Genetic structure revealed in endangered Red Sanders of India’s Eastern Ghats

In the dry, rocky hill ranges of the Eastern Ghats in southern India grows one of the world’s most coveted trees. Red Sanders, Pterocarpus santalinus, produces a dense heartwood of deep reddish-brown hue and wavy grain that has been prized for centuries as timber, dye and traditional medicine. Its rarity on the international market has made it a target for smugglers, and decades of illicit felling have pushed the species onto the endangered list of the IUCN Red List in 1988 and into Appendix II of CITES in 1995. Now, a team of Indian researchers has delivered the most comprehensive genetic survey of the species to date, mapping the DNA diversity of 22 natural populations across nearly its entire range and identifying the populations that conservationists must save first if the species is to retain its evolutionary potential.

The study, published in the journal Discover Forests, was led by P. Mohana Kumara of the University of Horticultural Sciences in Bagalkot, together with colleagues from the Institute of Wood Science and Technology in Bengaluru and The University of Transdisciplinary Health Sciences and Technology. Working in close cooperation with the Andhra Pradesh Forest Department, the researchers sampled fresh leaf tissue from 361 individual trees distributed across 22 forest ranges in five districts of Andhra Pradesh, spanning latitudes from roughly 13°32’N to 15°29’N and altitudes from 124 to 838 meters. Sampling was deliberately stratified: at each site the team collected from four girth classes, ranging from seedlings and saplings with a girth at breast height of up to 10 centimeters to mature trees exceeding 30 centimeters, allowing them to compare the genetic health of different generations within the same stands.

At the molecular level, the team turned to simple sequence repeat markers, or SSRs, the short, tandemly repeated DNA motifs scattered across the genome. Because SSRs mutate rapidly, are codominant, and produce multiple allelic forms per locus, they are among the most sensitive tools available for detecting fine-scale population structure. The 16 loci used in this study were developed in the team’s earlier genomic work and were selected for extreme polymorphism, with polymorphism information content values exceeding 0.9. DNA was extracted from silica-dried leaves using the CTAB protocol, amplified via PCR with fluorescently labeled M13-tailed primers, and sized by capillary fragment analysis. Genotypes were then screened for null alleles using Micro-Checker, and the researchers applied the excluding-null-alleles correction implemented in FreeNA to confirm that null alleles, which ranged in frequency from about 0.14 to 0.45, did not materially distort their estimates of population differentiation.

The headline numbers paint a picture of a species under genetic strain. Across all 16 loci, the average number of alleles per population was 7.79 and the expected heterozygosity averaged 0.65, with observed heterozygosity substantially lower at 0.34, a hallmark of heterozygote deficiency. These figures are moderate by the standards of tropical hardwoods but fall short of widely distributed relatives such as mahogany (Swietenia macrophylla, He = 0.78), Eucalyptus globulus (He = 0.82) and teak (Tectona grandis, He up to 0.94). This pattern fits a general ecological rule: narrow-range endemics tend to carry less genetic diversity than their widespread congeners, and centuries of logging pressure on Red Sanders have likely accelerated the erosion.

Within that overall picture, however, the variation between populations was striking. The Tirupati base–Sadashiva Kona population (CPV) emerged as a genetic hotspot, with expected heterozygosity of 0.87, the highest allelic richness of any population at 13.44 alleles per locus, and an abundance of rare alleles. At the other extreme, the Chitaleti Pati base camp population showed the lowest diversity, with expected heterozygosity of just 0.44, a deficit the authors attribute to forest fragmentation, encroaching agricultural land and anthropogenic disturbance that have constricted gene flow and likely induced genetic bottlenecks. Aggregated by administrative divisions of the forest department, the Tirupati circle recorded expected heterozygosity of 0.93 and the Chittoor division 0.91, making this southern portion of the range the clear center of the species’ genetic wealth.

Perhaps the most consequential finding concerns how that diversity is partitioned. Analysis of molecular variance revealed that while most variation, around 73 percent, resides within individual populations, the differentiation among populations was nonetheless considerable, with an average Fst of 0.31. Gene flow was correspondingly meager: the mean Nm value of 0.82 falls below the threshold of one migrant per generation below which genetic drift is expected to dominate. Pairwise Fst values ranged from 0.049 between two populations separated by only 34 kilometers to 0.427 between populations roughly 150 kilometers apart. The species’ reproductive biology helps explain the isolation. Red Sanders depends on insect pollination, chiefly by the giant rock bee Apis dorsata, which visits its yellow flowers that open at midnight during the dry season, and fewer than six percent of flowers ever develop into fruit. Wind-dispersed seeds and a limited pollinator foraging range combine with habitat fragmentation to keep populations reproductively disconnected.

Bayesian clustering in STRUCTURE, with the optimal number of clusters determined by the Evanno ΔK method, resolved the 361 individuals into three main genetic groups. Cluster I grouped populations from the Tirupati and Chittoor circles, Cluster II united populations from the Kadapa and Rajampet regions, and Cluster III gathered those from the Guntur and Nellore circles. Unweighted neighbor-joining dendrograms and principal coordinate analysis broadly corroborated this three-cluster architecture. Notably, a Mantel test found no significant correlation between genetic and geographic distance, indicating that differentiation is driven less by physical separation alone than by barriers such as forest fragmentation, soil discontinuities and pollinator limitation. The finding echoes patterns seen in Amazonian trees such as Protium and Inga, where edaphic heterogeneity rather than distance generates genetically distinct populations.

The study also searched for environmental fingerprints on the genome. The team compiled four decades of climatic data from NASA and overlaid the sampling sites on FAO soil maps. Twenty of the 22 populations grow on shallow Lithosols, while single populations occupy Chromic Luvisols and Vertic Cambisols, and genetic diversity parameters trended upward along that soil gradient. More telling were the climatic correlations: the number of alleles per population increased significantly with relative humidity (R² = 0.30) and annual precipitation (R² = 0.43), and expected heterozygosity and private allele counts were likewise positively associated with rainfall. Temperature, altitude and latitude, by contrast, showed no significant relationships with most genetic parameters, suggesting that moisture availability, not heat, shapes the evolutionary potential of these stands.

Private alleles, variants unique to a single population, added another layer of conservation significance. Across the 16 loci the researchers detected 236 private alleles, concentrated in populations such as NAG, TTP, CSM and CPV. When populations were grouped by latitude into three zones, each zone harbored its own suite of latitudinally restricted alleles, with the southernmost group, between 13.0 and 13.9 degrees north, carrying the most, at 142. Such population-specific variants represent distinct evolutionary lineages, and their bearers function as irreplaceable genetic reservoirs. The authors argue that these unique allelic pools could prove invaluable in future tree-breeding and restoration programs, particularly for traits tied to heartwood quality that make the species so commercially valuable.

One reassuring result emerged from the age-class analysis. Comparing the four girth classes, the researchers found no significant differences in allelic richness, heterozygosity or fixation indices, and observed a slight increase in diversity from larger to smaller classes, indicating that recent recruitment has not suffered detectable genetic erosion despite the selective removal of adult trees by loggers. He remains uniformly high at 0.94 across classes, suggesting a shared gene pool across generations. The team credits the sustained enforcement efforts of the Andhra Pradesh Forest Department, which auctions only seized wood, for helping maintain this demographic and genetic continuity, even as surveys show the proportion of harvestable trees over 70 centimeters in girth has plummeted from 7.8 percent to 3.9 percent in just six years.

Taken together, the findings argue for a geographically targeted conservation strategy rather than a uniform one. Protecting the high-diversity populations of the Tirupati circle and Chittoor division, particularly the Sadashiva Kona stand, should be a priority, as should safeguarding populations rich in private alleles in the Nellore and Rajampet divisions to preserve their unique evolutionary legacies. The authors also recommend further investigation of soil chemistry, pH and microbial associations to fully understand the edaphic underpinnings of the observed diversity gradients. For a species whose heartwood commands extraordinary prices on global black markets, and whose reproductive ecology leaves it dangerously isolated in fragmented hill forests, the genetic map now in hand offers something the species has lacked until now: a rational, data-driven blueprint for deciding which forests must be defended most fiercely.