Coffee seeds contain 11S storage proteins

A legumin‐like seed protein was purified from the endosperm of coffee ( Coffea arabica L. cv. Colombia). In contrast to legumes, where efficient storage globulin extraction requires buffered saline solutions well above the acidic pK I of the globulins, coffee legumin is readily extracted with acidic aqueous buffers. The coffee legumin migrates like other 11S storage globulins in sucrose gradients. Subunits of coffee legumin have an apparent molecular mass of about 55 kDa after one‐dimensional SDS‐polyacrylamide gel electrophoresis in the absence of a reducing agent. In the presence of 2‐mercaptoethanol, two polypeptides appear that have apparent molecular masses of 33 and 24 kDa. Two full‐length cDNAs were generated from mRNA of developing seeds that were more than 98% homologous. They had open reading frames of 1 458 and 1 467 bp. Each encoded legumin precursors of 486 and 489 amino acids, respectively (M r =54 136 and 54 818). Examination of a 5′ promoter region from a coffee legumin gene revealed a putative legumin‐box. Genomic DNA from C . arabica was digested with six different restriction endonucleases. After separation of the fragments by electrophoresis, single discrete fragments on DNA blots hybridized strongly to a cDNA probe for the acidic chain. Other fragments that hybridized weakly with this probe were visible after hybridization at very low stringency. DNA from other species and commercially important cultivars that comprise the genus Coffea produced similar results. Immunocytochemical studies revealed that some legumin was detected in the cytoplasm in mature coffee seeds, but that the majority of it was in large storage vacuoles that accounted for most of the cell volume.