1. Introduction
The Murgese belongs to the group of baroque-type horse breeds and is additionally characterized by a black or blue roan coat (
Figure 1). Originally being bred as a working horse, suitable for both military and civilian use, a shift in breeding objectives in the 1990s resulted in an emphasis on performing traits [
1,
2]. The Murgese is appreciated today in equestrian tourism, with increasing interest also in horse riding sports, mainly dressage. The history of the Murgese horse (MH) is strictly connected with that of the arid and rocky hills of Apulia, Le Murge, from which it takes its name. The breed has a rustic nature [
3], necessary for survival in such a difficult environment characterized by harsh climate (cold in the winter, hot and dry in the summer), poor pastures and the presence of enzootic pathogens [
4,
5].
At the dawn of the twentieth century, the horse in Italy was reduced to a mere source of food and, even worse, almost all the national equestrian resources were lost in World War I. Due to the geographical isolation and the production of very valuable mules derived by crossing Murgese mares with Martina Franca jacks, the Apulian horse was not involved in cross-breeding programs which affected most of the native Italian breeds.
During the Fascist regime, the substantial failure of the hippo-technical measures taken by previous governments came to light and a package of legislative actions to encourage equine breeding was set out. In 1925, the Murgese, a mesomorphic horse with wide transverse diameters, was officially distinct from the rest of the equine population present in Apulia. However, in 1926, when the official denomination of the breed, as well as the first individuals’ registration in the Murgese Stud Book, were established, the MH population counted only 46 founder brood-mares and 9 stallions [
3]. Since the 1970s, the revival of equestrian leisure culture has led to a long-lasting interest in the MH, whose population has progressively increased. In 1990, the Italian Ministry of Agriculture and Forestry established the “Anagraphic Register of the Equine Populations identifiable as Ethnic Groups” [
1], among which the MH was included. In 2008, the Murgese Stud-book was established, and it was entrusted to the Italian breeders association (Associazione Italiana Allevatori, A.I.A.). In 2019, it was transferred from A.I.A. to the Italian national association of Haflinger breeders (Associazione Nazionale Allevatori Cavallo Razza Haflinger Italia, A.N.A.C.R.HA.I.) and, in 2022, from A.N.A.C.R.HA.I. to the “National association of the Murgese horse and Martina Franca ass breeders” (A.N.A.M.F.). Over time a number of additional bodies have been involved in the MH conservation policy, among which the regional “Service for the Valorization of Natural Resources and Biodiversity” (evolved from the Apulia regional “Institute for the Improvement of Horse Populations). The MH population has now recovered to 6437 breeding animals [
6].
A precondition for decisions in livestock selection and breeding programs is the robust knowledge of breed attributes, including demography and genetic variability [
7]. Particularly pedigree analysis enables the estimation of genetic variability and its evolution over time using information from genealogical records [
8]. Moreover, demographic analyses may help to understand the factors that have affected the genetic history of a population [
9]. In the last years, a number of studies have been carried out on the population structure and genetic variability of horse breeds by analyzing pedigree data [
10,
11,
12,
13,
14,
15,
16,
17,
18,
19,
20,
21,
22,
23,
24,
25,
26]. Classically, monitoring of genetic diversity via the use of genealogical information has been carried out by assessing inbreeding (F) and relationships (R) in the population of interest, converted into effective population size (N
e), which is regarded as a good indicator to estimate genetic diversity for conservation purposes.
The main objective of this study was, therefore, a complete evaluation of the demographic and genetic status of the MH breed, combining pedigree information with data about the temporal evolution of basic descriptive morphometric traits. The knowledge generated by our research will be beneficial for orienting future genetic management and breeding strategies of the MH.
4. Discussion
Depth and completeness of pedigrees are of fundamental importance for accurate genetic diversity inferences, as incomplete pedigree information can bias, generally underestimating, inbreeding and relationship coefficients. In the MH breed, the analysis of the pedigree completeness parameters suggests that, while a relatively high number of maximum generations are traced on average (10.09), thus ensuring a good pedigree depth, each animal can count, on average, on a number of equivalent generations that is almost half the average number of maximum generations (5.88); this implies the presence of a certain number of unknown ancestries in the MH pedigree. As predicted, the lack of information was mainly concentrated in the earlier generations, with pedigree completeness falling below 50% in the seventh generation, while being higher than 95% in the first generation. As such, the MH pedigree, while being poorly able to capture ancient inbreeding phenomena, can still accurately detect recent inbreeding. The latter is expected to exhibit larger unfavorable effects than ancient inbreeding [
39]; thus, the MH pedigree can be considered a valuable tool for orienting genetic management practices and optimizing conservation strategies [
21]. Not surprisingly, percentages of pedigree completeness were higher, at any given generation, in the reference population compared to the total population.
In other horse populations, such as the Slovak Sport Pony [
15], the two endangered Spanish horse breeds, Asturcón and Mallorquí [
11], the recently revived, yet still “vulnerable” Croatian Posavina Horse [
22], the Brazilian Creole horse [
19] and the Trakehner [
40], pedigrees were not as deep-rooted/complete as in the Murgese horse. Notably, the latter represent an ancient breed, with a partially closed population since the year 1732 and a post-Second World War recovery using the few rescued mares and stallions. Similar pedigree depth/completeness was reported for the Lusitano horse [
16]; despite the long tradition of this breed in Portugal for several centuries, when the stud-book was established in 1967, the number of remaining animals was rather small, so, much alike the MH in Italy, the current population is considered to have derived from a reduced number of founder animals. Furthermore, the Maremmano horse breed from Italy [
21] was reported to have values for complete and equivalent complete generations very close to those observed in the Murgese. On the other side, for several horse breeds, most of them having a long breeding tradition, such as the Arab [
20,
41], the Pura Raza Español [
10], the Lipizzan [
9], the Slovack Hucul [
15], the American Shire [
17], the Noriker [
14] and the Hanoverian warmblood horses [
13], deeper pedigrees were reported.
The trends in demography shown in this study help to understand important circumstances affecting the genetic history of MH. In fact, due to the low initial numbers of registered animals, the considered time frame has been subdivided into five periods, each covering two decades, with the exception of the period 2011–2020 (
Figure 2). Particularly, the first period, including animals born from 1930 to 1950, gives an account of the bottleneck experienced by the MH population during the first half of the 20th century. As a consequence, more than 60% of the available genealogical records belong to animals born in the last 15 years, supporting the fact that the current population experienced a population expansion in the last decades, as clearly depicted by the time series of registered males and females per year. The similarity between the two sexes in the observed demographic trends may suggest that these were driven by general contingent socio-economical phenomena affecting the entire population (renovated international interest toward the horse species, as well as a revived interest in this specific breed for its good attitude in equestrian sports, mainly dressage), rather than by sex-focused selection practices. This scenario of recent demographic expansion is reported for other horse breeds that had formerly experienced, mainly in Europe, severe population size contractions [
41] and followed the present position of the European horse sector, which is growing again thanks to the green assets of equines in the European context of the ecological transition of agriculture [
42].
A close-to-unity sex ratio is theoretically possible in the reference population, although the numbers of dams and sires used to generate this population seem to suggest a scenario where, on average, one stallion would serve about five females. This figure is not dissimilar to the total population, likely suggesting that no significant changes have occurred over time for this parameter. However, the effective number of founders in the total Murgese population was about 1/7th of the total number of founders, thus supporting their unbalanced use, confirmed by the distribution of male animals in the total population per class of offspring numbers (
Figure 3). When comparing the distribution of male animals per class of offspring numbers in the total population and in the reference population, a higher number of males with no progeny was observed in the latter. This could possibly account for the increased use of the MH in horse riding sports.
Our generation interval (GI) values were longer than the 8.20 years found by Ivanković et al. [
22] in the Croatian Posavina, but similar to the 10.1 years reported for the Lusitano horse by Cervantes et al. in 2009 [
43] and confirmed by da Silva Faria et al. in 2018 [
44]. GIs are higher for dams than sires. This reflects the use of mares for longer periods in the herds than the sires. Murgese horses can be expected to live a long natural lifespan of 25 to 30 years, similar to the general horse population, particularly those partly bred in the woodland. However, in general, there is a large variation in the age of horses because some are killed at a young age owing to illness, breeding problems or injury, whereas some survive until senescence. The age class suffering most from premature mortality is 0 to 1 year of age [
45] for both sexes, but because of breeding policies in domestic horses, few stallions are graded for the breeding pool and most of the stallions born are castrated before the age of 3. The fact that 74% of the MH male population (88% in the MH male reference population) have no progeny is a confirmation of the above statement.
The effective number of founders in the MH was 36, while the effective number of ancestors was 19. The rate of the effective number of founders compared to the effective number of ancestors can be used to determine the bottleneck in the population [
29], and if the ratio is 1, the population is stable. Higher or lower values of the
fe/fa ratio always indicate an imbalanced use of sires, which poses a risk to the original genetic diversity. When a population suffers a demographic decrease, such as that experienced by the MH at the beginning of the last century,
fe is overestimated by ignoring some genetic bottleneck effects, but the ratio (
fe/
fa) allows us to consider the loss of genetic variation promoted by the overuse of few sires. The bottleneck ratio (
fe/
fa) in the MH population was 1.89, which is similar to that observed in the Czech-Moravian Belgian horse (1.86), more favorable compared to the Austrian Noriker (4.00) [
14] and American Shire (3.65) [
17], but lower than that found in the Slovak and Hungarian Hucul horse and the Croatian Posavina, that were 1.6, 1.41 and 1.29, respectively [
15,
22,
37]. The value of the
fa/
fe ratio in
Table 1 indicates a 50% reduction of the genetic diversity as expected by the fact that for the MH current population, the effective founders are 34 and the effective ancestors only 17. In addition, as far as the ancestor contributions to the current population, 50% of the genetic diversity could be accounted for by only six individuals. Basically, these statistics reflect the substantial loss of genetic diversity within the breed over the generations since its official founding.
The effective population size (N
e) is a key parameter in conservation and population genetics because it has a direct relationship with the level of inbreeding, fitness and the amount of genetic variation loss due to random genetic drift [
24]. In the MH population, the N
e obtained from the Wright equation was 1118 (data not shown), while that from the increase in inbreeding was 47.48. The estimates of N
e based on the individual increase in inbreeding more accurately reflect the genetic history of the populations, namely the size of their founder population, their mating policy or bottlenecks due to abusive use of reproductive individuals for the period in which the genealogies are known. Thus, figures close to 50 appear more acceptable as estimates of the effective population size for the MH, suggesting that the breed has experienced a significant loss of genetic diversity.
Likely, as a consequence of the narrow breed gene pool, the average relatedness was not negligible; this possibly points to the complexity in mating among unrelated animals (as attested by the mean inbreeding value in the MH, possibly underestimated due to the presence in the reference population of some individuals with unknown parents), although the limited mating between known close relatives seems to suggest certain care in mating planning. Notwithstanding, the estimated increase in inbreeding per generation was roughly around 1%, a value corresponding to the critical threshold generally accepted as the safe limit to assure population survival in the long run.
Effective and resilient population genetic management should be regarded as a balancing act between conservation and genetic improvement. The analysis of the basic morphometric descriptors did not highlight major conformational changes during the last 30 years. This would indirectly suggest that no high selection pressure has been applied in the considered temporal interval to conformational traits in the MH. In view of the growing request for boosting genetic improvement of the MH morpho-functional traits, the implementation of precision genetic breeding practices looks essential to assure a long-term potential for genetic gain and survival.