Constrained evolution of a core winter proteome across independently cold-adapted PACMAD grasses

Early maize planting requires cold tolerance in temperate regions, which elite maize lacks. Extending the growing season could boost productivity along with more efficient nutrient use. Wild PACMAD grasses have each independently evolved freezing tolerance. To uncover the molecular basis of this convergence, we performed tandem mass tag labeling with shotgun mass spectrometry to measure protein abundance in rhizome tissues sampled during winter dormancy and summer activity. Our study examined five species: <i>Panicum virgatum</i>, <i>Andropogon gerardii</i>, <i>Miscanthus giganteus</i>, <i>Sorghastrum nutans</i> and hybrids of <i>Tripsacum floridanum</i> and <i>Tripsacum dactyloides</i> (maize's sister genus), all grown in Ithaca, NY (where winter lows reach -29 °C). Of ~3500 proteins per species, 330 families showed consistent upregulation in winter, but only three-late embryogenesis abundant 3 (LEA3), aldose reductase, and phosphatidylethanolamine-binding protein (PEBP)-were shared across all five lineages. LEA3 proteins display conserved hydrophobicity patterns in cold-tolerant species that are disrupted in maize. Functional enrichment highlighted recurrent use of lipid transfer proteins, heat shock proteins and other drought-associated cryoprotectants. We next explored whether these conserved signatures extended to other tissues within <i>Tripsacum</i> and their orthologous genes in cold-sensitive maize. Comparisons with mRNA studies revealed that rhizome proteomic responses more closely resemble those of seedling leaves than roots, consistent with their shoot-derived anatomical identity, a pattern less pronounced in maize seedlings. Overall, the independent evolution of rhizome frost tolerance in PACMAD grasses appears to be governed by similar mechanisms driven primarily by expression-level changes complemented by protein structural adaptations offering candidate targets for improving freezing tolerance in maize.