Potential of Baled Silage to Preserve White Grape Pomace for Ruminant Feeding

Abstract

The use of agro-industrial by-products in animal feed represents a useful alternative to enhance the sustainability of the agri-food chain. Grape pomace represents an environmental problem mainly for wine-producing countries. Because of the high water content and the seasonality of this feedstuff, ensiling might be a technology to preserve its nutritional quality for a long time, and this must be considered and studied on a commercial scale. This study aimed to characterise the ensiling process of white grape pomace, evaluate its suitability for inclusion in the ruminant diet and compare its shelf life to untreated storage conditions. White grape pomace silos were made with baled silage (300 kg approx.). Samples were analysed at days 0, 7, 14, 35, 60 and 180 of conservation to determine microbial populations, fermentation metabolites, nutritional components and bioactive properties. The collected data were analysed using a general linear model, considering the effect of the treatment, sampling days and their interaction (Proc. GLM, SAS v9.4). White grape pomace showed good suitability for ensiling, and stabilisation was achieved on day 35. The microbial populations and fermentative components observed in silage treatments adhered to the expected standards for high-quality ensiling processes. There were no significant losses of dry matter, and no significant differences were observed in the nutritional composition for ruminant feeding. A small reduction in antioxidant potential was observed and considered irrelevant in terms of the bioactive properties of the silages. Additionally, the cost analysis demonstrated that white grape pomace silage could serve as a more economical feedstuff compared to conventional forages, considering its nutritional value. So, the ensiling of white grape pomace in baled silage is a suitable and cost-effective technique that allows its preservation over a long period.

1. Introduction

The global vineyard area is estimated at 7.2 million hectares, and Spain has the largest winegrowing area, with 945,000 hectares (14% of the global area). In the past year, global wine production was 237 million hectolitres. Europe leads global wine production, with France, Italy and Spain collectively accounting for 48% of the world’s total wine output. The United States follows, accounting for 10% of global wine production [1]. On the other hand, the wine industry is associated with a serious environmental impact due to the generation of enormous amounts of waste, primarily produced during the harvest. It is estimated that for every 100 kg of grapes processed, around 25 kg of by-product is generated [2], leading to amounts that can reach 1200 tonnes per year in the main wine-producing countries [3]. The by-product of the winemaking process is known as grape pomace and consists of a heterogeneous mixture of seeds, skins and pulp residues [4]. Grape pomace holds significant relevance in terms of use in animal nutrition, as many of the bioactive substances present in grapes, which have beneficial functional properties and positive technical applications, remain in this by-product without being transferred to the wine [3]. Grape pomace is a rich source of phytochemicals (phenols, flavonoids and tocopherols), antioxidants and polyunsaturated fatty acids [5,6,7].

The global demand for animal-derived products is increasing in the face of diminishing resources in the agri-food sector [8]. The valorisation of agro-industrial by-products for animal feeding is a useful solution to reduce the environmental impact of the agricultural sector while lowering livestock production costs and facilitating the recirculation of nutrients and the provision of bioactive compounds to animals, ensuring efficient energy expenditure [9]. Ruminants can tolerate a higher percentage of fibrous material and bioactive compounds with low digestibility, which are commonly abundant in agro-industrial by-products [10,11,12], producing food of high biological value for humans while also promoting the circular economy and environmental improvement. Previous studies [13,14,15] have demonstrated that if diets including by-products are balanced, their use should not compromise production indices or output quality. Furthermore, some publications have shown that incorporating by-products into diets, in addition to meeting nutritional standards, would be economically viable [16,17]. However, the marked seasonality and high water content of by-products make them perishable in the short term, limiting their use, as treatment is needed for their preservation in order to maintain their organoleptic and nutritional characteristics over time and enable their off-season inclusion in animal diets.

Ensiling is a method of preserving forages and raw materials rich in lignocellulose and water that is widely used around the world, mainly due to its low dependence on climate, unlike hay [18]. It involves a natural fermentation process under anaerobic conditions carried out by lactic acid bacteria, which transform soluble carbohydrates into organic acids, primarily lactic acid. As a result, the pH decreases, thereby inhibiting the growth of harmful microorganisms that degrade organic matter, ensuring its viability over time. Agro-industrial by-products are a fibrous raw material with moisture and sugar contents that make them ideal as substrates for fermentative reactions. Numerous studies [19,20,21] have shown that the product of this fermentation maintains its dry matter percentage and nutritional values, which is essential for ensuring productive yields when included in ruminant diets. Furthermore, during ensiling, bioactive substances such as phenolic compounds undergo transformations that enhance their associated antioxidant capacity [22], which could have positive implications for animal health and performance while improving the nutritional value of animal-derived products [23]. Several studies have shown that including fermented foods like silage in the diet improves growth, health parameters and ruminal ecosystems in cows [24] and goats [25]. Grape pomace is a fibrous raw material with moisture and sugar contents that make it ideal as a substrate for fermentative reactions, and, in addition, it is rich in bioactive compounds [26]. Massaro et al. [27], in a 60-day laboratory-scale study, observed that grape pomace silage exhibited good fermentative and chemical qualities, a linear increase in crude protein content, and improvements in the feed, especially regarding fibre quality. However, no studies have been undertaken to explore its application on a commercial scale, over extended periods, as well as its effect on other functional characteristics.

The aim of this study was to examine the evolution of white grape pomace preserved through baled silage (at commercial scale) over 6 months for its application in ruminant feeding. To achieve this, changes in variables related to the fermentation process, the dynamics of microbial populations, the nutritional composition and the antioxidant potential were studied and compared with those of material stored without any preservation treatment. This study hypothesises that the baled silage technique for this wine industry by-product will maintain the nutritional and organoleptic characteristics of this raw material for at least 6 months. The advantages of this type of ensiling over others include not requiring construction, having a high degree of compaction and airtightness that extends the useful life of the material, and the ease of handling and transport that facilitates its commercialisation.

2. Materials and Methods

2.1. By-Product

The study focused on the by-product of the white grape winemaking process, white grape pomace (WGP), sourced from wineries with the Protected Designation of Origin (PDO) “Vino de Alicante”. All the material was transported to the facilities of the Institute of Agricultural and Environmental Research and Innovation (CIAGRO) at the Miguel Hernández University, Elche (UMH), where the silos were manufactured. This process took place in September, following the start of the grape harvest. At the same time as the pomace was ensiled (silage), a proportional part of the same raw material was stored without undergoing any preservation treatment (untreated).

Following the description of the patent by Díaz et al. [28], on the same day, 20 silage bales were produced, each weighing approximately 300 kg and with a volume of 0.64 m³, using an Agronic MR 820 rotary baler (Agronic Oy, Haapavesi, Finland). To achieve suitable anaerobic conditions and ensure the proper compaction and firmness of the bales, 5 layers of internal mesh (Karatzis SA, Heraklion, Greece) and 13 layers of plastic film (Karatzis SA, Heraklion, Greece) were used. Untreated WGP was stored in bulk under a covered barn with limited ventilation. It was indirectly exposed to the region’s environmental conditions, characterised by a Mediterranean climate, simulating not-controlled storage conditions. During the storage period, mean environmental temperatures fluctuated between 10 °C and 23 °C, with moderate rainfall and a relative humidity of 58–65% RH, as is usual in the location. The experiment was conducted from the 2022 harvest (September) until 6 months later.

2.2. Experimental Design

An experimental design was planned in which the sample collection days were 0 (day of ensiling), 7, 14, 35, 60 and 180 of storage (silage vs. untreated). The effect of conservation days was considered to assess the evolution of WGP, as well as its nutritional and quality variables over time.

For the sampling on day 0, three samples were collected from three different parts of the entire batch of raw material. On each of the following sampling days, two different bales were sampled: 1 kg of sample was collected using a manual auger from three distinct points of the bale (silage): in the middle, at the top and 20 cm from the base, and the collected material was then mixed. In the untreated material (untreated), two samples of 1 kg were also collected each day from three distinct points in the trench. Thus, on each sampling day, two representative samples of different bales (silage) and two representative samples of the fresh by-product (untreated) were obtained.

In the laboratory, each sample was divided into three aliquots: one was kept fresh for pH analysis and microbiological cultures; another was dehydrated in an oven at 60 °C for 48 h for the determination of nutritional composition; and the last one was frozen at −80 °C for the determination of total phenolic compounds (TP), antioxidant activity (ABTS and DPPH) and the concentration of sugars and volatile organic compounds.

2.3. Microbiology

The samples for microbiological monitoring were collected each day of sampling and transported to the laboratory in sterile containers. For the cultivation of enterobacteriaceae, aerobic bacteria, moulds and yeasts, 25 g was weighed for each sample repetition and homogenised with 225 mL of peptone water in Stomacher bags (BagMixer 400; Interscience, Saint-Nom-la-Bretèche, France). The microbiological cultures for the counting of enterobacteriaceae were incubated directly on Petrifilm EB plates (3MTM Microbiology, Flemington, NJ, USA) at 37 °C for 24 h, while the aerobic bacteria were incubated on Petrifilm AC plates (3MTM Microbiology, Flemington, NJ, USA). Moulds and yeasts were incubated on Petrifilm YM plates (3MTM Microbiology, Flemington, NJ, USA) for 72 and 120 h, respectively. The samples for the counting of lactic acid bacteria (LAB) were diluted and homogenised in 255 mL of MRS 28 culture medium (Liofilchem, Roseto degli Abruzzi, Italy) and incubated on Petrifilm LAB plates (3MTM Microbiology, Flemington, NJ, USA) in an anaerobic jar at 37 °C for 48 h. The results were expressed as log10 cfu/g of fresh sample, following the AENOR guidelines (Spanish Association for Standardization and Certification, 2015). Spore counting of Clostridium genus was performed on days 0, 60 and 180, using the most probable number technique (MPN) and Bryant and Burkey broth (BBB; Merck, Darmstadt, Germany), following the methodology indicated in Arias et al. [29].

2.4. Fermentation Metabolites

To study the fermentative changes during the silage process, some of the metabolites from this reaction were measured from the frozen sample. The sugars present in the fruit (sucrose, fructose and glucose), which serve as substrates in the fermentations that occur during silage, and the short-chain organic acids (VFAs) (acetic, butyric, propionic and tartaric acid), lactic acid and ethanol were determined according to the method proposed by Feng-Xia et al. [30]. The results were obtained through analysis using HPLC (Agilent 1200, Santa Clara, CA, USA) with a C610H column of 30 cm and 7.8 mm ID (Supelcogel, Bellefonte, PA, USA), employing 0.1% orthophosphoric acid as the mobile phase. The contents of these compounds in the sample were expressed as g/kg DM.

2.5. Physico-Chemical Parameters and Nutritional Composition

On the same day of sampling, the pH (GLP 21, Crison, Alella, Spain) and dry matter (DM; g/kg, AOAC [31], 1990, 948.12) were determined.

The Flieg scores were calculated for each sample to determine the quality of the silage, according to the equation provided by Kilic [32]:

Flieg Scores = 220 + (2 × DM (%) − 15) − 40 × pH

According to this index, silage is classified as follows: <20 points corresponds to very low-quality silage; between 21 and 40 points indicates low-quality silage; from 41 to 60 points indicates medium-quality silage; high-quality silages are scored between 61 and 80 points; and silages with >81 points are considered very high quality.

Starting from samples dehydrated at 60 °C and subsequently ground (1 mm), the following variables were analysed using AOAC [29] procedures: ash (g/kg DM, 934.01), crude protein (CP; g/kg DM, 988.05), ether extract (EE; g/kg DM, 920.39) and total sugars (g/kg DM, 974.06). The contents of neutral detergent fibre (NDF, g/kg DM), acid detergent fibre (ADF, g/kg DM) and acid detergent lignin (ADL, g/kg DM) were analysed according to Van Soest et al. [33]. The non-protein nitrogen fraction (NPN, g/kg DM) was determined using the Cornell method for nitrogen fractionation in feeds described by Licitra et al. [34]. The starch content was determined using the polarimetric method of Ewers [35].

2.6. Fatty Acid Profile

The lipid profile determined by the separation of isomers of polyunsaturated fatty acids was performed through direct methylation of the lyophilised sample, without prior fat extraction, according to Kramer et al. [36]. Identification of the methyl esters of fatty acids (FAMEs) was carried out using a flame ionisation detector (FID) coupled to a GC-17A gas chromatograph (Shimadzu, Kyoto, Japan) equipped with a 30 m DB23 capillary column, 0.25 mm ID and 0.25 mm internal coating (Agilent, Santa Clara, CA, USA). A mix of FAMEs (18912-1AMP; Sigma-Aldrich, Saint Louis, MO, USA) was used as a standard for identifying the peaks in the fatty acid profile of the sample. The nutritional and health indices for evaluating fatty acids were calculated according to Chen and Liu [37].

2.7. Bioactive Properties

The frozen sample was used for the analysis of total polyphenol content (TP, mg GAE eq/g DM) using the Folin–Ciocalteu method described by Kim et al. [38]. The DPPH analysis (reduction of the 1.1-diphenyl-2-picrylhydrazyl radical) was conducted following the modified protocol by Cheng et al. [39], developed initially by Brand-Williams et al. [40]. For the ABTS analysis (reduction of the 2.2′-azino-bis(3-ethylbenzothiazoline-6-sulfonate) radical), the method described by Leite et al. [41] was followed. The results of these reactions were measured by spectrophotometry using a UV–visible spectrophotometer (Zuzi 4255/50) at wavelengths of 750, 515 and 734 nm for TP, DPPH and ABTS, respectively. The antioxidant capacity value of the sample was expressed in mg Trolox eq/g DM.

2.8. Statistical Analysis

All the determined variables were statistically analysed following a generalised linear model (Proc. GLM, SAS V 9.4, 2022), according to the following equation:

Y = μ + Di + Tk + (Di × Tk) + e

where Y is the dependent variable, μ is the mean, Di is the fixed effect of the conservation time (i = 0, 7, 14, 30, 60 and 180 days), Tk is the fixed effect of treatment (k = silage or untreated), Di x Tk is the interaction between both effects and e is the residual error.

To interpret the differences between levels of the fixed effect, least-squares means were calculated, representing model-adjusted estimates. Mean comparisons between levels of the fixed effect were performed by testing the null hypothesis:

H0: LSMean (h) = LSMeann (j)

where (h) and (j) represent different levels of the fixed effect. The significance of these comparisons was evaluated using Pr > t.

If p < 0.05, the null hypothesis is rejected, indicating that the differences between levels are statistically significant.

2.9. Other Key Aspects

An approximate calculation of the costs of the tested silos was made, including raw materials, the inner mesh and plastic film, labour, transport, and other production and marketing costs. The Milk Forage Units (UFLs), used for comparing alternative feedstuffs and common forages, were calculated based on their nutritional composition using the following equation described by INRA [42]:

UFL = (4.3 × PB) + (9.37 × EE) + (4.2 × CH) + (3.6 × FB) × 0.62/1.7

A multiresidual analysis of the main plant protection products used in grape cultivation was also conducted using chromatography on the raw material samples (WGP) on day 0, following the consolidation of Regulation (EC) No. 396/2005 of the European Parliament and of the Council.

3. Results

3.1. Microbiology

The ensiled and untreated raw material (day 0) had 6.57 log10 cfu/g FM and 6.23 log cfu/g FM of LAB (Figure 1a), respectively. The populations of LAB increased significantly (p < 0.05) during the first week in both treatments (8.22 log cfu/g FM of silage; 7.81 log10 cfu/g FM of untreated FM). From this point on, there was a slight decrease in silage (p < 0.0107) until day 14, after which it remained stable until the end of the study (7.63 log10 cfu/g FM on day 180). In untreated, there was stability in the growth of lactic populations until day 60, after which an exponential growth was observed (p < 0.05) until the end (7.92 log10 cfu/g FM on day 60 and 9.00 log10 cfu/g FM on day 180). The enterobacteriaceae population increased markedly (p < 0.05) in both treatments from day 0 (2 log10 cfu/g FM and 3.96 log10 cfu/g FM, for silage and untreated, respectively) to day 14 (5.02 log10 cfu/g FM for silage and 8.06 log10 cfu/g FM for untreated). During the following two weeks, the presence of these bacteria in silage decreased (p < 0.05), and they completely disappeared from day 35. In contrast, in untreated, the enterobacteriaceae colonies remained stable, observing the maximum at day 60 (8.07 log10 cfu/g FM) and a slight decline at day 180 (6.65 log10 cfu/g FM) (Figure 1b). The evolution of the total aerobic population (Figure 1c) was different between the treatments. Silage showed no differences over time (p = 0.54), although untreated exhibited a high increase (p < 0.05), the population doubling over the same period (7.04 log10 cfu/g FM at day 0; 12.60 log10 cfu/g FM at 180 days). Moulds showed an opposite evolution between both treatments (Figure 1d). Initially, the mould count was 5.92 log10 cfu/g FM in silage and 6.32 log10 cfu/g FM in untreated. By the end of the study, there were significant differences (p < 0.001) between the treatments, with counts of 3.01 log10 cfu/g FM and 8.50 log10 cfu/g FM in silage and untreated, respectively. The presence of yeasts decreased markedly (p < 0.05) in silage compared to untreated, starting from day 35 of production (Figure 1e). Regarding the spore count of the Clostridium genus, a treatment effect was observed (p < 0.05), with counts of 2.99 espores/g FM and 3.27 espores/g FM in silage and untreated samples, respectively, on day 180 of storage.

3.2. Sugars and Fermentation Metabolites

During the first two weeks, there was a rapid reduction (p < 0.05) in sugar content in both treatments (Figure 2). Additionally, it was observed that during the first 7 days, the effect of the silage treatment on the reduction in fructose and sucrose concentrations was twice as great (p < 0.05). The fructose concentration at day 0 (6.68 g/kg DM silage; 6.96 g/kg DM untreated) decreased by up to 2.42 g/kg DM at day 7 in silage and 4.43 kg/kg DM in untreated (Figure 2a). The sucrose concentration at day 0 (1.04 g/kg DM silage; 1.10 g/kg DM untreated) was reduced by up to 0.51 g/kg DM at day 7 in silage and 0.88 g/kg DM in untreated (Figure 2b). The glucose content (Figure 2c) also decreased more rapidly in silage than untreated (p < 0.05) between the second week (4.25 g/kg DM silage vs. 4.31 g/kg DDM untreated) and 60 days (1.72 g/kg DM vs. 2.67 g/kg DM, respectively). Nevertheless, the sugars after 14 days of fermentation were similar in both treatments, and the sugar contents of both treatments were considered residual.

The ethanol concentration in silage for up to 60 days ranged from 14.12 to 17.10 g/kg DM (Figure 2d). At six months, its presence had decreased (p < 0.05) to 9.96 g/kg DM. In untreated, ethanol underwent a high decrease in the first week of conservation and was negligible after that sampling time.

The lactic acid content in silage increased (p < 0.05) throughout the study period, from a minimal concentration on day 0 (8.10 g/kg DM) to a maximum concentration at day 180 (26.02 g/kg DM). In untreated, its evolution was the opposite (p < 0.05), and its concentration was residual after day 35 (Figure 3a). The concentration of acetic acid was initially different (p < 0.05) between silage (15.39 g/kg DM) and untreated (26.52 g/kg DM) and evolved differently: the concentration became residual from the second week in untreated, whereas it increased (p < 0.05) during the six months of study in silage (Figure 3b). The initial concentration of tartaric acid decreased (p < 0.05) in both treatments similarly in the first 7 days, remaining constant (p >0.05) in silage the rest of the time (7.52 g/kg DM on day 7; 6.75 g/kg DM on day 180), whereas in untreated the concentration decreased (p < 0.05), becoming residual by day 180 (Figure 3c). Propionic acid was initially detected (p < 0.05) in silage (1.16 g/kg DM) and untreated (1.99 g/kg DM), although its concentration became residual from day 7 (Figure 3d). Butyric acid was not detected in either silage or untreated.

3.3. Physico-Chemical Properties and Nutritional Composition

Figure 4a shows how the initial pH of silage (4.25) and untreated (4.01) evolved differently (p < 0.05) from the first week of the study. In silage, the pH remained consistently close to 4. In contrast, the pH of untreated increased (p < 0.05) from the first week, reaching a maximum value of 8.42 on day 35 that remained similar until the end of the study, maintaining an alkaline pH (pH > 7). The Flieg score (Figure 4b) at day 0 (higher than 120) indicated the high quality (Flieg score > 81) of this raw material. Silage slightly increased (p >0.05) the Flieg score over time and maintained a very high-quality score for the 6 months of conservation studied. The untreated material’s Flieg score dropped continuously (p < 0.05), and from days 35 to 180 presented a very low quality (<40).

Table 1. Effects of treatments (silage vs. untreated) and conservation time on nutritional composition in WGP.

Treatment WGP	Days of Conservation						SEM
	0	7	14	35	60	180
Dry Matter (g/kg)
Silage	445.71 bc	455.80 bc	440.04 bc	437.65 bc	439.57 bc	430.55 c	1.06
Untreated	451.09 bc	474.87 b	418.11 c	445.66 bc	518.48 a	533.02 a
Organic Matter (g/kg DM)
Silage	884.00 ab	896.00 a	878.50 ab	898.00 a	889.50 a	888.50 a	1.59
Untreated	877.50 ab	877.00 ab	875.50 ab	858.50 ab	840.50 b	838.50 b
Ash (g/kg DM)
Silage	52.00 b	50.00 b	51.50 b	55.50 b	55.00 b	69.00 b	0.92
Untreated	55.50 b	54.00 b	63.00 b	72.00 ab	99.00 a	81.00 a
Neutral Detergent Fibre (g/kg DM)
Silage	449.50 c	488.00 bc	551.50 abc	553.00 abc	499.00 bc	405.50 c	5.58
Untreated	547.00 abc	565.00 abc	436.00 c	685.00 ab	685.50 ab	715.50 a
Acid Detergent Fibre (g/kg DM)
Silage	458.00 bcd	453.50 bcd	463.00 bcd	482.50 bcd	420.00 cd	337.00 d	6.24
Untreated	430.50 bcd	482.00 bcd	514.00 abcd	571.50 abc	598.00 ab	683.50 a
Acid Detergent Lignin (g/kg DM)
Silage	303.85 cd	314.30 bcd	328.65 bc	343.70 bc	299.55 cd	189.45 d	3.88
Untreated	282.65 cd	334.25 bc	370.00 bc	394.25 abc	443.75 ab	527.80 a
Ether Extract (g/kg DM)
Silage	66.00 ab	69.00 ab	74.00 a	67.50 ab	69.50 ab	63.00 ab	0.45
Untreated	65.00 ab	70.50 ab	70.00 ab	61.50 ab	68.00 ab	56.00 b
Crude Protein (g/kg DM)
Silage	104.50 bc	100.00 c	106.50 abc	107.50 abc	105.00 bc	121.50 ab	0.65
Untreated	107.00 abc	106.00 bc	115.50 abc	122.50 ab	125.50 a	125.50 a
Non-Protein Nitrogen (g/kg DM)
Silage	10.50 abc	14.00 a	12.50 ab	9.00 abc	8.00 abc	10.50 abc	0.23
Untreated	10.00 abc	10.50 abc	4.00 c	5.50 bc	7.00 abc	8.50 abc
Starch (g/kg DM)
Silage	34.00 abc	32.00 abc	30.50 abc	33.00 abc	38.00 ab	46.50 a	0.47
Untreated	38.50 ab	32.00 abc	24.50 bcd	24.50 bcd	18.50 cd	12.50 d
Total Sugar (g/kg DM)
Silage	17.00 a	6.50 b	4.00 b	4.00 b	4.00 b	3.00 b	0.28
Untreated	11.50 ab	10.50 ab	4.00 b	3.00 b	4.00 b	3.00 b

a–d Different letters for the same variables indicate a significant difference between days per treatment (p < 0.05); SEM: Standard Error of the Mean.

shows the changes in nutritional composition during the 180 days of follow-up. The dry matter (DM) of silage remained constant from day 0 to day 180, with an average value of 441.2 ± 1.06 g/kg. The DM of untreated (451.09 g/kg) initially showed no significant differences compared to silage, but an increase was observed (p < 0.05) from day 60 to day 180 (533.02 g/kg). The organic matter (OM) remained with no significant changes over time in both treatments, although a greater decrease was observed in untreated at day 180. The ash content significantly increased (p < 0.05) only in untreated. No differences were observed in the NDF content of silage (491 ± 5.58 g/kg DM), whereas in untreated it increased (p < 0.05) between days 60 and 180 to 715.50 g/kg DM. The ADF content showed a similar trend to that of NDF, remaining constant in silage (435 ± 6.24 g/kg DM), while it increased (p < 0.05) by day 180 in untreated (527.80 g/kg DM). Between 60 and 180 days, silage had its lowest ADL content, while untreated had its highest ADL content (p < 0.05) during this period. The EE content in both treatments showed no significant changes during the study period in terms of CP and NPN levels, which were very stable in both treatments. The starch content in silage remained unchanged (35.67 ± 0.47 g/kg DM) but decreased (p < 0.05) starting from day 35 in untreated. Sugars were consumed more quickly in silage than in untreated, but ultimately both ended the study with a residual content of this component (3 g/kg DM).

3.4. Fatty Acid Profile

The characterisation of the fatty acid profile of WGP (

Table 2. Average values and standard deviations of percentages of the methylated fatty acid area in the fatty acid profile and nutritional and health indices in WGP (silage vs. untreated).

	White Grape Pomace		SEM
Fatty Acid Profile	Silage	Untreated
SFA	19.33	19.53	0.44
MUFA	17.54	18.56	0.18
PUFA	63.13	61.90	0.36
SCFA	1.08	0.45	0.20
MCFA	13.15	13.19	0.36
LCFA	85.77	86.34	0.48
C18:2	58.10	59.12	0.64
C18:1	15.86	16.95	0.20
C16:0	11.66	10.78	0.19
C18:0	4.53	5.20	0.06
C18:3	3.41 b	2.31 a	0.18
Nutritional Health Indices
n6n3	18.07 a	77.69 b	3.42
LAALA	18.16 a	81.02 b	3.32
UI	148.03	144.85	0.81
TFA	0.03	0.08	0.04
IA	0.16	0.16	0.01
IT	0.34	0.35	0.00
HH	6.67	7.07	0.26

a,b Different letters for the same variables indicate a significant difference between treatments (p < 0.05); SEM: Standard Error of the Mean; SFA: Saturated Fatty Acid; MUFA: Monounsaturated Fatty Acid; PUFA: Polyunsaturated Fatty Acid; SCFA: Short-Chain Fatty Acid (C6:0-C10:0); MCFA: Medium-Chain Fatty Acid (C11:0-C17:0); LCFA: Long-Chain Fatty Acid (C18:0-C24:0); n6n3: ω-6/ω-3 Fatty Acid Ratio; LAALA: Linoleic Acid/α-Linolenic Acid Ratio; UI: Unsaturation Index, TFA: Trans Fatty Acid; IA: Index of Atherogenicity; IT: Index of Thrombogenicity; HH: Hypocholesterolemic/Hypercholesterolemic Ratio.

), expressed as the mean over the study period for each treatment (silage vs. untreated), showed a higher unsaturated fatty acid (UFA) content compared to saturated fatty acids (SFAs), with no differences between treatments. The polyunsaturated fatty acid (PUFA) content predominated over monounsaturated fatty acids (MUFAs). Linoleic acid (C18:2) was observed to be the principal fatty acid (58.10% in silage; 59.12% in untreated), followed by oleic acid (C18:1) (15.86% in silage; 16.95% in untreated), palmitic acid (C16:0) (11.66% in silage; 10.78% in untreated) and stearic acid (C18:0) (4.53% in silage; 5.20% in untreated). The alpha-linolenic acid content (C18:3) was lower (p < 0.05) when WGP was not subjected to any preservation treatment.

Table 2 also presents nutritional and health-promoting indices calculated from the fatty acid profiles of silage and untreated. The effect of silage treatment was positive for the ω-6-to-ω-3 fatty acid ratio (n6n3) (18.07 in silage vs. 77.96 in untreated). A similar effect (p < 0.05) was observed for the linoleic acid-to-alpha-linolenic acid ratio (LAALA). The Unsaturation Index (UI) indicated a high degree of unsaturated fatty acids in both treatments. Among unsaturated fatty acids, the content of trans fatty acids (TFAs) was negligible. The health-promoting indices, the Atherogenic Index (AI) and the Thrombogenic Index (TI) were below 0.5, with no significant differences between treatments. The Hypocholesterolemic/Hypercholesterolemic (HH) Ratio was 6.67 and 7.07 (p > 0.05) for silage and untreated, respectively.

3.5. Bioactive Properties

The TP obtained by the Folin–Ciocalteu colorimetric method (Figure 5a) in silage varied from 7.96 to 9.11 mg GAE eq/g DM (p > 0.05) throughout the monitoring period. In contrast, untreated lost (p < 0.05) half of the TP after the first week, the content being residual on day 180 (0.92 mg GAE eq/g DM).

The results for the antioxidant activity of silage measured by DPPH (Figure 5b) ranged from 423.40 to 595.76 mg Trolox eq/g DM over 180 days. In contrast, untreated only maintained the initial values (492.22 mg Trolox eq/g DM) during the first week, undergoing a continuous and intense reduction until day 35 of conservation, and the value kept decreasing until the end of the experiment (p < 0.05).

The antioxidant capacity measured by ABTS (Figure 5b) showed a preservation of the reducing capacity of silage until day 60. At the end of the study, the ABTS values had slightly decreased (p < 0.05) to 751.88 mg Trolox eq/g DM. In untreated, ABTS decreased (p < 0.05) after the first week until the end of the conservation time, presenting the minimum at day 180 (56.41 mg Trolox eq/g DM).

3.6. Other Key Aspects

Table 3. Estimates of manufacturing costs (EUR/t) of WGP in baled silage (commercial scale) and selling price of WGP silage compared to alfalfa hay.

	Feedstuffs
Manufacturing Cost	WGP Silage	Alfalfa Hay
EUR/t fresh WGP	25.00	-
EUR/t baled silage manufacturing	28.80	-
EUR/t FM WGP silage	53.80	-
Selling Price
EUR/t DM	120.89	307.69
EUR/kg CP	1.63	1.60
EUR/UFL	122.12	350.00

WGP: White Grape Pomace; FM Fresh Matter; DM: Dry Matter; CP: Crude Protein; UFL: Milk Forage Unit.

shows the different costs of fresh WGP and the manufacturing process for baled silage of WGP. Additionally, this study compared the final prices, expressed as EUR/kg CP and EUR/UFL, of WGP silage (alternative feedstuff) and alfalfa hay (910 g/kg DM; 175 g CP/kg DM; 0.80 UFL), assuming the manufacturing costs previously mentioned. The unit cost of CP in WGP silage was similar to that of alfalfa hay; however, the unit cost of UFL for WGP silage was considerably lower.

No results equal to or exceeding the maximum residue limit (MRL) for the most commonly used phytosanitary products for grape (Permethrin, Diflubenzuron, Azadirachtin, Hexythiazox, Fenhexamid, Metalaxyl, Iprovalicarb and Tebuconazole) were detected in the analysed samples.

4. Discussion

Naturally, plant material hosts homofermentative and heterofermentative lactic acid bacteria (LAB), with the former exhibiting greater potential for achieving high-quality silage and minimising DM losses, as they solely produce lactic acid from sugar fermentation [43,44]. The acidic environment generated by these populations further promotes the growth of other LAB species with higher tolerance to such conditions [45]. The LAB populations remained dominant in silage throughout the experiment, similarly to other studies using grape pomace as part of the ruminant diet [46,47]. The presence of other microbial colonies was controlled in silage due to the acidic and anaerobic conditions [48] created by LAB, which was not the case in untreated samples. Müller et al. [49] and more recently Fang et al. [50] highlighted the potential of LAB to inhibit the growth of other harmful microorganisms during the process by producing antimicrobial substances such as bacteriocins. The population of enterobacteria drastically decreased in silage after the second week of the trial, disappearing completely within one month. Enterobacteria counts in silage were significantly lower compared to untreated samples and those observed in studies with other by-products [51,52,53]. The reduction in yeast counts observed in silage from day 35 onward is likely attributable to decreased substrate availability and increased acetic acid concentration in baled silage, which has antifungal and bactericidal properties [54]. During this period, yeast counts remained below 6 log10 CFU/g FM, a threshold suggested by Kung et al. [55] to ensure high stability upon silo opening and favourable animal acceptance [56]. The mould population initially observed in silage, higher than that reported in other plant-based by-products [52,57], also declined as the acetic acid concentration increased, further supporting its antifungal role, as noted by Kung et al. [54]. The results of the microbiological evolution study demonstrate that concentrations of microorganisms with unfavourable metabolic activity for maintaining hygiene and nutritional quality in silage were below the reference levels reported in other publications [58,59,60] in the absence of a regulation that sets maximum permissible limits in such cases.

Several factors influence the synthesis of fermentation products, including the predominant populations of microorganisms, the available fermentable substrates and the types of fermentation occurring throughout the ensiling process [18]. These fermentation metabolites are closely linked to the organoleptic and nutritional quality of the silage, affecting voluntary intake by animals [54], as well as the production, composition and quality of their derivatives [61]. The production of lactic acid during fermentation depends on the presence of LAB populations and the availability of easily fermentable substrates (monosaccharides and disaccharides) under anaerobic conditions [62]. The reduction in fermentable sugars (fructose, sucrose and glucose) as the lactic acid concentration increased observed in the silage has also been reported in other studies on grape pomace silages [46,63]. The lactic acid concentration detected in silage was lower than the values reported by the same authors for mixed silages of alfalfa and grape pomace (55.7–64.6 g/kg DM), which may be attributed to the monoculture nature of the WGP used. Up to day 60, the lactic acid concentration in silage was higher than that of other organic acids and fermentation products, a factor that supports good macronutrient preservation, as well as good palatability and a pleasant smell, facilitating animal acceptance [64]. However, by the end of the six months, the acetic acid concentration in silage exceeded that of lactic acid as a result of prolonged fermentation that favoured a higher proportion of heterofermentative bacterial colonies [43]. The shift in the ensiling process toward fermentation dominated by heterofermentative lactic acid bacteria is beneficial for maintaining the aerobic stability of baled silage after opening [65]. Additionally, a moderate concentration of acetic acid can be advantageous for the animal, as it is absorbed ruminally and converted into energy or fat that is incorporated into milk or body reserves [66]. The lactic/acetic ratio reflects the dominant type of fermentation during the ensiling process and its quality, with the ideal ratio reported by Kung et al. [54] being 3:1. In silage, although the lactic acid concentration was higher than that of acetic acid for most of the process, it did not approach this ratio, with an average acetic acid concentration exceeding the range considered optimal (10–30 g/kg DM) by Kung et al. [54]. The lactic fermentation carried out by LAB colonies during the ensiling process maintained the initial ethanol concentration inherent to WGP (first ethanol fermentation), reflecting the predominance of this energy-conserving pathway over less efficient routes [67]. The presence of this alcohol, together with lactic acid, contributed to maintaining a hostile environment for other undesirable microorganisms [43]. The tartaric acid naturally present in WGP remained stable, fostering a suitable environment for ensiling while preserving its own integrity [68]. The absence of propionic and butyric acids after the ensiling process indicates good fermentation quality.

The silage preserved its nutritional characteristics while minimising losses of dry matter (DM) and energy over time. The DM content of the silage aligns with the values reported by De Bellis et al. [47] and Monlosse et al. [16]. Additionally, the DM losses observed in silage (3.54%) in this experiment align with the findings of Robinson et al. [69], who stated that the major variations in DM occur in the form of effluents and gases, and with McDonald et al. [43], Chen et al. [70] and Benjamin da Silva et al. [71], who observed that DM losses due to the activity of lactic acid bacteria range between 2 and 5%, as it is converted into CO₂ when fermentation is driven by heterofermentative bacteria. In silage, during the first 7 days, a pH below 4.2 was reached and maintained until day 180, at which level proteolytic activity was significantly reduced [72], coinciding with an increase in the population of lactic acid bacteria and a significant rise in the lactic acid content, which has a high potential for reducing pH [63]. The pH of the ensiled WGP agrees with the level of a high-quality silage of 3.6–4.1 stated by Kulyk et al. [73]. The rapid pH decline and its stability over 6 months observed in the silage are indicative of high-quality silage [74], also supported by the Flieg score obtained, exceeding that achieved with other ensiled agro-industrial by-products [75,76] and aligning with the scores referenced by D’Alessandro et al. [77] for grape pomace silages, reaffirming the suitability of this by-product for the ensiling process. In contrast, the increase in DM (15.12%) observed in untreated WGP might be attributed to moisture losses in the form of effluents [78], explained by the growth of undesirable microbes. The loss of acidity observed in untreated WGP is due to the generation of alkalising metabolites as the degradation of organic matter progresses [79].

These results indicate that ensiling reduces pH and maintains DM for at least 180 days, both of which are key factors for the stability of other nutritional components [80]. Thus, during this period, silage preserved nutritional values for OM, ash, NDF, ADF, ADL, EE, CP, NPN, and starch. Total sugars in silage were consumed as substrates for the lactic acid fermentation during the ensiling process [81]. The determinations of these nutritional components using AOAC procedures corresponded to the fructose, glucose and sucrose levels observed by the HPLC methodology. In contrast, untreated material exhibited changes indicative of organic degradation over the 180-day period, with the inorganic fraction increasing by 47.3% as OM decreased. This material underwent mineralisation, similar to the processes observed in low-quality composts [79]. These authors further noted that during the decomposition of plant material, nitrogen and dissolved organic carbon were lost through leaching, accompanied by the release of volatile compounds such as ammonia. These losses explain the observed increase in the fibrous, protein and ether extract contents of the untreated material. Non-structural carbohydrates were the most rapidly and efficiently degraded macronutrients, serving as readily available energy sources during the decomposition of the untreated material [82].

The fatty acid profile of WGP aligns with most of the literature related to this by-product [83,84,85]. This matrix exhibits a high unsaturation index (UI) that is particularly rich in polyunsaturated fatty acids (PUFAs), which have been reported to induce several health benefits upon consumption [86,87]. Specifically regarding C18:2, it has been demonstrated in dairy cows that feeding strategies enriched with this compound are effective in inducing an increase in the concentration of conjugated linoleic acid (CLA) in milk and its derived dairy products [88]. CLA is endogenously produced in ruminants as an intermediate of the ruminal biohydrogenation of dietary C18:2. These compounds are found almost exclusively in milk and derivatives, and the positive health impact of their beneficial properties has driven the development of experimental feeding strategies for ruminants aimed at increasing their concentration in animal products [89]. The only observed difference between treatments lies in α-linolenic acid (C18:3 ω-3, ALA) and the health-promoting indices directly associated with it (n6n3, LAALA). Sealls et al. [90] reported that this fatty acid is highly susceptible to oxidation compared to other fatty acids, which could justify the lower presence in the untreated WGP compared to the silage. The IA, IT and HH indices are more useful than fatty acid composition for nutritional assessment [37]. The IA and IT values for WGP are considered low, being comparable to those of other plants, such as borage and hemp seed, respectively [91]. The HH index obtained for WGP falls within the range of a vegetable oil like olive oil [92], thus being considered healthy.

Phenolic compounds exhibit strong antioxidant activity that influences the antioxidant potential of food [22]. Grapes are attributed significant antioxidant potential due to their concentration of phenolic compounds, as well as flavonoids, anthocyanins and vitamins C and E [93]. These bioactive substances remain in the by-product after grape processing [3], highlighting the importance of studying this variable in grape pomace. Significant differences in polyphenol content in grape pomace have been reported depending on several factors, including grape varieties, crop locations, winemaking processes and even the distribution of these cultivars, which may vary based on geographic regions, harvest year and vintage [94,95]. The results from this study clearly demonstrate that, across all antioxidant activity evaluation methods applied (TP, DPPH and ABTS), the silage treatment had a significant effect. Ensiling favoured the preservation of antioxidant capacity for at least 6 months, as reported by Kammerer et al. [96], Fitri et al. [97] and De Bellis et al. [47], in contrast to the loss observed in untreated WGP. Additionally, some studies observed strong correlations between the hydrophilic antioxidant activities measured by DPPH and ABTS, suggesting that these methods have a similar predictive capacity for determining this bioactive property [22]. These authors further added that the correlation between the Folin–Ciocalteu (TP) method and DPPH and ABTS indicates that total phenolic content can be used as an indicator of the hydrophilic antioxidant activities of silage. In future research exploring the inclusion of silage in ruminant diets, it would be interesting to study the transfer of these ingested compounds to physiological factors in animals and their production, as cited by Caetano et al. [98] in beef cattle, Martin Flores et al. [13,99] and Mu et al. [100] in lambs, Bennato et al. [101] in Pecorino sheep cheese, and Akter et al. [87] in dairy cattle.

Considering the nutritional composition and energy contribution of each feedstuff, WGP silage has been shown to be more economical than other forages commonly included in conventional ruminant diets. Based on price data from the Lonja de Albacete [102] and the composition and nutritional value of feedstuffs used for compound feed manufacturing [103], the price of top-quality alfalfa hay is approximately 1.60 EUR/kg CP, which is comparable to the calculated price of WGP silage when considering its manufacturing costs. However, when accounting for the UFL provided by each of these feedstuffs to animals, WGP silage is 65% cheaper than top-quality alfalfa hay. Despite the production costs associated with baled silage manufacturing, the economic profitability of these by-product silages emerges as a key consideration. Muck [104] additionally highlighted other advantages that enhance the profitability of the baled silage format: it facilitates the easy handling and transport of ensiled feed, simplifies logistics, and enables commercialisation. Furthermore, it provides storage flexibility, optimising the use of available on-farm space.

5. Conclusions

The ensiling of white grape pomace (WGP) stabilised by day 35, marked by microbial and chemical indicators of fermentation stability. From this point onwards, WGP silage remained stable over six months, with minimal changes in dry matter, organic acid profile and antioxidant capacity, indicating preserved nutritional quality. The results demonstrate that ensiling is an effective method for preserving WGP, supporting its use as a ruminant feed ingredient. This approach offers a sustainable and cost-effective strategy for managing winery by-products and recirculating bioactive compounds within the livestock sector. Further in vivo studies are recommended to assess its impact on animal performance, health and product quality.