Genetic Diversity and Structure of Higher-Resin Trees of Pinus oocarpa Schiede in Mexico: Implications for Genetic Improvement

Abstract

Pinus oocarpa Schiede is the most widely distributed conifer in the Americas. In Mexico, it inhabits diverse environments and is the primary pine species utilized for resin production, prompting the establishment of a genetic improvement program (GIP). Preserving a broad genetic diversity is fundamental to the success of the GIP. This study aimed to assess the genetic diversity and structure of trees selected for their high resin yield. A total of 146 trees from 15 provenances within three populations (MX-MIC, MX-MEX, and MX-OAX) constituting the selection population of the GIP were evaluated. Five SSR microsatellite markers (PtTX3013, NZPR1078, PtTX2146, PtTX3107, and PtTX3034) were used to determine key indicators of genetic diversity and structure. All three populations exhibited high genetic diversity; however, the heterozygosity observed was lower than the expected heterozygosity. Genetic structure analysis revealed the presence of two distinct genetic groups: the Transverse Volcanic Axis (MX-MIC and MX-MEX) and the Sierra Madre del Sur (MX-OAX). Most of the genetic diversity (87.42%) was found within provenances. Gene flow was high among provenances within the same genetic group but limited between provenances from different groups. The findings suggest that GIPs should be tailored to each genetic region, with a focus on within-provenance selection to maintain genetic diversity.

1. Introduction

Genetic diversity is a fundamental component for the effective conservation of perennial species, such as forest trees, as it directly influences the adaptability of biological systems and significantly enhances population fitness [1]. This enhancement occurs by mitigating short-term inbreeding depression and facilitating the development of local adaptations in response to long-term environmental changes [2,3]. Furthermore, genetic diversity plays a critical role in various ecological processes, including primary productivity, population recovery from disturbances, interspecific competition, community structure, and energy and nutrient flow dynamics [4].

Anthropogenic activities such as land-use change, alterations in demographic structures, forest fragmentation, species exploitation, the introduction of competitors and exotic pests, and environmental degradation substantially affect the genetic diversity and structure of forest trees [5]. Forest management, driven primarily by human demands, alters forest succession, original composition, and structure, inadvertently impacting the genetic makeup of forest species [6,7]. These impacts are mediated through processes such as genetic drift, selection, gene flow, mutation, reproductive systems, fertility and viability selection, and migration [5,8], with the degree of influence varying according to the forest management system, stand structure, demographic dynamics, and ecological attributes [9].

Silviculture, as a core aspect of forest management, is intrinsically linked to forest genetic improvement [10]. Forest genetic improvement programs involve recurrent cycles of selection, progressing through the base population, selection population, propagation population (e.g., seed orchards), and breeding population [11]. Maintaining broad genetic diversity is essential in these programs; however, as breeding cycles advance, genetic gain typically increases at the expense of genetic diversity [11,12,13]

Within the forest breeding cycle, seed orchards—constituting the propagation population—are designed to produce genetically improved seeds to support operational plantation programs [11,14]. Nevertheless, these orchards often encompass only a subset of the species’ total genetic variation, with a bias favoring genotypes exhibiting commercially desirable traits [4]. In Mexico, information on the genetic diversity and structure of selection populations (superior trees) and seed orchards for forest species included in genetic improvement programs remains scarce. For example, in an asexual seed orchard of Pinus patula Schiede, genetic diversity was found to be moderate, with a heterozygote deficit and low differentiation levels [15]. Conversely, a sexual seed orchard of the same species exhibited high genetic diversity and an excess of heterozygous individuals [16].

Pinus oocarpa is among the most widely distributed conifer species, ranging from northwestern Mexico (28°10′ N) to northern Nicaragua (12°40′ N) [17,18]. In Mexico, it is the leading species for resin production [19,20], with yields surpassing those of other Mexican and resin-producing pine species [21,22,23,24]. Recognizing its economic and ecological importance, genetic improvement efforts for P. oocarpa began in Michoacán, Mexico, in 2010, focusing on enhancing timber production and resin yield through tree selection and genetic trials [23,25,26,27,28]. In 2019, a broader program was initiated to select superior P. oocarpa resin trees and establish sexual seed orchards, incorporating provenances from central and southern Mexico [29,30,31,32].

Evaluating the genetic diversity of superior P. oocarpa resin trees is critical for the continuity of forest genetic improvement programs. Evidence indicates that the selection of superior trees can reduce genetic diversity, potentially leading to the fixation of deleterious alleles and diminished adaptive potential [33]. Intensive selection in conifers has been associated with inbreeding depression and the fixation of deleterious minor alleles, compromising the genetic variability required for adaptation to environmental changes and pathogen resistance [12]. However, early selection in some species has been shown to have minimal effects on genetic variation [34,35], although it may lead to the loss of rare and localized alleles in domesticated populations [35].

The objective of this study was to evaluate the genetic diversity and structure of P. oocarpa trees selected in natural stands for their superior resin production. The working hypothesis posits that the genetic diversity of selected trees is reduced, as selection based on phenotypic traits may decrease genetic diversity relative to natural populations. The implications of the observed genetic diversity and structure are discussed to inform strategies for advancing the genetic improvement cycle of resin-producing trees.

2. Materials and Methods

2.1. Selection of Superior Trees

In 2019, superior Pinus oocarpa trees were selected for resin production across three natural populations in Mexico: San José de Cañas, Michoacán (MX-MIC); San Gabriel Cuentla, Estado de México (MX-MEX); and Santo Domingo Coatlán, Oaxaca (MX-OAX). Each population comprised four to six provenances (Figure 1; Table S1). In the MX-MIC and MX-MEX populations, where resin harvesting was actively conducted, tree selection was performed using the own merit method, based on the expertise of resin collectors who identified the most productive trees in each stand [25]. Conversely, in the MX-OAX population, where resin harvesting was not practiced at the time of selection, the regression or baseline method was applied [30,31].

2.2. Collection of Biological Material

Needle samples from P. oocarpa trees were collected during the active growing season to ensure optimal physiological conditions for DNA analysis. Samples were taken from the middle portion of the crown of each tree to maintain consistency. The needles were cleaned on-site to remove surface residues and stored at 4 °C to minimize enzymatic degradation.

Biological material was transported to the laboratory under controlled refrigeration to preserve DNA integrity. Once in the laboratory, samples were stored at −80 °C to ensure optimal preservation until processing.

2.3. DNA Extraction

Pine needles (300 mg) were homogenized using a mortar and pestle under liquid nitrogen, and the resulting powder was transferred to 2 mL tubes. A pre-warmed CTAB buffer (1 mL; 2% CTAB, 8% NaCl, 20 mM EDTA, 0.1 M Tris-HCl, 4% polyvinylpyrrolidone, 0.2% β-mercaptoethanol) was added, and samples were incubated at 65 °C for 1 h. Cell debris was pelleted by centrifugation at 18,000× g, and 700 µL of the supernatant was transferred to a new tube containing RNAase A at a final concentration of 100 µg/mL. After incubation at 37 °C for 30 min, 1/5 volume of 5 M NaCl and an equal volume of chloroform/isoamyl alcohol (24:1) were added. The mixture was gently inverted and centrifuged at 18,000× g for 20 min.

The aqueous phase was re-extracted with chloroform/isoamyl alcohol, and after a second centrifugation at 18,000× g for 20 min, the supernatant was transferred to a new tube. Cold isopropanol was added in equal volume, and DNA was precipitated at −20 °C overnight. DNA was pelleted by centrifugation at 18,000× g for 30 min, washed with 70% ethanol, air-dried, and resuspended in 50 µL of sterile water [36].

2.4. Microsatellite Amplification and Fragment Analysis

Microsatellite loci (PtTX3013, NZPR1078, PtTX2146, PtTX3107, and PtTX3034) previously reported for P. oocarpa [18] were amplified using primers labeled with specific fluorophores (FAM, HEX, or TAMRA).

Polymerase chain reaction (PCR) amplifications were conducted in 50 µL reaction volumes containing 100 ng of genomic DNA, 1X AllTaq PCR buffer, 0.2 µM fluorophore-labeled forward primer, 0.2 µM reverse primer, 0.2 mM dNTPs, and 1 unit of AllTaq DNA polymerase (Qiagen, Valencia, CA, USA). Reactions were performed on a Techne Touchgene thermal cycler using the following protocol: an initial denaturation at 95 °C for 15 min; three cycles of 30 s at 94 °C, 30 s at 60 °C, and 1 min at 72 °C; three cycles of 30 s at 94 °C, 30 s at 57 °C, and 1 min at 72 °C; followed by 30 cycles of 30 s at 94 °C, 30 s at 55 °C, and 1 min at 72 °C. A final extension was performed at 72 °C for 15 min, with an indefinite hold at 4 °C [18].

PCR products of PtTX3013 (FAM), NZPR1078 (HEX), and PtTX2146 (TAMRA) were combined into a single mixture for simultaneous capillary electrophoresis analysis. Similarly, PtTX3107 (FAM) and PtTX3034 (HEX) products were pooled for a second run. Capillary electrophoresis was performed using an ABI Prism 3130xl Genetic Analyzer (Applied Biosystems, Foster City, CA, USA), and fragment analysis was conducted through an external service (Macrogen Fragment Analysis Service, Republic of Korea) with a 400 HD size standard.

2.5. Data Analysis

Allelic profiles for the five microsatellite loci were used to assess genetic variability within the study populations (Table S1). Key genetic parameters, including the number of alleles (Na), effective number of alleles (Ne), Shannon’s information index (Ii), observed heterozygosity (Ho), expected heterozygosity (He), and fixation index (F), were calculated using GenAlEx 6.503 software [37], commonly used in genetic diversity studies.

Genetic structure was assessed by calculating genetic differentiation (FST) values among subpopulations, performing an analysis of molecular variance (AMOVA), and obtaining F-statistics for all loci. These analyses were conducted using Arlequin 3.5 software [38], a robust tool for population genetic analyses.

Clustering analysis was performed using the UPGMA (Unweighted Pair Group Method with Arithmetic Mean) method based on Nei’s genetic distances among subpopulations. This analysis was conducted in R (version 4.2.2) using the ‘hierfstat’ package (version 0.5-11) [39] in RStudio. To assess the robustness of the clustering, a bootstrap analysis with 1000 replicates was performed using the ‘ape’ and ‘phangorn’ packages, and bootstrap values were added to the nodes of the dendrogram. This approach provides a hierarchical representation of genetic relationships and is widely adopted in genetic diversity studies to illustrate clustering and differentiation patterns.

3. Results

3.1. Genetic Diversity and Allelic Variability

The number of alleles (Na) and the effective number of alleles (Ne) per locus ranged from 4 to 13 and 1.96 to 7.68, respectively, across the loci analyzed in the three populations. The Shannon information index (Ii) varied from 1.03 to 2.35 (

Table 1. Genetic diversity parameters of higher trees from three Pinus oocarpa Schiede populations.

Population	Locus	N	Na ± SE	Ne ± SE	Ii ± SE	Ho ± SE	He ± SE	F ± SE
MX-MIC	PtTX3013	50	12 ± 0.57	3.01 ± 0.77	1.56 ± 0.11	0.50 ± 0.05	0.67 ± 0.04	0.25 ± 0.03
	NZPR1078	50	11 ± 0.33	3.11 ± 0.81	1.60 ± 0.10	0.48 ± 0.02	0.68 ± 0.04	0.29 ± 0.03
	PtTX2146	50	9 ± 1.20	3.36 ± 0.19	1.50 ± 0.04	0.68 ± 0.06	0.70 ± 0.01	0.03 ± 0.07
	PtTX3107	50	5 ± 0.57	2.60 ± 0.29	1.17 ± 0.05	0.58 ± 0.10	0.61 ± 0.05	0.06 ± 0.11
	PtTX3034	50	13 ± 0.33	7.63 ± 1.24	2.25 ± 0.15	0.82 ± 0.07	0.87 ± 0.05	0.06 ± 0.04
	Average		10 ± 1.41	3.94 ± 0.93	1.62 ± 0.17	0.61 ± 0.06	0.71 ± 0.04	0.14 ± 0.05
MX-MEX	PtTX3013	50	11 ± 0.57	4.48 ± 0.77	1.84 ± 0.11	0.58 ± 0.05	0.78 ± 0.04	0.25 ± 0.03
	NZPR1078	50	10 ± 0.33	4.40 ± 0.81	1.79 ± 0.10	0.54 ± 0.02	0.77 ± 0.04	0.30 ± 0.03
	PtTX2146	50	5 ± 1.20	3.43 ± 0.19	1.36 ± 0.04	0.56 ± 0.06	0.71 ± 0.01	0.21 ± 0.07
	PtTX3107	50	4 ± 0.57	2.97 ± 0.29	1.19 ± 0.05	0.58 ± 0.10	0.66 ± 0.05	0.12 ± 0.11
	PtTX3034	50	13 ± 0.33	6.82 ± 1.24	2.17 ± 0.15	0.84 ± 0.07	0.85 ± 0.05	0.02 ± 0.04
	Average		8.6 ± 1.75	4.42 ± 0.66	1.67 ± 0.17	0.62 ± 0.05	0.75 ± 0.03	0.18 ± 0.05
MX-OAX	PtTX3013	46	10 ± 0.57	5.68 ± 0.77	1.92 ± 0.11	0.70 ± 0.05	0.82 ± 0.04	0.16 ± 0.03
	NZPR1078	46	10 ± 0.33	5.91 ± 0.81	1.97 ± 0.10	0.50 ± 0.02	0.83 ± 0.04	0.40 ± 0.03
	PtTX2146	46	6 ± 1.20	3.98 ± 0.19	1.48 ± 0.04	0.78 ± 0.06	0.75 ± 0.01	−0.05 ± 0.07
	PtTX3107	46	6 ± 0.57	1.96 ± 0.29	1.03 ± 0.05	0.28 ± 0.10	0.49 ± 0.05	0.42 ± 0.11
	PtTX3034	46	12 ± 0.33	3.56 ± 1.24	1.75 ± 0.15	0.61 ± 0.07	0.72 ± 0.05	0.15 ± 0.04
	Average	46	8.8 ± 1.2	4.22 ± 0.73	1.63 ± 0.17	0.57 ± 0.09	0.72 ± 0.06	0.22 ± 0.09

N = population size; Na = number of alleles; Ne = number of effective alleles; Ii = information index; Ho = observed heterozygosity; He = expected heterozygosity; F = fixation index.

). Among the populations, MX-MIC exhibited the highest average Na, but lower average Ne and Ii. Conversely, the MX-MEX population had the highest average Ne and Ii, but the lowest average Na. The MX-OAX population showed intermediate averages for all three indicators (Table 1).

Observed heterozygosity (Ho) and expected heterozygosity (He) ranged from 0.28 to 0.84 and 0.49 to 0.87, respectively, across the loci in the three populations (Table 1). The MX-MEX population exhibited the highest average Ho and He, while MX-OAX and MX-MIC presented the lowest averages, respectively (Table 1). The fixation index (F) ranged from −0.05 to 0.42 across loci in the three populations. On average, the MX-OAX population exhibited the highest levels of fixation or heterozygote deficiency, whereas MX-MIC displayed the lowest heterozygote deficiencies (Table 1).

Among loci, PtTX3034 exhibited the greatest variability and diversity in the MX-MEX and MX-MIC populations, whereas PtTX3107 showed the lowest variability and diversity values in the MX-OAX population (Table 1).

3.2. Genetic Structure and Differentiation

The majority of total genetic variability (87.42%) was attributed to differences within provenances, while 10.24% was explained by differences between populations and only 2.34% by differences among provenances within populations (

Table 2. Analysis of molecular variance (AMOVA) for genetic diversity in Pinus oocarpa Schiede populations (Average based on five microsatellite loci).

Source Variation	Sum of Squares	Variance Components	Percentage of Variation	F-Statistics (p-Value)
Between populations	46.84	0.21	10.24	FCT = 0.10238 (0.00000)
Between provenances within populations	32.07	0.05	2.34	FSC = 0.02607 (0.00684)
Within provenances	496.27	1.80	87.42	FST = 0.12578 (0.00000)
Total	575.18	2.06	100.00

). Total genetic differentiation (FST = 0.12578) and genetic differentiation between populations (FCT = 0.10238) were moderate, whereas differentiation among provenances within populations (FSC = 0.02607) was low.

Pairwise FST values between provenances indicated subtle but significant genetic differences, ranging from values indicating no differentiation to moderately high differentiation (FST = −0.020 to 0.279) (

Table 3. Genetic differentiation matrix (FST) between provenance pairs of superior Pinus oocarpa Schiede resin trees in Mexico.

		MX-OAX				MX-MEX					MX-MIC
		El Nanche	Las Tejas	El Tizne	Llano San Pedro	El Salto	Cerro Gachupín	La Laja	Tenería	El Coyul	Joya de Uva	Canalejas	Piedras Paradas	Rancho Nuevo	Cañas	El Banco
MX-OAX	El Nanche	0.000
	Las Tejas	−0.004	0.000
	El Tizne	−0.020	−0.020	0.000
	Llano San Pedro	0.061	0.074	0.078	0.000
MX-MEX	El Salto	0.108	0.092	0.098	0.136	0.000
	Cerro Gachupín	0.205	0.184	0.178	0.250	0.021	0.000
	La Laja	0.098	0.096	0.097	0.150	0.000	0.035	0.000
	Tenería	0.190	0.167	0.200	0.237	−0.001	0.014	0.012	0.000
	El Coyul	0.181	0.138	0.153	0.192	0.019	0.019	0.045	0.016	0.000
MX-MIC	Joya de Uva	0.161	0.117	0.145	0.177	0.011	0.071	0.036	0.042	0.020	0.000
	Canalejas	0.215	0.160	0.197	0.222	0.038	0.096	0.071	0.103	0.036	−0.002	0.000
	Piedras Paradas	0.261	0.212	0.267	0.279	0.064	0.142	0.073	0.161	0.086	0.026	−0.038	0.000
	Rancho Nuevo	0.256	0.208	0.236	0.270	0.045	0.028	0.076	0.027	0.037	0.022	−0.002	0.035	0.000
	Cañas	0.166	0.132	0.160	0.194	0.022	0.083	0.048	0.028	0.029	0.024	0.032	0.061	0.020	0.000
	El Banco	0.188	0.169	0.199	0.220	0.025	0.112	0.032	0.067	0.077	0.037	0.036	0.006	0.036	0.002	0.000

). Genetic differentiation was low among provenances within the same population (FST ≤ 0.078), whereas higher differentiation was observed among provenances from different populations. Specifically, FST values ranged from 0.011 to 0.161 between MX-MEX and MX-MIC provenances, 0.096 to 0.250 between MX-OAX and MX-MEX provenances, and 0.117 to 0.279 between MX-OAX and MX-MIC provenances (Table 3).

The clustering analysis (UPGMA) based on Nei’s genetic distances revealed two main groups of provenances. One group consisted of the MX-OAX population provenances, while the second group included provenances from the MX-MIC and MX-MEX populations (Figure 2). Within the MX-OAX population, the Llano San Pedro provenance exhibited slight differentiation from the rest. In the group comprising MX-MIC and MX-MEX populations, the Tenería and Piedras Paradas provenances showed greater differentiation compared to other provenances (Figure 2).

4. Discussion

4.1. Genetic Diversity

The results revealed extensive allelic variability among the superior Pinus oocarpa trees selected for high resin production. The average number of alleles per locus (Na) across the three populations was higher than previously reported for natural populations throughout the species’ range (Na = 4.09–6.27) [18]. Furthermore, the average Na values exceeded those reported for other Mexican pines with restricted distributions, such as Pinus strobiformis Engelm. (Na = 3.80), Pinus flexilis E. James (Na = 1.75), Pinus ayacahuite Ehrenb. ex Schltdl. var. veitchii (Roezl) Shaw (Na = 4.60), and Pinus chiapensis (Martínez) Andresen (Na = 4.00) [40,41]. Even among widely distributed species, P. pseudostrobus Lindl. (Na = 4.00), P. montezumae Lamb. (Na = 3.40), and P. patula Schiede ex Schltdl. & Cham. (Na = 4.58) showed lower Na values than P. oocarpa in this study [18,42]. Notably, only Pinus radiata D. Don (Na = 8.18), another species with a wide range, presented a value close to the allelic richness observed in P. oocarpa.

The high Na values reported for P. oocarpa in this study may be attributable to the larger sample sizes analyzed per population (46–50 individuals) compared to previous studies (10–37 individuals per population) [18,42]. Another contributing factor could be the favorable conservation status and management of the evaluated populations, which likely preserve high allelic variability. For instance, the Na values determined here are comparable to those of P. patula populations from the Sierra Juárez in Oaxaca, Mexico (Na = 9.5), which are characterized by high conservation levels and effective management [43].

The observed (Ho) and expected heterozygosity (He) values across loci indicated high genetic diversity in the three populations of P. oocarpa. The He values aligned with those reported in 27 populations of P. oocarpa using microsatellite markers (He = 0.437–0.698) [18] and were notably higher than those obtained using enzyme markers in Michoacán populations (Ho = 0.115, He = 0.102) [44]. This difference is consistent with the greater sensitivity of SSR markers in detecting genetic variability [18]. Similarly, the He values obtained near Uruapan, Michoacán (He = 0.666), were higher with SSR markers than with enzymatic markers (He = 0.115) [18,44].

In comparison with other Mexican pines evaluated using SSR markers, the average Ho and He values for P. oocarpa were higher than those reported for P. strobiformis (Ho = 0.246, He = 0.274), P. flexilis (Ho = 0.276, He = 0.287), and P. montezumae (Ho = 0.282, He = 0.294) [41]. However, P. ayacahuite var. veitchii and P. radiata presented higher He values (0.811 and 0.734, respectively) [41]. Among Mexican pines, only P. patula from the Sierra Juárez showed Ho (0.30) and He (0.659) values comparable to those of P. oocarpa [43]. This highlights the relationship between high genetic diversity and effective conservation and management, as inadequate management has been linked to decreased genetic diversity in overexploited populations [45,46,47].

The high genetic diversity observed in the studied P. oocarpa populations may result from a combination of biological, ecological, and geographical factors. The species’ outcrossing reproductive system facilitates genetic recombination, which contributes to maintaining high levels of genetic diversity within populations [48,49]. Moreover, P. oocarpa has one of the widest geographic distributions among Mexican pines, ranging from northwestern Mexico to northern Nicaragua. This extensive range, coupled with its ability to thrive in diverse environments, likely promotes local adaptations driven by varying selection pressures across its distribution [17,18].

Anemophilous pollination also plays a critical role in maintaining genetic diversity by enabling gene flow over long distances. Although the exact pollen dispersal range of P. oocarpa is not documented, it is expected to be significant, potentially reaching up to 41 km, as observed in other Pinus species [50]. This long-distance gene flow can homogenize genetic variation across populations while allowing rare alleles to persist.

Environmental heterogeneity within populations further enhances genetic diversity. In this study, the three populations analyzed included provenances with substantial variation in altitude, mean annual temperature, and precipitation (Table S1). These ecological gradients likely impose distinct selection pressures, fostering genetic differentiation and the maintenance of diverse allelic combinations within populations. Such variation underscores the adaptability of P. oocarpa to diverse climatic and geographical conditions, which is crucial for its resilience under changing environmental scenarios.

Despite the high Ho values observed, they were consistently lower than He across loci, reflecting heterozygote deficiencies and positive fixation index (Fi) values. These patterns suggest deviations from Hardy–Weinberg equilibrium, likely driven by genetic drift, selection, or inbreeding [51,52]. Such deviations have been documented across the species’ range, with allele fixation levels (F = 0.011–0.431) similar to those found in this study [18].

Locus-specific variability also highlighted important patterns. PtTX3034 exhibited the highest Na, He, and Ho values, along with lower F in MX-MEX and MX-MIC populations, reaffirming its role as one of the most variable loci in P. oocarpa, P. patula, and P. tecunumanii [18]. Conversely, PtTX3107 showed consistently lower Na, Ne, and He values, particularly in the MX-OAX population, aligning with previous findings [18]. These loci offer insights into genetic adaptability under selective pressures and environmental changes.

4.2. Genetic Structure

The variance components and fixation indices (F-statistics) obtained in this study revealed that most genetic variability (87.42%) in P. oocarpa is found within provenances, while 10.24% is attributable to differences between populations and only 2.34% to differences among provenances within populations (Table 2). These results are consistent with previous studies reporting that between 86.9% and 91.39% of the genetic variability in P. oocarpa is concentrated within provenances [18,53]. This distribution emphasizes the intrinsic genetic diversity of the species and highlights the need for conservation strategies targeting local provenances to protect this variability.

The genetic differentiation between populations (FCT = 0.10238) and among provenances within populations (FSC = 0.02607) was moderate to low, with a total genetic differentiation of FST = 0.12578. These results align with previous findings for P. oocarpa, which reported similar values of FST = 0.131, FCT = 0.101, and FSC = 0.033 when analyzing the species across its entire natural range [18]. However, genetic differentiation among Central American populations of P. oocarpa was lower (GST = 0.088) [53], suggesting regional differences in genetic structure.

The high percentage of genetic variation within provenances of P. oocarpa is consistent with the distribution of phenotypic trait variation observed in the species. For instance, 64.96%–90.96% of the variation in traits such as seed emergence, resin components and quality, needle and cone morphology, seed size, and wood density is located within provenances [30,31,54,55]. This concordance highlights the genetic complexity within provenances and reinforces their importance as focal points for conservation and management strategies.

The moderate total genetic differentiation (FST) observed in this study indicates that while genetic differentiation exists primarily between populations, provenances still share a common genetic background to some extent, likely due to gene flow between geographically proximate provenances. The FST values between provenance pairs in this study suggest higher gene flow than previously estimated for the species (Nm = 2.9) [18], possibly because the provenances sampled within each population in this study are closer in geographic and environmental terms. Genetic differentiation in P. oocarpa is strongly influenced by geographic, climatic, and ecological factors [56,57]. Differences in precipitation, temperature, and soil type further contribute to the observed genetic divergence among populations [58].

Recent phylogeographic studies of other plant and insect species in Mexico reinforce the importance of the Trans-Mexican Volcanic Belt (TMVB) and the Sierra Madre del Sur (SMS) as both biological corridors and barriers. For example, Peñaloza-Ramírez et al. [59] demonstrated long-term fragmentation and isolation of suitable habitats for Quercus castanea in the TMVB and SMS, with models indicating discontinuities in the species’ distribution between these regions during the Last Glacial Maximum. These findings align with our observations of genetic differentiation in P. oocarpa, where the Río Balsas depression acts as a significant barrier to gene flow between the TMVB and SMS regions. Similar patterns have been observed in Lycianthes moziniana [60] and Dendroctonus mexicanus [61], where the TMVB not only separates genetic lineages but also influences the phylogeographic structure of species distributed across these regions.

The genetic differentiation between populations (FCT) observed in this study supports the presence of significant genetic barriers for P. oocarpa, potentially driven by geography, ecology, or human activities that restrict gene flow. These findings align with the clustering analysis based on Nei’s genetic distances, which corroborated the results of AMOVA and pairwise FST comparisons. Provenances from MX-OAX showed higher differentiation from those in MX-MEX and MX-MIC (FST = 0.096–0.279), while the genetic differentiation between MX-MIC and MX-MEX provenances was lower (FST = 0.011–0.161). This clustering identified two genetic groups: one corresponding to the Eje Volcánico Transversal (MX-MIC and MX-MEX) and another to the Sierra Madre del Sur (MX-OAX) [18].

The distinction between these genetic groups is likely shaped by historical climatic fluctuations and the complex physiographic landscape of Mexico, as proposed by Peñaloza-Ramírez et al. [59]. Their study highlights periods of high dispersal and contact between pine-oak forests during the Pleistocene, followed by upward contractions driven by climate change, which may have contributed to the genetic structure observed in P. oocarpa. Additionally, Mastretta-Yanes et al. [62] emphasized the interaction of climate and volcanism as critical factors influencing genetic structure in Mexican highland species, which could also explain the observed genetic patterns in P. oocarpa. Despite the Río Balsas depression acting as a major barrier, the historical connectivity suggested by overlapping haplotypes in other species (e.g., Q. castanea) between the TMVB and SMS further underscores the need for more quantitative assessments of gene flow in this region.

The differentiation of these two genetic groups is likely a result of limited gene flow between them. While several factors may contribute to this restricted gene flow [56,57,58], the Río Balsas depression represents the most significant barrier. This physiographic barrier spans over 170 linear kilometers, separating the Eje Volcánico Transversal (MX-MIC and MX-MEX) from the Sierra Madre del Sur (MX-OAX) [63]. The tropical forests and distinct environmental conditions within the Río Balsas depression interrupt the continuity of coniferous forests, where P. oocarpa thrives, further limiting genetic exchange between these regions [63].

4.3. Implications for Genetic Improvement

The evaluation of genetic diversity in the selected population of P. oocarpa trees is critical for initiating a successful genetic improvement program. Genetic diversity represents the raw material for any breeding program, as it ensures adaptability and resilience to changing environmental conditions [11,14]. In forest tree breeding, genetic gain typically increases as breeding cycles progress, but this is often accompanied by a reduction in genetic diversity [11,12,13]. Recurrent selection may result in the loss of rare and unique alleles, which are essential for long-term adaptation [35]. Therefore, starting the breeding cycle with a selection population that maintains high genetic diversity is essential [11,14].

In this study, the high genetic diversity observed in P. oocarpa suggests that selection based on phenotypic traits, such as resin production, did not lead to significant genetic erosion. This could be attributed to the large number of individuals selected per population, which helped maintain allelic diversity, albeit with potentially lower genetic gain [15]. Similar patterns have been observed in other tree species, such as Pseudotsuga menziesii (Mirb.) Franco and Picea sitchensis (Bong.) Carr., where genetic diversity in selection, production, and breeding populations exceeded that of the base population [34,35]. Likewise, in sexual and asexual seed orchards of P. patula, genetic diversity values were higher than those in natural populations [15,16].

Silvicultural and forest management practices also play a pivotal role in preserving genetic diversity during the selection process. Studies like those by Gautam et al. [64] and Aravanopoulos [47] emphasize the need for balanced management practices that maintain genetic variability while achieving breeding objectives. This aligns with our findings, as the selection strategies employed in this study aimed to maximize genetic diversity while selecting for high resin production.

In individual selection, increasing the number of selected trees tends to preserve genetic diversity, although this comes at the expense of genetic gain. Genetic gain depends on selection intensity and the heritability of the trait of interest [11,65]. In this study, the selection intensity for resin production, based on the number of trees evaluated, is estimated to be approximately 1.0 [11]. Consequently, while genetic gain in resin production relative to the base population may be modest, the heritability and genetic gain for resin yield in P. oocarpa are expected to be high. For instance, in progeny trials with selection intensities of 1.4 and individual heritability (h² = 0.20), genetic gain in resin production reached 25% [28].

The results also suggest that selection for high resin production did not alter the genetic structure of the populations. The genetic differentiation and distribution of genetic variation observed in this study were consistent with those reported for the entire natural range of P. oocarpa [18]. Given the presence of two distinct genetic groups—Eje Volcánico Transversal and Sierra Madre del Sur—it is recommended to establish separate breeding programs for each region. Within each region, provenance and progeny trials should be conducted to identify the best-performing families for resin production [11,14].

The wide geographic and environmental variability within the genetic regions of P. oocarpa must also be considered. Provenance and progeny trials should account for these gradients to ensure that genetic improvement strategies capture the full spectrum of genetic diversity [11,14]. This is particularly relevant in the Sierra Madre del Sur, where resin composition and yield are influenced by geographic and climatic factors [31]. Similar trends have been observed in other pine species, such as P. halepensis Mill., P. merkusii Jungh. & de Vriese, P. peuce Griseb., and P. pinaster Ait., where altitude and climatic conditions significantly affect resin yield and quality [66,67,68,69].

To accelerate genetic improvement, asexual seed orchards (ASOs) using clones of the best-selected trees are recommended [11,14]. These ASOs should be region-specific, reflecting the genetic structure of P. oocarpa, with distinct orchards for the Eje Volcánico Transversal and Sierra Madre del Sur. Within each genetic region, selection should prioritize within-provenance diversity to preserve genetic variability [70]. Additionally, as the breeding cycle progresses and recurrent selection reduces genetic diversity, genetic infusions from the other genetic region should be incorporated into the breeding population [11]. For example, genotypes from Piedras Paradas and Rancho Nuevo provenances could be introduced into the Sierra Madre del Sur breeding population, while San Pedro and El Nanche genotypes could enhance the Eje Volcánico Transversal program. These infusions would help maintain genetic diversity and increase the frequency of favorable alleles for resin production.

Recent advancements in tree-breeding methodologies, such as those discussed by Degen and Müller [71], demonstrate the potential of genomic selection (GS) to optimize breeding outcomes. Integrating genomic tools like those applied in Tilia amurensis [72] and Pinus radiata [73] could complement traditional approaches, enhancing selection precision while maintaining genetic diversity. As shown by Li et al. [73], correcting documented pedigrees with genomic data ensures accurate trait selection, which is critical for traits like resin yield and drought resistance.

While ASOs provide a rapid path to genetic improvement, technical challenges in cloning and grafting Mexican pines remain [74,75,76]. As an alternative, sexual seed orchards (SSOs) could be established. These SSOs would initially serve as progeny trials, allowing for genetic evaluation based on resin production, before undergoing genetic thinning to retain the best-performing genotypes. To maximize genetic diversity, SSOs should integrate as many families as possible from both genetic regions [11,14]. During genetic thinning, selection within progenies should take precedence over selection between provenances or individuals to preserve within-population diversity [70].

Finally, genetic improvement programs for resin production must integrate strategies that balance genetic gain with the conservation of genetic diversity. The long-term viability and ecological functionality of P. oocarpa populations depend on preserving their genetic adaptability. By minimizing inbreeding, maximizing genetic diversity, and maintaining a broad adaptive potential, breeding strategies can ensure the sustainable use and conservation of this economically and ecologically important species [35,77,78,79].

5. Conclusions

The genetic diversity of P. oocarpa trees selected for high resin production is substantial, with the majority of this diversity concentrated within provenances. Selection based on the phenotypic trait of high resin production does not significantly reduce genetic diversity nor alter the genetic structure relative to the base population. This diversity within the selection population provides a robust foundation for designing strategies that balance increased genetic gain with the conservation of genetic variability throughout the breeding cycle.

The selection population comprises two distinct genetic groups associated with geographic regions: the Eje Volcánico Transversal and the Sierra Madre del Sur. To preserve genetic diversity, breeding programs should be tailored to each genetic region, with a focus on within-provenance selection. This approach ensures the retention of local adaptations while optimizing resin production traits.

Understanding the genetic diversity and structure of the selection population is essential for informed decision making in the establishment of production and reproductive populations. This knowledge guides the design of breeding strategies that integrate both genetic improvement and conservation, ensuring the long-term sustainability and adaptability of P. oocarpa populations.