Dairy production in Algeria, particularly in the fertile Mitidja region, relies heavily on Holstein cow farming, a breed prized for its high milk yields. However, facing a massive forage deficit (Meklati et al., 2020), optimizing feed rations represents a major challenge to enhance the nutritional quality of milk, particulary its fatty acid profile, which directly impacts human health and farm profitability.

Varied diets combining base forages, concentrates and local by-products such as brewery draffs or corn silage can modulate these characteristics, but their specific effects in a local context remain underexplored.

This study aims to assess the impact of three distinct feed rations (R₁, R₂ and R₃) on the physicochemical composition and fatty acid profile of milk from Holstein cows, which are particularly sensitive to dietary changes, with significant alterations in the milk fatty acid profile occurring in just 15 days (Ouchene-Khelifi et al., 2017; Techera et al., 2023). Conducted from June to December 2023 across three farms in Tipaza, Blida and Algiers, it compares:
-   R1 (Tipaza): 1/3 concentrate feed (including 33% corn) and 2/3 base ration (oat hay and straw).
-   R2 (Blida): 50% brewery draffs, 21.43% concentrate feed (20% corn) and reduced base ration (28.58% oat hayand straw).
-   R3 (Algiers): High-energy diet with 65.22% corn/silage, 17.39% oat hay and 17.39% concentrate feed (corn, soybean meal, wheat bran), supplemented by abundant pasture (64 kg purple clover/cow/day).

Distribution and composition of feed rations
 
This study gathered 174 Holstein dairy cows (dc), distributed across three farms located in three regions: Tipaza (74 cows under ration R₁), Blida (51 cows under ration R₂) and Algiers (49 cows under ration R₃), in their 1^st, 2^nd and 3^rd lactation (Table 1).



This approach minimized climatic variability by selecting farms in the Mitidja plain, characterized by a Mediterranean climate and standardized the Holstein breed as well as lactation stages (1^st, 2^nd and 3^rd) to best isolate the specific effect of feeding under real farming conditions (management) without an adaptation period.

The experiment was conducted from June to December 2023. Table 2 presents the composition of the daily rations distributed: The first ration (R₁, Tipaza) consisted of 1/3 of concentrated feed and 2/3 of basic ration (oat hay and straw). The second ration (R₂, Blida) aimed to show the value of the draffs of the brewery (50%), easily obtained by the farmer, while reducing both the share of the basic ration (oat hay 14.29% and straw 14.29%) and the concentrated feed (21.43%). However, the 3^rd ration (R₃, Algiers) aimed to maximize the individual performances by feeding cows a high-energy diet, containing corn (65.22%) and oat hay (17.39%) and completing by 17.39% of concentrated feed (corn, wheat bran and soybean meal). The concentrate formulations were proposed and prepared by the farmers on site (Table 3).





In addition, the cows consumed an average of 50 kg of green forage per day (offered ad libitum in all farms), with each site receiving a specific type (alfalfa at Tipaza, clover at Blida, local grass at Algiers).

Table 4 presents the chemical composition of feed rations.


 
Milk sampling and physicochemical analysis
 
Milk samples were collected once a week in the evening. Each sampling corresponded to a composite sample prepared from a mixture of milk from the morning and the evening milkings, collected in clean bowls, identifided and preserved at 4^oC. pH was measured using a pH meter. The milk acidity and density were respectively measured according to the AFNOR standards (NF V04-206, 1969 and NF V04-204, 2004). The total dry extract (TDE) was determined by desiccation in infrared humidity analyzer. Protein levels (PL) were obtained after measuring the total amount of nitrogen following Kjeldahl digestion (AFNOR, 1986; NF V04-211), while fat levels (FL) were measured using Gerber method (NF 04-210, AFNOR 1986).
 
Fatty acids profile
 
The fatty acids of milk fat, extracted using the Röse-Gottlieb method (AFNOR 1986; NF EN ISO 1211), were analyzed in the form of methyl esters in order to avoid the drawbacks associated with the high polarity of these acids.

Methyl esters were obtained from triglycerides according to the AFNOR method (AFNOR 2000; NF T60-233), using 10 mL of chromatography-grade heptane and 0.5 mL of a methanolic potassium hydroxide solution for a 1 g milk fat test portion.

The fatty acid profile of milk fat was determined by analyzing the methyl esters thus obtained using gas chromatography equipped with an FFAP column and an FID detector (Thermo-Finnigan), according to LOCK et al., (2013): The carrier gas was helium at flow rate 0.5 mL/min. The injected sample volume was 1 µL. The injector temperature was maintained at 250^oC. The initial oven temperature was held at 40^oC for 4 min, increased to 175^oC at a rate of 13^oC/min (held for 27 min) and then increased by 4^oC/min to a final temperature of 215^oC (held for 35 min). The detector temperature was 215^oC.

Analyses of milk samples were carried out three times at the Nationale Higher Agronomic School (Algiers) and the average mean values were calculated.
 
Statistical analysis
 
The data were analyzed and the standard variance was obtained with criterion ANOVA 1, using Statistica ® version 6.1 (Statsoft, France). In case of ANOVA’s results were significant, Duncan’s test was utilized to compare average means. For this comparison, only one significant figure of 5% was retained.

According to the type of ration, a very highly significant effect (p<0.05) on the physico-chemical parameters was found, except for the pH, acidity and density (Table 5).



Rations R₂ and R₃ exerted a similar effect on milk TDE, which showed comparable values for these two diets. In contrast, ration R1 induced a significantly lower TDE (p<0.05), with differences of 4.75 and 8.05 g/L compared with rations R₂ and R₃, respectively.

Ration R₃ resulted in the highest protein content (p<0.05), with differences of 0.92 and 2 g/L compared with rations R₂ and R₁, respectively. Conversely, ration R1 produced the lowest protein content. The amount of protein can vary considerably under the influence of diet factors (Gupta et al., 2020; Bakshi and Wadhwa, 2023; Mishra et al., 2025). Hence, the increase in the amount of protein is more related to the quantity of the energy absorbed than to its source (Broderick, 2003). The high energy provided by the silage of corn (65.22%) (Table 2 and 4) used in Algiers farm (R₃), justifies the high protein level in milk. On the other hand, adding fat to the ration to increase its energy density has a depressive effect on the protein level (PL) (Martínez Marín et al.,  2013). The fat content of rations R₁ and R₂ was indeed higher than that of ration R₃ (Table 4). Meanwhile, rations based on corn (poor in these amino acids) lead to a decrease in protein levels. That way, the low protein level recorded with ration R₁ can be explained by the high proportion of corn included in the concentrated feed (33%, Table 3) which is rich in fat (Table 4), as opposed to what is used in the concentrated feed of ration R₂ (20% of corn, Table 3).

Furthermore, it can be observed that ration R3 significantly increased fat content (p<0.05) compared with rations R₂ and R₁. Fat level is the most sensitive element of the diet (Lock et al., 2013; Wang et al., 2024). An increase in the concentrated feed in the total ration negatively affects the fat level. In fact, an increase in the starch in the ration provided by the concentrated feed can cause the fat level to drop significantly (Rulquin et al., 2007), incriminating the glucose of the Corn.

Meanwhile, the increased fat level in the milk from the Algiers farm (R₃) compared to the milk of the otherfarms (gap of +2.87 and +1.22 g/L compared to R₁ and R₂ respectively, Table 5) could be explained by the low proportion of the concentrated feed (17.39%, Table 2) in the total ration and the excellent quality of pasture (approximately 64 kg of purple clover was consumed by cow per day). This purple clover provides abundant lipogenic precursors, particularly polyunsaturated fatty acids such as linoleic and á linolenic acids, which are hydrogenated in the rumen to form long chain fatty acids and conjugated linoleic acid, thereby stimulating milk fat synthesis (Lashkari et al., 2021). In the study conducted by Bony et al., (2005), where the proportion of concentrated feed reached 55% of the ingested dry weight, a slightly low-fat level was observed. That way, the ratio of feed/concentrated feed, which determines the cytoplasmic content in fibers and carbohydrates of the ration, is a crucial variation factor of the fat content in the cow’s milk. However, it is only with very high proportions of concentrated feed (more than 40% of the dry matter of the ration) that the fat level drops considerably; this decrease can vary from 3 to 10 g/kg of milk according to the type of additional feed or the nature of the feed used in the process. In this study, the proportion of concentrate in ration R₁ is 38.46%, whereas in ration R₂ it clearly exceeds 40%, due to the combination of 21.43% concentrated feed and 50% Draffs of brewery (Table 2).

The profile of fatty acids (Table 6) in cow’s milk fat was significantly different according to the feed ration. Rations R₂ and R₃ significantly decreased saturated fatty acid (SFA) levels compared with ration R₁, with differences of approximately -1.79% and -2.97%, respectively. In contrast, these two rations (R₂ and R₃) significantly increased monoun- saturated fatty acid (MUFA) levels, with differences of +1.71% and +2.80%, respectively compared with ration R₁.

Conversely, the variation in polyunsaturated fatty acids (PUFA) remained non significant across the three rations (p>0.05), even though rations R₂ and R₃ included pasture and corn silage, which are naturally rich in C18 PUFA such as linoleic and α linolenic acids. In ruminants, most dietary PUFA undergo extensive ruminal biohydrogenation: after lipolysis of plant lipids, 70-90% of released PUFA are isomerized and hydrogenated by rumen microorganisms into saturated end products, mainly stearic acid (C18:0), thereby markedly limiting the direct transfer of PUFA to milk fat (Dewanckélé et al.,  2020). Under the conditions of this study, the intensity of ruminal biohydrogenation appears to have offset the higher PUFA supply from clover based pasture and corn silage, so that no significant increase in milk PUFA could be detected in rations R₂ and R₃. Instead, the main lipid response to these diets was a redistribution within the fatty acid profile, characterized by an increase in MUFA secretion accompanied by a decrease in SFA, rather than a measurable rise in PUFA, compared with ration R₁ (Sun et al., 2022).

Then, our results have shown that the amount of SFAs represents between 60 and 63% of total milk fat. It is generally recognized that SFAs represent a risk factor for atherosclerosis by increasing the total cholesterol level and the cholesterol LDL level.

The fatty acids of the short carbonic chain in milk (C4 in 16) are synthesized intra-mammary. In contrast, the rest of the fatty acids, which have 18 carbon atoms and more, are transported by the chylomicrons (exogenous origin), the  VLDL (endogenous origin) and albumin (Rulquin et al.,  2007).

Generally, the decrease in the fatty acids content of both short and average carbonic chains results from the decrease in their mammary synthesis. The decrease in the de novo synthesis of the fatty acids of both short and average carbonic chains is compensated by an increase in the level of fatty acids with 18 carbon atoms in milk, which causes a slight decrease or preservation of the fat level (Glasser et al., 2008). However, the synthesis of the stearic acid (C18) is achieved by an elongation of the palmitic acid in the endoplasmic reticulum of all the cells.

For the last fifty years, the nutritional recommendations made in favor of preventing atherosclerosis urge the intake of PUFAs, in particular w6 and w3 and reduce the intake of SFAs. 

The grass diet is classically associated with an increase in the fatty acids of the milk, so the milk produced as a result of a grass diet showed a high SFA level (Martin et al., 2002). Besides, a production system based on the corn silage, concentrated feed obtained from the market or the by-products, is associated with a milk content high in linolenic acid (C18:2) and in MUFAs (Slots et al.,  2009). 

On the other hand, the contribution of the concentrated in the pasture significantly modifies the make-up of fatty acids, resulting in an increase in the SFAs and a decrease in UFAs; this is due to the reduced ingestion of the grass rich in UFAs. Concentrated feed low in fibers lowers the ruminal pH, causing a deviation in the fermentation pathways, which gives rise to the modifications of the ruminal biohydrogenation and, thus, the fatty acids profile of the milk (Peyraud and Apper-Brossard, 2006). The draffs are also rich in UFAs and also make UFAs increase in milk (Meribai et al., 2015). In light of these contrasting trends, further research is needed to confirm these observations.

Concretely, the high content of the unsaturated fatty acids observed with ration R2 (Table 6) can be attributed to the inclusion of draffs of brewery in the ration given to lactating cows, which supplies additional long chain C18 unsaturated fatty acids that are transferred directly into milk fat and partly inhibit de novo mammary synthesis of saturated fatty acids. Moreover, the draffs promote a favorable ruminal biohydrogenation pathway: Their high content of starch and soluble sugars supports fibrolytic bacteria (e.g., Fibrobacter succinogenes), maintaining a higher ruminal pH that favors the production of MUFAs like cis-9, trans-11 CLA over complete hydrogenation to SFAs (Guo et al., 2024). In contrast, the unsaturated fatty acid fraction recorded with ration R₃ reflects the combined effect of a higher proportion of pasture and the presence of corn silage. Both increase dietary C18 fatty acids, enhancing exogenous MUFA transfer into milk and limiting de novo SFA synthesis in the mammary gland. Specifically, corn silage’s lipid profile (rich in C18:2) interacts with ruminal bacteria (e.g., Butyrivibrio fibrisolvens) to shift biohydrogenation toward trans-10 C18:1 isomers, which inhibit mammary Δ9-desaturase and acetyl-CoA carboxylase, leading to the observed decrease in palmitic acid (C16:0) (Guo et al., 2024).



Despite these consistent trends, relatively few studies have accurately quantified milk fatty acid composition while systematically comparing different forages, concentrates, lipid supplements and their interactions, which makes it difficult to define precise response models for individual fatty acids of interest in relation to dietary changes.

In this experiment, the diet has a highly significant effect on the proteins. Indeed, the highest value of the protein level was found under diet R₃, explained by the higher contribution of energy (corn silage). The fat level increased with diet R₃, this fact could be explained by the low portion of the concentrated feed (17.39%) in the total ration and the excellent quality of pasture. Besides that, a decrease in the SFA, compensated by an increase in the secretion of the MUFA, was recorded as a result of the ingestion of diets R₂ and R₃. On the reverse, the variation of the UPFA remained insignificant according to the diet as a factor. Indeed, the high content of the unsaturated fraction associated with diet R₂ may be explained by the presence of draffs of the brewery, while the fraction recorded under diet R₃ may be explained by the higher portion of the pasture and the presence of corn silage rich inunsaturated fatty acids.

In the Algerian context, characterized by a structural forage deficit and a strong dependence on imported concentrates for feeding dairy cows, the valorization of local co-products emerges as a strategic lever to enhance the sustainability of livestock systems. Brewery spent grains, produced in substantial quantities by processing units located particularly in coastal and urban areas, represent an abundant fibrous and protein-rich resource, while simultaneously reducing the reliance on more expensive concentrate feeds.
 
Disclaimers
 
The views and conclusions expressed in this article are solely those of the authors and do not necessarily represent the views of their affiliated institutions. The authors are responsible for the accuracy and completeness of the information provided, but do not accept any liability for any direct or indirect losses resulting from the use of this content.
 
Informed consent
 
All animal procedures for experiments were approved by the Committee of Experimental Animal Care and Handling Techniques were approved by the University of Animal Care Committee.

The authors declare that there are no conflicts of interest regarding the publication of this article. No funding or sponsorship influenced the design of the study, data collection, analysis, decision to publish, or preparation of the manuscript.