This study was conducted during the summer season (25°C–35°C range) at Kaffirstad Farm (Mpumalanga Province, South Africa), situated at 26.34°S and 29.18°E. Sorghum (Table 1) was acquired from the Agricultural Research Council (Potchefstroom, South Africa). SBM, corn, molasses, lime, and premix were obtained from Opti Feeds (Lichtenburg, South Africa), while canola meal was purchased from Southern Oil (Pty) Ltd., South Africa.

Two iso-nitrogenous and iso-energetic treatment diets were formulated as follows: 1) a maize–soybean diet = control diet; and 2) a sorghum–canola meal (SCM) diet totally replacing maize with sorghum and soybean with canola meal in rabbits’ diets (Table 1), to meet the nutritional requirements recommended by the National Research Council (NRC, 1977). Experimental diets were milled and analyzed for dry matter, organic matter, and crude protein, as per method nos. 930.15, 924.05, and 984.13, respectively (AOAC, 2005), while crude fiber (ANKOM Technology, New York, USA) and minerals (calcium, phosphorus, sodium, chloride, and potassium) were analyzed in accordance with the AgriLasa (1998) guidelines.

Thorough cleaning of the house and equipment was done with water and detergent. Thereafter, disinfection was done with F10 (Health and Hygiene (Pty) Ltd., Roodepoort, South Africa). After a 2-week downtime, the house was warmed using an infrared bulb, prior to commencing the experiment. A total of 40 New Zealand white rabbits (20 bucks and 20 does), with an average initial weight of 613 ± 6.59 g, were randomly allotted to two treatment diets in a two-way factorial arrangement in a CRD with a pen (0.07 m² per weaned rabbit) as an experimental unit, holding one rabbit, replicated five times. The house was fitted with an automatic nipple drinker, while feed was offered ad libitum, with constant lighting (24 h) for a period of 42 days.

Every morning, feed was offered, and refusals were weighed using a digital weighing scale (Explorer EX224, 0.01g, OHAUS Corp., Parsippany, NJ, USA). All the rabbits were weighed for initial weight at the start of the experiment, and thereafter on a weekly basis, to determine weekly weight gain (WWG). The feed conversion ratio (FCR) was then computed as the average feed intake (FI) over average weight gain. During slaughtering, blood was collected from three rabbits randomly selected from each treatment and sex from the neck, into a 4-mL red top anticoagulant, then stored for 45 min at room temperature to coagulate, before being refrigerated at 4°C (Washington and Van Hoosier, 2012). Next, 48 h after blood collection, serum was made by centrifuging the clotted blood for 20 min at 1,500 rpm. The serum supernatant settled on top, while blood cells were deposited at the bottom. The serum was sieved into a 0.5-mL conical centrifuge tube and then stored at −20°C. Finally, the serum biochemical parameters (total protein, albumin, creatine, urea, sodium, potassium, calcium, magnesium, glucose, cholesterol, triglycerides, alanine transaminase, and aspartate aminotransferase) were analyzed using an automated Idexx Vex Test Chemistry Analyser (IDEXX Laboratories, Inc., Westbrook, ME, USA).

All rabbits were starved for 13 h at the end of the feeding trial, then weighed (slaughter weight) using a digital weighing scale, and transported to a local poultry abattoir for slaughtering. The rabbits were electrically stunned and hung upside down on a movable metal rack. The jugular vein was cut with a sharp knife for bleeding, and they were then skinned.

Muscle color was determined 24 h after slaughter. The fillet and leg color, including lightness (L*), redness (a*), and yellowness (b*), were determined by a spectrophotometer (CM 2500c model, Konica Minolta, Japan), replicated three times to obtain the average value of the color. A pH meter (CRISON pH25, CRISON Instruments SA, Spain) was used to determine the pH of each rabbit at 45 min (pH_i) and 24 h (pH_u) post-mortem, as described by Manyeula et al. (2020).

From the 24 sampled rabbits (12 males and 12 females), approximately 30 g of muscle strips were cut from the fillet muscle, parallel to the fiber direction, then weighed (W_i) on a weighing scale (Explorer EX224, OHAUS Corp.) of 0.01 g sensitivity. The muscle strips were suspended in a cold room (4°C) for 72 h and reweighed (W_F) to determine drip loss, following the method proposed by Honikel and Hamm (1994), as indicated in Equation 1:

Raw muscle cubes from 24 sampled rabbits were cut from the fillet muscles, initially weighed (W_i), and oven-heated at 75°C for 45 min. Thereafter, the oven-heated muscle was cooled down at room temperature for 15 min and dried with soft tissue and reweighed (W_F) (Sanka and Mbanga, 2014). Cooking loss was calculated using Equation 2:

A total of 50 untrained consumers were recruited from among the participants in a 1-day conference held at the North-West University (Mafikeng, North-West Province, South Africa). Information regarding sensory evaluation was collected via a questionnaire before the sensory assessment of the samples, which was conducted in the class, before lunch at noon. Each individual panelist was issued with an eight-point hedonic scale to evaluate each meat sample from the two treatments and deionized water to cleanse the mouth after each sample was offered. Verbal instructions on the test were highlighted prior to commencing. Approximately 5-cm portions of meat were cut and baked for 30 min in an oven (105°C). Immediately after cooking, the meat was cut into 1 cm³, randomly placed (one/treatment) on a plastic form pack, and divided into two equally sized wedges. Each panelist assessed two samples, one from each of the dietary treatments, and numbers were used to identify the samples. The panelists were asked to score each sample using an eight-hedonic scale, as follows: extremely bland = 1; very bland = 2; fairly bland = 3; slightly bland = 4; slightly intense = 5; fairly intense = 6; very intense = 7; and extremely intense = 8, following the procedure proposed by William et al. (2023).

The data on average weekly weight gain were analyzed using repeated measures analysis (SAS, 2010), employing the following statistical model:

Where, Yijkl = respond variables, μ = population mean, Di = effects of diets, Sj=effects of sex,  Wk= time effect in week,  (D*s)ij = interaction effect (diet and sex), (D*W)ij = diets and weekly interaction effect, and Eijkl = random error.

Overall growth performance, serum parameters, and meat quality data were analyzed using the general linear model (GLM) procedure of the Statistical Analysis System (SAS, 2010), with the dietary treatment and sex as the factors, using the model below:

Where, γijk = dependent variable,  μ = population mean, Di= diets effect, Sj=Sex effect,  DSij = interaction of diet and sex, and Eijk= random error associated with observation,  ij = assumed to be normally and independently distributed. Significance was declared at P ≤ 0.05, and the means were separated using the probability of difference option in the LSMEANS statement of SAS.

The main rabbits’ performance parameters, with values of FI, FCR, and AWG, are summarized in Table 2. They showed that there was a diet and sex (male or female) interaction (P < 0.05) effect on AWG (P = 0.01) but not on FI (P = 0.96) and FCR (P = 0.89). Likewise, significant diet was also noted on AWG (P = 0.03), but not on FI (P = 0.98) and FCR (P = 0.36). Diet and sex did not affect overall FI and FCR but influenced overall weight gain.

In weeks 3 and 6, only the diet had significant effects (Table 3). However, in weeks 4 and 5, the effects (P < 0.05) were on diet and sex only, whereas in weeks 7 and 8, only sex had a significant effect. Rabbits fed SCM were the heaviest in weeks 1, 3, 4, 5, and 6. In week 1, the bucks were heavier (P < 0.05) than the does, whereas in weeks 4, 5, 7, and 8, the opposite was true. No significant effects, based on sex, were detected in weeks 2 and 6.

The findings in Table 4 showed that the diet significantly affected sodium and potassium, while magnesium was affected by both diet and interaction. Protein, creatinine, and cholesterol were significantly affected (P > 0.05) by interaction (diet * sex). With regard to diet effects, potassium and magnesium were higher in rabbits fed the CON diet, compared to those on the SCM diet, whereas sodium was higher in the rabbits reared on the SCM diet than those on the CON diet. Sex did not have any significant effect (P > 0.05) on the plasma protein, minerals, energy, and enzyme profiles of the rabbits.

The interaction (diet * sex) was significant on leg b*, pH_u, and drip loss compared with the other experimental groups (Table 5). Fillet L* and b*, pH_u, cooking, and drip loss were affected by the diet, whereas sex affected (P < 0.05) pH_u and drip loss only in terms of meat quality. Rabbits fed the SCM diet had the highest fillet L*, b*, and pH_u, while those fed the control diet had the highest values on cooking and drip loss. Nonetheless, sex significantly affected pH_u and drip loss only (Table 5). Other meat quality parameters were not significantly (P > 0.05) affected by the main effects. Does recorded the highest value of pH_u but had the lowest value of drip loss.

There was no diet and sex (male or female) interaction in the heart, liver, kidneys, lungs, and small intestine (Table 6). Likewise, no significant dietary effects were observed on the fillet, heart, kidney, lungs, and small intestines. Notably, sex significantly affected the hot carcass weight (HCW), cold carcass weight (CCW), hind legs, heart, liver, and large intestine (Table 6). Rabbits fed the SCM diet had the highest values (P < 0.05) of HCW, CCW, hind leg, heart, liver, and large intestine weights. Does had significantly higher HCW, CCW, and hind leg weight. On the other side, the liver and large intestine weights were higher (P < 0.05) in bucks, while other parameters were not affected.

Table 7 shows that diet did not affect (P > 0.05) meat quality. The consumers classified the meat from rabbit-fed treatment diets as slightly intense in all attributes.

Sorghum offers a nutritive value comparable to that of maize and other conventional feed grains, confirming its suitability as a feed ingredient (Zarei et al., 2022). Similarly, canola meal ranks second only to soybean meal as a high-quality protein source in animal diets and is rich in essential minerals, vitamins, and beneficial phenolic compounds (Newkirk, 2009). In this study, previous results showed that the interaction (diet * sex) affected weight gain, suggesting that the weight gain of rabbits is affected by age and sex. A lack of interaction (diet * sex), observed in Tables 2-6, indicates that diet and sex acted independently on those parameters. However, significant interaction in the tables revealed that those parameters were acted on jointly by diet and sex. A lack of interaction effects (diet * sex) on FI concurs with the findings of Hussein and Abdelfattah (2020), who reported no interaction (diet * sex) in FI in growing rabbits fed a concentrate meal. The interaction (diet * sex) on FCR contradicts the findings of the study of Hussein and Abdelfattah (2020): diet and sex did not have effects on FI and FCR, implying that the feed was consumed and converted to muscle in the same way as in the control diet. This aligns with what was reported by Gugołek et al. (2018), who found similar FI and FCR in rabbits fed rapeseed meal, white lupine, and pea seeds as SBM substitutes. Also, Hussein and Abdelfattah (2020) reported that sex had no significant effect on the FI of California rabbits fed a commercial diet. The lack of significant diet and sex effects on FI suggests that substituting maize with sorghum and SBM with canola meal as energy and protein ingredients in rabbit diets does not change the palatability of the diets, irrespective of the sex of the animal. The non-significant interaction (diet * sex) on weight gain aligns with the findings of a study by North et al. (2018). Rabbits have been reported to possess well-developed GITs when it comes to utilizing fibrous diets and tolerating antinutritional compounds such as tannin (Wolf and Cappai, 2020). This may explain why rabbits fed SCM diets had similar (weeks 2, 7, and 8) or higher in other weeks in BWG, compared to those reared on the control diet. These observations match those of Njoku et al. (2013), who reported an increased average daily weight gain with increasing sorghum milling dust in the diets of growing rabbits. Indeed, it was noted by Hamlin (2012) that rabbits are anatomically and physiologically developed for a high-fibrous (20%–25%) diet. The faster weight gain of does (more than bucks) in week 1 was surprising, and there is no evidence as yet to show that this was reported in previous rabbit- or livestock-related research. The current results in weeks 4, 5, 7, and 8 match what is reported in the literature (Ologbose et al., 2017; Lamptey et al., 2022), where bucks were reportedly slightly heavier than does at all ages, possibly due to better feed utilization. Contrary to these results, Birolo et al. (2020) reported similar weights in male and female rabbits, in line with the results from weeks 2, 3, and 6, suggesting that in those weeks rabbits utilized feeds to the same degree of efficiency, regardless of their sex. This could be because, 6 weeks after weaning, bucks and does frequently use diets similarly because they have not reached sexual maturity, when physiological and metabolic disparities depending on sex become more noticeable.

Interaction (diet * sex) was observed on protein, creatinine, magnesium, and cholesterol, suggesting that these serum metabolite parameters are controlled jointly by diet and sex. The plasma proteins and energy were not affected by diets and sex, suggesting that the intake of the sorghum–canola-based diet had no negative post-ingestive feedback, irrespective of sex. The lack of post-ingestive feedback was further explained by similar (P < 0.05) plasma enzymes (ALT and AST) profiling, which are indicators of hepatocellular damage (Ambrosy et al., 2015), and suggests that the combination of sorghum and canola has the potential to be used in the diets of rabbits, as indicated by values falling within the normal ranges for healthy rabbits (Jenkins, 2008). The higher plasma sodium in rabbits reared on an SCM diet may be a reflection of the high concentration of this mineral reported in sorghum (9.04 mg/100 g), compared to maize (3.74 mg/100 g) (Jocelyne et al., 2020). Rabbits are more tolerant of high sodium, but a plasma sodium concentration above 185 mmol/L can induce hypernatremia (Patson and Sladky, 2020), which is why rabbits fed SCM did not show any sign of hypernatremia at plasma concentrations of 141.0 mmol/L in this study. Sorghum contains antinutritional factors such as tannins, kafirin, ferulic acid, endosperm, and phytate that hinder its utilization by monogastric animals and result in a lowering of the digestibility of carbohydrates, proteins, and minerals (Masenya et al., 2021), which could explain why rabbits fed a sorghum–canola-based diet had lower plasma potassium and magnesium levels. Moreover, canola is known to contain low levels of potassium and magnesium (Mahan et al., 2005).

Interaction (diet * sex) had significant effects on meat yellowness, pH_u, and drip loss, implying that these traits are controlled jointly by diet and sex, which contradicts the findings of the study of Daszkiewicz and Gugołek (2020). The decline in the pH value hinders the growth of undesired bacteria, rates of conversion of color, and the formation of undesired flavors in the meat, which was observed in the meat from rabbits reared on the CON diet in this study. It may be deduced that combining sorghum and canola in the diet of rabbits reduced the conversion rate of glycogen levels to lactic acid after slaughter, hence improving meat quality. This finding concurs with the literature reporting the effects of sorghum- (Hernández-Martínez et al., 2018) and canola-based diets (El-Medany and El-Reffaei, 2015) on the femoris and biceps muscles of rabbits. Investigators (Yalçın et al., 2006; North et al., 2018) reported higher muscle ultimate pH in bucks, whereas other studies found no sex-related effects on meat pH (Dalle Zotte et al., 2016; Frunza et al., 2019). Cooking and drip loss were higher in rabbits fed the control diet, which corroborates the findings reported by El-Medany and El-Reffaei (2015). Those authors reported significantly lower cooking loss in rabbits fed canola than in those fed a control diet. During meat processing (cooking and drying), juices are released together with the nutrients, due to protein denaturation and muscle atrophy (Purslow et al., 2016), resulting in poor-quality products. In this study, low cooking and drip loss observed in rabbits fed the SCM diet imply that utilizing sorghum and canola (SCM) in the diet improved the meat quality in terms of tenderness, juiciness, and flavor by promoting muscle structure to reduce the loss of juices (nutrients). Furthermore, drip loss can also be influenced by secondary metabolites (methylglyoxal), changing the properties of muscle proteins (Przybylski et al., 2022). Loss of water and the application of heat to the meat led to the formation of pores and shrinkage, which can affect the texture and taste, leading to poor meat quality (Koc et al., 2008). The significantly higher drip loss on the meat from does contradicted the findings of North et al. (2019), who reported similar drip loss values from the meat of males and females. The current results suggest that bucks possess better meat quality than does, as drip loss determines the appearance, texture, nutritional value, and attractiveness of the meat (Filho et al., 2016).

The heavier HCW, CCW, heart, hind leg, liver, and large intestine in rabbits reared on SCM could be associated with final weight. Indeed, Michalik et al. (2009) reported a strong positive relationship between rabbit body and carcass weight, which is indicative of an adequate diet and nutrition (Cardoso et al., 2011). However, the presence of toxins in the diet is reflected by hypertrophy or hypotrophy of the liver (Ewuola et al., 2003), which could be another reason why rabbits fed the SCM diet had heavier livers, as mentioned in the study of Masenya et al. (2021). The kidneys, heart, and large intestine were not affected by diet, suggesting that replacing maize with sorghum and canola with soybean in rabbit diets did not show any anatomical adaptation responses in the rabbits. The present results align closely with similar findings reported by many researchers (Hernández-Martínez et al., 2018; Wafar et al., 2020). The authors reported similar carcass traits and internal organs when weaner rabbits were fed a malted sorghum cultivar. In this study, it was clear that the sorghum–canola-based diet (SCM), when compared to the control diet, can improve the profitability of sustainable rabbit farming, as it is associated with heavier internal and external organs that determine market prices in the commercial milieu. The higher (P < 0.05) HCW, CCW, fillet, and hind leg weights of does contradict the findings of North et al. (2019), who found similar HCW, CCW, and hind leg weights in male and female rabbits. Bucks had heavier livers and hearts and longer large intestines than does, which correlate with the findings of Hussein and Abdelfattah (2020), who reported significantly heavier internal organs in bucks than in does, which could explain why males grew faster than females (Hussein and Abdelfattah, 2020). The results reported by Tůmová et al. (2022) match the findings of the current study, i.e., the internal organ weights of rabbits were not affected by their sex.

The consumers classified the meat from rabbits fed with all treatment diets as slightly intense (5) in all attributes, implying that rabbit meat from an SCM diet would equally be patronized as the meat from rabbits fed the control diet. This indicates that the combination of sorghum (maize replacer) and canola (soybean replacer) in the rabbits’ diet did not have an adverse effect on meat quality. However, meat juiciness is an important parameter for consumer choice, as it is associated with fat content (Wognin et al., 2018), and this led the researchers to deduce that the lack of a significant effect in this study could be due to similar lipid meat contents. Similar to these current results, Kpehe et al. (2020) reported a lack of difference (P < 0.05) in the organoleptic quality of the meat of rabbits fed diets containing graded levels of rice offal.