Introduction
Animal feeding contributes the greatest variable cost to dairy production, and careful planning is required when choosing the dietary ingredients (Nogueira et al., Reference Nogueira, Beber, Silva, Diniz, Peixoto, Silveira, Rosa and Bezerra2021). Most of the energy in supplements for animal production comes from grains such as corn and sorghum, which have experienced large price fluctuations worldwide in recent years (Keim et al., Reference Keim, Mora, Ojeda, Saldías and Bedenk2022). This has driven the search for alternative feedstuffs that can be used to feed cows and reduce production costs without compromising the nutritional quality or profitability of milk and dairy products (Araújo et al., Reference Araújo, Fernandes Da Silva, Do Nascimento Junior, Silva Damasceno, Costa and da Silva2019).
Because of its biodiversity, Brazil has a wide availability of products with the potential for use in animal feeding (Pinto et al., Reference Pinto, Magalhães, Teodoro, Gois, Véras, Campos, Oliveira and Campos2020). Among these products, cassava (Manihot esculenta) stands out for its production capacity and suitability to different soil types and ecosystems (Fathima et al., Reference Fathima, Sanitha, Tripathi and Muiruri2023). The roots of M. esculenta can supply large amounts of energy from its digestible carbohydrates, making it a good option for animal feeding (Wanapat and Kang, Reference Wanapat and Kang2015; Howeler, Reference Howeler and Hershey2017; Jiwuba et al., Reference Jiwuba, Jiwuba, Ogbuewu and Mbajiorgu2021).
Galvão et al. (Reference Galvão, Vargas, Mezzomo, Oliveira, Dos Santos, Maciel, Soares, Farias, Caldas, Faciola and Alves2024) evaluated the effects of replacing ground corn with cassava root silage (CRS) in the supplement of grazing dairy cows, on the intake, digestibility and milk production, and reported that both feedstuffs produce similar responses. In addition, there is a direct relationship between the nutritional composition of alternative feedstuffs and the milk production and composition in ruminants (Lima et al., Reference Lima, Vargas, Gomes, Maciel, Alves, Oliveira, Aguiar, de Carvalho Reis, Oliveira and Mezzomo2021; Castro et al., Reference Castro, Maciel, Vargas, Gomes, Maciel and Mezzomo2021), as well as the production of cheese and its attributes (Cerutti et al., Reference Cerutti, Viegas, Barbosa, Oliveira, Dias, Costa, Nornberg, de Carvalho, Bezerra and Silveira2016; Oliveira et al., Reference Oliveira, de Lima, de Freitas, Dos Santos, Reis, Gomes, Alves, Maciel, Vargas and Mezzomo2020; Maciel et al., Reference Maciel, Vargas, Mezzomo, Gama, Leite, Castro, Oliveira, Farias, dos Santos Luz and Alves2022). Nevertheless, the impacts of replacing ground corn with CRS in the supplement of grazing dairy cows on physicochemical characteristics, sensory attributes, and economic viability in dairy products remains unexplored. Thus, this study evaluated the effects of replacing ground corn with CRS in the supplement of grazing dairy cows on yield, physicochemical characteristics, sensory attributes, and profitability of Minas Frescal cheese. We hypothesized that the replacement of ground corn with CRS would maintain cheese yield, physicochemical characteristics, sensory attributes, while increasing profitability due to feed costing savings. Minas Frescal cheese was selected as a dairy product model because of its economic importance and high consumption in Brazil (Rocha et al., Reference Rocha, Calvalcanti, Silva, Guimarães, Balthazar, Pimentel, Esmerino, Freitas, Granato, Costa and Silva2020). According to Felicio et al. (Reference Felicio, Pinto, Oliveira, Lempk, Pires and Lelis2015), this cheese is categorized as semi-fat and unripened, featuring high moisture levels, low salt content, and a pH above 5.0, produced through the enzymatic coagulation of milk using rennet.
Materials and methods
Facilities, experimental design, and treatments
This study was conducted during the Amazon rainy season at the Animal Production Station of the Universidade Federal Rural da Amazônia (UFRA), in the city of Parauapebas, Pará State, Brazil. During the experiment, the cumulative rainfall was 1582 mm, equivalent to a monthly average of 395.5 mm. The average relative humidity and temperature were 77.50% and 28.14°C, respectively.
Ten primiparous Girolando dairy cows (¾ Holstein and ¼ Dairy Gir) with 76 days in lactation were used. The animals weighed 373.45 ± 63.55 kg and had a mean milk production of 12.48 ± 1.58 kg/d. The experimental design consisted of two simultaneous 5 × 5 Latin squares, taking the lactation period into account to balance residual effects. Before the experiment began, animals were allowed to adapt for 30 days. The study was divided into five experimental periods consisting of 20 days each, and each period had 14 days for adaptation and 6 days for data collection.
Cassava root was harvested in the rural area of Parauapebas and transported to the dairy cattle station at UFRA. Each cassava root was cleaned with running water, and after drying, it was crushed with a forage crusher without the use of sieves and ensiled in 200 L plastic drums with lids. The roots were compacted in the cylinders with wooden tampers.
The silos were opened following a minimum of 150 days of ensiling. When each silo was opened, the upper portion was discarded, and the temperature was measured. Then, samples were collected for performing pH measurements and laboratory analyses. For pH determination, 9 g samples of silage were taken and placed in a 250 mL beaker, to which 60 mL of distilled water was added. The pH was then measured at three consecutive times at 30-minute intervals (Silva and Queiroz, Reference Silva and Queiroz2002). The average pH of CRS was 4.04, with its mean chemical composition consisting of 510 g/kg DM, 112.4 g/kg DM of neutral detergent fiber (NDF), 22.3 g/kg DM of ether extract (EE), 778.8 g/kg DM of non-fibrous carbohydrates (NFC), and 34.5 g/kg DM of crude protein (CP).
The cows were managed in a rotational grazing system that utilized Megathyrsus maximus cv. Mombasa and were assigned to five treatments:
I. Grazing without supplementation (WOS group);
II. Grazing with 5 kg DM of supplement without the inclusion of CRS (0 g/kg DM CRS);
III. Grazing with 5 kg DM of a supplement containing 26% CRS (260 g/kg DM CRS);
IV. Grazing with 5 kg of DM of a supplement containing 52% CRS (520 g/kg DM CRS); and
V. Grazing with 5 kg of DM of a supplement containing 78% CRS (780 g/kg DM CRS).
The CRS was included in the supplement to gradually replace the ground corn until 100% replacement was reached (Table 1). The supplement was composed of soybean meal, urea, mineral salt, and corn grain. This was supplied twice a day, with half provided in individual stalls after each milking. Supplements across treatments were isonitrogenous and formulated to meet nutritional requirements of primiparous dairy cows weighing 400 kg and producing 14 kg of milk per day with 4.0% fat (NRC, 2001).
Proportion and chemical composition of the ingredients used in the experimental supplements, containing different levels of cassava root silage (CRS) as a replacement for ground corn, for grazing dairy cows
1 Mineral supplement composition – Calcium: 200 (g/kg); Sodium: 74 (g/kg); Phosphorus: 100 (g/kg); Sulfur: 12 (g/kg); Magnesium: 15 (g/kg); Zinc: 4896 (mg/kg); Manganese: 6285 (mg/kg); Iron: 2000 (mg/kg); Copper: 1650 (mg/kg); Fluoride: 1000 (mg/kg; max.); Cobalt: 200 (mg/kg); Iodine: 195 (mg/kg); Selenium: 32 (mg/kg); Chromium: 20 (mg/kg).
2 WOS: Without supplementation.
3 DM: dry matter; OM: organic matter; CP: crude protein; EE: ether extract; NDFap: neutral detergent fiber, corrected for ash and protein; NFC: non-fiber carbohydrates, corrected for ash and protein; NDFi: indigestible neutral detergent fiber.
4 Chemical composition referring to the simulated grazing sample.
For the animals fed on pasture only (WOS), only mineralized salt (100 g/animal/day) was supplied after milking, 50 g in the morning and 50 g in the afternoon. All animals were given their specific treatments and allowed to graze in the same area. Water was available ad libitum through drinking fountains.
Chemical analyses
Forage samples and supplement ingredients were analyzed according to AOAC (1990) methods for DM (method 934.01), ash (method 942.05), CP (method 984.13) and EE (method 920.39). The NDF content was corrected for residual ash (Mertens, Reference Mertens2002) and residual nitrogen compounds (Licitra et al., Reference Licitra, Hernandez and Van Soest1996).
Milk collection and analysis
Cows were milked twice a day at 07:00 h and 16:00 h. On the 18th day of each period, milk samples (approximately 300 mL) were collected from each animal during morning milking and afternoon milking to generate a composite milk sample. A 50 mL aliquot of each composite sample was transferred into a plastic bottle that contained the preservative Bronopol® (2-bromo-2-nitropropane-1,3-diol; Ashland, Wilmington, DE, USA). Samples were kept between 2 and 6°C and processed at the Milk Quality Laboratory of the Universidade Federal de Goiás (UFG), to analyze the concentration of casein, protein, lactose, fat, non-fat milk solids (NMFS), total milk solids (TMS), and milk urea nitrogen (MUN) (Table S1).
Minas Frescal cheese production
Milk collection for Minas Frescal cheese production was performed separately on the 18th day of each experimental period. Three cheeses were produced from each cow during each period (each cheese was derived from exactly 1 kg of milk), with two cheeses used for physicochemical analyses and one cheese for sensory analyses. Therefore, 50 experimental replications were conducted using 10 cows across 5 periods, resulting in the production of 150 cheeses.
Milk physicochemical characteristics for the production of Minas Frescal cheese were evaluated as described in the Normative Instruction No. 76/2018 (Brasil, 2018). Milk was analyzed for Dornic acidity (using an acid-base titration with sodium hydroxide N/9; Sigma Aldrich), pH (using a digital pH Meter, model: mPA – 210P, MS Tecnopon, Piracicaba, SP, Brazil), and density at 15 °C (using a thermolactodensimeter, model: 5784.1.L, Incoterms, Sao Paulo, SP, Brazil; Table S1).
Cheese was produced on the 19th day of each period, following the procedure outlined by Silva (Reference Silva2005). Briefly, the milk from each experimental diet was weighed, filtered through a 100-mesh nylon cloth, and pasteurized at 65°C for 30 min. The milk was then cooled to 39°C, at which point 40 mL of 50% calcium chloride (Rica Nata, Piracema, MG, Brazil) and 30 mL of rennet (Rica Nata, Piracema, MG, Brazil) with standard coagulation strength for Minas Frescal cheese production, were added to initiate coagulation. The rennet was diluted in an equal part of filtered water before being added to the milk. After 50 min, the coagulated curd mass was cut and drained, and thereafter, the whey was drained, the mass was placed in plastic molds and salted with 700 g of refined white salt per 100 L of curd mass. The cheeses were cooled between 6 and 10°C for approximately 24 hours, removed from molds on the following day, weighed to determine the mass yield. A cheese sample was stored at − 20 °C for later physicochemical analyses at the end of the experiment. Sensory analyses were performed immediately after cheese manufacture.
Analyses of Minas Frescal cheese
Physicochemical parameters
Physicochemical analyses of the cheeses were performed following the methodologies described in the Normative Instruction no. 68 of 12/12/2006 of the Ministry of Agriculture, Livestock and Food Supply–MAPA (Brasil, 2006). The cheeses were analyzed in the Food Analysis Laboratory of UFRA for pH, titratable acidity, moisture, total dry extract (TDE), non-fat-dried-extract (NFDE), fat concentration in total dry extract (FTDE), and CP content according to Maciel et al. (Reference Maciel, Vargas, Mezzomo, Gama, Leite, Castro, Oliveira, Farias, dos Santos Luz and Alves2022) protocols. The Gerber‒Van Gulik method (ISO, 2008) was used for determining the fat content in cheeses.
Yield and recovery of fat and protein
The following values were calculated: gross yield (Equation (1)), as outlined by Andreatta et al. (Reference Andreatta, Fernandes, Santos, Mussarelli, Marques, Gigante and Oliveira2009); moisture-adjusted yield (Equation (2)), using a moisture reference value of 57% for Minas Frescal cheese, based on Oliveira (Reference Oliveira1986); and individual yield (Equation (3)). Additionally, the recovery of fat and protein from the produced cheese was assessed following the methods of Cipolat-Gotet et al. (Reference Cipolat-Gotet, Bittante and Cecchinato2015) (Equation (4)). The equations used for calculating the aforementioned parameters, are presented below:
\begin{equation}Gross\,yield = \,\frac{{Mass\,of\,milk\,used\,\left( {kg} \right)}}{{Mass\,of\,cheese\,obtained\,\left( {kg} \right)}}
\end{equation}
\begin{align}
Moisture - & adjusted\,yield = \nonumber \\ & \quad \frac{{\left( {gross\,yield} \right) \times \,(100 - \left( {cheese\,moisture\,\left( \% \right)} \right)}}{{\left( {desired\,moisture\,\left( \% \right)} \right)}}\end{align}
\begin{align}
Individual\,yield = \,\frac{{Milk\,production\,\left( {\frac{{kg}}{{day}}} \right)}}{{Gross\,yield}}
\end{align}
\begin{align}
& Recovery\,\left( \% \right) = \nonumber \\
& \quad \,\frac{{\left( {{{Mass\ of\ cheese }}\left( {{g}} \right)} \right) \times { }\left( {{{Fat\ content\ in\ cheese }}\left( {{\%}} \right)} \right)}}{{\left( {{{Mass\ of\ milk }}\left( {{g}} \right)} \right){ } \times { }\left( {{{Fat\ content\ in\ milk }}\left( {{\%}} \right)} \right)}} \times { }100{ }\end{align}Colour analysis
The colour analysis of the cheeses was conducted using a Chroma Meter (model CR-400, Minolta, Ramsey, NJ, USA) with the CIE L*a*b* system. In this system, the L* parameter represents luminosity, ranging from 0 to 100, with 0 being black and 100 being white; the a* parameter indicates the balance between green (-) and red (+); and the b* parameter reflects the balance between blue (-) and yellow (+) (HunterLab, 1996). The device was precalibrated with D65 lighting and measurements were performed in triplicate, using the inner part of the cheese for evaluation immediately after cutting (Queiroga et al., Reference Queiroga, Santos, Gomes, Monteiro, Teixeira, de Souza and Pintado2013). Saturation index (C*) was calculated to represent the hue and colour intensity of the cheeses via the following equation:
\begin{equation}{\text{C}}* = \left[ {{{\left( {{{\text{a}}^{*2}} + {{\text{b}}^{*2}}} \right)}^{0.5}}} \right]\end{equation}Sensory evaluation
Sensory analysis of Minas Frescal cheese was conducted with 100 untrained judges, comprising 43 males (aged 18 to 60) and 57 females (aged 18 to 56), none of whom had any allergies or intolerances to the cheese's chemical components. The analysis took place over five experimental periods, with 20 tasters in each period. Prior to each evaluation, the panelists signed an informed consent form to participate in the study.
The sensory acceptance test was based on the method described by Meilgaard et al. (Reference Meilgaard, Civille and Carr2006), using a hedonic scale from 0 to 10, where 0 indicates “very disliked,” 5 means “neither liked nor disliked,” and 10 represents “liked very much.” The samples were cut into approximately 25 g cubes, coded, and provided in disposable cups along with water and crackers between samples. Each taster received 10 samples presented simultaneously and evaluated by the judges. The assessed attributes included appearance, colour, aroma, texture, flavor, and overall perception. Additionally, purchase intention was measured (Dutcosky, Reference Dutcosky2015) using a structured hedonic scale from 1 to 5 points, where 1 indicates “would certainly not buy,” 3 means “might or might not buy,” and 5 represents “would certainly buy.”
Economic analysis
Economic analysis was performed using partial budgets of the production costs and returns of Minas Frescal cheese following the methods described by Dillon and Hardaker (Reference Dillon and Hardaker1980) and Martínez-García et al. (Reference Martínez-García, Rayas-Amor, Anaya-Ortega, Martínez-Castañeda, Espinoza-Ortega, Prospero-Bernal and Arriaga-Jordán2015). The prices of the dietary ingredients and the amounts of cheese sales were obtained from market values in the state of Pará, Brazil, and all values were expressed in US dollars (US$; 1 US$ = 5.4502 BRL).
The cost of the supplements was calculated from the price of each ingredient multiplied by the amount included in the supplement formulation to produce 100 kg of supplement (Santos et al., Reference Santos, Modesto, Souza, Ítavo, Jobim, Kazama, Valloto and Massuda2009), with the mineral salt market in the state of Pará, Brazil versus the amount supplied. The calculation was as follows (Equation (6)):
\begin{equation}{C_{supplement}} = {\text{ }}{P_{supplement}} \times {\text{ }}{Q_{supplement}}\end{equation}where:
Csupplement is the cost of the supplement (US$/animal/day),
Psupplement is the price per kg of supplement (US$/kg), and
Qsupplement is the amount (in kg) of the supplement given (kg/animal/day).
The daily cost of supplementation was calculated as follows (Equation (7)):
\begin{equation}{C_{daily}} = {\text{ }}{C_{supplement}} + {\text{ }}{C_{labor}}\end{equation}where:
Cdaily is the daily cost of supplementation (US$/animal/day),
Csupplement was calculated from Equation (6), and
Clabor is the cost of labor for supplementation (US$/animal/day).
The total cost per cow per day was calculated as follows (Equation (8)):
\begin{equation}{C_{total{\text{ }}\,= }}\,{C_{supplement}} + {\text{ }}{C_{labor}} + {\text{ }}{C_{grazing}}\end{equation}in which:
Ctotal is the total cost (US$/animal/day),
Csupplement (from Equation (6)),
Cgrazing is the cost of grazing (US$/animal/day).
The cost of the milk was calculated as follows (Equation (9)):
\begin{equation}{C_{milk}} = {\text{ }}{P_{milk}}/{C_{total}}\end{equation}where:
Cmilk is the cost of the milk (US$/kg of milk),
Pmilk is milk production (kg/day) (Galvão et al., Reference Galvão, Alves, Dos Santos, Oliveira, Mezzomo and Vieira2023), and
Ctotal was calculated via Equation (8)
The cost of making cheese was calculated as follows (Equation (10)):
\begin{align}{C_{cheese{\text{ }}manufacturing}} & = {\text{ }}{C_{ingredients}}\,+ {\text{ }}{C_{milk}}\,+ {\text{ }}{C_{packaging}} \nonumber \\
& \quad + {\text{ }}{C_{energy}}\,+ {\text{ }}{C_{labor}}\end{align}wherein:
Ccheese manufacturing is the manufacturing cost per kg of cheese (US$/kg of cheese),
Cingredients is the cost of the ingredients used in the manufacturing process (US$/kg of cheese),
Cmilk (calculated via Equation (9)),
Cpackaging is the cost of packaging (US$/kg of cheese),
Cenergy is the cost of the energy used (US$/kg of cheese), and
Clabor is the cost of labor to produce the cheese (US$/kg of cheese).
The total cost per kg of cheese was calculated as follows (Equation (11)):
\begin{equation}{C_{cheese{\text{ }}\,= }}\,{C_{supplement}} + {\text{ }}{C_{grazing}} + {\text{ }}{C_{labor}} + {\text{ }}{C_{cheese{\text{ }}manufacturing}}\end{equation}wherein:
Ccheese is the total cost per kg of cheese (US$/kg of cheese),
Csupplement (calculated via Equation (6)),
Cgrazing is the pasture cost (US$/animal/day),
Clabor is the cost of labor for supplementation (US$/animal/day), and
Ccheese manufacturing (calculated via Equation (10)).
The gross animal revenue (GR) per day was calculated based on the amount of cheese produced from each treatment (kg of cheese/animal/day) versus the cheese sales price at that time (US$ 4.74/kg of cheese).
The net profit from cheese (US$/kg of cheese) was computed as the difference between the GR from the sale of each kg of cheese and the total cost of the cheese (US$/kg of cheese).
Statistical analysis
The data were analyzed using a duplicated 5 × 5 Latin square design (Kaps and Lamberson, Reference Kaps and Lamberson2017) using the MIXED procedure in the Statistical Analysis System (SAS) software (Version 9.4). The statistical model used was as follows:
\begin{equation}
{{\text{Y}}_{ijkl}} = {{\unicode{x00B5} }} + {{\text{S}}_i} + {{\text{P}}_{j(i)}} + {{\text{A}}_{k(i)}} + {{\text{T}}_l} + {{{\unicode{x03B5} }}_{ijkl}}
\end{equation}
where Yijkl represents the response variable, μ denotes the overall mean, Si indicates the fixed effect of square i, Pj(i) represents the fixed effect of period j (within square i), Ak(i) is the random effect of animal k (within square i), Tl is the treatment effect l, and εijkl is the random error.
Subsequently, an orthogonal contrast analysis was conducted, comparing the treatment without supplementation against all supplemented treatments (WOS versus inclusion levels of 0, 260, 520, and 780 g/kg DM CRS in the supplement). The analysis also examined the potential linear and quadratic effects of CRS inclusion levels on the physicochemical characteristics and sensory attributes of Minas Frescal cheese. A significance level of 5% was adopted for the type I error.
Results
Physicochemical characteristics and cheese yield
The inclusion level of CRS in the supplement had no effect (P > 0.119) on the physicochemical characteristics (fat, protein TDE, moisture, NFDE, FTDE, acidity, and pH) of the cheese (Table 2). Additionally, fat and protein recovery rates were not affected (P > 0.240; Table 2). Similarly, there were no differences (P > 0.419) in the gross (kg of milk/kg of cheese), moisture-adjusted (kg of milk/kg of cheese), and individual (kg cheese/animal/day) yields of cheese, when CRS was included in the supplement as a ground corn replacer (Table 2).
Physicochemical characteristics, fat and protein recovery, and production yield of minas Frescal cheese from grazing dairy cows supplemented with different levels of cassava root silage (CRS) in the supplement
1 TDE: Total dry extract (%); NFDE: non-fat-dried-extract (%); FTDE: fat concentration in total dry extract (%).
2 WOS: Without supplementation.
3 SEM: Standard error of mean.
4 P value for L: linear effect; Q: quadratic effect; and WOS × Sup: contrast effect between WOS and supplemented treatments (regardless of the inclusion level).
5 Gross yield: kg of milk/kg of cheese; Moisture-adjusted yield: kg of milk/kg of cheese (with correction for cheese composition); Cheese production per cow: kg of cheese/cow/day.
When evaluating cheese from cows receiving supplementation versus cheese from cows WOS, it was found that cheeses from cows of the former group had greater protein contents (P = 0.025) and individual production (kg of cheese/animal/day; P < 0.001) than those observed in the latter group (Table 2). The other physicochemical and production variables did not differ (P > 0.134) between supplemented and WOS animals (Table 2).
Colour and sensory profiles of cheeses
There were no differences (P > 0.067) in the colour parameters (L*, a*, b*, and C*) of the cheeses made from milk from cows including different levels of CRS in the supplement (Table 3). Nevertheless, the yellow/blue coordinate (b*) and saturation index (C*) of the cheese from supplemented animals were greater (P < 0.001) than those of the cheeses made from WOS animals’ milk (Table 3). Other colour parameters (L* and a*) in the cheese did not differ (P > 0.260) between supplemented and WOS animals (Table 3).
Colour and sensory attributes of minas Frescal cheese from grazing dairy cows supplemented with different levels of cassava root silage (CRS) in the supplement
1 L*: luminosity; a*: green intensity; b*: in yellow intensity; C*: Chroma index: colour saturation and intensity.
2 WOS: Without supplementation.
3 SEM: Standard error of mean.
4 P value for L: linear effect; Q: quadratic effect; and WOS × Sup: contrast effect between WOS and supplemented treatments (regardless of the inclusion level).
5 Sensory analysis: the hedonic scale structured in the test had 10 points. varying among (0) = very disgusted; (5) = neither liked/ disliked. and (10) = I really liked it. Purchase intention: the hedonic scale structured in the test had 5 points. varying among (1) = certainly would not buy; (3) = maybe buy/maybe not buy. and (5) = certainly buy.
Sensory attributes of the Minas Frescal cheese, overall perception, and purchase intention were not affected (P > 0.107) by the inclusion level of CRS in the supplement of cows (Table 3). Cheeses from supplemented animals had higher scores for appearance (P = 0.035) and colour (P = 0.031) than those observed in the cheeses from WOS animals (Table 3). However, other sensory attributes, overall perception, and purchase intention did not differ (P > 0.452) between cheese derived from those groups.
Economic analysis
The inclusion of CRS in the supplement up to 520 g/kg DM maximized cheese production by 0.73 kg of cheese/animal/day and increased the gross revenue by 3.49 US$/animal/day, compared to WOS animals (Table 4).
Economic indices generated from the production of minas Frescal cheese from grazing dairy cows supplemented with different levels of cassava root silage (CRS) in the supplement
1 WOS: Without supplementation.
2 US $ = US dollar.
Costs of ingredients and others: US $/kg: Corn meal – US $ 0.2369; Cassava Root Silage – US $ 0.1585; Soybean meal – US $ 0.4374; Mineral Supplement – US $ 0.7891; Urea: ammonium sulfate (9:1) – US $ 0.5831; Pasture maintenance (including costs for irrigation. fertilization. and herbicides) – US $ 0.308.
Other production costs (US $/cow/day): Includes labor. milking. electricity. and other expenses (cleaning materials. health care. artificial insemination) – US $ 0.9549.
Data also revealed that milk cost to produce 1 kg of cheese, daily cost of supplementation, and cost of the supplement increased by approximately 27.9, 42.3, and 48.3% respectively, with the increasing in the CRS inclusion from 0 to 780 g/kg DM (Table 4). Milk cost and labor for supplementation remained relatively constant with the increasing in CRS inclusion in the supplement (Table 4). In general, all the above-mentioned variables were greater in supplemented animals than in WOS animals (Table 4).
Exploration of other economic indices revealed an increase in the total cost and cheese production cost and by ∼ 26.7 and ∼ 73.8% respectively, when CRS inclusion increased from 0 to 780 kg DM (Table 4). In addition, cheese production cost when using 0 and 780 g/kg DM, was 0.61 and 1.40 US $/kg of cheese respectively, more expensive than that derived from non- supplemented animals (Table 4). Similarly, total cost by using 0 and 780 g/kg DM increased by approximately 58.2 and 67.0%, respectively, compared to WOS animals (Table 4).
Finally, selling price did not change across supplementation levels of CRS (4.74 US $/kg of cheese), but its inclusion, up to 520 g/kg DM in the supplement, resulted in an increase in 3.38 US $/kg of cheese in the net profit compared to 260 and 780 g/kg DM CRS inclusion levels (Table 4). However, regardless of inclusion level, net profit for product derived from supplemented animals was lower than that observed for WOS animals (Table 4).
Discussion
The physicochemical properties of the Minas Frescal cheese produced in this study met the standards set by the Technical Regulation on the Identity and Quality of Dairy Products established by the Ministry of Agriculture, Livestock, and Food Supply (MAPA) (Brasil, 1996, 2004). With a moisture content exceeding 55%, the cheese was classified as high moisture, and the average moisture content recorded for the cheeses in this study was 62.30 ± 1.04%. In addition, according to Nunes and Caldas (Reference Nunes and Caldas2017), Minas Frescal cheese is classified as slightly acidic with pH values between 5 and 6; hence, the pH and acidy of cheese produced in this study met that classification.
Changes in the nutrient composition of the cows’ diets can affect the chemical composition of their milk and, consequently, that of the dairy products (Gulati et al., Reference Gulati, Galvin, Hennessy, McAuliffe, O'Donovan, McManus, Fenelon and Guinee2018). Hence, an effect of CRS inclusion level on physicochemical profile in the Minas Frescal cheese was expected. However, this was not observed in the current study, which could be due to milk composition being unaffected by CRS inclusion level in the supplement of grazing cows (Galvão et al., Reference Galvão, Vargas, Mezzomo, Oliveira, Dos Santos, Maciel, Soares, Farias, Caldas, Faciola and Alves2024). However, this is not a particular condition attributed to CRS supplementation, as other studies also have reported that inclusion of alternative feedstuffs in grazing dairy cow supplementation may not affect the composition of dairy products (Maciel et al., Reference Maciel, Vargas, Mezzomo, Gama, Leite, Castro, Oliveira, Farias, dos Santos Luz and Alves2022).
The protein content of Minas Frescal cheese can vary from 12 to 18% according to the current Normative Instruction (Brasil, 1996, 2004), and the values observed in the present study are within this range. In addition, cheese yield is directly related to the milk protein concentration (Villegas-Soto et al., Reference Villegas-Soto, Díaz-Abreu and Hernández-Monzón2017), and a direct relationship was observed between the protein concentration in cheese and the protein concentration of the milk used for cheese production when CRS was included in the diet of dairy cows (Galvão et al., Reference Galvão, Vargas, Mezzomo, Oliveira, Dos Santos, Maciel, Soares, Farias, Caldas, Faciola and Alves2024). Hence, a higher protein concentration may increase the amount of casein available for coagulation, contributing to a higher cheese yield (Paszczyk et al., Reference Paszczyk, Polak-Śliwińska and Zielak-Steciwko2022).
The results in this study revealed that supplemented animals produced cheeses with greater values of protein and production yield than those observed in WOS animals. This could be attributed to the greater moisture concentration (∼1.6%) in the cheese derived from WOS animals, compared to supplemented animals. It is important to note that compositional traits in cheese of cows are interrelated (Bittante et al., Reference Bittante, Amalfitano, Ferragina, Lombardi and Tagliapietra2024). Hence, changes in one compositional trait may influence others. In addition, a similar response was observed by Galvão et al. (Reference Galvão, Vargas, Mezzomo, Oliveira, Dos Santos, Maciel, Soares, Farias, Caldas, Faciola and Alves2024) who found greater levels of milk protein and yield in grazing cows including CRS in the supplement. Thus, data of the current study suggests that inclusion of CRS could be beneficial for enhancing cheese protein and yield, regardless of inclusion level. This may be interesting for the dairy industry, considering the concomitant increase in protein and casein levels in the cheese (Gomes et al., Reference Gomes, Vargas, Alves, Gomes, Medeiros, Rodrigues, de Freitas, Mezzomo and Maciel2024), and the role of casein in rennet coagulation, forming a network that retains fat, lactose and mineral salts in different proportions, which directly affects cheese properties (Furtado, Reference Furtado1999; Souza et al., Reference Souza, Caldeira, Rocha, Antunes, Ruas, Santana, Costa and Borges2015).
The colour of a dairy product is highly dependent on the concentration of carotenoids, especially β-carotene (Nozière et al., Reference Nozière, Graulet, Lucas, Martin, Grolier and Doreau2006). Carotene levels may vary among foods, with animals fed forage-based diets having higher concentrations of carotenoids (O'Callaghan et al., Reference O'Callaghan, Mannion, Hennessy, McAuliffe, O'Sullivan, Leeuwendaal, Beresford, Dillon, Kilcawley, Sheehan and Ross2017), whereas preserved and supplement ingredients have relatively low carotenoid concentrations. Although corn and CRS have low levels of carotenoids (Kimura et al., Reference Kimura, Kobori, Rodriguez-Amaya and Nestel2007; Luterotti and Kljak, Reference Luterotti and Kljak2010), the supply of carotenoids in the supplement may have contributed to the increase in the intensity of the yellow colour and the saturation index of the cheeses compared with those of the animals that were only pasture fed, suggesting that these effects on the cheese colour resulted from the supplementation (Frétin et al., Reference Frétin, Martin, Buchin, Desserre, Lavigne, Tixier, Cirie, Bord, Montel, Delbès and Ferlay2019; Panthi et al., Reference Panthi, Kelly, Hennessy, O'Sullivan, Kilcawley, Mannion, Fenelon and Sheehan2019).
The sensory attributes (aroma, appearance, colour, flavor and texture) of cheeses are often evaluated before purchase or consumption, among which colour is an important attribute in food identification because it creates expectations that influence consumers’ taste preferences (Drake and Delahunty, Reference Drake and Delahunty2017). In the sensory evaluation, the cheeses from the animals under supplementation received higher appearance and colour scores, which could be a direct result of the intensity of the yellow colour. The aroma, flavor, and texture of the Minas Frescal cheese were not affected by the diet, suggesting that those attributes were not affected by animal supplementation, which was also reported by Kilcawley et al. (Reference Kilcawley, Faulkner, Clarke, O'Sullivan and Kerry2018).
The economic analysis indicated that the inclusion of CRS up to 520 g/kg DM in the supplement resulted in greater cheese production with a lower total cost per kg of cheese produced, which could represent an increase in revenue for the dairy industry (Souza et al., Reference Souza, Caldeira, Rocha, Antunes, Ruas, Santana, Costa and Borges2015). Therefore, the results reported herein confirm that higher cheese yields translate into greater economic gains (Oliveira et al., Reference Oliveira, de Lima, de Freitas, Dos Santos, Reis, Gomes, Alves, Maciel, Vargas and Mezzomo2020; Sainz-Ramírez et al., Reference Sainz-Ramírez, Colín-Navarro, Estrada-Flores, Velarde-Guillen, López-González and Arriaga-Jordán2023). In addition, CRS can be a strategic energy source for supplement production, especially when corn costs are high. Thus, CRS can be an alternative energy source for dairy cows and can be partially included in the supplement with positive results from an economic standpoint.
Including the maximum amount of CRS (780 g/kg DM) in the supplement raised the cost per kg of supplement due to the increased acquisition/production and operational costs compared with the costs upon ground corn inclusion (Lima et al., Reference Lima, Veloso, Silva, Pires, Teixeira and Nascimento2015). This is related to the acquisition period of the ingredients, since their prices vary throughout the year (Keim et al., Reference Keim, Mora, Ojeda, Saldías and Bedenk2022). Moreover, since the corn used in this study was acquired at harvest, it was more affordable (US$ 0.27 kg/DM), whereas the CRS during this same period had the highest price (US$ 0.39 kg/DM), contributing to the higher costs. Santos et al. (Reference Santos, Modesto, Souza, Ítavo, Jobim, Kazama, Valloto and Massuda2009) replaced corn silage with aerial parts of cassava silage in dairy cow supplementation and reported a reduction in feeding costs due to the low cost of this cassava byproduct.
The gross revenues of the cheeses produced after CRS supplementation to the dairy cows were greater than the cheese revenue from non-supplemented cows because cheese production increased upon inclusion of CRS in the supplement. However, net profits with CRS supplementation were lower because of the higher production costs. The lower revenue upon treatment WOS alone was due to reduced cheese production; however, the cheese production costs using this strategy were significantly lower than those of supplementation, which was justified by the lower cost of using only mineral salts (Galvão et al., Reference Galvão, Alves, Dos Santos, Oliveira, Mezzomo and Vieira2023).
Conclusions
The use of CRS may be a suitable alternative for replacing ground corn in the supplement of grazing dairy cows and can be included at concentrations of up to 520 g/kg DM without negatively affecting production, composition or quality of Minas Frescal cheese. Particularly, inclusion of CRS in the supplement may enhance the protein content of the cheese and cheese production, regardless of inclusion level. In addition, it is necessary to pay special attention to the current and local market value of cassava root and the strategy of its utilization, in order to increase the profitability of Minas Frescal cheese production.
Supplementary material
The supplementary material for this article can be found at https://doi.org/10.1017/S0022029925101015.
Acknowledgements
The authors thank the support of the members, including professors and students, of the Grupo de Estudos em Produção de Animais Ruminantes de Carajás (GEPARC) during field data collection and laboratory analyses.
Funding
This study was funded by the Programa Nacional de Cooperação Acadêmica na Amazônia (PROCAD/Amazônia/project number: 88887.200488/2018–00).
Conflict of interest
None.
Ethics approval
The authors confirmed that the sensory study involving human participants was carried out in compliance with the World Medical Association's code of ethics (Declaration of Helsinki) and was approved by the Ethics Committee for Research at the Universidade Estadual do Pará (UEPA; protocol number: 4.106.18). Additionally, all animal procedures received approval from the Ethics Committee on the Use of Animals at UFRA (protocol number: CEUA 015/2018).
Data availability
Data of this research can be shared upon reasonable request.
Authors’ contributions
Cláudia Siqueira Caldas: data collection and curation, laboratory analyses, writing of the original draft, review, and editing. Julián Andrés Castillo Vargas: Data curation, investigation, writing of the original draft, review, and editing. Rafael Mezzomo: Data curation, statistical analysis, writing of the original draft. Luís Rennan Sampaio Oliveira and Raylon Pereira Maciel: Data curation and investigation, writing of the original draft. Luckas Thiago Oliveira Galvão, Perlon Maia dos Santos: data collection and curation, writing of the original draft. Fernanda Lopes da Silva, Ellen Vitória da Silva, and Açucena Guedelha Bonfin: Data collection and laboratory analyses. Antonio Pinheiro Faciola: data analysis, review and editing. Kaliandra Souza Alves: conceptualization, data curation, funding acquisition, project administration, and writing of the original draft, review, and editing.
