The giant panda (Ailuropoda melanoleuca) and red panda (Ailurus fulgens) independently evolved an obligate bamboo diet and a 'pseudothumb' — an enlarged radial sesamoid wrist bone that serves as a sixth digit for grasping bamboo. Comparative genomics (Hu et al., PNAS 2017) identified 70 adaptively convergent genes, proposing that convergent substitutions in the limb-development/ciliogenesis genes DYNC2H1 (R3128K, K3999R) and PCNT (S2327P, Q2458R) may drive pseudothumb development via altered ciliogenesis and Sonic Hedgehog/GLI3 signaling, and that convergent substitutions in digestive and metabolic enzymes (PRSS1, PRSS36, CPB1, GIF, ADH1C, CYP3A5, CYP4F2) may compensate for bamboo's poor nutrient profile; of the DYNC2H1/PCNT pseudothumb proposal the paper stated that it 'warrants further experimental validation in the future', and none of the proposed mechanisms was experimentally tested. Morphological, paleontological, transcriptomic, epigenomic, and population-genomic follow-ups corroborate the phenotype and expression-level convergence but leave the causal-mechanism core open: whether the identified substitutions alter protein function, ciliogenesis, skeletal development, or enzyme kinetics, and whether they causally contribute to the convergent phenotypes, remains entirely untested.

Background

The giant panda (Ailuropoda melanoleuca, family Ursidae, the bears) and the red panda (Ailurus fulgens, family Ailuridae) are distant relatives within the mammalian order Carnivora that independently abandoned carnivory for an almost exclusive bamboo diet. Both species also independently evolved a 'pseudothumb': a greatly enlarged radial sesamoid — a wrist bone — that acts as an opposable sixth digit for manipulating bamboo stems. Because the two species last shared a common ancestor roughly 40 million years ago, this pair is a textbook example of convergent evolution, the independent evolution of similar traits in separate lineages. The genetic basis of the pseudothumb and of the shared bamboo-diet physiology was unknown until the comparative-genomic study at the center of this problem (Hu et al. 2017), which sequenced a de novo red panda genome and an improved giant panda genome and searched for convergent amino-acid substitutions — positions in a protein where both panda lineages independently acquired the same or similar amino-acid state relative to ancestral carnivores. Applying conservative filters (a nonrandom-convergence test under two substitution models, removal of genes with gene-tree discordance, and intersection with statistical signals of positive selection), the study identified 70 'adaptively convergent' genes and highlighted two functional groups. First, limb development: DYNC2H1 encodes cytoplasmic dynein-2 heavy chain 1, the motor protein driving retrograde intraflagellar transport (IFT) — the transport of material from the tip to the base of the primary cilium, a tiny antenna-like sensory organelle present on most vertebrate cells. Primary cilia are required for Sonic Hedgehog (SHH) signaling, a key developmental pathway; defective cilia impair processing of GLI3, a major SHH downstream transcription factor, and GLI3 dysfunction produces extra digits (polydactyly) in mice and humans. Human and mouse mutations in DYNC2H1 or in PCNT (pericentrin, a centrosomal coiled-coil protein that anchors IFT proteins at the base of the cilium and is required for primary-cilia assembly) cause skeletal ciliopathies (disorders arising from defective primary cilia) with abnormal skeletogenesis and polydactyly. Hu et al. found two convergent substitutions in DYNC2H1 (R3128K, in a stalk domain between AAA domains 4 and 5 — the dynein motor's ATPase domains — a region implicated in microtubule binding, and K3999R) and two in PCNT (S2327P, Q2458R) that are otherwise absent from polar bear, dog, human, and mouse variation data, and proposed that these substitutions 'may work synergistically for pseudothumb development in both pandas'. Second, bamboo nutrient utilization: convergent substitutions in the digestive proteases PRSS1 (D119N, I140V), PRSS36 (R57S), and CPB1 (V218F, T271I, M310L) — proposed to improve release of the essential amino acids lysine and arginine, which are scarce in bamboo — and in GIF (gastric intrinsic factor, required for vitamin B12 absorption; M212L, D406K, H407D), ADH1C and CYP3A5 (vitamin A/retinol metabolism; V187I and T363S/V393L/M395I/T400S respectively), and CYP4F2 (arachidonic-acid metabolism; M92F). The same study also showed that TAS1R1, a component of the umami (savory) taste receptor, was independently pseudogenized (converted into a nonfunctional pseudogene) in both pandas, with relaxed functional constraint (weakened purifying selection) in the red panda dated to roughly 1.58 million years ago. The study was explicit that its central inferences were hypotheses: the ciliopathy-based argument for pseudothumb development 'warrants further experimental validation in the future', the enzyme effects were framed as possibilities ('may', 'seems'), and the authors acknowledged that gene-tree discordance and model biases could still affect inference of adaptive convergence.

Problem Statement

Determine whether the convergent amino-acid substitutions identified between the giant and red pandas alter protein function and causally produce their convergent adaptations — the question Hu et al. 2017 left open. The core questions are jointly linked by one scientific objective, a causal genotype-to-phenotype account of the pandas' convergent bamboo-adaptation phenotype: (1) pseudothumb mechanism — test whether the DYNC2H1 substitutions R3128K and K3999R and the PCNT substitutions S2327P and Q2458R alter at least one of: dynein-2 retrograde intraflagellar transport, primary-cilia assembly, or SHH/GLI3 signaling output, and whether they causally contribute to enlargement of the radial sesamoid (pseudothumb); the source noted as a caveat that the substitutions' effects on other aspects of skeletal development remain elusive; (2) nutrient-utilization mechanism — test whether the convergent substitutions in the digestive proteases (PRSS1 D119N/I140V, PRSS36 R57S, CPB1 V218F/T271I/M310L) and in GIF (M212L/D406K/H407D), ADH1C (V187I), CYP3A5 (T363S/V393L/M395I/T400S), and CYP4F2 (M92F) quantitatively change at least one of: substrate specificity, catalytic efficiency, or nutrient absorption in the direction predicted (improved lysine/arginine release, vitamin B12 and vitamin A utilization) as adaptations to bamboo's nutrient profile. The source proposed both mechanisms as hypotheses — the pseudothumb inference with an explicit call for further experimental validation in the future, the enzyme effects as hedged possibilities ('may', 'seems') — and experimentally tested neither.

References

  1. Hu Y, Wu Q, Ma S, Ma T, Shan L, Wang X, Nie Y, Ning Z, Yan L, Xiu Y, Wei F (2017). Comparative genomics reveals convergent evolution between the bamboo-eating giant and red pandas. Proceedings of the National Academy of Sciences of the United States of America 114(5):1081–1086. doi:10.1073/pnas.1613870114. https://doi.org/10.1073/pnas.1613870114
  2. Wang X, Su DF, Jablonski NG, Ji X, Kelley J, Flynn LJ, Deng T (2022). Earliest giant panda false thumb suggests conflicting demands for locomotion and feeding. Scientific Reports 12:10538. doi:10.1038/s41598-022-13402-y. https://doi.org/10.1038/s41598-022-13402-y
  3. Barnett B, Lyu Y, Pichney K, Sun B, Wu J (2023). Mechanical Evidence for the Phylogenetic Origin of the Red Panda's False Thumb as an Adaptation to Arboreal Locomotion. arXiv:2305.05086. https://arxiv.org/abs/2305.05086
  4. Wang D, Gao Z, Bottazzi J, Shao Q, Li Y, Wu K, Zhou W, Jiao F, Li S, Jiangzuo Q (2021). Significance of the preservation of 'pseudo-thumb' in fossil skeletons of giant panda (Ailuropoda melanoleuca) in Shuanghe Cave, Guizhou Province, southern China. Historical Biology 34(11):2188–2194. doi:10.1080/08912963.2021.2006195. https://doi.org/10.1080/08912963.2021.2006195
  5. Hu Y, Thapa A, Fan H, Ma T, Wu Q, Ma S, Zhang D, Wang B, Li M, Yan L, Wei F (2020). Genomic evidence for two phylogenetic species and long-term population bottlenecks in red pandas. Science Advances 6(9):eaax5751. doi:10.1126/sciadv.aax5751. https://doi.org/10.1126/sciadv.aax5751
  6. Fan H, Wu Q, Wei F, Yang F, Ng BL, Hu Y (2019). Chromosome-level genome assembly for giant panda provides novel insights into Carnivora chromosome evolution. Genome Biology 20:267. doi:10.1186/s13059-019-1889-7. https://doi.org/10.1186/s13059-019-1889-7
  7. Ma J, Zhang L, Shen F, Geng Y, Huang Y, Wu H, Fan Z, Hou R, Song Z, Yue B, Zhang X (2023). Gene expressions between obligate bamboo-eating pandas and non-herbivorous mammals reveal converged specialized bamboo diet adaptation. BMC Genomics 24:23. doi:10.1186/s12864-023-09111-z. https://doi.org/10.1186/s12864-023-09111-z
  8. Li L, Shen F, Jie X, Zhang L, Yan G, Wu H, Huang Y, Hou R, Yue B, Zhang X (2022). Comparative Transcriptomics and Methylomics Reveal Adaptive Responses of Digestive and Metabolic Genes to Dietary Shift in Giant and Red Pandas. Genes 13(8):1446. doi:10.3390/genes13081446. https://doi.org/10.3390/genes13081446
  9. Chen L, Ma J, Xu W, Shen F, Yang Z, Sonne C, Dietz R, Li L, Jie X, Li L, Yan G, Zhang X (2024). Comparative transcriptome and methylome of polar bears, giant and red pandas reveal diet-driven adaptive evolution. Evolutionary Applications 17(6):e13731. doi:10.1111/eva.13731. https://doi.org/10.1111/eva.13731
  10. Chen L, Zhang L, Zhao Y, He M, Wu H, Wang J, Chen Z, Zhao Y, Shen F, Zhang X (2025). Impact of DNA methylation on digestive and metabolic gene expression in red pandas (Ailurus fulgens) during the transition from milk to bamboo diet. BMC Genomics 26:404. doi:10.1186/s12864-025-11606-w. https://doi.org/10.1186/s12864-025-11606-w
  11. Wolsan M, Sato JJ (2022). Role of feeding specialization in taste receptor loss: insights from sweet and umami receptor evolution in Carnivora. Chemical Senses 47:bjac033. doi:10.1093/chemse/bjac033. https://doi.org/10.1093/chemse/bjac033
  12. He Z, Xu S, Shi S (2020). Adaptive convergence at the genomic level—prevalent, uncommon or very rare? National Science Review 7(6):947–951. doi:10.1093/nsr/nwaa076. https://doi.org/10.1093/nsr/nwaa076
  13. Dong Z, Wang C, Qu Q (2024). WGCCRR: a web-based tool for genome-wide screening of convergent indels and substitutions of amino acids. Bioinformatics Advances 4(1):vbae070. doi:10.1093/bioadv/vbae070. https://doi.org/10.1093/bioadv/vbae070