The Effects of Replacing Dry Forage With Corn Silage on Milk Yield, Composition and Fatty Acids’ Pro les, Blood Metabolites, Nitrogen Balance, and Rumen Fermentation Parameters in Mahabadi Lactating Goats


 The present study was performed with the aim of investigating the effects of replacing dry forage with corn silage on performance, milk yield, composition and fatty acids’ profiles, blood metabolites, nitrogen balance, and rumen fermentation parameters in Mahabadi lactating goats. For this purpose, a total of 20 lactating goats aged between 2 and 5 years old and with body weight of 45.3 ± 7.20 (DS) kg were categorized into two groups, each one containing ten goats. Experimental treatments consisted of controlling diet and dietary substitution of forage part with corn silage at 20% of dry matter. The obtained data were analyzed by the mixed model for a randomized completed design using statistical packages of SAS (2002). The replacement of dry forage with corn silage did not affect the dry matter intake (DMI), live body weight, and milk yield. Feeding corn silage, rather than dry forage, had no significant effect on milk unsaturated fatty acids, monounsaturated fatty acids, C18:1C9 percentages, and non-esterified fatty acids concentration, and it caused a decline in milk polyunsaturated fatty acids percentage. As well, dietary inclusion of corn silage significantly decreased plasma urea nitrogen content in goats (P<0.05). Furthermore, feeding corn silage led to a significant increase in antioxidant capacity of rumen liquor (P<0.05), plasma (P<0.01), and milk (P<0.01) of lactating goats. Of note, the dietary addition of corn silage caused no significant effects on microbial nitrogen and nitrogen balance in lactating goats. Feeding corn silage did not affect the ruminal total volatile fatty acids, acetic acid, propionic acid, and butyric acid at both periods before and after feeding. In general, results indicated that replacing dry forage with corn silage could consequently improve the antioxidant status in Mahabadi lactating goats.


Introduction
The provision of foodstuffs in the ruminant's industry leads farmers to incur a high cost due to the high foodstuffs' prices (USDA-ERS 2010). So, using cheaper foodstuffs is very important to modulate the foodstuffs' price (Shi et al. 2015). On the other hand, ruminants need to consume forage to maintain the optimum production and health via the ruminal environment stability. It is noteworthy that about 40 to 90% of livestock requirements in ruminant production systems are provided by forage intake (Mahanna and Chase 2003). Therefore, the decreased brous feedstuffs intake, especially NDF amount, in ruminants leads to the lower ruminal pH level, consequently resulting in the digestive ineffectiveness. In this regard, this then causes the decreased dry matter intake, milk yield and fat contents, and ber digestibility (Plaizier et al. 2008). While feeding diets containing forage are necessary to maintain the maximum health and production of ruminants, dietary inclusion of ber-rich roughage also increases methane emissions (van Gastelen et al. 2015). So, feeding starch-rich forage is recommended to reduce methane production and emissions (Hassanat et al. 2013).
Although alfalfa hay and corn silage are the two main forages used in dairy diets (NRC 2007), in developing countries, the inconsistency of forage quality as well as the shortage of high quality-forage limit the developments in the agriculture industries, particularly ruminant's industry (Khaing et al. 2015).
In this regard, feeding conserved forage like ensiling is an important strategy used to overcome the inconsistent quality of forage (Khaing et al. 2015). Corn silage is considered as a major forage component used in most dietary rations of ruminants (Khan et al. 2012). Because it is rich in starch (between 25 and 35% DM basis; NRC 2007), it provides a remarkable amount of fermentable energy to the ruminal microbial population resulting in the rises in feed intake, milk yield, and milk protein (Khaing et al. 2015), and it can also reduce methane emissions (Khan et al. 2012; Wattiaux and Karg 2004).
Similarly, corn silage, which is rich in polyunsaturated fatty acids (PUFA), especially n-3 and n-6 types, has been used to alter and modulate milk fatty acid components (Dewhurst et al. 2006). Furthermore, corn silage relative to other forages is readily mechanized, stored, and nally mixed with diets (Charmley 2001; Neylon and Kung 2003). Nevertheless, the nutritive value of corn silage mostly relies on corn density, degree of maturity, hybrid, different growing conditions, the moisture level, and ensiling conditions (Satter and Reis 2012). Many factors, including aerobic deterioration, moisture content, forage composition, and the elevated silage pH level can affect corn silage quality due to the produced mycotoxins concentration leading to the abortion in dairy herds (González-Pereyra et al. 2008; Schmidt et al. 2015).
In regard to the low facilities used to dry corn hay as well as the high cost related to the produced dry forage and its storage in Iran, it seems that using silo technology is an e cient strategy. Furthermore, in the present study, we hypothesized that feeding corn silage, which is high in anthocyanin content, could improve the antioxidant status in milk of Mahabadi lactating goats. So, this study was performed to evaluate the effects of replacing dry forage with corn silage on feed intake, milk yield, composition and fatty acids pro les, blood metabolites, nitrogen balance, antioxidant capacity, and rumen fermentation parameters in Mahabadi lactating goats.

Animals and treatments
The present study was conducted in the Small Ruminant Station of Agricultural Research, Education and Extension Organization, Animal Science Research Institute of Iran, Karaj, Iran. Moreover, all the protocols were approved by the Agricultural Research, Education and Extension Organization, Animal Science Research Institute of Iran, and Karaj Care and Use Committee. A total of 20 lactating goats aged between 2 and 5 years old with body weight of 45.3 ± 7.20 (SD) kg were selected and then divided into two groups, each one containing ten goats in terms of their age and body weight. Thereafter, every 10 goats were randomly housed into two boxes (3m × 10m) each one containing 3 goats and one box containing 4 goats during a 56-day period. The experimental treatments were controlling diet and dietary substitution of forage part with corn silage at 20% of dry matter. Accordingly, the experimental diets were prepared in TMR ration, and formulated using the exact values obtained in terms of NRC (2007), in order to meet all the nutritional requirements of small ruminants at lactation time ( Table 1). As well, the studied goats were fed with experimental diets as TMR twice, once at 8:00 AM and once at 20:00 PM, ad libitum. The amounts of offered feeds and orts were weighed daily. Additionally, each individual's feed intake was recorded by subtracting daily weight of offered feed from refused quantities for each group containing 3 or 4 animals. Next, it was divided by 3 or 4. Of note, all these goats had free access to water during the experiment. As well, they were individually weighed on both days 0 and 56 at 8:00 AM after 16h deprivation. Body weight gain for goats was calculated by the nal body weight minus the initial body weight. The chemical compositions, including dry matter, crude protein, calcium and phosphorous of forage parts, and concentrates used in this trial were determined using the AOAC methods (AOAC 2006

Milk yield and composition
Goats were handy milked twice daily, once at 06:00 and once at 18:00 h. Thereafter, Milk samples (50 ml) were collected from the sequential a.m. and p.m. The milking process was performed every two weeks after parturition during the experiment using prelabeled plastic vials treated with preservative (potassium dichromate, Merck) and the samples were then stored at 4C. Subsequently, these milk samples were analyzed in terms of pH, somatic cell counts, fat, crude protein, solids, solids non-fat, and lactose by infrared spectroscopy (Milk Oscan, 134 BN Foss Electric, Hillerød, Denmark; AOAC 2006). Furthermore, De novo fatty acids, mixed fatty acids, and free fatty acids were evaluated as g/100g of milk using Fourier transform infrared spectroscopy (FTIR) of CombiScope FTIR 600 HP (Delta Instruments, Drachten, The Netherlands) based on the partial least squares prediction models, as explained in a study by Woolpert et al. (2016). In addition, as previously described by Bach et al. (2019), de novo relative percentages were assessed by calculating the proportions of de novo fatty acids to total milk comprising of de novo, mixed, and preformed fatty acids. Next, Milk urea nitrogen content was assessed using a differential pH technique (Luzzana and Giardino 1999). As well, Lipids were extracted from the obtained milk samples using the method explained in a previous study by Folch et al. (1957). Notably, fatty acids pro les were determined according to a study by Feng et al. (2004). Finally, the levels of beta hydroxyl butyrate, nonesteri ed fatty acids, and acetone were measured using a clinical autoanalyzer (Hitachi 7020, Tokyo, Japan).

Plasma metabolites
To determine some biochemical plasma parameters, ve goats from each treatment group were randomly selected at 7:30 AM and then bled from jugular vein, at the end of the trial period. Thereafter, the obtained blood samples were centrifuged (Sigma-16-P-Germany) at 3000 × g for 20 min at 4c and the plasma samples were kept at -2 0C. The concentrations of triglyceride, cholesterol, glucose, total protein, albumin, and urea nitrogen were measured using commercial assay kits attained from Farasamed Diagnostics (Tehran, Iran) via a spectrophotometer (Genova; Jenway, Barloworld Scienti c Ltd., Dunmow, Essex, UK).

Ruminal parameters
At the end of the trial, by passing 56 days from kidding, ve goats were randomly selected and then rumen liquors (about 50 ml) were taken from the esophageal tube once before and once 3 h after the feeding time in the morning. Subsequently, the rumen pH level was measured using a mobile pH meter (Rocky Mount NC 27894-USA). To determine the ammonia nitrogen concentration, 5 ml of each one of the rumen uid samples was mixed with 5 ml of 0.2N HCL, and then centrifuged (Sigma-2-16-Germany). Finally, the ltered ruminal uid was kept at -2 0C, and Ammonia nitrogen concentration was also measured by micro kjeldahl apparatus, as explained earlier by Nasserian (1996).
To measure volatile fatty acid concentration, 5 ml of the ltered ruminal uid was mixed with 1 ml of 20% of meta-phosphoric acid, and then centrifuged at 4000 × g for 10 min. Afterward, 2-ethylbutyric acid was added to approximately 2 ml of the supernatant and stored at -2 0C. As previously described in a study by Bhandari et al. (2007), the concentrations of volatile fatty acids, acetic acid, propionic acid, butyric acid, valeric acid, and iso valeric acid were determined through gas chromatography (Agilent 6890 Silica Capillary Column BPX-70).

Antioxidant capacity
At this stage, the rumen liquor, blood plasma, and milk samples were collected to evaluate the total antioxidant capacity using the ferric reducing antioxidant power method (Benzie and Strain, 1996).
Accordingly, this was done via the reduced ferric tripyridyltriazine (Fe 3+ -TPTZ) complex to ferrous (Fe 2+ ) form, which consequently produced an intense blue color with an absorbance rate of 593nm using a spectrophotometer (Genova; Jenway, Barloworld Scienti c Ltd., Dunmow, Essex, UK).

Microbial N synthesis and N retention
In this phase, ve goats from each treatment group were randomly selected and then housed in individual metabolic cages (0.6m × 1.2m) during 21 days (14 days for being adapted and 7 days for sample collection) at the end of the trial period (started from day 57). From each animal, feces and urine samples were separately collected daily for one week. The feces collected from each animal were completely mixed and a sample weighing 100 g was obtained at last, which was kept in the freezer. The urine sample of each animal was collected daily in containers with 100 ml of 10% sulfuric acid during a 7-day period. The urine was weighed daily and 10% of the urine amount was then taken and diluted 4 folds (ratio of urine to water: 1:4) in order to prevent purine derivatives (e.g. uric acid) sedimentation. Finally, all these feces and urine samples were mixed and one sample per each goat was used for the chemical analyses.
Total purine derivatives excreted in the urine, including allantoin, uric acid, xanthine, and hypoxanthine, were evaluated using a spectrophotometer. As well, the urinary allantoin content was assessed at an absorbance rate of 522nm after conversion to phenyl hydrazine. Uric acid concentration was also determined in the optical density on 293nm after performing degradation of uric acid to allantoin via uricase (Product No. U-9375, Sigma-Aldrich, Darmstadt, Germany). Moreover, Xanthine and hypoxanthine contents were evaluated at an absorbance of 293 nm after conversion to uric acid via xanthine oxidase enzyme (Product No. X-1875, Sigma-Aldrich, Darmstadt, Germany). As explained in a study by Chen and Gomes (Chen and Gomes 1995), the amount of the absorbed exogenous purines was assessed by the excreted urinary purine derivatives using the following equation: Where X is the absorbed exogenous purine, Y stands for the excreted purine derivatives, and W is the live weight.
The microbial protein synthesis was calculated by determining the urinary excreted and the absorbed purine derivatives by the following equation (Chen and Gomes 1995): Where Y stands for the excreted purine derivatives, X shows the absorbed purine derivatives, (0.15W 0.75 e −0.25X ) shows purine derivatives with internal origin, and W is metabolic weight.
The nitrogen balance was also calculated as the differences between total nitrogen intake and the excreted nitrogen in both urine and feces samples (Philips and Rao 2001). The concentrations of nitrogen in feed, feces, and urine samples were measured by a micro kjeldahl apparatus (AOAC 2006).

Statistical analysis
All the obtained data were analyzed as a mixed model using the statistical package of SAS (2002) for completing the randomized design. The statistical signi cance level was set at p≤0.05. The statistical model for data analyses was as follows: Where Y ij is observation, µ is the general mean, T i is the effect of treatment, A j is the effect of age, W k is the effect of body weight, and e ijkl is the standard error of the mean.

Results
The effects of the dietary addition of corn silage on feed intake, milk yield, and milk composition analyses in Mahabadi goats during lactation phase are presented in Table 2. The dietary inclusion of corn silage caused no signi cant effects on feed intake, milk yield, fat, protein, lactose, solids, solids nonfat, total cell numbers, de novo relative fats, the mixed relative fats, performed relative fats percentages, and free fatty acids. The inclusion of corn silage in diet, instead of dry alfalfa/straw, signi cantly (P<0.03) decreased the milk urea nitrogen concentration in lactation goats ( Table 2).
In the present study, milk fat percentage by passing four weeks from the time of lactation signi cantly (P<0.01) decreased in goats fed with a diet containing corn silage compared to those fed with diet without any corn silage (as shown in supplementary Table 1 as shown in Online Resource).
As indicated in Table 3, although feeding with a diet containing corn silage had no signi cant effects on milk C 16:0 , and C 18:0 percentages, as well as beta hydroxy butyrate, and acetone concentrations during the present trial, it signi cantly caused the decreased saturated fatty acids, and PUFA (P=0.05) percentages,

Page 10/22
as well as the increased monounsaturated fatty acids, C 18 :1C 9 , and non-esteri ed fatty acids contents (P<0.05) in milk within six weeks after lactation (supplementary Table 2 as shown in Online Resource). The dietary inclusion of corn silage, instead of dry forages, did not affect plasma glucose, triglyceride, cholesterol, total protein, and albumin concentrations, as well as the albumin to globulin ratio in lactating goats (Table 4). The plasma urea nitrogen content was markedly (P<0.01) decreased in goats fed with corn silage compared to those fed with the control diet ( Table 4).
As shown in Table 5, the dietary inclusion of corn silage caused no signi cant effects on creatinine, total purine derivatives, the excreted purine derivatives, the absorbed purine derivatives, microbial nitrogen, nitrogen intake, feces nitrogen, urine nitrogen, milk nitrogen, and nitrogen retention contents in lactating goats when compared to those who were fed with the control diet. Feeding diet contains corn silage, dissimilar to the diet containing dry forages, had no signi cant effects on ruminal volatile fatty acids, acetic acid, propionic acids, butyric acid, iso valeric acid, valeric acid, and isobutiric acid percentages, as well as acetic acid to propionic acid ratio at both before and three hours after the feeding times (Table 6). Additionally, feeding corn silage in diet signi cantly (P<0.05) lowered pH level in rumen within three hours before the feeding time and also tended to decrease (P=0.07) the ruminal ammonia nitrogen concentration in lactating goats by passing three hours from the feeding time (Table 6). The effects of feeding corn silage on antioxidant capacity in lactating goats are shown in Table 7. In this study, the dietary addition of corn silage signi cantly enhanced the antioxidant capacity in rumen liquor (P<0.05), plasma (P<0.01), and milk (P<0.01) in Mahabadi lactating goats compared to the control diet.

Discussions
Milk yield and compositions including fat, protein, lactose, solids, solids non-fat, total cell numbers, de novo relative fats, the mixed relative fats, performed relative fats percentages, and free fatty acids were crude protein in the rumen is degraded to ammonia by microbes, then absorbed into the portal blood, and nally converted to urea in the liver; in turn, urea readily diffuses from the blood into the milk (Mccormick et al. 1999). Bacteria need the energy source, especially fermentable carbohydrate, in order to capture the ammonia in the rumen, which leads to microbial protein synthesis (Broderick et al. 2003).
Dietary addition of corn silage caused a decrease in milk fat percentage by passing four weeks from the time of lactation. In ruminants, both the concentration and composition of milk fat were affected by the diet (Palmquist 2006). Accordingly, this might be ascribable to the lower NDF content in diets containing corn silage (Table 1). Additionally, by including readily-fermentable carbohydrates (rich in starch) in diet, a decrease in milk fat concentration was observed (Yang et al. 2019). Our ndings are in line with those of the study by Brito and Broderick (2006), which showed that a decline in alfalfa silage to corn silage ratio in diet resulted in a signi cant decrease in milk fat percentage in dairy cows.
While feeding corn silage in diet didn't affect milk C 16:0 , and C 18:0 percentages, it led to a decrease in saturated fatty acids, and PUFA percentages and a rise in monounsaturated fatty acids, C 18 :1C 9 , and nonesteri ed fatty acids contents in milk within six weeks after lactation. Dewhurst et al. (2006) and Kliem et al. (2008) in their studies have reported that ensiled forages, known as the PUFA sources, can also modulate milk fatty acids composition. It is reported that feeding forage or concentrate rich in starch is related to the increased unsaturated fatty acids with 18 carbons and the decreased milk fatty acid with 6 to 16 carbons (Tripathi 2014). In another study, Glasser et al. (2013) have found that corn silage contains a low level of linolenic acid along with high levels of both α-linoleic acid and oleic acid. The reasonable explanation for the difference between milk fatty acids pro les may possibly be that the mobilization of body fat reserves resulted from the high energy demands for the high milk production (Yang et al. 2019).
In addition, Oleic acid is known as a marker of fat mobilization (Nogalski et al. 2012).
In this regard, several researchers have previously reported that the dietary inclusion of corn silages in dairy cows can increase Trans fatty acid contents as well as decreasing n-3 level, leading to high n-6 to n- Dietary inclusion of corn silage, instead of dry alfalfa/straw, had no effects on plasma glucose, triglyceride, total protein, and albumin concentrations in Mahabadi goats. Our ndings are in agreement with those reported in the study by Wang et al. (2016), which showed that the dietary inclusion of corn silage has no effect plasma glucose, total protein contents, and blood urea nitrogen contents, but it enhanced plasma triglyceride concentration in Saanen goats.
Feeding corn silage in diet instead of dry alfalfa/straw decreased plasma urea nitrogen content in lactating goats. Accordingly, this may probably be due to the simultaneous availability of both energy (corn silage as the energy source; Khaing et al. 2015) and protein in the rumen, which resulted in the decreased ruminal N content. As well, it was considered that the dietary N to energy ratio, forage intake level, protein degradability, and dietary carbohydrate level can affect blood urea nitrogen content in ruminants (Hammond et al. 1994). In addition, blood urea nitrogen, which is correlated with the solubility of fed nitrogen ingredients, is known as a good marker of rumen ammonia concentrations (Polat et al. 2009). Our observations are in contrary to those reported in the research by Polat et al. (2009), as they observed that replacement of corn silage had no signi cant effect on blood urea nitrogen content in lactating dairy cows.
Inclusion of corn silage in diet didn't affect creatinine, total purine derivatives, the excreted purine derivatives, urine nitrogen, milk nitrogen, and nitrogen retention concentrations in goats. In another study, Gregorini et al. (2010) have noticed that corn silage is used to dilute N intake, decrease N intake, and decrease urinary N excretion. Purine derivative excretion is also considered as a marker for microbial protein out ow from the rumen (Tas and Susenbeth 2007). In contrary to our ndings, Al-Marashdeh et al.
(2015) in their study have observed that feeding corn silage at 3 kg of DM lowered urinary N excretion compared to those fed with herbage only.
The ruminal volatile fatty acids, acetic acid, propionic acids, butyric acid, iso valeric acid, valeric acid, and isobutiric acid percentages, as well as acetic acid to propionic acid ratio at both before and three hours after the feeding times were unaffected in goats fed on corn silage instead of dry alfalfa/straw. In line with our ndings, van Gastelen et al. proportion as well as the decreased ruminal propionate, butyrate, and valerate proportions in dairy cows. Arndt et al. (2015) in their study have shown that feeding the high ratio alfalfa silage to corn silage led to the increased acetate proportion and acetate to propionate ratio, as well as the declined propionate proportion in rumen in dairy cows.
Dietary addition of corn silage instead of dry alfalfa/straw lowered the ruminal ammonia nitrogen content in goats by passing three hours from the feeding time. This might be due to the better simultaneous availability of N components and energy substrates, which consequently resulted in higher microbial protein production (Broderick 2003).
Antioxidant capacity in rumen liquor, plasma, and milk in Mahabadi lactating goats was forti ed by feeding corn silage instead of dry alfalfa/straw. Anthocyanin, which is a polyphenol, existed in corn

Conclusion
This study showed that replacing dry forage with corn silage resulted in the enhanced the antioxidant capacity in rumen liquor, plasma, and milk in Mahabadi lactating goats. Therefore, it is recommended corn silage be included in the diet of lactating goats to improve milk quality.