The use of industrial citrus wastes in dairy cattle nutrition: An in vitro approach on gas production and digestive parameters

2.1 Feed materials

In this study, by-products of grapefruit (Riored) (Citrus × paradisi (Macfad.)), lemon (Lamas) (Citrus limon (L.)), mandarin (Satsuma) (Citrus unshiu (Swingle) Marcow.), and orange (Navelina) (Citrus sinensis (L.) Osbeck) were obtained from a private fruit juice factory located in Mersin Province, Turkey. Along with feed ingredients supplied from a private dairy farm, samples were collected in accordance with official sampling regulations to ensure homogeneity. The collected samples (The procurement of raw materials considered the harvest period of the fruits, which occurred between October 2024 and January 2025) were shade-dried in the laboratory and subsequently ground to pass through a 0.5 mm screen before being subjected to chemical analyses.

Total mixed rations (TMRs) were formulated based on the average performance of a Montofon dairy cattle herd from the same farm. The herd had an average live weight of 635 kg, a daily milk yield of 29 L, and the milk composition of 3.5% fat, 3% protein, and 3% lactose. According to national research council (NRC) (2001) requirements, the animals’ dietary needs were estimated as 14.12% crude protein and 8.58 MJ/kg dry matter (DM) metabolizable energy. Rations were thus formulated to meet these requirements.

2.2 Chemical analyses

In the laboratory-prepared TMRs, grapefruit, lemon, mandarin, and orange by-products were included as partial substitutes for corn in iso-caloric and iso-nitrogenous proportions. The diets consisted of nine groups: control (without additives), G15 (15% grapefruit), G30 (30% grapefruit), L15 (15% lemon), L30 (30% lemon), M15 (15% mandarin), M30 (30% mandarin), P15 (15% orange), and P30 (30% orange).

The chemical composition of the TMR groups was analyzed for DM, EE (Ether Extract), CP (Crude Protein), and ash according to AOAC (1998), and for ADF (Acid Detergent Fiber), NDF (Neutral Detergent Fiber), and hemicellulose (HC) using the ANKOM2000 Fiber Analyzer following Van Soest et al. (1991). Formulation ratios, feed ingredient composition, and ration composition have been given in Tables 1-3.

For in vitro gas production and Daisy incubator assays, rumen fluid was collected according to (Kılıç and Abdiwali 2016) from three Montofon cattle (4–5 years, 510–565 kg). Samples were transported to the laboratory at ∼39°C in insulated containers under CO₂ (Carbon dioxide), filtered through cheesecloth to maintain anaerobic conditions, and then used for analyses.

The 24-h gas and methane production of TMRs (Control, G15, G30, L15, L30, M15, M30, P15, P30) were determined using the in vitro method of Menke et al. (1979). For each TMR, 0.5 g DM was incubated in triplicate with 40 mL buffered rumen fluid in 100 mL glass syringes for 24 h. Gas volumes were corrected with the Hohenheim standard (44.16 mL/g DM/24 h), and methane (%) was measured using an infrared analyzer (Goel et al., 2008).

True dry matter digestibility was determined by transferring syringe residues to beakers with NDF solution, boiling, and filtering through crucibles, following Van Soest et al. (1991) and Blümmel et al. (1997).

ME (Metabolizable energy), NEL (Net energy for lactation), and OMD (Organic Matter Digestibility) of feed ingredients and citrus by-products were calculated using Menke and Steingass (1988):

TDDM (True Dry Matter Digestibility), PF (Partitioning Factor), MY (Microbial Yield), TDD (True Dry Matter Degradability), and MPSE (Microbial Protein Synthesis Efficiency) were calculated according to Blümmel et al. (1997) and Vercoe et al. (2010):

In vitro digestibility (DM, NDF, OM) was measured with the Ankom DaisyII incubator (Van Soest et al., 1991; Ankom Technology, 2005). Bags with buffered rumen fluid were incubated for 48 h, washed, dried, and weighed to calculate %IVDMD (In Vitro Dry Matter Digestibility), %IVNDFD (In Vitro Neutral Detergent Fiber Digestibility), and %IVOMD (In Vitro Organic Matter Digestibility) using standard formulas.

2.3 Statistical analyses

Data were analyzed by ANOVA, and means were compared with Duncan’s multiple range test (Duncan, 1955) using SPSS 20.0 (IBM, 2017). Polynomial contrasts were applied to evaluate dose effects.

3.1 Effects of citrus by-products on ruminal fermentation parameters (Gas, Methane, True Dry Matter Digestibility, Partitioning Factor, Microbial Yield, Microbial Protein Synthesis Efficiency, True Dry Matter Degradability)

In this study, the inclusion of citrus by-products at different levels (15% and 30%) in dairy cattle rations resulted in significant differences in the in vitro gas production, methane yield, and digestibility parameters of the experimental groups (Table 4).

Total gas production was highest in Orange30 (112.94 mL), followed by Lemon15 (110.56 mL), Mandarin30 (111.48 mL), and Lemon30 (109.03 mL) (p < 0.05). The lowest value occurred in Grapefruit30 (83.68 mL), suggesting that high grapefruit inclusion limits fermentation. Methane production was greatest in Lemon15 (22.28 mL), Lemon30 (22.39 mL), and Orange30 (22.31 mL), at or above control levels (p < 0.05). The lowest methane was in Grapefruit30 (14.40 mL). Methane percentage showed a similar trend, highest in Orange15 (20.57%), Lemon30 (20.53%), and Lemon15 (20.15%). True dry matter digestibility increased in most citrus groups, with Orange15 (289.28 mg/g) and Orange30 (287.99 mg/g) showing the highest values (p < 0.01), likely due to their energy-rich composition. The partitioning factor was highest in Grapefruit30 (3.37 mg/mL), reflecting greater digestibility with lower gas output. Other citrus groups had PF values similar to the control (2.50–2.69). Microbial yield and microbial protein synthesis efficiency were also highest in Grapefruit30 (98.48 mg and 34.75 mg, respectively; p < 0.01), indicating efficient microbial biomass production with reduced gas. Other groups were comparable to the control.

Total digestibility score was highest in Orange15 (57.80%) and Orange30 (57.71%), showing that orange by-products improved overall digestibility.

3.2 Effects of citrus by-products on digestibility parameters (DMD, NDFD, OMD, ME, NEL)

Evaluation of the in vitro digestibility parameters obtained by incorporating citrus by-products at different levels into the TMR revealed statistically significant differences among the groups (p < 0.05) (Table 5).

In vitro Dry matter digestibility ranged from 36.30% to 49.10%, highest in Mandarin15 (49.10%) and lowest in Grapefruit15 (36.30%), showing species-related differences. Neutral detergent fiber digestibility was also highest in Mandarin15 (43.00%) and lowest in Grapefruit15 (18.99%), confirming Mandarin’s positive and Grapefruit’s negative effect on fiber digestibility. IVNDFD also showed marked variation, being highest in Mandarin15 (43.00%) and lowest in Grapefruit15 (18.99%), confirming Mandarin’s positive and Grapefruit’s negative effect on fiber digestibility. IVOMD% varied slightly (95.95–97.00%), with Lemon15 highest (97.00%) and Grapefruit15 lowest (95.95%). Metabolizable energy and net energy for lactation differed with species and inclusion levels. Orange30 showed the highest ME (9.00 MJ/kg DM) and NEL (5.33 MJ/kg DM), while Grapefruit30 showed the lowest ME (7.34 MJ/kg DM) and NEL (4.10 MJ/kg DM), highlighting orange as a better energy source. Organic matter digestibility score ranged from 54.81 to 65.26, with Orange30 highest, supporting orange by-products as enhancers of overall digestibility.

3.3 Evaluation of citrus by-products by multivariate statistical methods principal component analysis (PCA) Results

The PCA biplot showed clear structural differences among citrus by-products based on chemical composition, gas production, and digestibility. Using four variables (Net Gas, Net Methane, IVDMD%, TDMD), PC1 explained 51.81% and PC2 33.32% of the variance, together accounting for 85.13% (Fig. 1). This indicates strong explanatory power of the selected variables.

Close

Fig. 1.

PCA biplot of in vitro gas production, methane production, and digestibility parameters with citrus by-products at 15% and 30% inclusion levels in dairy cattle rations.

Export to PPT

PC1 was associated with gas production (Net Gas, Net Methane), while PC2 was linked to digestibility (IVDMD%, TDMD).

Orange30: High NET GAS and NET METHANE; strong fermentation but moderate digestibility.

Orange15: Average gas and digestibility, with variability among samples.

Lemon30: Balanced profile; medium-high gas and TDMD, with differences in digestibility observed.

Grapefruit30: Lowest gas and methane, highest TDMD; low fermentation and high fiber content.

Grapefruit15: Low gas, high TDMD, but reduced IVDMD; fibrous and less digestible structure.

Mandarin15 & Mandarin30: Average gas and IVDMD, higher TDMD; moderate energy with fiber-rich traits.

Lemon15: Highest IVDMD; easily degradable and highly digestible material.

Control: Center of the PCA; balanced and average values, serving as the reference group.

3.4 Hierarchical cluster analysis (HCA) results

In this analysis, nine different TMR samples containing citrus processing by-products were clustered using hierarchical cluster analysis (Ward’s method) based on 14 biochemical and fermentative parameters (Net Gas, Net Methane, %Methane, TDMD, PF, MY, MPSE, TDD, IVDMD%, IVNDFD%, IVOMD%, ME, NEL, OMD). As a result of the analysis, the samples were clearly separated into four distinct clusters (Groups A–D) (Fig. 2).

Close

Fig. 2.

HCA of in vitro fermentation parameters and digestibility characteristics with citrus by-products at 15% and 30% inclusion levels in dairy cattle rations.

Export to PPT

Group A (P15, P30 – Orange pulp): High NET GAS, OMD, ME, and NEL values, reflecting easy rumen fermentability, strong microbial activity, and high energy supply. Orange pulp’s soluble carbohydrates and low fiber supported high fermentation potential.

Group B (M15, M30, L15 – Mandarin & Lemon pulp): Moderate fermentative capacity with balanced energy and fiber. Provided consistent digestibility (TDMD, PF, DMD%), supporting both microbial activity and ration structure.

Group C (Control, L30 – Control & Lemon30): Stable, reference-like profile with average gas, IVOMD, and energy values. Suitable for long-term feeding strategies where stability is preferred.

Group D (G30 – Grapefruit30): Formed a distinct cluster with the lowest Net Gas and Net Methane, but the highest TDMD and PF. Low fermentability but high fiber density, potentially useful for rumen motility despite reduced digestibility.

Overall, clustering showed that citrus by-products behave differently in the rumen: Group A offered rapid fermentation and high energy; Group B balanced energy and fiber; Group C represented stability; and Group D displayed unique, fiber-rich traits requiring careful evaluation.

4.1 Effects of citrus by-products on in vitro fermentation parameters

Gas production: Orange30 (112.94 mL), Lemon15 (110.56 mL), and Mandarin30 (111.48 mL) showed the highest values, Grapefruit30 the lowest (83.68 mL). Similar results were reported by Giller et al. (2022) and Romero-Huelva et al. (2012). High soluble carbohydrates and pectin in orange and mandarin pulps increased fermentable substrate and microbial activity (Bampidis & Robinson, 2006; Romero-Huelva et al., 2013; Giller et al., 2022). Grapefruit pulp, with high lignin and phenolics, limited fermentation (Vastolo et al., 2022; Yu et al., 2024).

Methane production: Lemon15 (22.28 mL), Lemon30 (22.39 mL), and Orange30 (22.31 mL) had the highest methane, Grapefruit30 the lowest (14.40 mL). Vastolo et al. (2019) and Giller et al. (2022) reported similar values. Grapefruit’s lower methane relates to high fiber and phenolics.

Methane percentage: Orange15 (20.57%), Lemon30 (20.53%), and Lemon15 (20.15%) were the highest (Giller et al., 2022). This may reflect acetate-driven fermentation (Giller et al., 2022; Vastolo et al., 2019). Grapefruit30 (17.21%) confirmed polyphenol inhibition of methanogens (Gadulrap et al., 2023).

True dry matter digestibility (TDMD): Orange15 (289.28 mg/g) and Orange30 (287.99 mg/g) were highest (p < 0.05). Giller et al. (2022) and Romero-Huelva et al. (2013) also reported high values for orange pulp. Orange’s pectin and low neutral detergent fiber (NDF) explain faster digestion (Bampidis & Robinson, 2006; Romero-Huelva et al., 2012). Grapefruit30 combined high TDMD with low gas, suggesting phenolics reduce gas but not digestibility (Vastolo et al., 2019).

The higher PF, MY, and MPSE values observed in the grapefruit pulp groups indicate that, despite lower gas production, microbial protein synthesis was more efficient. Blümmel et al. (1997) and Vercoe et al. (2010) also emphasized that increased PF and MPSE are commonly associated with fibrous and polyphenol-rich feeds, reflecting a shift of fermented substrates toward microbial biomass production. This suggests that grapefruit pulp, although limited as an energy source, may play a supportive role in enhancing rumen microbial efficiency. On the other hand, the elevated TDD values in the orange pulp groups can be attributed to their higher soluble carbohydrate and pectin content, which promote improved digestibility (Romero-Huelva et al., 2012; Giller et al., 2022). Overall, these findings demonstrate that different citrus by-products exert distinct functional roles in the rumen: orange pulp enhances digestibility and energy contribution, while grapefruit pulp favors microbial protein synthesis efficiency.

4.2 Effects of citrus by-products on in vitro digestibility parameters

In vitro dry matter digestibility (%) (IVDMD%): Ranged from 36.30% (grapefruit) to 49.10% (mandarin). Higher digestibility with mandarin relates to soluble carbohydrates and moderate fiber (Romero-Huelva et al., 2012; Giller et al., 2022). Grapefruit’s low IVDMD is linked to naringin and limonin (Vastolo et al., 2019). Lashkari et al. (2013) and Heuzé (2018) confirmed higher values for mandarin and orange pulp.

In vitro neutral detergent fiber digestibility (%) (IVNDFD%): Varied from 18.99% (grapefruit) to 43.00% (mandarin). Mandarin’s less lignified fiber supported higher values, while grapefruit’s phenolics reduced cellulolytic activity (Vastolo et al., 2019). Similar results highlight the positive role of orange pulp (Bampidis & Robinson, 2006; García-Rodríguez et al., 2020).

In vitro organic matter digestibility (%) (IVOMD%): Narrow range (95.95–97.00%), highest in Lemon15, lowest in Grapefruit15. Lemon’s low lignin and high pectin improved breakdown (Romero-Huelva et al., 2012), while grapefruit’s phenolics reduced it (Vastolo et al., 2019). Giller et al. (2022) also found higher IVOMD for orange pulp.

ME: Ranged from 7.34 MJ/kg DM (Grapefruit30) to 9.00 MJ/kg DM (Orange30). Orange pulp’s soluble carbohydrates increased VFA and energy (Giller et al., 2022; Romero-Huelva et al., 2013). Grapefruit pulp reduced ME due to phenolics (Vastolo et al., 2019).

NEL: Varied between 4.10 MJ/kg DM (Grapefruit30) and 5.33 MJ/kg DM (Orange30). Orange pulp supported higher milk yield (Romero-Huelva et al., 2012; Bampidis & Robinson, 2006), and grapefruit showed a limited effect.

4.3 PCA

The PCA biplot showed clear differentiation of citrus by-products (orange, mandarin, lemon, and grapefruit at 15% and 30%) based on gas, methane, IVDMD%, and TDMD. PC1 explained 51.81% and PC2 33.32%, together 85.13%.

PC1 correlated positively with Net Gas and Net Methane, positioning Orange30 and Lemon30 on the positive side. This aligns with reports on the high soluble carbohydrate and pectin content of orange and lemon pulps, which accelerate fermentation and increase gas (Bampidis & Robinson, 2006; Romero-Huelva et al., 2012; Giller et al., 2022). Lower phenolics in orange pulp likely reduced methanogen inhibition, supporting higher methane.

PC2 was linked to IVDMD% and TDMD. Grapefruit30 and Grapefruit15, with high TDMD but low gas, clustered here. Phenolics like naringin and limonin suppressed methanogens and cellulolytic bacteria, reducing gas while allowing fiber digestion (Vastolo et al., 2019). This may have redirected substrates to microbial biomass, raising TDMD.

Orange30 separated clearly on PC1, confirming high gas and methane yields and strong ruminal fermentation, likely driven by its high content of readily fermentable sugars and pectin, which accelerate acid production and pH decline (Romero-Huelva et al., 2013; Gessner et al., 2020). Grapefruit30 was the most isolated group, characterized by low gas production and relatively high TDMD, a pattern that may be associated with the high polyphenolic content and lignin-rich structure of grapefruit pulp, which are known to limit rumen fermentation and gas production (Van Soest et al., 1991; Vastolo et al., 2019; Correddu et al., 2020; Yu et al., 2024).

Mandarin groups (M15, M30) were central, with balanced gas and digestibility, reflecting moderate fiber and soluble carbohydrates. Compounds like hesperidin showed limited inhibition of methanogens. Lemon groups, especially Lemon15, scored high on PC2 with elevated IVDMD%, explained by low lignin and high pectin. Its acidity slightly reduced methane but not digestibility.

Overall, PCA confirmed differentiation of citrus pulps by chemical composition and fermentation traits, consistent with previous reports.

4.4 HCA

HCA clustered the nine ratios into four groups using Ward’s method, reflecting citrus type and inclusion level effects.

Group A (P15, P30): High NET GAS, OMD, ME, and NEL due to soluble carbohydrates, high pectin, and low NDF. Orange pulp’s low phenolics reduced inhibition, yielding high methane (Romero-Huelva et al., 2012; Giller et al., 2022).

Group B (M15, M30, L15): Moderate gas, high digestibility, balanced profile. Less lignified fiber and pectin provided energy and structure. Phenolics partly restricted methane without suppressing fermentation.

Group C (Control, L30): Stable, reference-like profile. Lemon30 resembled control, maintaining pH and microbial balance. Its acidity slightly suppressed methane without reducing energy.

Group D (G30): Isolated cluster with the lowest gas and methane, but the highest TDMD and PF. High naringin, limonin, lignin, and peel proportion inhibited methanogens and cellulolytic bacteria (Vastolo et al., 2019).

In summary, HCA showed orange pulp with high fermentation and energy, mandarin and lemon with balanced profiles, and grapefruit with low gas but high fiber traits.