*4.1. Macronutrients*

Almost all the reports on meat from beef raised in the Tropics that include proximate analyses are from Latin America (Table 2). Compiled in tabulated form, it is readily apparent that there is generally little variation in the macronutrients of cattle grown in tropical regions. With the exception of three findings, the moisture ranges from 71% to 76%, and three quarters of the reported data are between 72% and 74%. The three excluded findings appear to be outliers, reporting very low moisture values [96,98]. Two of these data points are from a study comparing the use of anabolic steroid implants where low moisture contents (about 60%) and concomitant high protein contents (about 36%) were reported [96]. The third is from a study comparing meat from cattle raised on pasture with supplementation to feedlots, and the moisture contents of the latter were reported as 67.3%, likely, at least in part, a consequence of the high IMF content [98]. Indeed, in all of the studies reporting differences in moisture content, these data correspond with opposing differences in IMF content.


*Foods* **2021**, *10*, 1025



anabolic agent; Revalor (Rev) implants consisting of the anabolic steroid trenbolone acetate and the estrogen hormone, estradiol.

While almost all of the protein content values are in the range from 20% to 24%, some differences were observed between bulls and steers (22.9% vs. 23.8%, respectively [108]), breeds of Nellore and Nellore × European crosses (25.1% vs. about 23.8% respectively) [66] and pasture and grain finishing (21.4% vs. 18.4%, respectively [68]). However, these differences in the protein contents were relatively small, and the majority of the reports found no differences in the studies that covered a range of factors, including age at slaughter, fat class, carcass grade, sex, breed, feeding system and muscles.

The IMF content showed the greatest variation of the proximate analyses, ranging from 1.0% to 8.9%. However, the majority of the studies indicates a lean meat, with more than three-quarters of the data reporting an IMF of ≤3.6%. These findings are in accord with the low marbling and lean carcasses reported in tropical beef [62,63,66,67,84,92,94]. Given the importance of marbling score in the global marketplace, studies have been undertaken to better understand the low marbling scores in *Bos indicus*-influenced cattle compared to *Bos taurus* cattle [118–120]. In a review of these studies, it was noted that no strong relationship between the capacity to synthesize fatty acids *de novo* and the marbling score or adipocyte volume was reported, and it was concluded that the low marbling scores typically observed in *Bos indicus*-influenced cattle are mainly attributed to their smaller intramuscular adipocyte volume compared with *B. taurus* breeds [16]. Given that it is the volume of adipocytes, and not the quantity, that is of most consequence in explaining the difference in the genotypes, this reduced volume would also explain low IMF contents of tropical beef.

While generally low in cattle in tropical environments, there are reported differences of IMF contents with production characteristics. In three studies comparing slaughter ages, IMF content was observed higher in older animals. Cattle described as typical for Puerto Rico that were at least 3 years of age at slaughter had higher IMF than those of up to 2.5 years at slaughter (1.9% vs. 2.7%) [77]. In southern India, Kangayam bulls > 3 years of age had higher IMF contents than those 12–18 months old (2.89% vs. 2.09%) [110]. And, in Brazil, a progression of increasing fat contents with the age of Nellore steers was observed when grouped as 20–24 months (4.2%), 30–36 months (5.0%) and 42–48 months (5.7%) [67].

Differences in IMF content with sex are also reported in two Brazilian studies. One study found that pasture-finished Nellore × Aberdeen Angus steers had higher IMF content than bulls (1.96% and 0.95%) [108], while another reported that whole bulls had the lowest IMF content (1.23%), surgically castrated steers the highest (2.17%) and chemically castrated steers were intermediate and significantly different from both (1.61%) [107]. Increased accumulation of IMF through fat deposition induced by castration is primarily a result of an altered hormonal balance [121].

In terms of breed, the IMF content findings are inconsistent (Table 2). Lower IMF content was reported in Brahman than Charolais bulls (2.4% vs. 2.9%) [74], and in 25-month-old Nellore steers (2.65%) compared to Santa Gertrudis × Nellore crosses (3.64%), while the IMF content of Simmental × Nellore steers was intermediate and not different from either [105]. Two studies found no differences in IMF contents, both reporting low values (1.3–2.0% IMF) [66,78]. In one of these studies, the breeds were not reported [66], while in the other bulls of zebu, Holstein and European Brown Swiss purebreds and zebu crosses with Holstein, American Brown Swiss or European Brown Swiss were used. The lack of effect of genotype in the latter study was explained a result of the use of the combination of an anabolic implant and a β-agonist in the diet, which can significantly reduce the accumulation of fat in the muscle [122]. Lastly, one study reported a higher IMF content in *Bos indicus* than *Bos taurus cattle* (5.7% vs. 5.0%), but the breeds of cattle were not given. These IMF values are notably high compared to the other studies [68].

Finally, there are two reports on the impact of the feeding system on IMF content. Grain-finished bulls (*Bos indicus* and *Bos taurus*) produced meat with higher IMF contents than those finished on pasture (7.7% vs. 3.0%) [68]. Meat from steers (defined as "multiracial") raised in feedlot had higher IMF content than those raised on pasture with supplementation (8.9% vs. 5.58%), both values being relatively high [98]. These findings

are in accordance with the generally accepted conclusion that diets with a high content of concentrate cause rapid growth of cattle, and this is associated with a greater deposition of IMF [123].

### *4.2. Fatty Acids and Cholesterol*

Beef IMF, regardless of where the cattle is raised, is comprised of over 20 individual fatty acids, of which six contribute more than 90% of the total fatty acid (TFA) content, myristic (C14:0), palmitic (C16:0), palmitoleic (C16:1), stearic (C18:0), oleic (18:1 cis-9) and linoleic (C18:2) acids [20]. Fatty acid profiles can vary with factors such as breed, feed and sex of the animal.

Given the decades of research on crossbreeding as a means to improve production traits and meat quality, including through increased fatness and marbling of otherwise lean *Bos indicus* cattle, it is not surprising that many of the studies on fatty acids focus on the impact of breed. In a Cameroonian study, while some differences were found, overall, genotype had a limited effect on the fatty acid profile of crossbred Simmental × Gudali and purebred Gudali bulls [106]. All of the animals were raised on pasture and were slaughtered at 20–41 months of age. These workers suggested that similar diets explained the lack of differences in the fatty acid profiles between breeds. Indeed, the levels of α-linolenic acid (C18:3n-3) and linoleic acid (C18:2n-6), both of dietary origin, were similar between the genotypes. High levels of PUFA (17.8% of TFA) were explained [106], not only by the effect of the relatively high proportion of phospholipids in muscle expected in the very lean animals but, also, by the high content of PUFA, particularly PUFA n-3, that characterizes fresh forage from pasture [124].

In another Cameroonian study using three local *Bos indicus* breeds, it was also found that breed had a limited effect on the fatty acid profile of meat from Gudali and Red Mbororo bulls raised on pasture and slaughtered at 3–5 years of age [125]. The only difference, albeit small, was a lower concentration of stearic acid (C18:0) in the Gudali bulls (18.0% vs. 19.8% of TFA). Gudali also had a lower concentration of stearic acid compared to White Fulani (20.7% of TFA). It was suggested that these differences could be a consequence of differences in gastrointestinal tract and rumen volume among breeds, which can influence the ruminal microbial ecosystem [68]. Ruminal biohydrogenation of dietary fat was concluded to have occurred to a lower extent in Gudali compared to in White Fulani cattle, given that, in addition to the aforementioned lower proportion of stearic acid in the IMF of Gudali, a higher proportion of α-linolenic acid (C18:3n-3; 2.34% vs. 1.61% of TFA) was found. Furthermore, docosapentaenoic acid (C22:5n-3), which is derived from α-linolenic acid, was also higher in Gudali than in White Fulani beef (1.62% vs. 1.17% of TFA). As already mentioned, α-linolenic acid is exclusively of dietary origin, and in addition, stearic acid is an end product of the biohydrogenation of dietary unsaturated fatty acids (UFA) [125].

Other differences in the fatty acid profiles of the Gudali and White Fulani were evident, while the Red Mbororo was generally not significantly different from either. Gudali bulls had higher tridecanoic acid (C13:0) (0.25% vs. 0.11% of TFA) and lower pentadecanoic acid (C15:0) (0.29% vs. 0.37% of TFA) relative to White Fulani. The pentadecanoic acid findings were explained by genetic differences between the breeds related to *de novo* C15:0 syntheses from propionate in adipose tissue [126]. Total SFA was lower and PUFA and n-3 PUFA were higher in Gudali compared to White Fulani. The SFA and MUFA were positively correlated with IMF and the PUFA was negatively correlated, suggested a consequence of the decrease in the phospholipids/neutral lipids ratio that arises from an increase in the IMF [127]. Reported PUFA/SFA ratios of 0.29 [125] and 0.33–0.36 [106] are lower than the minimum PUFA/SFA ratio of 0.45 recommended for human health [128]. The inability to achieve the recommended PUFA/SFA ratio is well-documented in both *Bos taurus* [129–131] and *Bos indicus* [68,107] cattle, a consequence of the extensive biohydrogenation of the dietary UFA by rumen microorganisms.

In another African study, further differences in the fatty acid profiles of two *Bos taurus* breeds (Borgou and Lagunaire), and Zebu Fulani were observed [58]. These cattle were raised on pasture and slaughtered at about 5 years of age. Zebu Fulani had higher contents of myristic (C14:0; 2.08% vs. 0.37% and 1.42% of TFA, respectively) and palmitic acids (C16:0; 22.4% vs. 14.5% and 19.3% of TFA, respectively) than Lagunaire and Borgou breeds. The content of α-linolenic acid varied from 2.46% to 3.81% of TFA but did not differ with breed. Zebu Fulani had the highest proportion of SFA (49.7% vs. 35.6% and 43.0% of TFA, respectively) and lowest total n-6 fatty acids (10.2% vs. 20.3% and 16.4% of TFA, respectively) when compared to Lagunaire and Borgou bulls. The ratio of PUFA/SFA fatty acids varied from 0.04 to 0.57 and was higher in Borgou than in Lagunaire.

In a Brazilian study, it was found that Nellore × Santa Gertrudis steers had higher SFA (508 vs. about 474 g/kg total fatty acid) and conjugated linoleic acid (CLA) isomer C18:2 cis-9, trans-11 (9.9 vs. 8.4 and 9.2 mg/g fat, respectively) and lower PUFA (46 vs. about 70 g/kg of TFA) than Nellore or Simmental × Nellore [105]. No differences in n-6 fatty acids were observed, but Simmental × Nellore cattle had lower n-3 than Nellore or Santa Gertrudis. In a Mexican study, it was found that fatty acid profiles of IMF from crossbreeds of 3/4 zebu or 3/4 European (based on Holstein crosses) differed, regardless of pasture or feedlot finishing [114]. The IMF from steers with *Bos indicus* dominance contained more myristic (31 vs. 26 mg/g fat), palmitic (255 vs. 238 mg/g fat), linoleic (64 vs. 38 mg/g fat) and linolenic acids (13 vs. 6 mg/g fat), but less stearic acid (191 vs. 226 mg/g fat) than steers with *Bos taurus* dominance.

While these few studies illustrate the complexity of research on the impact of breed on the fatty acid composition of beef, recent work demonstrates that it is even more complex than these studies would suggest. Indeed, a Brazilian study of pasture- and grainfinished purebred Nellore and crossbred Simmental × Nellore bulls found that, among the 43 individual fatty acids and indices of fatty acids in the IMF, 14 were affected by an interaction between the genetic group and the finishing system [117]. For the major groups of fatty acids, the interaction of the genetic group and the finishing system influenced the totals of the SFA and PUFA, while, for MUFA, only the effect of genetic group was significant. Of the 29 fatty acids and indices of fatty acids where the interaction was not significant, 11 were influenced by the genetic group and 25 by the finishing system. Overall, with only the exceptions of cholesterol, 18:1 trans-6,-7,-8, 18:1 trans-12 and the ratio 22:5n-3/18:3n-3, all the individual fatty acids and indices were affected by at least one of the factors considered or their interactions. These findings serve to illustrate that the importance of studying animals in their production environment cannot be overstated. The most prominent feature was the impact of the finishing system on the IMF and the fatty acid profile, but differences among the genetic groups were important. Differences among the genetic groups were minor with pasture finishing, but in grain finishing, the *Bos indicus* showed higher amounts of SFA and stearic acid and lower concentrations of fatty acids synthesized from linoleic and α-linolenic acids. Generally, animals finished on pasture produced meat with lower IMF, trans fatty acids and SFA contents and higher CLA and long-chain PUFA (20:5 n-3 and 22:5 n-3).

A number of studies have evaluated the impact of feed on fatty acid composition, and in particular, the manipulation of animal feed has been used as a method to improve the nutritional quality of meat. In general, pasture systems (not necessarily tropical) lead to an increase in PUFA in bovine meat compared to grain-based diets [132,133]. Diets rich in forage favor the growth of fibrolytic microorganisms responsible for the rumen production of CLA [134], and cattle fed forage have higher concentrations of linoleic, stearic, arachidonic (20:4n-6), eicosapentaenoic (20:5n-3) and docopentanaenoic (22:5n-3) acids in meat than those fed concentrates [135]. However, it must be kept in mind that pastures in climatically different environments can vary enormously. Furthermore, climatic variations in tropical regions can greatly impact pasture. For example, a Venezuelan study concluded that variations in linoleic acid and CLA contents in IMF could be explained by slaughter in two different seasons [136]. Brahman crosses were slaughtered at 17 and 19 months of age

in April and June 1998, which coincided with the "El Niño" climatic effects, characterized by a prolonged dry season. A 24-month-old group was slaughtered between November and December 1998 under less severe environmental conditions. During drought periods, the fiber contents in plants in pastures increase. In addition, a higher quantity of soluble fiber creates an environment that promotes a greater production or a decreased utilization of CLA by rumen bacteria [137], explaining, at least in part, the higher CLA content in meat from the younger animals (1.76 mg/g IMF at 17 months, 1.98 mg/g IMF at 19 months and 1.20 mg/g IMF at 24 months). It should be noted that these workers also concluded that, considering the sparingly low lipid concentrations (<2 g/100 g of fresh muscle), none of the meat could be considered a significant source of CLA. And, while one might think that feedlots would reduce variation among climatic regions given the global trade of grain, factors other than those related to climate may be significant. For example, it is reported that due to limitations of grain processing, the utilization of starch in feedlots in Brazil is not optimal, and levels fed are much lower than those used in North American feedlots [138]. Furthermore, there is a significant incorporation of forage and byproducts in the feeds of feedlot-raised beef.

In Mexico, a study found that steers of 3/4 zebu and 3/4 European cattle in a feedlot system consumed greater amounts of fatty acids compared to those on pasture (157 vs. 116 g/day SFA, 154 vs. 77 g/day MUFA and 189 vs. 135 g/day PUFA), but the latter consumed more alpha-linolenic acid [114]. Pasture-fed animals deposited lesser proportions of myristic and palmitic acids (32.4 and 255.4 mg/g fat, respectively) in the IMF than feedlot animals (21.8 and 236.5 mg/g fat, respectively), even though the consumption of these two fatty acids was similar with production system. No differences in the n-6/n-3 ratio (about 7.2) or total content of CLA were observed (14.4–16.8 mg/g fat). It was suggested that CLA is more likely located in the subcutaneous fat than in the IMF.

In another Mexican study, the concentration of the CLA isomer, C18:2 *cis*-9, *trans*-11, in beef from grazing cattle was slightly more than double that in meat from cattle-raised in a feedlot system [98]. In this experiment, steers defined as "multiracial" were raised on pasture to a final live weight of 320 kg or in feedlot to 480 kg. In the pasture-fed beef, concentrations of pentadecanoic (C15:0), heptadecanoic (C17:0) and stearic acids and CLA isomers (C18:2, n-6, C18:3, n-3, C18:2 *cis*-9 and *trans*-11) were higher, while beef from the feedlot system had higher concentrations of myristic, myristoleic (C14:1), oleic (C18:1 *cis*-9), elaidic (C18:1, n-9) and *cis*-10 heptadecanoic acids (C17:1*cis*-10).

In a Colombian study, meat from zebu cattle raised in four production systems (two silvopastoral systems, improved pasture and a traditional grazing system) was compared [139]. Myristic and palmitic acids were higher in meat from the traditional (3.65 and 32.6 g/100 g of TFA, respectively) than the improved pasture system (2.82 and 28.9 g/100 g of TFA, respectively), while linolenic acid was lower (0.96 vs. 2.30 g/100 g of TFA). The results from the silvopastoral systems differed, with one system showing a similar fatty acid profile to the traditional pasture system.

Aside from breed and feed, the potential of castration as a means to improve the meat quality in tropical beef is a research area that has piqued some interest. Higher levels of stearic acid were found in the IMF of meat from castrated Nellore cattle (around 21% of TFA) than from intact bulls (18.5% of TFA) in a Brazilian study comparing two methods of castration [107]. The linoleic acid concentration was lower in surgically castrated cattle than chemically castrated or intact bulls (3.5% vs. 4.5% and 5.1% of TFA, respectively). Regardless of the type of castration, the content of linolenic (about 0.9% vs. 1.8% of TFA), eicosapentaenoic (C20:5n-3; about 0.13% vs. 0.19% of TFA) and docosapentaenoic acids (C22:5n-3; 0.7–1.1% vs. 1.6% of TFA) was higher in IMF of meat from bulls compared to steers. The IMF from the surgically castrated animals contained less PUFA than that from chemically castrated cattle (8.2% vs. 9.7% of TFA), which was less than that from bulls (12.0% of TFA). This was explained by the fact that bulls exhibit a greater musculature development than steers [109], and PUFA is a major component of phospholipids in the cellular membranes of muscle tissues [140]. Steers also had lower SFA contents (49.0% and 52.2% of TFA for chemical and surgical castration, respectively) than bulls (47.3% of TFA), resulting in a smaller PUFA/SFA ratio of surgically castrated cattle than the bulls (0.16 vs. 0.25) and that from chemically castrated cattle being no different from either (0.20). The PUFA/SFA ratio increased as the fat content decreased, given that, at low levels of IMF, the contribution made by phospholipids is proportionately greater, and these are more unsaturated than the triacylglycerols, which themselves increased in proportion as the total lipid increased [141]. The n-6/n-3 ratios were higher in castrated animals (about 3.0) than intact bulls (2.1), but all were low.

Similar trends were observed in a Brazilian study of crossbred zebu x Aberdeen Angus on supplemented pasture [108]. The myristic (1.3% vs. 1.7% of TFA), linoleic (6.6% vs. 4.2% of TFA) and linolenic acids (1.2% vs. 0.6% of TFA) were all higher in bulls than steers. In addition, PUFA were higher in bulls (14.1% vs. 8.0% of TFA), resulting in a higher PUFA/SFA ratio than in steers (0.29 vs. 0.19). And, the n-6/n-3 ratios were lower in bulls than steers (2.4 vs. 3.0). Also observed were lower palmitic acid (23.5% vs. 25.0% of TFA) and MUFA (36% vs. 39% of TFA) in the IMF of bulls than steers.

The cholesterol content of IMF has also been measured in some studies. In a Venezuelan study, feed supplementation in pasture finishing had no impact on the cholesterol content (around 29 mg/100 g muscle tissue) of meat from Criollo Limonero steers slaughtered at 36 months of age after finishing on one of three systems: pasture, pasture supplemented with concentrate or pasture supplemented with legumes [112]. Another Venezuelan study found that cholesterol in meat from Brahman-influenced bulls and steers raised on pasture increased from 54.5 mg/100 g tissue at 19 months of age to 69.04 mg/100 g at 24 months of age [109]. A Brazilian study reported a lower cholesterol content in Nellore (46.6 mg/100 g muscle) and Nellore × Simmental steers (46.9 mg/100 g muscle) than in Nellore × Santa Gertrudis steers (48.3 mg/100 g muscle) [105].

Finally, it must be noted that there are a number of studies aimed at identifying the means to genetically select traits in tropically adapted cattle, including a couple on fatty acid composition. Some workers have suggested that the demonstrated existence of genetic variation in the IMF from feedlot-finished Nellore steers allows the possibility to increase the proportion of healthy and favorable beef fatty acids through selection [142]. Others have found that the selection for decreased subcutaneous fat resulted in decreased proportions of oleic acid with concomitant increases in stearic, myristic and palmitic acids [143]. It was suggested that selection for decreased fat at a given weight will result in a decrease in the proportion of MUFA in the subcutaneous fat in the carcass, with a corresponding increase in the proportions of less healthy SFA.

### *4.3. Minerals*

In a recent review on minerals in meat, particularly from animals raised in tropical regions, Ribeiro, Mourato and Almeida [144] described how concentrations are influenced by a vast array of factors, including species, sex, genotype, production stage, region, climate, tissue characteristics and animal managemen<sup>t</sup> practices, namely rearing systems and nutrition. However, there are few studies on mineral contents of meat from beef raised in tropical environments.

Investigating the use of trace elements as fingerprints for traceability of Brazilian beef, mineral content was analyzed in meat from cattle raised in five different environments, of which three were tropical [145]. While relatively few differences were observed, these differences were sufficient to conclude that chemical traceability of Brazilian beef according to the biome of origin was feasible. Differences in mineral contents of the muscle tissues included the largest mass fractions of bromine and selenium in the Amazônia biome (tropical rainforest) and the lowest mass fraction of zinc in the Pantanal biome (tropical savanna).

In Venezuela, no differences in mineral contents of meat from Criollo Limonero steers fattened on three different feed systems were observed [112,113], nor was any difference found in mineral content with carcass grade of bulls, steers and heifers obtained from a commercial slaughterhouse in Venezuela [146]. In another study on carcass grade of Venezuelan beef, subprimal cuts were obtained from graded carcasses [115,116]. Again, mineral content did not vary with the carcass grade, but did show lower concentrations of Ca, Fe and Zn and higher P and K than meat imported from the US.

Also in Venezuela, no impact of sex on mineral content was observed in meat from zebu-influenced steers and bulls slaughtered at 17, 19 and 24 months [147]. However, an age effect was found for all of the minerals studied. As animal age increased from 17 to 24 months, concentrations of Na (59.5 vs. 71.2 mg/100 g fresh tissue), K (321 vs. 400 mg/100 g fresh tissue), Ca (3.54 vs. 8.17 mg/100 g fresh tissue), Mg (21.8 vs. 23.6 mg/100 g fresh tissue) and Zn (3.66 vs. 3.83 mg/100 g fresh tissue) increased, while those of P (203 vs. 195 mg/100 g fresh tissue), Mn (0.02 vs. 0.01 mg/100 g fresh tissue), Fe (2.34 vs. 1.75 mg/100 g fresh tissue) and Cu (0.16 vs. 0.09 mg/100 g fresh tissue) decreased. When the values at 19 months of age were included, it was noted that variations in mineral contents did not always show an apparent trend. While statistical differences were not identified, Na, P, Mn, Fe and Cu contents were similar at 17 and 19 months of age, and K and Zn were intermediary between 17 and 24 months, but Ca and Mg were the highest at 19 months. It was suggested that an effect of feed (produced by a drought period) could have impacted the findings. The amount and quality of grasses were reduced for the animals slaughtered at 17 and 19 months of age, while the slaughter of the 24-month-old animals occurred under less severe environmental conditions, a phenomenon that could have produced changes in the intramuscular accumulation of some minerals.

Indeed, Ribeiro, Mourato and Almeida [144] describe how the impact of climatic conditions on the soil and pasture is a determinant factor for the mineral concentrations of edible tissues. In tropical countries, pastures suffer a seasonal effect on their quantities and qualities, and fluctuations of feed quality may cause mineral imbalances throughout the year, with dry season being critical due to the lignification of natural pastures and water/feed scarcity. In addition, in the wet season, heavy rains may cause nutrient leaching and, consequently, reduce the mineral contents of forage and the availability to grazing ruminants [144].
