The recent increase in demand for buffalo milk and its derivatives has increased the value of the buffalo species in dairy farming worldwide (Santillo et al., Reference Santillo, Caroprese, Marino, Sevi and Albenzio2016). One of the most important functions performed by buffaloes is certainly the production of milk. This is usually the objective, above all, in some Asian countries where religion does not allow the consumption of meat. The buffalo has desirable production characteristics such as high fertility, productive longevity, low morbidity and mortality, greater capacity in the digestion of fibres and the use of urea as a source of nitrogen (Mintoo et al., Reference Mintoo, Zhang, Chen, Moniruzzaman, Deng, Mahbud, Huque, Guang, Wang, Zhong, Han, Khatun, Awal, Gao and Liang2019; Sallam et al., Reference Sallam, Attia, Nour El-Din, El-Zarkouny, Saber, El-Zaiat and Zeitoun2019; Batista et al., Reference Batista, Pereira, Pereira Filho, de Lima, dos Santos, Araújo, Moura, Pereira, Oliveira and Bezerra2020). Buffalo milk has good nutritional value with high levels of fat, protein and minerals (mainly calcium) and can be used both for fresh consumption and in the production of dairy products. In Brazil, buffalo milk is traditionally used to made cheeses of the Mozzarella type, as well as provolone, ricotta and mascarpone types.
To achieve higher levels of milk production, the diet that the buffaloes receive is an important factor. In this sense, it is important to understand the efficient use of protein and energy and their utilisation for ruminal microorganism growth and microbial protein production (Paul et al., Reference Paul, Mandal, Kannan, Mandal and Pathak2003; Santillo et al., Reference Santillo, Caroprese, Marino, Sevi and Albenzio2016; Reddy et al., Reference Reddy, Kumar, Rao, Seshiah, Sateesh, Rao, Reddy and Hyder2019) Hence it is necessary to evaluate the pattern of diets offered to the buffaloes (Paul et al., Reference Paul, Mandal, Kannan, Mandal and Pathak2003; Santillo et al., Reference Santillo, Caroprese, Marino, Sevi and Albenzio2016; Agustinho et al., Reference Agustinho, Zeoula, Santos, Machado, Yoshimura, Ribas, Bragatto, Stemposki, dos Santos and Faciola2020). Regarding nitrogen (N) compound metabolism and the interaction of protein and non-protein nitrogen (NPN) for the N-balance for lactating ruminants, there are two essential metabolite parameters. The first is milk urea nitrogen (MUN) and the second is blood urea nitrogen (BUN), both of which are efficient measures in this regard. However, MUN is in sme respects preferable due to its ease of collection (Hagawane et al., Reference Hagawane, Shinde and Rajduru2009; Malik et al., Reference Malik, Gandotra, Brar, Honparkhe and Uppal2013). Dietary sources of NPN are less costly than sources of dietary protein, an example being soybean meal (Broderick and Clayton, Reference Broderick and Clayton1997). However, the most widely used source of NPN, urea, has a rapid release of ammonia in the rumen and, depending on the quantity, may exceed the capacity for microorganism use (Fox et al., Reference Fox, Tedeschi, Tylutki, Russell, Van Amburgh, Chase, Pell and Overton2004). In this regard, the use of urea supplements can help microorganisms to use ammonia and improve the production of microbial protein. However, the excess ammonia is absorbed by the rumen wall and, once in the bloodstream, ammonia can be toxic to the animal (Swenson and Reece, Reference Swenson and Reece1996). In addition, the sulphur (S) content is normally low in feeds with high levels of NPN, especially in diets with a high proportion of grain, or based on silages from grain-producing plants. Thus, the microbial synthesis of S-amino acids (methionine, cysteine and cystine) may be impaired, limiting the use of urea (Broderick and Clayton, Reference Broderick and Clayton1997). This study was conducted to test the hypothesis that a blend of ground corn (GC) with added urea-ammonium sulphate (U-S) can replace soybean meal, providing nutritional support for ruminal microorganisms and the necessary amino acids for protein by pass, thus improving milk production without compromising quality.
Material and methods
All animal procedures were conducted in accordance with the regulations of the animal ethics committee of the Federal University of Rio Grande do Norte, Natal, Brazil (5°37′5″ South latitude and 33°35′44″ West longitude) with Protocol No 078/2015 (Brazil, 2008).
Animals, diets and general procedures
The experimental work was conducted at the dairy farm of Tapuio Agropecuaria LTDA. Twelve pluriparous, lactating Murrah buffaloes with an average of 100 ± 4 d of lactation and weight of 650 ± 45 kg, were treated to control internal and external parasites. The experimental phase lasted 72-d and was divided into four periods of 18-d each. The animals were adapted to the diets during the first 13-d of each period, and data were collected during the last 5-d. The animals were housed in a covered shed in 6 × 3-m2 individual pens and were provided with access to water and feed ad libitum. The experimental diets were provided by Tapuio Agropecuaria LTDA and were formulated to meet the nutritional requirements of lactating buffaloes to produce milk at 10 kg/d with 7.0% fat according to the recommendations of Paul and Lal (Reference Paul and Lal2010). The chemical composition of the dietary ingredients and the make up of the experimental diets is given in online Supplementary Tables S1 and S2, respectively. Principal ingredients were soybean meal, ground corn, urea-ammonium sulphate blend (GCU-S) and mineral mixture as concentrates combined with sugarcane and cactus pear as roughage in a concentrate:roughage ratio of 60:40%. The animals were randomly distributed in a triple Latin square (4 × 4) comprising four inclusions of GCU-S blend replacing soybean meal at levels of 0 (control), 8.0, 16.4 and 24.1 g/kg total DM (online Supplementary Table S2). Care was taken to ensure that the four diets were isonitrogenous. The cactus material was chopped by a modified machine into pieces approximately 5.0 to 10 cm in size. Sugarcane (Cynodon sp.) was cut in a stationary forage machine with a 5.0 cm sieve. The total mixed ration was fed twice daily at 6:00 and 15:30 h and afterwards, at the time the buffaloes were milked, the allocation was adjusted to allow 10% refusals.
Intake, digestibility and nitrogen (N) balance
Dry matter intake (DMI) and nutrients intake were obtained through the records of the feed offered and refusals and the collection of diet and refusals samples performed during the last five days of each experimental period. Details are given in the online Supplementary File. Daily faecal samples were obtained for analysis according to the schedule described in Casali et al. (Reference Casali, Detmann, Valadares Filho, Pereira, Henriques, Freitas and Paulino2008) and detailed in the online Supplementary File. Indigestible neutral detergent fibre (iNDF) was used as an indicator to estimate faecal excretion (Van Soest et al., Reference Van Soest, Robertson and Lewis1991; Casali et al., Reference Casali, Detmann, Valadares Filho, Pereira, Henriques, Freitas and Paulino2008; Valente et al., Reference Valente, Detman, Queiroz, Valadares Filho, Gomes and Figueiras2011). Samples of the diet, faces and refusals were used to measure digestiblities through incubation for 12 d in the rumen of two adult buffaloes, previously adapted to the diet for 7 d (detailed in online Supplementary File). The faecal output of dry matter (FODI) was calculated using the following equation: FODI (kg) = (indicator intake (kg)/ % of faecal indicator) × 100. Digestibility coefficients were calculated for dry matter (DM), crude protein (CP), neutral detergent fiber (NDF), total carbohydrates (TC), non-fiber carbohydrates (NFC) and crude energy (CE) using the equation digestibility coefficient = [(kg ingested – kg excreted)/(kg ingested)] × 100 (Sniffen et al., Reference Sniffen, O'Connor, Van Soest, Fox and Russell1992). The intake of total digestible nutrients (TDN) was derived from these values using the equation TDN = [(CPI – CPf) + 2.25 (EEI – EEf) + (TCI – TCf)], where ‘I’ is respectively nutrient intake and ‘f’ is respectively faecal excretion. Concentrations of dietetic TDN were calculated with the equation TDN = (TDN intake/DM intake) × 100. The nitrogen (N) balance (NB) was obtained by the difference between the N-intake and that present in the urine (Reed et al., Reference Reed, Moraes, Casper and Kebreab2015), faeces and milk. Further details are given in the online Supplementary File.
Chemical analysis
Samples of diets, refusals and faeces were analysed for DM, ash, CP and ether extract (EE) following methods described in AOAC (2012) and detiale din the online Supplementary File. NDF) and acid detergent fibre (ADF) were determined according to the methodology of Van Soest et al. (Reference Van Soest, Robertson and Lewis1991) with ash and protein correction (Licitra et al., Reference Licitra, Hernandez and VanSoest1996) as detailed in the onlien Supplementary File. The content of non-fibre carbohydrates (NFC) was calculated as NFC = 100 – [(CP – CP from urea + urea) + NDF + EE + ash (Hall, Reference Hall2000) and total carbohydrate (TC) was calculated according to Sniffen et al. (Reference Sniffen, O'Connor, Van Soest, Fox and Russell1992), where: TC = 100 – (CP + EE + ash).
Blood collection
Blood samples were collected from all animals using jugular venipuncture in the morning before feeding on experimental day 18 of each experimental period following the protocols for blood collection of Nexus Academic Publishers (2013) and Uhart (Reference Uhart2016). Sera were frozen at −20 °C until analysis using a semiautomatic biochemical analyser (BioPlus 2000®, São Paulo, Brazil) and commercial kit tests for total protein, albumin, BUN and cholesterol (detailed in online Supplementary File).
Milk production and composition
Animals were manually milked twice a day (at 6:00 and 15:30 h) after feeding. Daiuly milk production (MP) was obtained by weighing morning and afternoon milk each day between days 13 and 18. Milk production correction at 6% fat (MPC6%Fat) was performed according to Rice et al. (Reference Rice, Andrews, Warnwick and Legates1970) where MPC6%Fat (kg/d) = 0.308 × total milk yield (kg) + 11.54 × total fat yield (kg). Samples of morning and afternoon milk were mixed to form a single sample for each animal. Fat, protein, lactose, casein, urea, total solids and nonfat dry extract were determined using the infrared spectroscopy method (Bentley, 1995; Bentley Instruments Incorporated®, USA) and expressed in g/d. Feed efficiency (FE) was determined using the formula FE = MPC6% Fat /DMI
Statistical analysis
Data were analysed using a triplicated 4 × 4 Latin square design, with four periods, four levels of ground corn and urea + ammonium sulphate blend [0 (control); 8.0, 16.4 and 24.1 g/kg in total DM] and 12 animals. Urea + ammonium sulphate blend level (1 to 4) and square (1 to 3) were included in the statistical model as fixed effects, and period (1 to 4) and buffalo nested within square were the random effects. Polynomial contrasts evaluated the linear, quadratic and cubic effects of substitution of soybean meal by GCU-S blend, to test whether higher levels of inclusion might compromise production. Results are presented as least square means ± sem. All data were analysed using the MIXED procedure of SAS (SAS, 2003). Differences were considered to be significant when P ≤ 0.05. The mathematical model used was:
$${\rm Y}ijkl = \mu + {\rm A}i + {\rm P}j + {\rm S}k + {\rm T}l + ( {\rm A \times S}) \,ik + ( {\rm T \times S}) \,lk + eijkl$$where: Yijkl is the observed variable, μ is the mean, Ai is the effect of the animal; Pj is the effect of the period; Sk is the effect of the square; Tl is the effect of the ground corn + urea-ammonium sulphate blend at levels 0 (control); 8.0, 16.4 and 24.1 g/kg in total DM; (A × S)ik is the interaction between animal and square; (T × S)lk is the interaction between treatment; and square and eijkl is the experimental error.
Results
Intakes of nutrients and digestibility values are shown in Table 1. Replacing soybean meal with the blend of ground corn, added urea + ammonium sulphate (GCU-S) resulted in a significant linear decrease in absolute DM intake and intake of crude protein. Correction for body weight reduced the effect on DM intake such that it was no longer statistically significant. Intake of carbohydrates and non-fibrous carbohydrates demonstrated numerical (non-significant) decreases, whereas the ratio of total digestible nutrients to DMI showed a significant linear increase. (Table 3). Other intake parameters including NDF (as kg/d and %BW), ether extract and TDN were not affected (P > 0.05). There was also no effect of the GCU-S diet for the digestibility coefficients of CP, NDF, TC, and NFC, however, DM digestibility was increased linearly and significantly (Table 1). Cubic and quadratic analyses were done but showed no evidence of inhibitory effects at higher GCU-S inclusion (data not shown).
Table 1. Daily nutrient intake and digestibility of dairy buffaloes fed diets containing a blend of ground corn together with urea-ammonium sulphate (GCU-S) replacing soybean meal
NDF, Neutral detergent fibre; NFC, Non-fibrous carbohydrates; TDN, Total digestible nutrients
Significance for Linear effect. Quadratic and cubic analyses produced no significant effects.
Data for nitrogen balance and concentration of various metabolites in blood are in Table 2. Absolute values for N-intake and N-urinary excretion were decreased significantly and linearly by the GCU-S diet. Other nitrogen balance parameters demonstrated a numerical but non significant decrease (N-total excretion, N-retained) or no effect. The one exception was N-transfer into milk when expressed as a percentage of total N intake, which increased in response to the GCU-S diet (Table 2). There were no effect of the dietary treatment on serum concentrations of blood urea nitrogen (BUN), total protein, albumin, globulin (and albumin: globulin ratio) and total cholesterol of dairy buffaloes (Table 2). Cubic and quadratic analyses were done but showed no evidence of inhibitory effects at higher GCU-S inclusion (data not shown).
Table 2. Nitrogen (N) balance and blood metabolites of dairy buffaloes fed diets containing a blend of ground corn together with urea-ammonium sulphate (GCU-S) replacing soybean meal
BUN, blood urea nitrogen.
Significance for Linear effect. Quadratic and cubic analyses produced no significant effects.
Data production parameters and derived feed efficiency are given in Table 3. The dietary treatment did not significantly affect milk production or any aspect of milk composition. Feed efficiency (calculated as kg of 6%FCM per kg of DMI) of the dairy buffaloes presented linear and quadratic increases (P = 0.01) in response to GCU-S diet, and the 24.1 g/kg DM level was the most efficient level (Table 3). Similarly, the efficiency of protein utilisation was increased when expressed on a milk output basis (kg milk per kg CPI), but not when expressed as milk protein output.
Table 3. Milk production and composition of dairy buffaloes fed diets containing a blend of ground corn together with urea-ammonium sulphate (GCU-S) replacing soybean meal
FCM, Milk production corrected for 6% fat; MUN, milk urea nitrogen.
a kg FCM produced per kg DMI.
b kg milk produced per kg CPI.
c kg of milk protein produced per kg CPI.
*Significance for Linear (L), Quadratic (Q) and Cubic (C) effects.
Discussion
The use of urea has allowed the use of roughage ingredients with low quality by ruminants, ones that, under normal conditions, are little used. Mertens (Reference Mertens1987) proposed that the mean value of NDF intake that affects DMI by physical mechanisms in lactating animals is 1.20% BW. In our case the NDF intake was considerably lower, 0.44% BW on average. Inclusion of GCU-S casued DMI to decrease, presumably as a consequence of reduced palatability as urea content increased. However, the use of urea would probably have promoted greater synchronicity in the availability of energy and nitrogen in the rumen (Van Soest, Reference Van Soest1994; Sharma et al., Reference Sharma, Singh, Roy and Thakur2016), which thus increased the soluble carbohydrates intake, albeit to a modest extent. NFC and TC showed numerical rather than significant increases, but the increase in TDN as a proportion of DMI was significant. This is also reflected by the greater DM digestibility.
CPI also decreased linearly as the levels of urea increased in the diets, and this is related to the decrease in DMI, which influences directly and quantitatively the other nutrients (Naveed-ul-Haque et al., Reference Naveed-ul-Haque, Akhtar, Munnawar, Anwar, Khalique, Tipu, Ahmad and Shahid2018). The lower N-intake in response to GCU-S inclusion was probably because of DMI and CPI intake reductions since the diets were isonitrogenous. In addition, there was some evidence of a modest reduction in N excretion, promoting N-retention, but since N transfer into milk was increased the overall N balance was reduced. Naveed-ul-Haque et al. (Reference Naveed-ul-Haque, Akhtar, Munnawar, Anwar, Khalique, Tipu, Ahmad and Shahid2018) reported a negative influence of high protein diets (>142 g/kg DM) on milk production, concluding that a diet with 120 g CP per kg DM allowed greater production and efficiency. This is also the level that we used. A characteristic of rumen function in buffalo is more rapid protein fermentation and degradation (Bartocci et al., Reference Bartocci, Tripaldi and Terramoccia2002). In addition, the feed remains in the rumen longer and spends less time in the intestines, so relatively little intact dietary protein leaves the rumen to be used directly at the intestinal level (Di Lella et al., Reference Di Lella, Infascelli and Cutrignelli1995).
The efficiency of the use of dietary NPN by the rumen microorganisms depends on factors such as the availability of soluble carbohydrates and sulphur and synchronisation between the release of ammonia, resulting from the hydrolysis of urea, and the presence of energy for microbial protein synthesis (Fox et al., Reference Fox, Tedeschi, Tylutki, Russell, Van Amburgh, Chase, Pell and Overton2004; Brun-lafleur et al., Reference Brun-Lafleur, Delaby, Husson and Faverdin2010; Sharma et al., Reference Sharma, Singh, Roy and Thakur2016; Naveed-ul-Haque et al., Reference Naveed-ul-Haque, Akhtar, Munnawar, Anwar, Khalique, Tipu, Ahmad and Shahid2018). Thus, the relative increase in TDN intake may have promoted greater use of N due to the greater availability of soluble sugars, reducing N-urinary excretions that result in greater energy losses for ruminants (Tedeschi et al., Reference Tedeschi, Fox and Russell2000; Fox et al., Reference Fox, Tedeschi, Tylutki, Russell, Van Amburgh, Chase, Pell and Overton2004).
Despite N-intake being lower, N-faeces excretion did not differ among treatments, which may have been due to the greater use of N in the synthesis of microbial protein, compared to the soybean meal that has a greater amino acids composition (Fox et al., Reference Fox, Tedeschi, Tylutki, Russell, Van Amburgh, Chase, Pell and Overton2004). In other words, the addition of U-S to the blend promoted a biological value for the protein similar to that of soybean meal. These results are in agreement with Owens and Zinn (Reference Owens, Zinn and Church1988), who stated that the rumen nitrogen recycling system easily adapts to the rapid release of ammonia by sources of non-protein nitrogen, as long as the concentrations do not reach toxic levels. This explains the similar BUN values among treatments. In addition, the positive ruminal N-balance in all treatments demonstrates a similar microbial protein synthesis of the U-S blend when added to ground corn compared to soybean meal diet.
The blood protein metabolites (total protein, albumin and globulin) and cholesterol were not influenced by the replacement levels, which is similar to previous studies (Hagawane et al., Reference Hagawane, Shinde and Rajduru2009; Malik et al., Reference Malik, Gandotra, Brar, Honparkhe and Uppal2013; Arif et al., Reference Arif, Al-Sagheer, Salem, Abd El-Hack, Swelum, Saeed, Jamal and Akhtar2019). Despite the inclusion of urea, the isonitrogenous diets together with the efficient utilisation of nutrients by microorganisms could be responsible for the non-significant alterations in blood metabolites values (Saleem et al., Reference Saleem, Zanouny and Singar2018; Arif et al., Reference Arif, Al-Sagheer, Salem, Abd El-Hack, Swelum, Saeed, Jamal and Akhtar2019). There was no increase in absolute or fat-corrected milk production in response to GCU-S inclusion, but there was an increase when expressed relative to CPI. This is probably due to improve synchronisation between protein and energy supply, which reduces the metabolic energy expenditure by the animal and then can improve the microbial efficiency and availability of microbial protein to be absorbed in the intestine (Arif et al., Reference Arif, Al-Sagheer, Salem, Abd El-Hack, Swelum, Saeed, Jamal and Akhtar2019). In earlier studies, greater availability of N improved fat corrected milk yield in buffaloes (Paul and Lal, Reference Paul and Lal2010; Bezerra Junior et al., Reference Bezerra Junior, Silva, Hongyu, Couto, Medeiros, Araújo, Lima and Fraga2018), but we did not observe this. Increased feed efficiency was offset by decreased DMI, with an overall milk production level of 0.65 kg per kg of DMI, which is greater than the daily averages observed by Bartocci et al. (Reference Bartocci, Tripaldi and Terramoccia2002: 0.4 kg/kg) and similar to Arif et al. (Reference Arif, Al-Sagheer, Salem, Abd El-Hack, Swelum, Saeed, Jamal and Akhtar2019: 0.7 kg/kg).
Milk fat concentration was not changed by GCU-S inclusion, which is important to the dairy industry (Teixeira et al., Reference Teixeira, Bastianetto and Oliveira2005; Bezerra Junior et al., Reference Bezerra Junior, Silva, Hongyu, Couto, Medeiros, Araújo, Lima and Fraga2018). Buffalo milk, due to its high levels of protein, fat and total solids, has been increasingly attracting the interest of the dairy industry, which is willing to invest in the manufacture of differentiated and high value products such as mozzarella (Teixeira et al., Reference Teixeira, Bastianetto and Oliveira2005; Soares et al., Reference Soares, do Nascimento Rangel, Novaes, de Lima Júnior and Bezerra2013). We saw no change in MUN which is to be expected since the diets all contained the same level of CP: Fox et al. (Reference Fox, Tedeschi, Tylutki, Russell, Van Amburgh, Chase, Pell and Overton2004) demonstrated the existence of a positive correlation between CP content and MUN. MUN values obtained were within the reference interval which ranges from 12 to 24 mg/dl (Saleem et al., Reference Saleem, Zanouny and Singar2018).
In conclusion, the use of a blend of ground corn (34.5%) together with 24% urea- ammonium sulphate (9:1 ratio) as a replacement for soybean meal can be recommended in the diet of lactating buffaloes because, although it decreases DMI and N-urinary excretion, this diet improves DM digestibility and feed efficiency whilst completley maintaining production of milk with a normal fat and total solids content. The high cost of soybeans compared to corn and urea-ammonia sulphate blend could allow reduced concentrate cost, and the substitution works well with high energy content roughage, such as cactus pear.
Supplementary material
The supplementary material for this article can be found at https://doi.org/10.1017/S002202992200036X
Acknowledgments
We thank CNPq-Brazil and CAPES (Brazil) for financial support and student scholarships, respectively. We express particular gratitude to Tapuio Farm for providing the animals and feed ingredients used in this research. In doing so we confirm that no conflict of interest exists with Tapuio Agropecuária LTDA.
