Enhancing Sensory Complexity in Porter-Style Beer via Sequential Inoculation with Non-Saccharomyces Yeasts

Abstract

The diversification of craft beer styles has stimulated interest in innovative yeast-driven strategies to enhance sensory complexity while maintaining process robustness and stylistic integrity. In this context, non-Saccharomyces yeasts represent promising biotechnological tools for modulating fermentation performance and flavor development in brewing systems. This study evaluated the application of Lachancea thermotolerans and Torulaspora delbrueckii in the production of a Porter-style beer using sequential inoculation with Saccharomyces cerevisiae. All fermentations were conducted in triplicate from a wort with an original gravity of 1042. The final alcohol content ranged from 4.82 to 4.99% (v/v), and apparent attenuation varied between 84.1 and 88.9%, with no significant differences among treatments (p > 0.05). Color (92–94 European Brewery Convention (EBC) and bitterness (~18 International Bitterness Units (IBU) remained within Porter-style parameters across all fermentations. Total acidity ranged from 0.19 to 0.21% (lactic acid equivalents), while volatile acidity was significantly higher in the L. thermotolerans treatment (0.55 g L⁻¹) compared with the control (0.22 g L⁻¹) (p < 0.05). Sequential inoculation influenced early fermentation kinetics and modulated selected sensory attributes. Quantitative Descriptive Analysis (n = 18 panelists) indicated higher aroma intensity and foam quantity in beers produced with L. thermotolerans, whereas T. delbrueckii was associated with moderate increases in foam persistence. The roasted character and overall stylistic perception remained stable across treatments. These findings indicate that sequential inoculation with selected non-Saccharomyces yeasts enables measurable sensory differentiation in dark beer matrices without compromising fermentative performance or stylistic integrity. The results support their controlled integration as technological tools for sensory innovation in Porter-style beers.

1. Introduction

The brewing industry has traditionally relied on Saccharomyces cerevisiae as the primary fermentative microorganism due to its robustness, predictability, and high ethanol production efficiency. While this technological dominance has ensured process reliability, it has also contributed to relatively standardized sensory profiles across beer styles, limiting aromatic diversity and product differentiation. In recent years, changing consumer preferences and the expansion of the craft beer sector have driven interest in innovative fermentation strategies aimed at enhancing sensory complexity and product differentiation [1,2].

From a global perspective, the craft beer market has experienced sustained growth over the past decade, driven by the diversification of beer styles, the premiumization of consumption, and a growing consumer preference for products associated with authenticity and origin. This expansion has intensified competition within the sector, prompting craft breweries to explore technological innovations beyond conventional brewing practices. In parallel, more consumers are seeking beers with lower alcohol content, and this is motivated by health-conscious consumption patterns and lifestyle choices that favor moderation without compromising sensory quality. In this context, the development of beers combining distinctive sensory profiles with reduced alcohol levels has become a key factor for market positioning and competitiveness. Market intelligence reports project continued global growth of the craft beer segment over the coming years, reflecting stable and positive trends across different regions [3,4]. Beyond alcohol modulation, a major technological challenge in craft brewing lies in achieving sensory differentiation while maintaining process robustness and stylistic integrity. In this regard, yeast selection has emerged as a critical lever for innovation, as metabolic diversity among yeast species directly shapes volatile composition, organic acid production, mouthfeel, and foam stability. Therefore, strategies that expand yeast biodiversity in controlled fermentation systems represent a rational approach to sensory diversification without fundamentally altering beer identity.

While the application of non-Saccharomyces yeasts has been extensively explored in winemaking [1,2,5] and, more recently, in brewing systems including hop-forward and specialty styles [1,2,6], most experimental studies have focused on pale malt-based worts. Their behavior in dark, malt-forward beers remains comparatively undercharacterized. Porter-style beers are defined by the use of roasted specialty malts, leading to elevated concentrations of Maillard-derived compounds, melanoidins, and phenolic constituents [7,8,9], which influence fermentability, redox balance, and flavor stability. These matrix components may interact with yeast-derived volatile esters, higher alcohols, and organic acids, potentially modifying both fermentation performance and sensory perception [6,10,11,12,13,14]. Consequently, targeted evaluation of non-Saccharomyces yeasts under porter-style brewing conditions is warranted.

In this context, non-Saccharomyces yeasts have gained attention as valuable biotechnological tools capable of modulating aroma, acidity, mouthfeel, and foam-related attributes. Initially explored in winemaking, these yeasts were long considered spoilage organisms; however, extensive research has demonstrated their positive contributions when applied in a controlled manner [2,5]. Importantly, recent studies conducted directly under brewing conditions have confirmed that native non-Saccharomyces yeasts can significantly influence beer volatile composition and aroma complexity, either in pure or mixed fermentations with S. cerevisiae [11,12]. Their ability to produce distinctive volatile compounds, organic acids, and secondary metabolites positions non-Saccharomyces yeasts as effective tools to diversify beer sensory profiles beyond traditional boundaries [5,13].

One of the most effective approaches to harness the metabolic potential of non-Saccharomyces yeasts is sequential inoculation. This strategy allows non-Saccharomyces species to dominate the early stages of fermentation, contributing to aroma and structure, while S. cerevisiae ensures fermentation completion and process stability [14,15,16]. Sequential inoculation has been shown to improve sensory complexity while mitigating risks such as sluggish fermentations or excessive residual sugars, making it particularly attractive for brewing applications [16,17,18]. From a biotechnological perspective, sequential inoculation allows temporal partitioning of metabolic activities, reducing direct competitive pressure between species and facilitating the expression of early-stage metabolites that may otherwise be suppressed in co-inoculation systems. This approach is particularly relevant in brewing, where fermentation kinetics, attenuation degree, and physicochemical stability must be preserved. However, experimental data evaluating sequential strategies under Porter-style brewing conditions remain limited.

Among the non-Saccharomyces species studied, Torulaspora delbrueckii and Lachancea thermotolerans stand out due to their increasing relevance in brewing research and their potential applicability under controlled fermentation conditions. Although these species have been extensively studied in wine fermentation, their application in brewing—particularly in dark beer styles—remains comparatively underexplored [1,18,19,20]. Torulaspora delbrueckii and Lachancea thermotolerans were selected for this study based on prior evidence indicating their technological compatibility with brewing conditions and their growing interest as non-Saccharomyces candidates for mixed or sequential inoculation strategies in craft beer production [21,22,23,24].

Therefore, the present study aims to evaluate the effects of sequential inoculation with Torulaspora delbrueckii and Lachancea thermotolerans on fermentation kinetics, physicochemical parameters, and descriptive sensory attributes of a Porter-style beer, compared with conventional single inoculation using Saccharomyces cerevisiae. By integrating fermentation monitoring, chemical analyses, qualitative yeast presence assessment, and quantitative descriptive sensory evaluation, this work seeks to provide experimental evidence on the feasibility and sensory impact of sequential non-Saccharomyces strategies in dark beer production.

2. Materials and Methods

2.1. Yeast Strains and Inoculation Strategy

Three yeast strains were used in this study: a commercial Saccharomyces cerevisiae ale strain, SafAle^TM US-05, from Fermentis Marcq-en-Barœul, France and two wine non-Saccharomyces yeasts, Torulaspora delbrueckii Level 2 Biodiva^TM and Lachancea thermotolerans Level 2 Laktia^TM, from Lallemand. Inc., Montréal, QC, Canada. Non-Saccharomyces strains were selected based on their documented technological compatibility with brewing conditions, their increasing relevance in mixed or sequential fermentation strategies for craft beer production, and their inclusion in commercial yeast catalogs traditionally used for wine fermentation [6,21].

The experimental design comprised three fermentation treatments performed in triplicate, resulting in a total of nine independent fermentations (three fermenters per treatment). All fermentations were conducted under identical brewing conditions to ensure technical consistency and comparability among treatments (Figure 1). Two treatments were based on a sequential inoculation strategy involving a non-Saccharomyces yeast followed by Saccharomyces cerevisiae, whereas a single inoculation with S. cerevisiae served as the control. The control treatment employed SafAle™ US-05 (Saccharomyces cerevisiae), a commercial ale strain. Sequential treatments (Treatment 1 and Treatment 2) included the wine yeast strains LEVEL² BIODIVA™ (Torulaspora delbrueckii; and LEVEL² LAKTIA™ (Lachancea thermotolerans)).

Dry yeast preparations contained a minimum of 1 × 10¹⁰ colony forming units (CFU) g⁻¹ according to manufacturer specifications. Non-Saccharomyces strains were inoculated into cooled wort at 25 g hL⁻¹ (≈2.5 × 10⁶ cells mL⁻¹). After 48 h, S. cerevisiae was inoculated at 50 g hL⁻¹ (≈5.0 × 10⁶ cells mL⁻¹). The 48 h interval allowed initial metabolic activity and partial establishment of the non-Saccharomyces strains prior to the introduction of S. cerevisiae [6,21].

Treatment 1 consisted of sequential inoculation with T. delbrueckii followed by S. cerevisiae, whereas Treatment 2 consisted of sequential inoculation with L. thermotolerans followed by S. cerevisiae. The control consisted of a single inoculation of S. cerevisiae at 50 g hL⁻¹ (≈5.0 × 10⁶ cells mL⁻¹) at the beginning of fermentation.

Each treatment was performed in three independent fermenters under identical brewing conditions (Figure 1).

2.2. Wort Production and Brewing Conditions

A Porter-style wort was produced using malted barley according to standard craft brewing procedures. The grist composition consisted of 19 kg of Weyermann Pale Ale malt, 3.6 kg of Castle Malting Cara Crystal 120 malt, 1.1 kg of Patagonia Malt^® Black Pearl malt, and 170 g of Patagonia Malt roasted barley. This formulation was selected to produce a wort suitable for Porter-style beer fermentation, providing the characteristic color, body, and flavor profile associated with dark beers. Malted grains were mashed using a single-infusion method at 67 °C for 60 min. The mashing water was preheated 4–5 °C above the target temperature to compensate for heat loss upon grain addition [25]. A total of 71 L of water was used, corresponding to a liquor-to-grist ratio of 3 L kg⁻¹ [26]. Wort lautering was performed to separate the spent grains from the sweet wort. The wort was recirculated over the grain bed until a clear wort was obtained. Grain sparging was then carried out with 50 L of water at 70 °C to maximize sugar extraction, and sparging was terminated when the runoff density reached 1005 (specific gravity × 1000) [25]. A total volume of 150 L of the wort was boiled for 60 min. Hop addition was performed at the beginning of the boil stage using 186 g of Challenger hops (UK; 7–10% α-acids) to achieve bitterness through α-acid isomerization [27]. The original gravity of the wort at the end of boiling was 1042 (Specific Gravity × 1000). After boiling, the wort was rapidly cooled to 20 °C using a copper immersion chiller (purchased from Prost, Santiago, Chile). The cooled wort (150 L) was transferred to fermentation vessels under atmospheric conditions. No active aeration or oxygen injection was applied; oxygen incorporation occurred passively during wort transfer. The dissolved oxygen concentration was not measured. This procedure was consistently applied across all (nine) treatments. The wort (16.6 L per fermenter) was inoculated according to the treatment design: Saccharomyces cerevisiae at 50 g hL⁻¹ (≈5.0 × 10⁶ cells mL⁻¹). Fermentations were conducted using one commercial brewing yeast and two commercial wine yeast strains. The control treatment employed SafAle™ US-05 (Saccharomyces cerevisiae; a widely used ale yeast strain. Non-Saccharomyces treatments included the wine yeast strains LEVEL² BIODIVA™ (Torulaspora delbrueckii; and LEVEL² LAKTIA™ (Lachancea thermotolerans; both typically used in winemaking).

All dry yeast preparations contained 1 × 10¹⁰ CFU g⁻¹ according to manufacturer specifications. Non-Saccharomyces strains were inoculated at 25 g hL⁻¹, corresponding to approximately 2.5 × 10⁶ cells mL⁻¹. After 48 h, Saccharomyces cerevisiae was inoculated at 50 g hL⁻¹ (≈5.0 × 10⁶ cells mL⁻¹). The control treatment received only Saccharomyces cerevisiae at 50 g hL⁻¹ (≈5.0 × 10⁶ cells mL⁻¹). All fermentations were conducted in triplicate.

2.3. Fermentation Monitoring and Yeast Viability

Fermentation progress was monitored daily in each individual fermenter by the measuring wort density and temperature throughout the fermentation process. Yeast growth and viability were assessed using a combination of direct and indirect cell counting methods in order to characterize yeast population dynamics under the different inoculation strategies. Indirect viable cell counts were determined by plating serial dilutions of fermentation samples on Yeast extract Peptone Dextrose (YPD) agar, followed by incubation under aerobic conditions until colony formation, as previously described [28]. Cellular concentration was expressed as CFU·mL⁻¹. In parallel, the yeast cell population was determined using a 1:100 dilution of the brewing wort, which was placed in a Neubauer counting chamber (Boeco Bright-Line, Merck-Millipore, Santiago, Chile) and mixed with a 1% (m/v) aqueous methylene blue solution (Sigma-Aldrich, Santiago, Chile). Cell counting was performed by optical microscopy using a 40× objective lens (Leica DM500 trinocular BA310, Leica Microsystems, Galénica S.A., Santiago, Chile). Unstained cells were considered viable, whereas blue-stained cells were considered non-viable. Microscopic counts were used to complement plate counts and to monitor changes in yeast viability during fermentation.

2.4. Yeast Presence Analysis

Yeast implantation during fermentation was assessed using molecular fingerprinting based on amplification of the ribosomal internal transcribed spacer (ITS) region. Genomic DNA was extracted from fermentation samples collected at defined stages of the process. Fermentation samples for ITS analysis were collected at an early fermentation stage (wort density approximately 1040–1035 (SG × 1000)) and at a later stage of fermentation (densities below approximately 1010 (SG × 1000)), corresponding to the main attenuation phase and near completion of primary fermentation, respectively. The ITS region, including the 5.8S rRNA gene, was amplified by Polymerase Chain Reaction (PCR) using universal fungal primers, following established protocols for yeast discrimination in mixed fermentations [29,30]. PCR amplicons were separated by agarose gel electrophoresis, and band separation and visualization were performed according to the protocol described by Vargas-Bello-Pérez et al. [31]. Banding patterns were used to verify yeast implantation and persistence throughout fermentation.

2.5. Bottling and Bottle Conditioning

After completion of primary alcoholic fermentation, each treatment was bottled in 1 L bottles. Bottles and caps were sanitized using peracetic acid (purchased from Prost, Santiago, Chile) at a concentration of 1 mL L⁻¹ to prevent microbial contamination. Bottle conditioning was performed to achieve natural carbonation. Viable yeast cells remaining after primary fermentation carried out the secondary fermentation, and dextrose (purchased from Prost, Santiago, Chile) was added at 6.2 g L⁻¹ to each bottle containing green beer. The progress of secondary fermentation was monitored by weight loss, with bottles weighed daily until a stable weight was maintained for at least 48 h. Carbonation levels were expressed as volumes of Carbon dioxide (CO₂). Dissolved CO₂ was estimated using standard carbonation tables relating CO₂ volumes and temperature according to the beer style. For this Porter-style beer, a target carbonation level of 2.3 volumes of CO₂ was established. Residual dissolved CO₂ from primary fermentation was estimated at 0.88 volumes of CO₂ at 20 °C. Considering that each gram of added dextrose produces approximately 0.23 volumes of CO₂, the required priming sugar addition was calculated according to [26].

2.6. Physicochemical Analysis of Beers

At the end of fermentation, the beers were analyzed for alcohol content, pH, color, bitterness, total acidity, and volatile acidity. Alcohol content (% v/v) was determined experimentally by distillation following the method described by Schmidt-Hebbel [32], allowing a direct quantification of ethanol independently of density-based estimations. pH was measured using a calibrated digital pH meter at 20 °C after calibration with standard buffer solutions (pH 4.0 and 7.0 Merck, Darmstadt, Germany). Total titratable acidity was determined by titration with 0.1 N sodium hydroxide (NaOH; Merck, Darmstadt, Germany) to an endpoint of pH 8.2 and expressed as g L⁻¹ lactic acid equivalents. Volatile acidity was determined by steam distillation and subsequent titration of the distillate, and the results were expressed as g L⁻¹ acetic acid equivalents. All analytical determinations were performed according to standard AOAC methods [33]. Beer color was determined by ultraviolet–visible (UV–Vis) spectrophotometry at 430 nm according to the method described in [34]. Color was expressed as EBC units (EBC = 25 × A₄₃₀). Bitterness was expressed as International Bitterness Units (IBUs) and determined following the methodology reported in [34]. Wort density was measured prior to inoculation (day 0), corresponding to the original gravity (OG), and monitored daily using a densimeter until stabilization, when the final gravity (FG) was recorded. All fermentations were conducted from the same wort batch, with an initial gravity of 1042 (SG × 1000). Apparent attenuation (AA, %) was calculated to assess fermentation performance based on OG and FG values using the standard brewing equation [35].

2.7. Sensory Analysis

Sensory evaluation was conducted in three sequential phases: a focus group, a triple triangular test, and a Quantitative Descriptive Analysis (QDA).

2.7.1. Focus Group and Attribute Definition

A preliminary focus group session was conducted to establish and standardize the sensory evaluation framework. Seven assessors with previous experience in beer sensory evaluation analyzed one representative beer sample from each treatment and the control under standardized tasting conditions. The session followed a structured format to facilitate systematic attribute identification, discussion, and consensus. Through moderated discussion and agreement, eight sensory attributes characteristic of Porter-style beer were defined: color, foam quantity, aroma intensity, sherry-like character, fruity, lactic, acidity, and persistence. Clear operational definitions were established to ensure consistent interpretation among panelists during subsequent evaluations. Commercial beers were used as external references to support attribute identification and panel calibration: Budweiser (non-fruity aroma reference) and Royal Guard IPA (fruity aroma reference). These references were used exclusively for aroma calibration purposes to assist panelists in recognizing and anchoring specific attributes and were not intended to represent Porter-style beers. This approach allowed alignment of sensory perception and improved reproducibility of the descriptive analysis. The selected descriptors were aligned with established beer sensory classification systems, particularly the Beer Flavor Wheel, and widely recognized beer sensory evaluation frameworks [34]. Although a formal standardized lexicon protocol was not implemented, attribute generation followed consensus procedures among trained assessors with prior experience in beer sensory analysis, consistent with descriptive sensory methodology guidelines [36].

2.7.2. Panel Selection by the Triple Triangular Test

Panelists were selected using a triple triangular test to assess their ability to discriminate fruity aromas in beer. Twenty regular beer consumers without prior formal sensory training participated in this stage. In each repetition, panelists received three beer samples served in opaque glasses and were asked to identify the sample exhibiting a fruity aroma. Royal Guard IPA was used as the fruity aroma reference, while Budweiser served as the non-fruity reference. Panelists who correctly identified the fruity sample in all three repetitions were selected to participate in the QDA. Eighteen out of twenty assessors (90%) met this criterion and were included in the descriptive panel. Although these assessors did not have previous formal sensory training, those who met the discrimination criteria participated in a calibration session before the QDA. During this session, attribute definitions and scale usage were standardized to ensure consistent interpretation and evaluation among assessors, thereby constituting a trained descriptive panel for the quantitative analysis.

2.7.3. Quantitative Descriptive Analysis (QDA)

Quantitative Descriptive Analysis (QDA) was conducted to evaluate the sensory attributes of the control and treatment beers. Eighteen trained panelists assessed eight attributes previously defined in a focus group, grouped into visual, olfactory, and gustatory phases.

Foam-related attributes (intensity and persistence) were evaluated exclusively as perceptual descriptors within the QDA framework. No instrumental measurements of foam stability were performed, as the objective of this study was to investigate fermentation-driven sensory differences modulated by yeast metabolism rather than to conduct a physicochemical characterization of foam properties. All evaluations were carried out under standardized serving conditions (controlled temperature, uniform glassware, and standardized pouring procedure) to ensure experimental reproducibility.

Prior to evaluation, panelists participated in a calibration session to ensure consistent understanding of attribute definitions and scale usage. Attribute intensity was measured using a 15 cm unstructured line scale, where 1 corresponded to the absence of the attribute and 15 corresponded to extremely high intensity. Samples were served at 12 ± 1 °C in clear glasses that were coded with three-digit numbers to ensure blinding. Water and unsalted crackers were provided to cleanse the palate between samples and minimize carryover effects. The presentation order of the samples was randomized and balanced across panelists to control for order effects. Each panelist evaluated all samples, including the three independent fermentation replicates for each treatment. Replicate evaluations were averaged within each panelist to reduce intra-assessor variability.

2.8. Statistical Analysis

All fermentations were conducted in independent replicates. Yeast presence capacity was evaluated using descriptive analysis. The yeast population (direct and indirect counts) and wort density were analyzed using a mixed-effects model, considering treatment (T1, T2, and control) and fermentation time (days) as fixed factors and biological replicates as random effects. Prior to statistical analysis, data normality was assessed using the Shapiro–Wilk test, and homoscedasticity assumptions were verified. When significant effects were detected, multiple comparisons between treatments were performed using Tukey’s post hoc test.

Physicochemical results are presented as means ± standard deviations (n = 3 independent fermentations).

A triple triangle test was conducted in accordance with ISO 4120 standards (International Organization for Standardization, Geneva, Switzerland) [37] to assess the panelists’ ability to discriminate between samples prior to the Quantitative Descriptive Analysis (QDA). In each session, assessors received three coded samples, two identical and one different, and were asked to identify the odd sample. The minimum number of correct responses required to establish statistical significance was determined using critical values for triangle tests based on the binomial distribution (probability of correct guess = 1/3). Statistical significance was established at a confidence level of 95% (α = 0.05).

For the Quantitative Descriptive Analysis (QDA), replicate evaluations for each sensory attribute were averaged within each panelist prior to statistical analysis to reduce intra-assessor variability. The panelist was considered the experimental unit, reflecting the crossover design of the study in which all 18 trained panelists evaluated all samples. Mean intensity values for each treatment were calculated across the 18 assessors. Differences among treatments were evaluated using one-way analysis of variance (ANOVA) while considering treatment as a fixed effect. Because the analysis was conducted on panelist-averaged data, panelist effects were not modeled and were not included as either fixed or random factors in the ANOVA. Since all panelists evaluated all samples, the data present a repeated measures structure that was not explicitly modeled in the statistical analysis, and panelist effects were not included as fixed or random factors. This aspect is acknowledged as a limitation when interpreting the results. Multiple comparisons were performed using Fisher’s least significant difference (LSD) test at a significance level of α = 0.05. All statistical analyses were performed using InfoStat software (Version 2020, InfoStat Group, Universidad Nacional de Córdoba, Córdoba, Argentina).

3. Results

3.1. Yeast Population Dynamics During Fermentation

The evolution of wort density differed among inoculation strategies throughout primary fermentation (Figure 2a). The control fermentation inoculated solely with Saccharomyces cerevisiae exhibited the fastest decrease in density, reaching lower values earlier in the fermentation process, consistent with a rapid and efficient sugar consumption. In contrast, both sequential inoculations showed a more gradual reduction in density during the initial fermentation stages. The Torulaspora delbrueckii–S. cerevisiae sequential inoculation displayed intermediate fermentation kinetics, with density decreasing steadily and reaching final values later than the control. The Lachancea thermotolerans–S. cerevisiae treatment showed the slowest initial density decline, particularly during the first days of fermentation, followed by an accelerated decrease after S. cerevisiae inoculation. Despite these differences in fermentation dynamics, all treatments achieved a substantial reduction in density by the end of primary fermentation, indicating successful attenuation under all inoculation regimes.

Yeast population dynamics during primary fermentation were monitored using both direct microscopic counts and indirect viable counts (CFU) (Figure 2b). Across all inoculation strategies, yeast populations increased during the early stages of fermentation, reaching maximum values around day 3. The sequential inoculation treatments exhibited higher total cell concentrations during the initial fermentation phase compared with the Saccharomyces cerevisiae control, as evidenced by both direct counts and CFU measurements.

Following the peak population, a gradual decline in yeast abundance was observed in all treatments from day 6 onward, coinciding with fermentation progression. The control fermentation showed a more pronounced decrease in viable counts after day 6, whereas sequential inoculations maintained higher cell numbers for a longer period. By the end of fermentation, yeast populations decreased markedly in all treatments, with both direct and indirect methods showing convergent trends.

Yeast presence was further supported by electrophoretic separation of PCR-amplified ITS regions (Figure 3). The Saccharomyces cerevisiae control showed a single band with an apparent size of approximately 850–900 bp. In contrast, reference lanes corresponding to non-Saccharomyces yeasts displayed distinct ITS amplicon sizes, with Torulaspora delbrueckii showing a band around 780–820 bp and Lachancea thermotolerans exhibiting a smaller fragment of approximately 650–700 bp.

In sequential inoculation treatments, the presence of bands matching these approximate sizes was observed, resulting in composite banding patterns compared with the control. Mixed samples displayed combined profiles consistent with the co-occurrence of non-Saccharomyces yeasts and S. cerevisiae, confirming the presence of the inoculated strains under the applied fermentation strategies.

In Panel A, sequential inoculation treatments showed additional ITS bands relative to the control, consistent with the presence of non-Saccharomyces yeasts during early fermentation. The T. delbrueckii–S. cerevisiae treatment displayed bands matching the T. delbrueckii reference profile, whereas the L. thermotolerans–S. cerevisiae treatment showed distinct bands corresponding to the L. thermotolerans reference.

Panel B illustrates the ITS profiles obtained at later fermentation stages. In sequential inoculations, banding patterns reflected the coexistence or transition between non-Saccharomyces yeasts and S. cerevisiae, while the control maintained a single, stable ITS profile throughout fermentation. Mixed samples exhibited combined banding patterns consistent with both non-Saccharomyces and S. cerevisiae reference lanes, confirming the presence of the inoculated yeasts under the applied fermentation strategies.

3.2. Physicochemical Parameters

The physicochemical parameters of Porter-style beers produced under different inoculation strategies are summarized in

Table 1. Physicochemical parameters of Porter-style beers produced under different inoculation strategies.

Inoculation Strategy	Color (EBC)	pH	Bitterness (IBU)	Total Acidity (% Acid Lactic)	Alcohol (% v/v)	Volatile Acidity (g L−1)	Final Gravity (SG × 1000)	Apparent Attenuation (%)
Td/Sc (T1)	94.3 ± 0.11 a	4.20 ± 0.01 a	18.15 ± 1.24 a	0.19 ± 0.0014 a	4.99 ± 0.15 a	0.28 ± 0.12 a	1006.7 ± 1.2 a	88.89 ± 2.75 a
Lt/Sc (T2)	92.27 ± 0.11 a	4.22 ± 0.04 a	18.10 ± 2.45 a	0.21 ± 0.0019 c	4.82 ± 0.15 a	0.55 ± 0.06 b	1005.3 ± 1.2 a	84.12 ± 2.75 a
Sc (control)	91.92 ± 0.11 a	4.33 ± 0.2 a	18.45 ± 0.8 a	0.20 ± 0.01 b	4.87± 0.13 a	0.22 ± 0.07 a	1004.7 ± 1.2 a	87.30 ± 2.75 a

Values are expressed as means ± standard deviations (n = 3). Different letters within the same column indicate significant differences (p < 0.05).

. Color, pH, bitterness (IBU), alcohol content, final gravity (FG), and apparent attenuation did not differ significantly among treatments (p > 0.05). All groups shared the same statistical designation, indicating that sequential inoculation with non-Saccharomyces yeasts did not alter the core technological or style-defining parameters. Alcohol content ranged from 4.82 to 4.99% v/v, and apparent attenuation varied between 84.12% and 88.89%, confirming comparable fermentative performance across treatments. In contrast, significant differences were observed in total acidity and volatile acidity (p < 0.05). Lt/Sc (T2) exhibited the highest total acidity (0.21 ± 0.0019% lactic acid) and differed significantly from both Td/Sc (T1) and the control. Similarly, volatile acidity was significantly higher in T2 (0.55 ± 0.06 g L⁻¹) compared to T1 (0.28 ± 0.12 g L⁻¹) and the control (0.22 ± 0.07 g L⁻¹).

3.3. Preliminary Sensory Results

3.3.1. Focus Group Outcomes

The focus group revealed preliminary sensory differences among the experimental treatments and the control. All beers exhibited a comparable brown color intensity consistent with the Porter style. However, differences in color hue were observed, with Treatment 1 differing from Treatment 2 and the control. Foam stability also varied, as Treatment 2 showed the highest foam persistence and consistency. Treatment 1 and the control presented moderate persistence and low consistency. Aroma intensity was medium in Treatment 1 and high in Treatment 2, while the control showed low intensity. A sherry-like character and fruity notes differentiated the samples, with Treatment 2 standing out due to mango peel descriptors. Lactic aroma, described as coffee with milk, further distinguished the beers, reaching its highest intensity in Treatment 2, while Treatment 1 and the control showed low and medium intensities, respectively. In the gustatory phase, Treatment 2 exhibited the highest perceived acidity, described as balanced and refreshing, and had the greatest mouth persistence. In contrast, Treatment 1 and the control showed low acidity and medium persistence, with retronasal descriptors mainly associated with roasted malt and toffee, while fruity notes predominated in Treatment 2 (Supplementary Table S1).

3.3.2. Panel Discriminative Performance

The triple triangular test conducted for panel selection demonstrated a statistically significant ability to discriminate beers with and without fruity aromas. Eighteen out of twenty assessors (90%) correctly identified the odd sample, exceeding the minimum number of correct responses required for significance (n = 11, p < 0.05, Table S2). Consequently, two assessors were excluded from the subsequent QDA due to incorrect discrimination, and the remaining 18 assessors constituted the final trained panel used for quantitative sensory characterization.

3.4. Quantitative Descriptive Analysis (QDA) Results

The sensory profiles of Porter-style beers produced under different inoculation strategies are shown in Figure 4 (see Table S3). No significant differences were observed among treatments for color perception, sherry-like character, fruity notes, or perceived acidity, indicating a comparable baseline sensory profile across beers. Aroma intensity differed among inoculation strategies, with the Lachancea thermotolerans–Saccharomyces cerevisiae treatment exhibiting higher scores compared with the Torulaspora delbrueckii–S. cerevisiae treatment, while the S. cerevisiae control showed intermediate values.

Foam-related attributes were influenced by the inoculation strategy. The L. thermotolerans–S. cerevisiae beer displayed significantly higher foam quantity, whereas foam persistence was highest in the T. delbrueckii–S. cerevisiae treatment, with the control presenting intermediate scores. The perceived lactic character showed differences among treatments, with higher scores observed in the L. thermotolerans–S. cerevisiae beer compared with the control, while the T. delbrueckii–S. cerevisiae treatment showed intermediate values.

Overall, sequential inoculation modulated specific sensory attributes related to aroma intensity and foam characteristics while preserving the typical sensory features of the Porter style.

4. Discussion

Sequential inoculation with non-Saccharomyces yeasts influenced fermentation dynamics and selected sensory attributes of Porter-style beers without significantly altering their final ethanol concentration, apparent attenuation, or core physicochemical parameters. Alcohol content ranged between 4.82 and 4.99% v/v, and apparent attenuation exceeded 84% in all treatments, confirming comparable fermentative performance. These findings demonstrate that, under the brewing conditions tested, sequential inoculation functioned primarily as a strategy for sensory modulation rather than ethanol reduction.

From a microbiological perspective, yeast population monitoring revealed higher early cell concentrations in the sequential treatments compared with the Saccharomyces cerevisiae control. This trend is consistent with wort oxygenation conditions that transiently favor biomass development of non-Saccharomyces species prior to S. cerevisiae predominance [20,38]. Nevertheless, because ITS-PCR provides qualitative rather than quantitative resolution, these findings should be interpreted as evidence of detectable implantation and persistence rather than numerical dominance [31].

Importantly, and in contrast to interpretations frequently extrapolated from oenological systems, the present analysis is grounded in the brewing-specific literature and in the compositional characteristics of malt worts, particularly within dark beer matrices such as Porter. The presence of melanoidins, roasted-derived compounds, and distinct nitrogen and carbohydrate profiles creates a metabolic environment that differs substantially from that of grape must and may condition yeast performance and interspecies interactions.

Sequential fermentations exhibited a slower initial decrease in wort density, which may be attributable to the distinct metabolic behavior of Torulaspora delbrueckii and Lachancea thermotolerans, as described in recent beer fermentation studies [20,39]. However, since no direct analysis of residual sugars was performed in this study, interpretations regarding specific sugar utilization should be considered with caution.

This delayed early attenuation did not translate into differences in final gravity or apparent attenuation, indicating that S. cerevisiae effectively completed fermentation following its delayed inoculation. Unlike previous studies reporting partial alcohol reduction in mixed fermentations [19,40], the present results demonstrate that in a Porter-style wort of moderate original gravity (1042), sequential inoculation modulated fermentation kinetics without significantly altering ethanol yield.

Porter-style beers constitute a technologically distinct fermentation matrix compared with pale malt-based systems. Roasted and specialty malts increase the concentration of Maillard reaction products, melanoidins, and phenolic compounds, which influence buffering capacity, redox balance, and protein–polyphenol interactions [7,8,9]. These matrix characteristics may modulate yeast metabolism and volatile perception, particularly in dark beer systems [6]. The preservation of color (≈92–94 EBC), bitterness (~18 IBU), and roasted character across treatments confirms that sequential inoculation did not disrupt the malt-driven sensory backbone typical of Porter-like beers.

Significant differences were observed in total acidity and volatile acidity, especially in the L. thermotolerans/S. cerevisiae treatment. The elevated acidity is consistent with the documented capacity of L. thermotolerans to produce organic acids, particularly lactic acid [18,41]. Although volatile acidity reached 0.55 g L⁻¹ (acetic acid equivalents), values remained within acceptable sensory thresholds for beer and were not associated with vinegar-like defects in QDA evaluation. Future studies incorporating HPLC-based quantification of lactic and acetic acids would allow mechanistic clarification.

Sensory analysis revealed modulation primarily in aroma intensity, foam quantity, and lactic character while preserving roasted perception and stylistic integrity. Enhanced aroma intensity in the L. thermotolerans treatment may be linked to modulation of higher alcohol and ester biosynthesis pathways, including regulation of alcohol acetyltransferases [42,43]. Nevertheless, extrapolation from wine systems should be approached cautiously due to compositional differences between wort and grape must [1,42].

Foam quality and CO₂ perception were selectively modulated by the inoculation strategy. The higher amount and persistence of foam in beers produced sequentially with L. thermotolerans could be explained not only by increased mannoprotein release during yeast autolysis but also by slightly enhanced CO₂ retention in the beer matrix, which could interact with proteins and polysaccharides to stabilize the bubbles [41,42,43,44,45]. In contrast, the T. delbrueckii/S. cerevisiae treatment showed lower foam retention, likely reflecting both a reduced contribution of mannoproteins and differences in dissolved CO₂ levels, suggesting a combined biochemical and physical basis for foam modulation. These observations support a physicochemical interpretation in which early yeast metabolism influences both protein–polysaccharide interactions and gas solubility, thereby affecting the sensory perception of foam. Because foam assessment was sensory rather than instrumental, future incorporation of objective methods such as NIBEM or Rudin testing would provide quantitative validation [46].

The sensory modulation observed in the present study—particularly the differences in aroma intensity, foam-related attributes, and perceived lactic notes—may be interpreted in light of the distinct metabolic behavior of non-Saccharomyces yeasts during the early stages of fermentation. Species such as Torulaspora delbrueckii and Lachancea thermotolerans have been reported to exhibit differences in higher alcohol and ester biosynthesis compared with Saccharomyces cerevisiae, thus contributing to distinct aroma profiles in beer fermentations [20]. These differences may reflect variations in metabolic flux distribution during fermentation, where the production of aroma-active compounds is enhanced even under conditions of comparable or reduced ethanol formation [38] and has been linked in previous studies to metabolic pathways such as the Ehrlich pathway and alcohol acetyltransferase activity (e.g., ATF1) [47]. However, since no direct analysis of volatile compounds or sugar consumption was performed in this study, these interpretations should be considered as general explanations based on previous studies rather than direct evidence from the present results.

These metabolic differences can influence the formation of acetate and ethyl esters, thereby modulating fruity and aromatic intensity even when the final ethanol concentration and attenuation remain comparable [20,42,48,49]. In the case of L. thermotolerans, its documented ability to produce organic acids—particularly lactic acid—provides a plausible biochemical basis for the higher perception of the lactic character observed by sensory analysis, despite the absence of dramatic shifts in total acidity [41,42,43,45]. This highlights the non-linear relationship between analytical acidity and sensory perception, where modest changes in acid composition may influence mouthfeel and freshness without exceeding stylistic thresholds.

Additionally, recent brewing-focused studies indicate that non-Saccharomyces yeasts may affect foam quantity and stability through differences in nitrogen metabolism, mannoprotein release, and polysaccharide–protein interactions within the beer matrix [20,42,48]. In dark beer systems such as Porter, where melanoidins and high-molecular-weight compounds derived from roasted malts are abundant, these interactions may further influence bubble stability and foam persistence through physicochemical mechanisms involving protein cross-linking and surface activity.

Sequential inoculation is particularly relevant in this context because it allows early metabolic expression of non-Saccharomyces traits before S. cerevisiae predominance ensures fermentative completion. This temporal separation facilitates selective modulation of aroma- and foam-related attributes while preserving attenuation, ethanol yield, and overall technological reliability [20,43,49,50]. Rather than inducing large compositional shifts, the strategy appears to fine-tune secondary metabolite production within the constraints imposed by the Porter-style matrix.

Overall, the present results indicate that sequential inoculation with selected non-Saccharomyces yeasts enhances sensory differentiation in Porter-style beers without compromising fermentative efficiency or ethanol yield. This study provides brewing-specific evidence supporting the controlled integration of L. thermotolerans and T. delbrueckii—strains also widely recognized in commercial wine fermentation catalogs—as tools for sensory innovation in dark beer matrices. Further research should integrate quantitative population dynamics, targeted organic acid profiling, and volatile compound analysis to refine mechanistic understanding and industrial applicability.

5. Limitations

This study was performed at the pilot scale using a single Porter-style beer and one sequential inoculation protocol with two commercial non-Saccharomyces wine strains. Fermentation behavior and sensory effects may therefore differ at the industrial scale, with other beer styles, or when alternative strains and inoculation strategies are applied. In addition, sensory evaluation was conducted by a trained panel, which supports descriptive accuracy but does not directly reflect consumer preference. Accordingly, extrapolation of these results beyond the experimental context should be made with caution, and further studies are needed to confirm their broader applicability.

6. Conclusions

Sequential inoculation with non-Saccharomyces yeasts modulated fermentation performance and specific sensory attributes in Porter-style beers without significantly altering their physicochemical parameters. Both Torulaspora delbrueckii and Lachancea thermotolerans were detected during fermentation by ITS-PCR profiling, confirming their presence throughout the process. Fermentations involving L. thermotolerans showed more pronounced effects on acidity-related and fruity attributes. Overall, these results support the application of selected non-Saccharomyces yeasts as practical biotechnological tools to modulate sensory complexity in craft beer production under sequential inoculation strategies.