Different grain samples representative of the 21 native races of maize (Zea mays ssp. mays) from Salta, Argentina, were used to study the storage proteins through different evaluations. The material was kindly supplied by Prof. Melchiore, who has also studied and classified it on the basis of traditional morphological traits in the Laboratory of Botanics, Faculty of Agronomy, Buenos Aires University.

The races studied are: Pisinga purpura (PP), Colorado (C), Chaucha (CH), Amarillo de ocho (AO), Chiriguano (CHO), Orgullo cuarenton (OC), Criollo amarillo (CA), Amarillo ancho (AA), Capia blanco (CB), Capia tabanito (CT), Capia purpura (CP), Capia rosado (CR), Chulpi (CHI), Socorro (S), Perla (P), Blanco ocho rayas (BOR), Ocho rayas (OR), Altiplano (A), Pinto (PO), Culli (CI) and Negro (N). These races constitute the 2 operational taxonomic units (OTU's).

Some morphological traits of these kernels were evaluated: 1) type (amylacea, amylaesaccharata, indurata, orysae), 2) pericarp (colourless, light coloured and dark coloured), 3) aleurone layer color (coloured or colourless), 4) amyloproteic parenchyma color (colourless, coloured), 5) kernel shape (rostrated, cuneiform, round or extended), 6) hard endosperm ratio, 7) kernel weight, 8) kernel length, 9) kernel width and 10) kernel thickness. The biochemical traits evaluated are: 11) protein content, 12) saline soluble protein content, 13) zein content, 14) glutelin content.

Polyacrylamide gel electrophoresis of zein (PAGE) let us obtain traits 15 to 51 using the criterion of presence-absence of bands (the electrophoretical patterns can be seen in Figure 1). Finally through SDS polyacrylamide gel electrophoresis of soluble proteins, we obtained traits 52 to 64 using the previous criterion (the electrophoretical patterns can be seen in Figure 2).

We have made a basic data matrix (BDM). Using traits 1 to 14 a mean taxonomic distance matrix (MTD) was built. Then, with this MTD we built the phenogram shown in Figure 3, by the unweighted pair group using arithmetic averages (UPGMA). On the basis of those traits derived from the electrophoretical studies, an association matrix (OTU x OTU) was built, using the simple matching coefficient (SMC), from which the resulting phenograms (see Figures 4 and 5) were built by the UPGMA method.

According to what we have previously analyzed, it can be deduced that those races studied can be clustered into 2 or 3 very well-defined racial complexes, in which there is a smaller distance among the different races. The different groupings show a partial congruence though statistically significant. Otherwise, the results obtained on the basis of protein storage traits are also partially congruent with those measured on the basis of traditional morphological traits (Melchiore, personal communication).

Finally these results point out that biochemical traits can be appropriate to make a clustering analysis and in addition they are demonstrative of the considerable variation among the different races. This fact is in accord with what we have previously communicated in similar studies and documents the enormous variability present in maize.

Figure 1. Polyacrylamide electrophoretic patterns of zein of races of maize from Salta.

Figure 2. SDS-polyacrylamide electrophoretic patterns of soluble proteins of races of maize from Salta.

Figure 3. Phenogram of 21 OTU's resulting from UPGMA cluster analysis of the OTU x OTU distance matrix, based on characters 1 to 14. r: cophenetic correlation coefficient.

Figure 4. Phenogram of 21 OTU's resulting from UPGMA cluster analysis of the OTU x OTU association matrix, based on zein electrophoretic patterns. r: cophenetic correlation coefficient.

Figure 5. Phenogram of 21 OTU's resulting from UPGMA cluster analysis of the OTU x OTU association matrix, based on SDS-PAGE soluble proteins. r: cophenetic correlation coefficient.

Angel A. Nivio and Jorge L. Magoja

Return to the MNL 62 On-Line Index
Return to the Maize Newsletter Index
Return to the MaizeGDB Homepage