Currently, skeletal and bone tissue assessment is performed by diagnostic imaging such as radiography, computed tomography, magnetic resonance and densitometry techniques, or invasive methods, such as bone biopsy, although these methods are still limited in many situations. Therefore, determination of the serum levels or urinary bone turnover biomarkers (BTMs) could be a good noninvasive analytical method for evaluation of bone metabolism, as it allows predicting bone cell activity in a quick and real time manner [1].

At present, some studies have been performed on BTMs in animal models for orthopedic research not only to improve the knowledge in animal and veterinary sciences, but also in human orthopedic research of metabolic bone diseases, such as post-menopausal osteoporosis. BTMs could be useful in diagnosing early-stage bone diseases, and helping to monitor evolution and efficiency of treatment.

BTMs are generally divided in two groups: the formation and the bone resorption markers, although there is a third group that is still poorly studied, the osteoclast regulatory protein [2]. During the metabolic process of bone formation by osteoblasts, formation markers are represented by serum alkaline phosphatase (ALP) and its bone-specific isoform (BALP), serum osteocalcin, and two other molecules that are released during synthesis of the type I collagen molecule – serum procollagen type I carboxy- and amino-terminal propeptides [3].

The ALP is a glycoprotein that is connected to the surface of cells. In humans ALP are expressed in four gene loci code: nonspecific, intestinal, placental and germ cells [4]. Nonspecific gene is synthesized in a variety of tissues (bone, kidney, liver and early placenta) [4]. BALP has been used due to its high sensitivity as a bone formation marker [4]. It is produced by osteoblasts [5] and is involved in the calcification of bone matrix [6].

In the bone resorption process, there is a breakdown of type I collagen, so resorption markers are represented by serum C-terminal telopeptide of type I collagen, urinary collagen type I cross-linked C- and N-telopeptide, urinary hydroxyproline, total and free urinary pyridinoline and deoxypyridinoline, as well as by the serum tartrate-resistant acid phosphatase (TRAP), which is produced by active osteoclasts [7].

TRAP is a glycoprotein produced by mature osteoclasts, activated dendritic cells, and macrophages, therefore TRAP is an indicator of osteoclast and macrophage activity [8, 9]. There are two known isoforms (TRAP5a,b). TRAP5b is a specific biomarker of osteoclastic resorption activity, while TRAP5a is a non-osteoclastic form [10]. In the future, the latter may be useful in clinical evaluation since it is expressed in bone pathologies, such as fracture healing in different mammals or osteoporosis in women [9, 11, 12].

The use of sheep as animal models in orthopedic research, also in studies focusing on BTMs, is based on different aspects such as easy handling [13, 14] and possibility of blood sample collection several times per day [15] due to a high blood volume when compared to other laboratory animal species. Nevertheless, bone markers in small ruminants are influenced by circadian and seasonal variation [16], among other possible factors causing serum and urinary variability such as diet, exercise, skeletal growth, as well as intra-individual variability [4]. As such, BTMs undergo variations throughout a lifetime, namely with age and different metabolic states. It is therefore important to assess BTMs variation during these phases and establish a reference range for BTMs in this species, to which the greatest limitation is the inter-individual variability [17].

Therefore, the aim of this study was to generate a reference range for some of the main BTMs – ALP and TRAP, throughout the lifespan and different physiological states in sheep and to evaluate possible correlations of these parameters with serum minerals.

The aim of this study was to assess the behavior of serum ALP, TRAP, and minerals in different ages and physiologic stages of the ewes life, therefore providing a dataset and information to assist in research studies focusing on BTMs in sheep.

In this study, the median ALP value was within the normal range reported for this species (68–387 U/L) [18] and the variations over the lifespan in sheep can explained by physiological changes. The interval range was wide, however, possibly due to such factors as seasonal influence, with decreased formation marker (BALP) during Autumn [16], or circadian influence, in which there is a variation of bone markers throughout the day. Because of the latter, blood samples sampling must be drawn at a standardized time [15]. Age was another influencing factor, with a higher level of ALP found in 1-month old animals, possibly due to synthesis in a variety of tissues, not only due to the bone isoenzyme [4], and high juvenile metabolism [19]. Lower levels of ALP were found in sheep over 8 years of age, possibly due to the influence of age in decreasing bone formation. In another study, Haversian remodelling in the caudal aspect of the femur, diaphysis of the radius and humerus was observed in sheep between 7 and 9 years of age [20], which justifies a link between advanced age and bone loss.

Previously reported values for P (1.62–3.36 mmol/L) and Mg (0.9–1.31 mmol/L) [18] were within the interval found in the present study, although Ca levels (2.88–3.2 mmol/L) [18] were higher in the aforementioned paper. This Ca value could be caused by a decrease in this mineral in pasture, as previously described [21]. In the referred works the lowest value of Ca was found in different samples from different grazing areas during Winter. Blood samples in the present study were drawn in the same season, however, pasture analyses were not performed which renders impossible to investigate a correlation.

In general, for serum minerals, there was a significant difference between the 1 month old group and the other groups, possibly due to the high bone modelling process that is occurring at this age. In fact, as the skeletal structure in these animals is still in growth, there is an increasing demand for Ca and P [22]. A statistical difference in serum minerals between 1-month-old and 6-months-old groups was not expected since both groups have growing animals. There was, however, a statistical difference in Ca and P between the two groups, possibly due to a change in feed, with the absence of milk and a diet based on grass and concentrate at 6 months old. The decrease in Ca in animals over 8 years of age when compared to the other groups may be associated with bone remodelling in older sheep, as has been reported [20], where the decline in Ca would thus be expected due to a decrease in bone mineralization.

During the analysis of the TRAP marker, the 1-month old animals had a higher range interval, probably related to an accelerated resorption during skeletal growth [9]. However, 6-months-old animals had a low level of this marker even though the sheep were undergoing skeletal growth, possibly because of a difference in diet, as previously mentioned. This marker showed an inversely proportional relationship with age, thus the animals with 6 to 8 years of age had lower values among all the groups. However, those over 8 years old had an increased TRAP, possibly due to osteoclastic activity [9] in geriatric animals. An explanation for this increase has not been investigated in this study in older sheep. Nevertheless, in older women, it is associated to osteoporosis [23]. Therefore, there may be an osteopenia in old animals related to an increase of this marker, but analyses of the bone, such as bone densitometry, would be necessary to prove that.

It should also be mentioned that the ideal TRAP markers to assess bone metabolism should determine the TRAP5b isoform by itself, since it has an osteoclastic origin, resulting from resorption activity, whereas the TRAP5a isoform is nonosteoclastic [24]. The occurrence of seven outliers (three in 6 months old, one in the 2 years and three in 8 years old group) in the determination of TRAP could be justified in this study by the marker being expressed in different tissues, such as muscles and heart, or pathological conditions. There are published studies in humans that correlated TRAP with leukemia and AIDS [25]. However, the present study used healthy animals and obtained normal values for the ovine species. In the present study, the values found are within the reported range of minimum value in adults and maximum values in juveniles (0,14–5,9 U/L) [11]. It is thus not possible to state which tissue was responsible for the occurrence of outliers and whether there was a pathological cause for these outliers. All TRAP variations can be explained by physiological or pathological changes as previously described.

This study presented a reasonable degree of correlation between P and Ca, previously described in sheep as being involved in bone mineralization [26]. In rats, the increase of these serum minerals were essential for mineralization of bone tissue developed in vitro [27]. The degree of correlation present between ALP and Ca is possibly caused by the role of Ca in enzymatic reactions involving ALP, a correlation which has been previously described [3, 12].

With regard to the three groups of 3- to 5-year old sheep in different physiologic stages, only the pregnant group showed values that were considerably lower than the dry and lactation groups for ALP analyses. A similar increase of ALP throughout pregnancy has been previously described [28]. However, Liesegang et al. (2006) [29] used the BALP isoform and obtained a decrease of the marker during gestation. Therefore, when the animals become pregnant, there is a significant reduction of ALP, which then increases between gestation and lactation, and is slightly decreased in dry sheep [28]. In this study, the same marker variation was observed among the three described physiologic states, as has been previously described. However, during the analysis of TRAP, the lactating group showed lower values when compared to the dry ones. This could be due to the abrupt drop in bone resorption during the 2 to 3 months of lactation, with values going back to basal values in dry animals. Another greatly important factor would be that the maximum peak of TRAP would occur closer to parturition, as described [28].

This study provided information on the variation of two of the most used BTMs – ALP and TRAP, throughout the different stages of life and metabolic stages in sheep. These bone turnover markers variations can be entirely explained by known physiology, confirming that unexplained changes do not occur during the lifespan of sheep. This information may be a useful tool in clinical orthopedic research studies both in animal and veterinary sciences, as well as when using sheep as an animal model for translational studies for humans.