μm nylon membrane for analysis of glucose, galactose and ethanol concentrations.
In all experiments, the yield (Y P/S ), productivity (P P ) and efficiency (η) of ethanol were calculated as described by Equations 2, 3 and 4, respectively.

= −
(2) where P max is the maximum ethanol concentration achieved during fermentation (g L −1 ), S 0 is the initial concentration of total monosaccharides (g L −1 ) and t fP is the fermentation time at which the maximum concentration of ethanol (h) is obtained.High performance liquid chromatograph (HPLC) The glucose, galactose, cellobiose, 5-HMF, furfural and ethanol concentrations in the hydrolyzed and fermentation assays samples were analyzed by a high performance liquid chromatograph (HPLC) (Shimadzu, Tokyo, Japan), equipped with an RID-10A refractive index detector (Shimadzu, Tokyo, Japan) and an Aminex HPX-87H column.The samples were filtered through a syringe filter with a 0.22 μm nylon membrane and 20 μL was injected for analysis.The mobile phase comprised 5.0 mM H 2 SO 4 at a flow rate of 0.6 mL min.−1 and a column temperature of 80°C.The analyzed molecules were determined using an RID-10A refractive index detector (Shimadzu, Tokyo, Japan) and quantified from the standard curves developed using standard reagents.
The content of carbohydrates in S. filiformis (65.8%) was similar to that related for the seaweed K. alvarezii (63.7%), a specimen of the same family (Solieriaceae) and a commercially important source of kappa-carrageenan (Hargreaves et al., 2013).The agarophyte seaweeds Gracilaria cervicornis (63.1%) and Gelidium amansii (67.3%) also presented high carbohydrate content (Marinho-Soriano, Fonseca, Carneiro, & Moreira, 2006;Park et al., 2012).However, the biochemical composition of the S. filiformis, showed in the corrent work, was different of the related in literature (Carneiro, Rodrigues, Teles, Cavalcante, & Benevides, 2014).Variations mainly in protein and lipid contents are associated at seasonality and algal life stage.The ash, protein, moisture and lipid content of S. filiformis corroborated the biochemical composition presented by other red seaweeds, showing that these organisms are good sources of carbohydrates, proteins and ash.Furthermore, S. filiformis was selected as the biomass for ethanol production in this study because there is potential for a mariculture system and it grows in a tropical sea, according to results obtained by our research group.

Acid hydrolysis of S. filiformis
The acid hydrolysis of the seaweed S. filiformis generated total monosaccharides maximum content of 18.1 g L -1 at the condition 0.5 M 20 min. - (Figure 1).The hydrolysates obtained from extreme hydrolysis conditions, 0.2 M 10 min. - and 1.0 M 30 min. - , showed the lower concentrations of total monosaccharides.The concentrations of cellobiose, disaccharide not fermentable that decreases the yield of the ethanol, ranged from 0.06 ± 0.08 to 2.72 ± 1.23 g L −1 in hydrolyzeds.The soft conditions contributed for the highest concentrations of the cellobiose.
The 5-HMF, a fermentation inhibitor, ranged from 0.33 ± 0.24 to 2.33 ± 1.35 g L −1 in hydrolyzeds.At 10 and 20 min., there was a similar trend for the generation of total monosaccharides; however, at 30 min., acid concentrations up to 0.2 M resulted in a decrease in total sugar concentration.The most severe conditions of hydrolysis, with longer exposure times and higher concentrations of acid, contributed to degradation of the cellobiose, dehydration of hexoses and degradation of 5-HMF.These reactions are undesirable because they produce organic acids, such as levulinic acid and formic acid, which inhibit fermentation (Larsson et al., 1999).
Galactose was the predominant monosaccharide in all conditions studied, with levels varying from 8.34 ± 1.4 g L −1 (0.2 M 10 min. - ) to 14.83 ± 1.02 g L −1 (0.5 M 20 min. - ).Glucose was also detected at all hydrolysis conditions performed, with concentrations varying from 2.36 ± 0.4 g L −1 (0.2 M 10 min. - ) to 3.30 ± 1.1 g L −1 (0.5 M 20 min. - ).The presence majority of the galactose in hydrolysates corroborated with the high contents of the pure iotacarrageenan of the S. filiformis (Murano et al,. 1997).Furfural was not detected, suggesting the absence of the pentose polymers.Meinita et al. (2012a) evaluated the hydrolysis of K. alvarezii (5%, w/v) using 0.2 M H 2 SO 4 solution at 130°C for 15 min.and they obtained similar effects of acid on galactose and 5-HMF concentrations and total sugar.An increase in acid concentration up to 0.2 M resulted in a sharp increase in galactose and glucose generation, and reduced the levels of 5-HMF and formation of levulinic acid.Similar to results obtained for S. filiformis, a longer time also contributed to the reduction of 5-HMF content in hydrolysates of K. alvarezii (3% w/v) with 0.2 M H 2 SO 4 , decreasing from 1.81 g L −1 at 10 min.to 0.25 g L −1 at 90 min.Jeong et al. (2015) related the glucose and galactose production optimized from Gracilaria verrucosa (66.6 g L −1 ) in H 2 SO 4 solution.The glucose concentration reached 5.29 g L -1 under conditions of 160°C (reaction temperature), 1.92% (catalyst concentration), 20 min.(reaction time), while the galactose concentration was 18.38 g L −1 under 160°C, 1.03%, 20 min.conditions.
The different HE verified in the hydrolysates from S. filiformis showed the influence of the time and acid concentration in the generation of fermentable monosaccharides (Figure 2).The hydrolysis extreme conditions showed the lower HE and the condition 0.5 M 20 min. - reached the highest HE value (59.7%), although it is similar to conditions 0.5 M 10 min. - , 1.0 M 10 min. - , 1.0 M 20 min. - , 0.2 M 30 min. - and 0.5 M 30 min. - .The absolute absence or near absence of lignin in seaweed (John, Anisha, Nampoothiri, & Pandey, 2011) makes the hydrolysis of algal polysaccharides simple when compared to land plants.Appropriate pretreatment methods are required to overcome the recalcitrance of lignocellulosic materials, such as dilute 0.5-2.5% H 2 SO 4 performed at temperatures from 100 to 200ºC (Sun, Sun, Cao, & Sun, 2016).In this work, the hydrolysis conditions selected for fermentation assays and denominated Sf-medium were 0.5 M 20 min. - because together, galactose, glucose and cellobiose corresponded to an HE of 63.21% of the carbohydrate content of S. filiformis and presented the highest fermentable carbohydrate concentration.

Ethanol productivity
Ethanol production utilizing seaweed hydrolysate (Sf-medium) as fermentation medium is shown in Figure 3. Sc Gal pre-cultured in B Glu and B Gal medium consumed galactose and glucose contained in the Sf-medium for ethanol production, but the consumption profiles were different.Sc Gal -B Glu showed a consumption velocity of monosaccharides lower, a diauxic behavior, preferring glucose and increasing the rate of consuming galactose after 10h of fermentation.Moreover, Sc Gal -B Glu utilized total glucose at 24 hours and used only 61.6% of the available galactose after 30h of fermentation, reaching a maximum ethanol concentration of 4.9 ± 0.2 g L −1 .Another relevant result was the increase in glucose concentration, in contrast to the decrease in galactose concentration in 2 hours of fermentation.This suggests the conversion of the galactose in glucose by enzymes of the Leloir Pathway.Although Sc Gal had been acclimatized to use galactose, it had this capacity reversed in the presence of glucose when cultured in B Glu medium.The expression of the enzymes involved in galactose metabolism is strictly down-regulated in the presence of glucose, called "glucose repression" in S. cerevisiae (Timson, 2007).Similar results were obtained by Keating, Robinson, Bothast, Saddler, & Mansfield (2004) when they observed that the preferential utilization of glucose and slow assimilation of galactose after depletion of glucose by S. cerevisiae Y-1528 results in reduced productivity of ethanol from a mixture of galactose and glucose, compared to that observed with glucose alone.Moreover, unexpectedly, endogenous glucose formation and the appearance of glucose in the culture medium was detected in galactose and mannose fermentations.
Non-acclimated S. cerevisiae, used for fermentation of Gelidium amansii hydrolysate, consumed the glucose in 24 hours; however, galactose was rarely consumed because of the repression of galactose uptake by glucose (Cho, Ra, & Kim, 2014).In contrast, the Sc Gal inoculated in B Gal medium (Sc Gal -B Gal ) did not show a preference to utilize glucose, and diauxic fermentation not was observed, with glucose and galactose simultaneously consumed; 93.4% of the carbohydrate content was consumed in 24 hours, reaching 6.5 ± 0.4 g L −1 of ethanol.The ethanol productivity was of 0.27 ± 0.04 g L −1 h −1 , 41.1% higher than Sc Gal -B Glu .The fermentative parameters of Sc Gal -B Glu and Sc Gal -B Gal are shown in Table 2. Y P/S , P P and η obtained using the inoculum Sc Gal -B Gal were higher than those obtained using Sc Gal -B Glu .
The initial 5-HMF concentration in Sf-medium (1.09 g L −1 ) decreased and reached 0.69 g L −1 using Sc Gal -B Glu and 0.35 g L −1 using Sc Gal -B Gal in 40 hours (data not shown).
Other strains, such as a S. cerevisiae mutant, produced equivalent results with P P = 0.32 g L −1 h −1 after fermentation of K. alvarezii acid hydrolysate (Khambhaty et al., 2012).In contrast, Cho et al., (2014) utilizing S. cerevisiae acclimated to high concentration of galactose reached P P = 0.15 g L −1 h −1 and Y P/S = 0.44 lower than Sc Gal -B Gal using acid hydrolysate from G. amansii.Park et al. (2014), using a mutant strain of S. cerevisiae (ATCC2341) to consume galactose, showed the yeast was able to take this monosaccharide at concentrations of 50 to 120 g L −1 , producing a maximum ethanol concentration of 54.0 g L −1 and lower η (88%) than Sc Gal -B Gal .Keating et al. (2004) demonstrated that S. cerevisiae Y-1528 was able to utilize galactose even in the presence of glucose, which was not observed in wild strains.Table 2. Fermentative parameters of ethanol production using acid hydrolysate from Solieria filiformis seaweed (Sf-medium) inoculated with Sc Gal -B Glu and Sc Gal -B Gal .Values represent the mean ± SD.

Inoculum
P max (g L −1 ) (g g −1 ) (g L −1 h −1 )  (%) Sc Gal -B Glu 4.9 ± 0.2 0.39 ± 0.03 0.16 ± 0.02 76.4 ± 2.6 Sc Gal -B Gal 6.5 ± 0.4 0.48 ± 0.02 0.27 ± 0.04 94.1 ± 3.0 Our efforts to use S. filiformis as an alternative source of biomass for ethanol production demonstrated that the complex polysaccharides of this species are homogenous, easily hydrolyzed, and an increase in efficiency of the fermentation process of the soluble monosaccharides (glucose and galactose) can be obtained when the yeast strain is metabolically adapted to galactose by simple and low cost method.The glucose directly enters glycolysis, which is the central metabolic pathway in ethanol fermentation, and simultaneously takes advantage of glucose and galactose which are available for fermentation.The search for better understanding of the principles of the cellular physiology involved in the fermentative activity of yeasts is the major challenge in the quest for ethanol derived from seaweed.

Conclusion
The tropical seaweed S. filiformis is a biomass rich in homogenous polysaccharides, easily converted into fermentable monosaccharides (glucose and galactose), with high efficiency and low generation of the fermentation inibitors by acid hydrolysis.Furthermore, the S. cerevisiae yeast galactoseacclimated was efficient in to enhance the ethanol productivity avoiding the diauxic behavior on consumption of the galactose and glucose present in acid hydrolysate from S. filiformis seaweed.

Figure 1 .
Figure 1.Effect of variable H 2 SO 4 concentration and reaction time on concentrations of cellobiose, total monosaccharides (galactose + glucose) and 5-HMF in the hydrolysates from Solieria filiformis seaweed at 121°C using 5 % w/v solid loading.Values represent the mean ± SD.Different letters indicate significant differences among hydrolysate conditions (ANOVA, Tukey's test; p < 0.05).

Figure 2 .
Figure 2. Effect of H 2 SO 4 concentration and reaction time on hydrolysis efficiency of the biomass from Solieria filiformis seaweed at 121°C using 5% w/v solid loading.Values represent the mean ± SD.Different letters indicate significant differences among hydrolysate conditions (ANOVA, Tukey's test; p < 0.05).

Table 1 .
Biochemical composition of red seaweeds.