4.1. Animal Performance and Beef Quality Traits
Animal performance and beef quality
per se were not a primary objective of this study, but we needed a preliminary analysis of them to understand and interpret the complex relationships with the mineral profile of beef. Results show that the differences due to the animal’s breed or sex relative to age at slaughter, carcass weight, and carcass gain did not reach significance. It is worth mentioning that the beef breeds of Central Italy have a common ancestry [
24], and their rearing in accordance with European Union PGI specifications for “Vitellone bianco dell’Appennino Centrale” is a traditional operation aimed at high quality production. The majority of the young bulls and heifers are reared on small-medium farms, and both the cow-calf and fattening phases have one of the highest gross margins per cow in the EU [
25]. The performance traits measured in this study are very similar to those reported in a previous large survey of more than 20,000 animals [
14], which also found that Chianina cattle have a greater carcass weight than the Romagnola (430 kg vs. 367 kg, respectively), as well as a greater carcass gain (0.64 kg/day vs. 0.54 kg/day) at similar ages at slaughter.
The results for the chemical and physical traits are in the range of those reported by [
26,
27,
28], except for the
b*, C* and
h* color parameters, which are little higher. This could be due to the fact the animals sampled for those studies came from single farms.
We have clearly shown that the effect of animal group, i.e., animals from the same farm slaughtered on the same day (farm/date effect), was the most important for almost all animal performance and beef quality traits (
Figure 1), accounting for between one to two thirds of their total variance. The only exceptions were for pH, ash and cholesterol contents, and beef shear force (5 to 25% of total variance represented by farm/date). It was not the objective of this survey to analyze in detail the effects of different management and feeding practices on the PGI farms, which would have required us to sample a much larger number of farms, but rather to obtain an overview of the average productive and qualitative traits, and of their variability and relationships with the detailed mineral profile of beef. In a very large survey carried out on another Italian beef breed (Piemontese) reared in north-west Italy in accordance with another set of PGI regulations [
29], the authors were able to disentangle the effect of farm from that of date of slaughter, and found that the latter was often more important than the former. They also noted that the variation between individual farms within a common beef farming system is often more important than the variation between different farming systems [
30]. Our results clearly show that a first level of analysis of the relationships between beef quality and the mineral profile should be the farm/date level.
Our results also show that the variation among individual animals within farm/date group is only slightly less important than the variation between different groups of animals (
Figure 1), confirming the results of [
29]. This reveals the need for a second level of analysis focusing on individual animals/carcasses. The third source of variation (among different samples within animal/carcass) was generally modest, with the exception of a few traits characterized by low reproducibility (pH, ash, and cholesterol contents, and beef shear force).
4.2. Mineral Profile of Beef and Its Relationship with Animal Performance and Beef Quality
Similarly to what we found for beef quality traits, the mineral profile was hardly affected by the animal’s breed and sex, as the only significant differences were among breeds for the content of Ca and B, and between young bulls and heifers for the content of K, Zn, Sn, and Pb [
23]. Very few of the many studies carried out on the mineral content of beef have compared different cattle breeds [
31,
32,
33,
34] or sexes [
35,
36,
37], and most have confirmed the modest effects of these sources of variation. From these results, it seems unnecessary to study the relationships between the mineral profile and beef quality within specific breeds or sexes.
As we observed for animal performance and beef quality traits, in the case of the mineral content of beef we also found considerable variability in the relative importance of farm/date, animal/carcass within farm/date, and beef sample within animal/carcass. In particular, farm/date was the most important source of variation for Na, Mg, P, S, Li, Al, Sn, and Ti; individual animal/carcass was the most important for Mn, B, and Sr; these two sources were equally important for Fe, Cu, Zn, and Pb; and, lastly, beef sample within animal/carcass was the most important for K, Ca, Cr, Ni, and Ba [
23].
The sources of variation in the beef mineral profile, like those in animal performance and beef quality, also confirm the need for at least two levels of analysis with respect to the co-variation between these two groups of traits: The farm/date level and the animal/carcass within farm/date level. The results summarized in
Table 3,
Table 4,
Table 5 and
Table 6 show the 320 correlations (52 of them significant) between the 20 minerals and the 16 animal performance and beef quality traits at the level of farm/date effects, and another 320 correlations (101 significant) at the level of individual animal/carcass within farm/date.
The scientific literature contains some studies that also report obtaining correlation coefficients [
38], often as Pearson correlations, among the raw data without making any distinction between different levels of analysis. While our data are too numerous to facilitate understanding of the interrelationships between mineral contents and beef quality, the few data that can be found in the literature are too heterogeneous and erratic to allow the same objective to be reached.
4.3. Latent Factors of the Beef Mineral Profile and Their Relationships with Animal Performance and Beef Quality
The fact that the content of one mineral in beef is not independent of the content of another, and that there are certain common genetic and physiological mechanisms of absorption, storage, mobilization and excretion, complicate the analysis. In the previous study [
23], we examined the 190 correlations among these 20 minerals at the farm/date level, and the 190 correlations at the animal/carcass within farm/date level, and obtained numerous high correlation coefficients within each group, both positive and negative. The farm/date correlations reflect the effect of different environmental conditions, facilities, management systems, and feeding strategies, whereas the animal correlations reflect the variability among animals in the same external conditions due to their genetics, physiology, and health.
We therefore carried out a multivariate analysis of the mineral dataset, which yielded 5 unmeasured latent explanatory factors, fully independent of each other, that summarized 69% of the co-variation in the minerals [
23]. Multivariate analyses have been used in previous studies, especially for authenticating beef origin or production systems [
39,
40], but we have found no other studies using a factor analysis of the mineral content of beef to investigate correlations with beef quality.
The most important latent factor, “
Quantity”, which related to the concentrations of almost half the minerals analyzed (9 out of 20), all with positive loadings except Sn, and explained 45% of their total co-variance, did not correlate well with the traits studied. Among the animal performance traits, it was positively correlated only with age at slaughter at the level of individual animals within farm/date groups. It is worth pointing out that age at slaughter also has a genetic aspect as it reflects the earliness or lateness of maturation of the animals [
14], and is negatively correlated with carcass weight, carcass gain and fat deposition, which could reflect greater mineral deposition in the beef of animals that require a longer fattening period to reach the level of maturation required by the market.
As we have seen, the only beef characteristic affected by the latent factor “
Quantity” was beef color. The beef samples with the highest scores for this latent factor were often also those with a greater intensity of color (
a*, b*, and
C*), and greater lightness (
Table 5). This cannot be attributed to the effect of myoglobin because Fe is not among the minerals characterizing this latent factor (its loading is −0.03).
Fe is one of the minerals characterizing the second latent factor, “
Na + Fe + Cu”, which explains almost 18% of the total factor variation, and also the fourth latent factor, “
Fe + Mn”, which explains almost 11% of factor variation. As expected, both were positively correlated with the redness index (a*), and negatively with the Hue index (H*) (
Table 5), while the factor “
Fe + Mn” was also negatively correlated with beef lightness (L*). All these effects are, of course, explained by the role Fe plays in the oxidative metabolism of the muscle [
41], and particularly the constituent role it plays in myoglobin, hemoglobin, and cytochromes, which could also explain the positive association of “
Na + Fe + Cu” with both carcass weight and carcass gain (
Table 3). A combined deficiency of Fe and Cu in the diet is known to reduce the growth rate of young cattle [
1], although this type of deficiency is not common. Excessive Fe intake, however, could result in the depletion of Cu in cattle and hence increase the dietary requirement of this mineral [
42,
43]. All these relationships were significant when individual animals within groups were compared, but not when different farm/date groups were compared. At this second level, we only found a high positive correlation (+0.64) between “
Fe + Mn” and the crude protein content of beef (
Table 4). Fe and Mn are known to interact: High levels of Fe have been shown to reduce the activity of the transporters involved in the metabolism of Fe and Mn, and to increase intestinal permeability in calves [
44].
The third latent factor, “
K-B-Pb”, explaining almost 16% of factor variation, was the most correlated with beef chemical composition (
Table 4). It is worth noting that K was negatively correlated with B and Pb at both the farm/date and animal/carcass within farm/date levels [
23], which explains why it has a positive loading (+0.76) in this factor, whereas B and Pb have negative loadings (−0.53 and −0.54, respectively). Moreover, K is the individual mineral most positively correlated with the chemical composition of beef, and B and Pb are the individual minerals most negatively correlated with the composition of beef (lipids excluded). The role of B and Pb in beef is not known, but the highest correlations of “
K-B-Pb” are with beef ash content, which could be explained by the fact that K is the most abundant mineral in beef (
Table 2). Moreover, K in beef is involved in muscle contraction and nerve impulses, and some enzymatic reactions [
1]. The animal’s body has a very limited capacity to store K, so deficiency can develop rapidly, causing, among other things, muscle weakness [
45].
The latent factor “
K-B-Pb” also correlated negatively with L* (
Table 5), and, in particular, with cooking losses (
Table 6). The fact that K is the major cation in intracellular fluids, hence its importance in acid-base balance, osmotic pressure and water balance [
1], could be the reason why it correlates with the loss of liquids during cooking.
Lastly, the fifth latent factor, the only one associated with just one mineral, “
Zn”, explaining just over 10% of total factor variance, also correlated positively with dry matter, protein and lipid content, and negatively with the ash content of beef (
Table 4), but it also seemed to be related to beef color (correlating positively with
a* and C*, negatively with H*,
Table 5). The highest correlations showed by this latent factor were not with beef quality but with animal performance traits, particularly carcass weight and carcass gain, at the farm/date and animal/carcass levels (
Table 3). These many and important correlations are testimony to the importance of Zn in several aspects of the animal’s metabolism: It is an essential component of many important metalloenzymes and also triggers other enzymatic activities that affect the metabolism of carbohydrates, proteins, lipids and nucleic acids [
46]. Zn deficiency, in addition to its well-known effects on the tegumental apparatus (swollen feet, parakeratotic lesions of the skin), also impairs growth, feed intake and feed efficiency [
1].