Raw material and juice preparation

The perennial ryegrass (Lolium perenne) was kindly provided by the Julius Kühn-Institute (Braunschweig, Germany). The production of the press-juice is described elsewhere (Varriale et al. 2022). The resulting juice was further processed before the use as fermentation medium. In particular, the juice was centrifuged twice at 4500 rpm for 20 min (Z38K, Hermle Labortechnik GmbH, Wehingen, Germany) to separate the slurry portion. To remove the remaining solid particles, the supernatant was filtered using filter paper (Macherey–Nagel, 185 mm). To prevent the loss of important nutrients, the sterilization process was carried out using a sterile filtration (Stericup®, 0.2 µm pore size, Merck Millipore, Massachusetts, USA). The obtained press- juice was stored at − 20 °C until further use.

Microorganism and medium

The yeast Saccharomyces cerevisiae 3799 was obtained from the German Collection of Microorganisms and Cell cultures (DSMZ GmbH, Braunschweig, Germany). Preculture and control experiments were carried out in yeast peptone dextrose (YPD) medium with the following composition: yeast extract 10 g/L, peptone 20 g/L, dextrose 20 g/L (pH 6.2). Unless otherwise stated, all the chemicals were purchased from Carl Roth + Co KG (Karlsruhe, Germany), except dextrose which was purchase from Sigma-Aldrich (Merck KGaA, Darmstadt, Germany).

Preculture and main culture preparation

Precultures were started from stock cryo-culture stored in 25% (v/v) glycerol at − 80 °C. They were grown aerobically in 100-mL baffled shake flasks for 24 h until the late-exponential phase was reached. The main cultures were inoculated to 0.1 OD (optical density) without pH adjustment. The main cultivations were performed anaerobically in 100-mL glass bottles with 50-mL working volume, tightly sealed with rubber septa. The bottles were sparged with pure nitrogen prior to inoculation. Both the precultures and the main cultures were incubated at 32 °C and 120 rpm (Ecotron, Infors AG, Bottmingen, Switzerland).

Sampling and cell growth determination

Samples were regularly taken for the OD measurement. One milliliter of each sample was further centrifuged (14,000 rpm, room temperature, 7 min; Eppendorf, Hamburg, Germany), and the supernatant was stored at − 20 °C for HPLC (high-performance liquid chromatography) analysis. The cell dry weight (CDW) was determined in a separate experiment using YPD medium and aerobic condition for the generation of an OD-CDW correlation. The experiment was performed as follows: 5 mL of cell suspension was harvested by centrifugation (4500 rpm, 10 min, 4 °C, Z38K, Hermle Labortechnik GmbH, Wehingen, Germany). The cell pellet was washed with 5 mL and resuspended in 5-mL NaCl solution (9 g/L). Finally, the cell suspension was transferred to 15-mL dried tubes and then dried (both at 50 °C, 48 h) in the oven (Memmert GmbH, Schwabach, Germany) until a constant weight was reached. The CDW/OD600 correlation was determined to be 0.32 ± 0.02, and it was established in triplicates. The maximum specific growth rate (µMAX) was determined in separate experiments in duplicates. Both the not adapted and adapted S. cerevisiae strains were cultivated in 100%(v/v) juice anaerobically. One milliliter sample was harvested and centrifuged (14,000 rpm, 7 min, room temperature, Eppendorf, Hamburg, Germany). The pellet was washed and resuspended in distilled water. The maximum specific growth rates were estimated from the slope of the linear regression between ln(OD) and time.

Analysis of press-juice components and metabolites

The analysis of the press-juice was carried out with regard to pH, protein content, sugars, amino acids, cations, and anions. Before any analysis, the samples were filtered through a 0.2-µm pore size nylon filter (KX Syringe Filter Nylon, Cole-Parmer GmbH, Wertheim, Germany). The pH value was measured using a pH meter (Microprocessor pH 211, Hanna Instruments Deutschland GmbH, Vöhringen, Germany). The protein concentration was determined using the Bradford assay (Pierce® Coomassie Bradford Protein Assay kit, Thermo Scientific, Massachusetts, United States) and the bovine serum albumin as internal standard (Thermo Fisher Scientific, Massachusetts, United States). The absorbances of the calibration curve and of the samples were measured at 595 nm (Cary 60 UV–Vis, Agilent Technologies, USA). Sugar and product concentrations were analyzed by HPLC [Autosampler AS 6.1L (Knauer GmbH, Berlin, Germany), Azura pump P 6.1L (Knauer GmbH, Berlin, Germany)], and a refractive index detector (RI 101 Shodex, Kawasaki, Japan). The HPLC was equipped with an Aminex HPX-87H column (Bio-Rad, 300 × 7.8 mm, Hercules, California, USA) at 80 °C. The mobile phase was 2.5 mM H2SO4 at a flow rate of 0.6 mL/min. The instrument control and data evaluation were carried out with a Clarity software system (Data Apex, Prague, Czech Republic). Amino acids were separated using a resolve C18 column (150 × 3.9 mm, Waters Corporation, Milford, USA) with a SecurityGuard Cartridge (C18, 4 × 3.0 mm ID, Phenomenex, Torrance, California, USA) placed in a column heater CT 2.1 (Knauer GmbH, Berlin, Germany) set at 30 °C. The detection was performed using Azura photodiode-array detector DAD 2.1L at 230 nm (Knauer GmbH, Berlin, Germany). The analysis included a precolumn derivatization with ortho-phthaldialdehyde (OPA). The derivatization process was automated using the Azura AS 6.1L autosampler (Knauer GmbH, Berlin, Germany) and the own “mix method” option. The OPA reagent was prepared by weighting 270 mg OPA in a 50-mL volumetric flask. The reagent was dissolved in 5-mL ethanol. Then, 200-µL 2-mercapto-ethanol was added, and the final volume was filled up with 0.4 M borate buffer. The mobile phase consisted of solvent A (0.025 M sodium-acetate anhydrous and 0.025 M NaH2PO4 monohydrate) and solvent B (50% methanol). The pH value of solvent A was adjusted to 7 with 10 M NaOH, and then 21 mL of both tetrahydrofuran and methanol was added. The gradient elution program was as follows: from 0 to 50 min, solvent B changed linearly from 0 to 100%; from 50 to 55 min, solvent B was set as isocratic at 100%; from 55 to 60 min, solvent B changed linearly from 100 to 0%; from 60 to 67 min, solvent B was set as isocratic at 0%. The flow rate was 1.0 mL/min. To analyze the sample, it must be protein and particle-free. For this reason, the proteins were firstly precipitated by adding four parts of ice-cold methanol and one part of the sample. The sample was placed at − 20 °C overnight. Then, it was centrifuged for 10 min at 10,000 rpm (Eppendorf, Hamburg, Germany). Finally, it was diluted with 0.4 M borate buffer. A dilution of at least of 1:1 is necessary as the sample has to have a pH 10 for the derivatization. The derivatization was performed by adding 80 µL of the sample or the standard and 50 µL of OPA reagent. After 1 min, 40 µL was injected for the measurement. Cations and anions were analyzed by ion chromatography (IC) (930 Compact IC Flex, Metrohm GmbH & Co. KG, Germany) with an inline system for dialysis (930 Compact IC Flex, Metrohm, Filderstadt, Germany) and an IC Conductivity Detector (Metrohom, Filderstadt, Germany). Cations were measured with a cation column (Metrosep C6-250/4.0, Metrohm) using 4 mM HNO3 and 0.7 mM dipicolinic acid as mobile phase at a flow rate of 0.9 mL/min. Anions were measured with an anion column (Metrosep A Supp5-250/4.0, Metrohom) using 1 mM NaHCO3 and 3.2 mM Na2CO3 as mobile phase at a flow rate of 0.7 mL/min. The oven temperature in both cases was 35 °C. For each analysis, the samples were diluted to a concentration inside the external calibration range (using 2 mM HNO3 for cation determination).

Propagation

Propagation was performed in sequential anaerobic processes (Fig. 1). S. cerevisiae was cultivated in increasing concentration of grass press-juice (10%(v/v), 25%(v/v), 30%(v/v) 50%(v/v), 60%(v/v), 75%(v/v), 100%(v/v)). Ten percent (v/v) was used as starting point and not considered in bioethanol production. The cells were transferred to the next juice concentration after 24 h, except from the passage to 60%(v/v) which occurred after 48 h due to the lower yeast growth. After the cultivation in 100%(v/v), the adapted cells were harvested to prepare cryo-cultures. They were used to re-cultivate the cells in 100%(v/v) fresh juice to analyze the physiological parameters and fermentation capacity. The results were compared with those obtained from the not adapted S. cerevisiae strain. In every passage of the propagation, the juice was inoculated to 0.1 OD using the cells from the previous concentration.

Fig. 1

Propagation scheme of S. cerevisiae during short-term adaptation. The figure was generated using BioRender program

Data processing and evaluations

The percentage of sugars utilization (Su) was calculated according to Eq. (1)

$$_=\frac_-_}_}\times 100$$

(1)

where S0 is the starting sugar concentration, and St is the sugar at the time.

Substrate consumption rates (rs) were calculated using the following equation:

where St is the substrate concentration at specific time, and S0 is the substrate concentration at the beginning of the fermentation, and t is the time. Accordingly, the sugar consumption rate was characterized by a negative value as it described the decrease of sugar over time. The volumetric ethanol productivity was expressed in g/Lh (QEtOH) and was estimated from the ratio between the ethanol concentration and the time at the end of the fermentations. The yields were calculated considering the g of product per g of total sugars consumed (glucose and fructose). Carbon balances included both biomass and metabolite yields, and it was assumed that 1 mol CO2 was formed per mol ethanol and acetate (Novy et al. 2017). For biomass yields, a C-molar weight of 26.4 g/Cmol was used (Lange and Heijnen 2001).