Does RFLP analysis of ancient Anasazi samples suggest that they utilized hybrid maize?

While hiking in S. Utah, I was intrigued by the abundance of corn cobs lying around ancient Anasazi Indian ruins that were abandoned when the culture dissipated in the 12th-13th century A.D. While very dry, these cobs were still otherwise in very good condition. Realizing that the first step in our RFLP analysis is a DNA preparative protocol which utilizes lyophilization of the material, I wondered whether we could prepare DNA from these samples that would then be analyzable through our normal procedures.

I obtained 8 cob samples which had been tentatively dated to the 13th century from Winston Hurst, Curator at the Edge of the Cedars State Park in Utah. I tried several procedures to prepare DNA from the samples and found that one, with modifications, was most effective (C. Lichtenstein and J. Draper, In: DNA Cloning Volume II, D.M. Glover ed., IRL Press, Oxford, pp.67-120). I was able to repeatedly obtain DNA from the same set of 5 of the original 8 samples. Examination of the DNA on an agarose gel revealed that there was high molecular weight DNA present in the samples with only moderate degradation when compared to modern samples.

The 5 sample DNAs were digested with HindIII, electrophoresed on an agarose gel, and Southern-blotted by established procedures. These blots were then hybridized with several different clones from our repertory of informative maize clones. The clones were selected on the basis of: 1) they hybridized to only a single locus, 2) they were known to be very informative in a selection of Corn Belt germplasm composed of 150 inbreds, and 3) the loci detected by this group of clones were distributed on several different chromosomal segments.

The first positive result was that a defined hybridization signal could be detected in all 5 samples. It was weaker than that seen in the modern samples, most likely due to the moderate degradation in the ancient samples. The second aspect of interest here was that all of the Anasazi samples revealed an identical hybridization pattern, unique to each clone tested; i.e. there was no variation amongst these samples revealed by any of the clones. This is somewhat surprising in that all of these 5 samples originated from different ruins, although all in the same county and from about the same historical period. Obviously with the discrimination we usually see with this procedure, we can conclude that the people at these sites were using similar, if not identical, cultivars.

The third point is even more intriguing, if not difficult to interpret. Not only were the hybridization patterns identical among the 5 samples, but a number of the clones revealed more than one hybridizing fragment with them. There are a number of possible explanations for this result. For instance on average, 1 in 4 enzymes should exhibit a site within the genomic segment revealed by any particular clone and this would then result in 2 fragments. I do not believe this to be the case here as these clones had been tested against a large number of inbreds with HindIII and there were few or no cases of alleles revealed as 2 fragments. More importantly, I tested some of the clones against one of these ancient samples with multiple enzymes and in each case, 2 fragments were revealed, arguing that 2 distinct genomic segments were being detected. A second explanation for this result might be that there were extra copies of these sequences on chromosomes besides the normal complement of 20, perhaps on addition chromosomes from Teosinte or Tripsacum or even B chromosomes. It is difficult to believe that the former could result in genetically stable lines that would be maintained. The latter, B chromosomal sequences, is more difficult to rule out. However, I have checked some of these clones against modern B-A translocation stocks and see no evidence that any of these sequences are also found on the B chromosomal segments. The only other explanation that I have come up with is that the 2 fragments represent a heterozygous situation at single loci. This is supported by the fact that the intensity of the 2 fragments is usually equal and that they both often line up with single fragments detected in modern inbreds. For this situation to always be true in each of several samples as we have found, we cannot accept chance as the basis but would have to conclude that these identically heterozygous isolates were purposefully created, i.e. they were produced as part of a hybridization program. While surprising, the prospect that ancient Indians were aware of and utilized hybrid maize is not completely out of the realm of possibility; it was relayed to me that modern Indians in the region are known to maintain inbreds in a homozygous state by isolation, as part of their religious tenets. I am concerned that I am overlooking other possibilities or that there are other means to test this thesis. I would truly welcome ideas concerning whether this is indeed possible, what other data exists that would support or refute such an idea, and how I might further test this proposal.

Tim Helentjaris

Please Note: Notes submitted to the Maize Genetics Cooperation Newsletter may be cited only with consent of the authors.

Return to the MNL 62 On-Line Index
Return to the Maize Newsletter Index
Return to the Maize Genome Database Page