Enhancing Single-Cell Protein Yield Through Grass-Based Substrates: A Study of Lolium perenne and Kluyveromyces marxianus

Abstract

This study evaluated Lolium perenne press juice as a sustainable substrate for Single-Cell Protein (SCP) production using Kluyveromyces marxianus. Key fermentation parameters were systematically optimized, including microbial reduction, dilution ratios, temperature, and nutrient supplementation. Pasteurization at 75 °C preserved essential nutrients better than autoclaving, resulting in a 27.8% increase in biomass yield. A 1:2 dilution of press juice enhanced fermentation efficiency, achieving 20.2% higher biomass despite a lower initial sugar content. Cultivation at 30 °C enabled sustained substrate utilization and outperformed 40 °C fermentation, increasing final biomass by 43.4%. Nutrient supplementation with yeast extract, peptone, and glucose led to the highest biomass yield, with a 71% increase compared to unsupplemented juice. Press juice from the tetraploid variety, Explosion, consistently outperformed the diploid Honroso, especially when harvested early, reaching up to 16.62 g·L⁻¹ biomass. Early harvests promoted faster growth, while late harvests exhibited higher biomass yield coefficients due to improved sugar-to-biomass conversion. Compared to a conventional YM medium, fermentation with L. perenne press juice achieved up to a threefold increase in biomass yield. These findings highlight the potential of grass-based substrates for efficient SCP production and demonstrate how agricultural parameters like variety and harvest timing influence both quantity and quality. The approach supports circular bioeconomy strategies by valorising underutilized biomass through microbial fermentation.

1. Introduction

With global population growth, the demand for sustainable, high-quality protein sources has risen. Traditional protein sources, such as soybean meal and fishmeal, are essential in food and feed industries but pose environmental challenges, including deforestation, overfishing, and greenhouse gas emissions [1]. In light of these issues, Single-Cell Protein (SCP) is a promising alternative, using microbial fermentation to generate protein-rich biomass from renewable resources [2]. SCP production offers several advantages, such as high protein content, rapid microbial growth, and the ability to utilize organic waste and agricultural residues [3]. Moreover, in contrast to conventional agriculture, SCP production requires far less land and water while maintaining high conversion efficiency, making it a viable solution for sustainable food and feed applications [4,5].

Yeast-based SCP production has recently gained attention for its superior protein quality, rapid growth, and well-established fermentation techniques [6]. Kluyveromyces marxianus shows strong potential for SCP production due to its thermotolerance, broad sugar metabolism, and GRAS (Generally Recognized As Safe) status [7,8]. Another major advantage of K. marxianus is its ability to metabolize glucose, xylose, and fructose, making it adaptable to diverse feedstocks [9,10]. Studies have investigated K. marxianus fermentation using dairy by-products [11]. Nayeem et al. reported that whey is an effective substrate, yielding up to 36 g/L biomass with 83.3% protein content. Sugar-rich waste streams [12,13] and lignocellulosic hydrolysates [7] have also shown promising potential. Molasses-based fermentation achieved protein yields up to 28.37 g/L [14].

A key factor in SCP production is the selection of a suitable substrate that is not only nutritionally rich and cost-effective but also readily available and environmentally sustainable. A variety of feedstocks have been investigated, including agricultural by-products, food waste, and lignocellulosic biomass [3,15]. However, many of these substrates present significant drawbacks. For example, fruit and vegetable waste and starch-rich residues, while high in carbohydrates, require enzymatic hydrolysis, as Kluyveromyces marxianus cannot directly utilize them [16,17]. Likewise, lignocellulosic materials require energy-intensive pretreatment to remove lignin, raising production costs and lowering process efficiency [7,18].

In contrast, Lolium perenne (perennial ryegrass) emerges as a particularly promising candidate. It is one of the most widely cultivated grasses in Europe, mainly valued for forage and turf applications [19], and is characterised by high biomass yields and adaptability to different climatic conditions [20]. Despite its agricultural importance, its potential for use in fermentation-based bioprocesses remains underexplored [21]. In particular, Lolium perenne press juice contains high concentrations of water-soluble carbohydrates (WSCs), amino acids, and minerals [22,23], which provide an excellent nutrient base for microbial growth. These characteristics make Lolium perenne a strong candidate for the sustainable production of SCPs, in line with both environmental objectives and industrial feasibility. Moreover, the press cake generated as a by-product during SCP production using Lolium perenne press juice, which retains a high proportion of solids, can be further utilized either as silage or directly as animal feed. This approach not only enhances overall resource efficiency but also contributes to the economic sustainability of the entire bioprocess.

This study examines Lolium perenne press juice as a sustainable medium for SCP production using Kluyveromyces marxianus. By optimizing fermentation parameters such as germ reduction methods, dilution ratios, and nutrient supplementation, the goal is to maximize microbial biomass yield while ensuring feasibility. Once optimal fermentation conditions were determined, they were used to compare Lolium perenne varieties, ploidy levels, and harvest timings. Real field yield data were incorporated to assess the feasibility of using Lolium perenne as an SCP feedstock. This study provides insights for cost-effective, scalable SCP production and improved agricultural waste valorisation. This research supports sustainability goals by reducing dependence on conventional proteins, minimizing agricultural waste, and enhancing resource efficiency. As interest in alternative proteins and the circular bioeconomy grows, this study highlights the potential of grass-based substrates for microbial fermentation, advancing SCP production and sustainable bioprocessing.

2. Materials and Methods

2.1. Raw Material

A field experiment was conducted at the Julius Kühn Institute in Braunschweig, Germany (80 m above sea level, 617 mm annual precipitation, 9.1 °C average temperature). The soil at the site is classified as silty sand with a topsoil depth of approximately 30 cm. The primary objective was to ensure a consistent and representative biomass supply of Lolium perenne for fermentation-based SCP production.

Several Lolium perenne varieties with different agronomic characteristics were grown in large field plots (7.0 m × 1.5 m) arranged in a randomised block design. Sowing took place on 25 August 2020 at a density of 1500 seeds/m². Standard agronomic management practices were applied throughout the experimental period. Basic fertilisation was carried out in 2021 using a PK fertiliser containing 14% phosphorus and 22% potassium (FOSAN SA, Szczecin, Poland), and a KornKali + Mg fertiliser with 40% potassium and 3.6% magnesium (K+S Minerals and Agriculture GmbH, Kassel, Germany). To support regrowth, calcium ammonium nitrate (KAS) with 27% nitrogen was applied in spring and after each cut (Lovochemie, Lovosice, Czech Republic). For weed control, a herbicide treatment with SIMPLEX at a rate of 2.0 L/ha was applied in early spring (Corteva Agriscience Germany GmbH, München, Germany).

In 2021, four uniform harvests were conducted, with no variation in harvest dates within the experiment. In contrast, the 2022 experiment included a variation of harvest dates for the first and second cuts to evaluate the influence of developmental stage on biomass properties relevant for SCP production. For each planned cut, two harvest dates were defined: the first took place when an early maturing reference variety reached BBCH stage 51 (beginning of inflorescence emergence) [24] and the second when a late maturing reference variety reached the same stage. All other varieties were harvested simultaneously, and their respective BBCH stages at harvest are listed in

Table 1. Lolium perenne varieties used in this study, including ploidy level, breeding company, and harvest time information.

Variety	Ploidy	Breeding Companies (City, Country)	BBCH	Harvest Time
Arvicola	4×	Freudenberger (Krefeld, Germany)	65	3 June 2022
Honroso	2×	DSV (Lippstadt, Germany)	31	early cut (E)	17 May 2022
			41	late cut (L)	31 May 2022
Explosion	4×	DSV (Lippstadt, Germany)	31	early cut (E)	17 May 2022
			41	late cut (L)	31 May 2022

.

After harvest, the biomass was vacuum sealed on site to preserve its composition and transported overnight to the FH Aachen. Upon arrival, samples were stored at −21 °C and thawed at room temperature prior to further processing and fermentation experiments.

2.2. Pretreatment and Press Juice Preparation

Frozen, vacuum-sealed grass was thawed at room temperature and cut into pieces approximately 10 cm in length. A screw press (Angel Juicer 8500S, Luba, Bad Homburg, Germany) was used to extract press juice, which was subsequently processed for use as a medium. The remaining press juice was stored at −21 °C for subsequent analyses and applications [25].

2.3. Microbial Reduction Methods

Since the press juice is a natural substance, it likely contained microorganisms that could contaminate subsequent fermentation processes. To address this, two microbial reduction methods were compared. The press juice was subjected to either autoclaving at 121 °C for 15 min using a laboratory autoclave (Systec VX-150, Systec, Osnabrück, Germany) or pasteurization at 75 °C for 1.5 h on a Stirring Hot Plate (88-1, Premiere, Sterling, VA, USA). Both methods resulted in the flocculation of substances. To remove the flocs, the cooled press juice was transferred to 50 mL centrifuge tubes and centrifuged at 10,000 rcf for 15 min using a centrifuge (MPW-380W, MPW, Warsaw, Poland). The clear supernatant was retained for further use, while the pellet was discarded [26,27].

2.4. Microorganism, Preculture

The yeast strain used was Kluyveromyces marxianus (DSM 5422), obtained from the German Collection of Microorganisms and Cell Cultures (DSMZ, Braunschweig, Germany). The starting material for this work came from cryocultures established from shake flask cultures and stored at −80 °C. For reactivation, a small portion of the thawed cryoculture was transferred onto YM agar plates using a sterile, disposable inoculating loop. Approximately 1 mL of the cryoculture was transferred under sterile conditions into a shake flask containing 100 mL of YM medium. The liquid culture was incubated at 30 °C for 48 h and then examined under a microscope to confirm the presence of Kluyveromyces marxianus and to check for contamination. The inoculated YM agar plates were incubated at 30 °C for 24 h in an incubator (UNE 500, Memmert, Schwabach, Germany) and then stored in a refrigerator at 4 °C.

For long-term storage, new −80 °C cryocultures were established from the cell suspensions of the shake flask cultures. After 24 h, 1 mL of the cell suspension was transferred into sterile 2 mL cryotubes and mixed with 1 mL of 50% (v/v) glycerol.

2.5. Preparation of Media and Agar Plate

In this study, the Universal Medium for Yeasts (YM) formulation provided by DSMZ [28] was used to prepare the agar plates and liquid media. The YM agar plates were prepared for strain maintenance. The composition of YM agar included 3 g·L⁻¹ malt extract, 3 g·L⁻¹ yeast extract, 5 g·L⁻¹ soy peptone, and 10 g·L⁻¹ glucose. The agar (15 g·L⁻¹) was dissolved in 1 L of deionized water, after which the remaining medium components were added. The medium was autoclaved at 121 °C for 15 min using a laboratory autoclave (Systec VX-150, Systec, Osnabrück, Germany). After autoclaving, the YM agar plates were poured and stored at 4 °C until inoculation [28].

When grass press juice was used as a medium, the germ-reduced juice was thawed at 4 °C 1 day before inoculating the main culture. Depending on the experiment, the press juice was either diluted 1:2 with autoclaved deionized water under sterile conditions or used directly as a medium.

Standard YM media were prepared by dissolving 3 g·L⁻¹ malt extract, 3 g·L⁻¹ yeast extract, 5 g·L⁻¹ soy peptone, and 10 g·L⁻¹ glucose in 1 L of deionized water. The medium was autoclaved at 121 °C for 15 min. The autoclaved medium was stored at 4 °C until use [28].

2.6. Optimization Experiments in Shake Flasks

Experiments aimed at optimizing media composition, temperature, and comparing the productivity of different grass varieties were conducted in shake flasks. For these experiments, 500 mL baffled flasks with triple baffles and a filling volume of 100 mL were utilized. The culture from a YM agar plate was aseptically inoculated into the flask using a sterile disposable loop. The flask was sealed with a sterile cotton plug and incubated in a shaking incubator (KS4000, IKA, Breisgau, Germany) at 120 rpm with a 100 mm shaking diameter and a temperature of either 30 °C or 40 °C for 12 h [29]. The optical density at 600 nm (OD600) of the preculture was measured, and sufficient medium and preculture were added to the main culture flasks to adjust the initial OD600 to 1 [30]. The main cultures were incubated under identical conditions for 48 h. Three replicate flasks were prepared for each experimental condition. Samples were collected hourly during the first day, and subsequently at 24 and 48 h. For each sampling, 3 mL of culture was aseptically transferred to a 15 mL centrifuge tube. Each sample was divided into two 1.5 mL microcentrifuge tubes; one was frozen at −18 °C as a backup. The optical density of the second tube was measured, after which it was centrifuged at 16,100 rcf for 15 min using a centrifuge (5415 D, Eppendorf AG, Hamburg, Germany). The supernatant was frozen for analysis, while the pellet was either discarded or used for dry cell mass determination.

2.7. Analytical Methods

To measure the water-soluble carbohydrates (WSC) in the liquid samples, the sample was first centrifuged at 16,100 rcf for 15 min (5415 D, Eppendorf AG, Hamburg, Germany). Afterward, the samples were diluted to a concentration within the calibration range and filtered through a 0.22 µm pore size polyethersulfone filter (Wicom Heppenheim, Germany). The sample was then analysed using HPLC (High-Performance Liquid Chromatography). The HPLC (1100 Series, Agilent Technologies, Santa Clara, CA, USA) was equipped with a Repromer H column (300 × 8 mm, Dr. Maisch, Ammerbuch, Germany) at 30 °C and a refractive index detector (1260 Infinity II, Agilent Technologies, Santa Clara, CA, USA) at 35 °C. The mobile phase was 5 mM H₂SO₄, and the flow rate was set at 0.6 mL·min⁻¹ [25,31,32].

To measure the optical density (OD600), a UV-Vis spectrophotometer (Ultrospec 2100 pro, Amersham BioScience, Amersham, UK) was used at a wavelength of 600 nm. First, 1 mL of the sample was taken using a pipette and diluted with deionized water until the absorbance was between 0.3 and 0.75. The absorbance was then measured, and the actual optical density was calculated by multiplying the measured absorbance by the dilution factor [33].

Protein concentration was determined indirectly via total nitrogen measurement. Nitrogen analysis was performed through combustion analysis using an Elementar NSX-2100H with an ND-210 nitrogen detector (Envirosciences, Düsseldorf, Germany). Nitrogen content was converted to protein using a standard factor of 6.25. For sample preparation, 1 mL was centrifuged at 16,100 rcf for 15 min using a 5415 D centrifuge (Eppendorf AG, Hamburg, Germany). The resulting pellet was dried at 50 °C in a UNE 500 oven (Memmert GmbH, Schwabach, Germany) until constant mass. Samples were then diluted with deionized water to fit the instrument’s measurement range before analysis [34,35,36].

3. Results and Discussion

3.1. Parameter Optimization for Fermentation

3.1.1. Optimizing Microbial Reduction Methods for Fermentation

To assess the impact of microbial reduction on fermentation performance, the two treatment methods described in Section 2.3 were compared directly. Prior to treatment, the press juice (PJ) was diluted 1:2 with deionized water (i.e., one part press juice mixed with one part water, resulting in a final dilution factor of 2).

After germ reduction, the diluted press juice was fermented at 30 °C for 48 h with Kluyveromyces marxianus to assess the impact of each germ-reduction method. Both germ-reduction methods led to a colour change from green to brown and flocculent substance formation, indicating the occurrence of the Maillard reaction. The results demonstrated that K. marxianus was successfully fermented in both types of diluted press juice without additional supplements. Notably, the growth rate of K. marxianus (Figure 1) was higher in the pasteurized PJ (0.457 h⁻¹) than in the autoclaved PJ (0.363 h⁻¹). This difference is likely attributable to the higher temperatures (121 °C) used during autoclaving, which may have damaged essential enzymes and vitamins required by K. marxianus for fermentation. Additionally, high temperatures may have caused sugar degradation into furfural and hydroxymethylfurfural (HMF), both known inhibitors of K. marxianus growth [37,38,39]. Furthermore, more extensive Maillard reactions in the autoclaved PJ may have further reduced its suitability as a fermentation medium [40]. After fermentation, the final biomass concentration in autoclaved PJ was 13.15 ± 0.76 g·L⁻¹, a 27.8% reduction compared to 18.20 ± 0.87 g·L⁻¹ in pasteurized PJ. Given these results, pasteurization at 75 °C for 90 min will be used as the sole germ reduction method in future experiments to preserve fermentation medium quality.

3.1.2. Optimizing Press Juice Dilution Ratios for Fermentation

The optimal concentration and dilution ratio of press juice for the fermentation of K. marxianus was investigated using undiluted press juice and a 1:2 dilution. Both solutions were used as media to culture K. marxianus at 30 °C for 48 h.

The experimental results revealed that the undiluted press juice had an initial total sugar concentration of 35.58 ± 1.55 g·L⁻¹ and a growth rate of 0.401 h⁻¹ and reached its maximum biomass concentration at 48 h, with a biomass increase of 13.81 ± 0.67 g·L⁻¹. In comparison, the 1:2 diluted press juice (Figure 2B) exhibited half the initial total sugar concentration (17.57 ± 0.35 g·L⁻¹) but demonstrated an 18.7% higher growth rate of 0.476 h⁻¹. Additionally, the diluted press juice achieved a biomass increase of 16.61 ± 0.49 g·L⁻¹ after 48 h, which was 20.2% higher than that of the undiluted press juice. Both media initially displayed an increase in fructose and xylose concentrations during the first few hours of fermentation, peaking at the third hour before gradually declining. The maximum sugar concentration in the undiluted press juice (Figure 2C) reached 22.73 ± 0.32 g·L⁻¹, whereas in the diluted press juice (Figure 2D), it peaked at 10.97 ± 0.11 g·L⁻¹. This initial increase was likely attributed to the release of fructose from fructans by K. marxianus.

Sugar consumption in the undiluted press juice slowed significantly after 24 h, leaving a residual sugar concentration of 3.57 g·L⁻¹ at the end of the 48 h fermentation period. In contrast, sugar consumption in the diluted press juice decelerated after 10 h but was fully depleted by 24 h, with the total sugar concentration dropping to 0 g·L⁻¹.

Overall, despite its lower initial substrate concentration, diluted press juice supported higher growth rates, yielding 2.8 g·L⁻¹ (20%) more biomass than undiluted press juice. This was particularly evident in the biomass yield coefficient relative to the consumed substrate, with the undiluted press juice yielding 0.42 ± 0.03 g·g⁻¹ compared to 1.00 ± 0.05 g·g⁻¹ for the diluted press juice. A possible explanation is the presence of phenolic compounds in the press juice, which can inhibit K. marxianus growth. Özköse et al. and Oliva et al. reported that phenolic concentrations above 0.3 g·L⁻¹ significantly suppress K. marxianus [23,41]. In Lolium perenne press juice, phenolic levels typically range from 0.17 to 0.4 g·L⁻¹ Dilution likely reduces these inhibitory compounds, contributing to the improved fermentation performance in diluted samples. Therefore, a 1:2 dilution was chosen for future experiments.

3.1.3. Optimizing Temperature for Fermentation

Temperature is a critical factor influencing biomass growth rate and final yield during fermentation. To identify the optimal temperature for maximizing K. marxianus biomass production, a 48 h experiment was conducted using a DSMZ-recommended YM medium at 30 °C and 40 °C.

A comparative analysis clearly demonstrated that at 40 °C (Figure 3), biomass concentration increased more rapidly than at 30 °C. The growth rate at 40 °C was 0.632 h⁻¹, significantly higher than the 0.417 h⁻¹ observed at 30 °C. These results align with studies by Grba et al. (2002) and Sampaio and Spencer-Martins (1989), who reported similar trends in deproteinized whey- and lactose-based media [42,43]. Their findings suggest that elevated temperatures enhance K. marxianus growth. Faster growth at 40 °C was also evident in glucose consumption, with depletion occurring significantly quicker than at 30 °C. At 40 °C, glucose concentration dropped to 0 g·L⁻¹ within 6 h, whereas at 30 °C, complete glucose depletion required 24 h.

At 30 °C, biomass concentration reached its maximum after 24 h, with a biomass increase of 3.12 g·L⁻¹. In contrast, at 40 °C, growth slowed significantly after 6 h, and the maximum biomass concentration was reached only after 48 h, with a minimal increase of just 0.61 g·L⁻¹. This was due to the complete depletion of available glucose at 40 °C after 6 h, leaving no substrate for further growth. In comparison, at 30 °C, 1.55 g·L⁻¹ of glucose remained available, supporting continued biomass production.

The maximum biomass yield at 30 °C was 5.22 ± 0.17 g·L⁻¹, which was 43.4 % higher than the 3.64 ± 0.13 g·L⁻¹ observed at 40 °C, representing an absolute increase of 1.58 ± 0.04 g·L⁻¹. Given that biomass production is the primary objective of this study, optimizing conditions to maximize biomass yield was essential. Therefore, 30 °C was selected as the fermentation temperature for all subsequent experiments.

3.2. Effects of Press Juice Supplementation on K. marxianus Growth Rate, Biomass Yield, and Substrate Consumption

Previous experiments have shown that Lolium perenne press juice is a promising medium for K. marxianus fermentation. To further examine how additional nutrient supplementation affects growth rate and biomass yield in K. marxianus fermentation using Lolium perenne press juice as the base medium, the following experiments were conducted. The press juice from the Agaska variety was chosen as the base medium, with various concentrations of yeast extract, malt extract, peptone, and glucose added (

Table 2. Composition of supplemented Lolium perenne press juice media for K. marxianus fermentation.

Abbreviation	Samples	Press Juice Dilution	Substance Concentration
PJ	Press juice	Diluted 1:2 with deionized water	-
PJ-Y	Press juice + Yeast extract		Yeast extract 3 g·L−1
PJ-P	Press juice + Peptone		Peptone 5 g·L−1
PJ-G	Press juice + Glucose		Glucose 10 g·L−1
PJ-YP	Press juice + Yeast extract + Peptone		Yeast extract 3 g·L−1 Peptone 5 g·L−1
PJ-YPG	Press juice + Yeast extract + Peptone + Glucose		Yeast extract 3 g·L−1 Peptone 5 g·L−1 Glucose 10 g·L−1

). These compounds are essential components of the DSMZ-recommended YM medium and play a crucial role in supporting K. marxianus growth. The fermentation conditions were maintained based on previously established optimal parameters.

First, three representative experiments were conducted to compare the effects of different supplements on the growth rate and biomass yield of K. marxianus. Specifically, the comparison involved the use of pure press juice, press juice supplemented with glucose, and press juice supplemented with yeast extract, peptone, and glucose, as these combinations showed the most significant improvements.

The growth curves, along with biomass concentrations and quantified substrates (Figure 4), indicate similar growth patterns across the three media. After approximately 7 h, all three media shifted from the exponential phase to the stationary phase. The differences in growth rates among the three groups were minor. The pure press juice sample had the highest growth rate (0.490 h⁻¹), whereas the sample supplemented with yeast extract, peptone, and glucose exhibited the lowest growth rate (0.422 h⁻¹). The glucose-supplemented sample exhibited a slightly lower growth rate (0.479 h⁻¹) than the pure press juice sample. For comparison, the growth rate of K. marxianus under different fermentation conditions typically falls within the range of 0.3–0.7 h⁻¹ [44].

In terms of substrate consumption, the initial glucose concentration in the pure press juice medium was 2.91 g·L⁻¹, while the initial fructose and xylose concentration was 11.40 g·L⁻¹. Glucose was completely consumed within 5 h, while fructose and xylose concentrations initially rose to 15.33 g·L⁻¹ at 3 h, before gradually decreasing to 1.6 g·L⁻¹ at 7 h and, eventually, to 0.65 g·L⁻¹ by the end of fermentation. The biomass concentration increased from 0.47 g·L⁻¹ to 4.41 g·L⁻¹ after 7 h, reaching a maximum of 9.42 g·L⁻¹ at the end of the fermentation.

When the press juice was supplemented with glucose, the initial glucose concentration was 11.52 g·L⁻¹, which was fully consumed within 7 h. The fructose and xylose concentrations followed a similar pattern to those in the pure press juice medium, starting at 12.82 g·L⁻¹, peaking at 16.57 g·L⁻¹ at 4 h, and then decreasing to 2.22 g·L⁻¹ at 7 h and 0.60 g·L⁻¹ by the end of fermentation. This suggests that glucose and fructose (plus xylose) were consumed simultaneously, with glucose being depleted more rapidly, consistent with findings in the literature. The biomass concentration increased from 0.44 g·L⁻¹ to 2.22 g·L⁻¹ after 7 h, eventually reaching a maximum of 11.58 g·L⁻¹.

For the press juice supplemented with yeast extract, peptone, and glucose, the initial glucose concentration was 12.20 g·L⁻¹, which was completely consumed within 6 h. The fructose and xylose concentrations followed a similar trend, starting at 13.33 g·L⁻¹, peaking at 16.91 g·L⁻¹ at 4 h, and being fully consumed by the end of fermentation. The biomass concentration increased from 0.50 g·L⁻¹ to 5.54 g·L⁻¹ after 5 h, reaching a maximum of 16.00 g·L⁻¹.

Experimental comparisons (

Table 3. Effects of different supplementations in press juice on growth rate and maximum biomass concentration, the biomass increase relative to pure press juice (PJ).

Samples	Growth Rate [h−1]	Maximum Biomass Concentration [g·L−1]	Increase in Biomass
PJ	0.490	9.42	-
PJ-Y	0.511	10.47	11%
PJ-P	0.512	10.25	8.8%
PJ-G	0.479	11.58	23%
PJ-YP	0.497	11.80	25%
PJ-YPG	0.422	16.00	71%

) indicate that supplementing press juice significantly enhances final biomass yield. The most substantial improvement was observed in the group supplemented with yeast extract, peptone, and glucose (PJ-YPG), which achieved a 71% increase in biomass yield compared to the control group using pure press juice. Statistical analysis confirmed glucose’s significant contribution (p = 0.036), indicating a strong correlation with biomass concentration. Although yeast extract (p = 0.132) and peptone (p = 0.106) were not statistically significant (p < 0.5), both positively influenced biomass production, with peptone showing a stronger effect.

Interestingly, the peptone-only group exhibited the highest growth rate (0.512 h⁻¹), but this did not result in a higher biomass yield—it reached only 10.25 g·L⁻¹, 56% lower than PJ-YPG. A similar trend observed in earlier trials reinforces that growth rate alone is not a reliable indicator of fermentation efficiency. Final biomass yield remains the most meaningful metric. While yeast extract and peptone individually offered modest improvements, their combined use with glucose further boosted biomass output. The PJ-YPG group ultimately achieved the highest biomass concentration (16.00 g·L⁻¹) across all shake flask experiments.

3.3. Effect of Lolium perenne Varieties and Harvest Time on Biomass Production

To investigate the influence of variety and harvest time on biomass production, press juice from two Lolium perenne varieties, Honroso (diploid) and Explosion (tetraploid), was used. Each variety was harvested at two developmental stages in 2022: an early cut (E) and a late cut (L), spaced 14 days apart, as described in Section 2.1.

During the 48 h fermentation period (Figure 5), all four samples (HonrosoE, HonrosoL, ExplosionE, ExplosionL) showed similar growth patterns in the first 7 hours, with slight variations in growth rates. Within each variety, early-harvested samples grew faster than late-harvested ones. The tetraploid variety, Explosion, consistently outperformed the diploid variety, Honroso, regardless of harvest timing, with a growth rate advantage of approximately 0.05 h⁻¹. Notably, only the samples with Explosion press juice showed a higher growth rate than the YM-medium control (0.460 h⁻¹), whereas samples with Honroso press juice had a lower growth rate than the control. Explosion (Early Cut) achieved the highest biomass concentration, reaching 16.62 ± 0.49 g·L⁻¹ after 48 h. Early-harvested varieties consistently yielded, on average, 10.8% more biomass than late-harvested ones. Overall, the tetraploid variety, Explosion, surpassed the diploid Honroso across both harvest times, showing an 8.8% increase in biomass. All samples exhibited a remarkable 260–320% biomass yield increase compared to the YM-medium control group. Statistical analysis based on multiple linear regression revealed that both the ploidy level of Lolium perenne varieties and harvest timing significantly influenced the maximum biomass concentration. Specifically, the tetraploid variety, Explosion (p = 0.007), and early harvest timing (p = 0.004) demonstrated statistically significant positive effects (p < 0.05) on enhancing maximum biomass yield.

Biomass yield was further evaluated based on actual sugar (glucose, fructose, and xylose) consumption, allowing for the calculation of the biomass yield coefficient for each sample (

Table 4. Effect of Lolium perenne varieties and harvest time on growth rate, maximum biomass concentration, and biomass yield coefficient. Error indicates standard deviations of the mean (n = 3).

Samples	Growth Rate [h−1]	Maximum Biomass Concentration [g·L−1]	Biomass Yield Coefficient [g·g−1]
ExplosionE	0.483	16.62 ± 0.49	0.996 ± 0.049
ExplosionL	0.482	15.05 ± 0.90	1.174 ± 0.102
HonrosoE	0.433	15.32 ± 0.23	0.958 ± 0.038
HonrosoL	0.429	13.79 ± 0.49	1.199 ± 0.060
YM Media	0.460	5.22 ± 0.17	0.510 ± 0.030

). An interesting trend was observed: early-harvested samples produced more biomass but also had a higher initial total sugar content than late-harvested samples. As a result, early-harvested samples had a lower biomass yield coefficient than late-harvested ones. Among the tested varieties, Honroso (late harvest) showed the greatest increase in biomass yield coefficient (25.1%), followed by Explosion (late harvest) with a 17.8% increase. In contrast, the YM-medium control group had the lowest efficiency, with a biomass yield coefficient of just 0.510 g·g⁻¹. This may be because the YM medium contains only glucose, while the press juice medium also provides fructose and xylose, both metabolized by K. marxianus. Limited sugar availability in the YM medium likely constrained biomass formation, reducing both total biomass yield and its yield coefficient.

3.4. Evaluation of Field Yield and Harvest Timing Effects on Biomass and SCP Production in Different Lolium perenne Varieties

In agricultural production, optimizing harvest time is a key factor for maximizing biomass yield. However, changes in harvest timing can also affect the quality of the harvested material, which in turn influences its suitability for downstream applications, such as fermentation-based SCP production. To evaluate these interactions, two Lolium perenne varieties were compared with respect to their field yields (

Table 5. Field yield and biomass production in different Lolium perenne varieties under varying harvest timing. Error indicates standard deviations of the mean (n = 2).

Samples	Field Yield [t WM·ha−1]	Press Juice Yield * [m3·ha−1]	Maximum Biomass Concentration [g·L−1]	Biomass Yield [kg·ha−1]
ExplosionE	15.9 ± 1.9	6.8 ± 0.8	16.62 ± 0.49	113.3 ± 13.9
ExplosionL	16.9 ± 0.4	7.3 ± 0.2	15.05 ± 0.90	109.3 ± 7.1
HonrosoE	12.3 ± 0.2	5.3 ± 0.1	15.32 ± 0.23	81.0 ± 1.8
HonrosoL	14.2 ± 0.2	6.1 ± 0.3	13.79 ± 0.49	84.3 ± 4.8

* Conversion of press juice from tons to cubic meters, based on a density of 1 ton per cubic meter.

) and SCP production (

Table 6. Field yield and SCP production in different Lolium perenne varieties under varying harvest timing. Error indicates standard deviations of the mean (n = 2).

Samples	Biomass Yield [kg·ha−1]	SCP Content [%]	SCP Yield [kg·ha−1]
ExplosionE	113.3 ± 13.9	19.1 ± 0.9	21.6 ± 3.7
ExplosionL	109.3 ± 7.1	18.9 ± 1.2	20.7 ± 2.7
HonrosoE	81.0 ± 1.8	18.6 ± 1.8	15.0 ± 1.8
HonrosoL	84.3 ± 4.8	20.9 ± 0.9	17.6 ± 1.8

) potential under early and late harvest conditions. Since this study used press juice, a conversion factor is needed to quantify the transformation of fresh raw material into PJ. Measurements showed that 43% of Lolium perenne wet mass (WM) was extracted as press juice. This conversion rate enables the calculation of press juice yield per unit of farmland.

A wet mass (WM) yield comparison showed that both varieties benefited from late harvesting, albeit to different degrees. The tetraploid variety, Explosion, increased its WM yield by 1.0 t WM·ha⁻¹ (6.5%) compared to the early harvest. In contrast, the diploid variety, Honroso, had a greater increase of 1.9 t WM·ha⁻¹ (15.7%). Despite Honroso’s larger relative increase, Explosion retained a higher total WM yield, exceeding Honroso by 23.8% on average. This highlights the biomass production advantage of tetraploid varieties. Notably, moisture content varied between harvest timings. Early-harvested samples averaged 26.3% moisture, while late-harvested samples reached 29.9%, a relative increase of 3.59%. The higher moisture content in late-harvested samples benefits press juice extraction, potentially enhancing juice yield.

A comparison of final biomass yield per hectare revealed distinct trends. The tetraploid variety, Explosion, had the highest biomass yield, reaching 113.3 kg·ha⁻¹ in the early harvest. However, a delayed harvest resulted in a slight yield reduction of 4%. In contrast, the diploid variety saw a 4% yield increase with late harvesting, reaching 84.3 kg·ha⁻¹. Despite yield variations, tetraploid varieties averaged 34.6% more biomass per hectare than diploid varieties, reinforcing their overall advantage. Statistical analysis reaffirmed that tetraploid Lolium perenne varieties significantly increased biomass yield (p < 0.001), while harvest timing had no significant effect (p = 0.914).

Combustion-based analysis revealed an average protein content of 19.4% in the produced biomass. This is lower than the approximate 40% protein content commonly reported for K. marxianus grown on molasses or whey-based substrates. However, Lolium perenne press juice has a more complex composition than the defined media used in those studies. Moreover, differences in fermentation strategies across studies further limit direct comparisons [11,14]. Thus, protein content alone should not be considered a definitive measure of substrate suitability or process performance in this context.

Statistical analysis indicated no significant correlation between SCP content and either ploidy level or harvest timing (p > 0.05). However, the SCP yield per hectare followed trends consistent with previously observed biomass production. The tetraploid variety, Explosion, showed a clear advantage, with its early cut yielding the highest SCP at 21.6 kg·ha⁻¹. On average, tetraploid varieties yielded 29.7% more SCP per hectare than diploids, independent of harvest timing. Harvest timing had no statistically significant effect on SCP yield.

In conclusion, field data confirmed that tetraploid Lolium perenne varieties consistently outperformed diploids in both biomass and SCP yield per hectare. Delayed harvesting had a slight, inconsistent impact on yield. Protein content in the biomass was unaffected by ploidy level or harvest timing. Therefore, in agricultural practice, harvest timing should be optimized alongside production planning and cost efficiency to maximize both economic and environmental benefits.

4. Conclusions

This study investigated Lolium perenne press juice as a sustainable medium for producing Single-Cell Protein (SCP) using Kluyveromyces marxianus. Key parameters, including germ reduction methods, dilution ratios, fermentation temperature, nutrient supplementation, and effects of different grass varieties and harvest timings, were systematically evaluated. Pasteurization at moderate temperatures (75 °C) proved superior to autoclaving, preserving nutrients essential for microbial growth and significantly enhancing biomass yields. Dilution of the press juice (1:2) also improved fermentation efficiency, producing approximately 20% more biomass than undiluted juice despite lower initial sugar levels, underscoring the importance of optimal nutrient concentrations for microbial productivity. Temperature significantly impacted fermentation outcomes; while K. marxianus initially grew faster at 40 °C, rapid glucose depletion limited biomass formation. In contrast, fermentation at 30 °C allowed sustained substrate utilization, resulting in greater overall biomass.

Supplementing press juice with yeast extract, peptone, and glucose notably enhanced biomass production, achieving approximately 71% higher yields compared to unsupplemented juice. Among these, glucose supplementation had the most pronounced effect and demonstrated clear statistical significance. Although supplementation only moderately improved growth rate, its primary benefit was evident in significantly increased final biomass yields, highlighting nutrient supplementation’s critical role in SCP production. Comparing Lolium perenne varieties and harvest timings revealed that the tetraploid variety, Explosion, consistently yielded higher biomass than the diploid, Honroso. Early harvests resulted in higher biomass concentrations, whereas late harvests showed improved sugar-to-biomass conversion efficiency. Field trials further validated the advantage of tetraploid varieties, which yielded approximately 35% more biomass per hectare and about 19% more SCP per hectare compared to diploid varieties. In terms of protein content within the biomass, statistical analysis showed no significant influence from either ploidy level or harvest timing. Notably, fermentation performance using Lolium perenne press juice significantly surpassed the traditional YM medium, achieving approximately three times higher biomass yields, clearly demonstrating the substantial advantages of grass-based substrates.

In conclusion, this research confirms Lolium perenne press juice’s suitability as a fermentation substrate for SCP production, presenting an effective strategy for sustainable protein production and agricultural waste valorisation. These findings underscore the potential of grass-based substrates within circular bioeconomy frameworks, addressing the growing demand for sustainable alternative proteins.