Fingerprinting Yields
Surprising
Findings on Wheat Diversity
Capturing diversity by collecting and storing wheat landraces is a tricky business. Should collectors go to many fields and obtain a single sample from each? Or should they go to a single field and collect a multitude of samples? Should sampling strategies be the same for regions that are a center of origin as for those that are not? And that's just the beginning. Storage and maintenance strategies also differ based on the variation in each landrace. Genetic fingerprinting can answer these questions.
CIMMYT molecular geneticist Marilyn Warburton and her team looked at about 150 landraces collected from various countries using a range of methods. "Passport data," when available, supplied information about where a sample was collected, how it was collected, and how it was stored. For example, a sample could be collected as a single spike (ear) of grain from one plant, as spikes from several plants from the same field that were conserved separately by the genebank, or as part of a "bulk" of seeds—seeds collected from a number of plants thought to be representative of a given landrace and maintained in the same sample. The analysis reinforced long-held views on collection but, surprisingly, contradicted others.
The team found tremendous genetic differences among landraces grown within a country considered a center of origin for wheat, even if the landraces went by the same name and were collected from adjacent villages. On the other hand, in countries not designated as centers of origin, even landraces going by different names appeared to be very similar genetically.
"These findings tell scientists to focus their collecting in centers of diversity," says Warburton. "While we knew that in theory, we now have the data to back it up."
However, the team came up with disconcerting results that showed that the amount of variation within a landrace sample did not necessarily correlate with how it was collected—a finding at odds with the conventional wisdom.
"If the sample was collected as a bulk," Warburton explains, "you'd expect to see several different alleles—forms of the same gene—at each marker, but all too frequently we saw only one or two alleles. Either the field where we collected the sample was planted to a single genotype, or some variation was lost after years in storage."
Such samples should not be treated as bulks but rather as a single inbred line. When the sample is regenerated, fewer seeds can be planted. In addition, breeders can be informed that there is limited variation in the line and it can be treated as an inbred.
A few samples collected from a single spike showed a lot of variation. There are several possible explanations for this unexpected result. The sample may have been an outcrossed hybrid (a rare but not impossible occurrence), seeds may have been mixed during some stage of handling, or the passport data were simply incorrect. Regardless of how the variation occurred, wheat breeders or genebank curators should not treat these samples as inbreds. They should treat them as bulks and conserve their diversity.
"This work provided an immediate practical payoff for the
genebank,"
says Bent Skovmand, head of the wheat genebank. "By
employing these techniques on a wider scale, we can help people collect and
store genetic resources more efficiently, avoid loss of variation, and
save money by growing only the number of plants needed to retain the
genetic diversity in a particular sample."
What Is Genetic
Fingerprinting?
Genetic fingerprinting is probably more widely known for its uses in people—where it is used to determine paternity or indicate whether a person was present at a crime scene—than for its uses with plants. However, just like fingerprinting in humans, fingerprinting in plants can clear up a few mysteries.
Known also as "DNA fingerprinting" and "DNA profiling," fingerprinting in plants is based on the assumption that every individual variety or population has a genetic profile, revealed through its DNA, that is unique to that variety.
Researchers obtain samples of DNA from plant tissue and use several techniques to produce a "fingerprint" that looks like a series of bands of varying size, much like a bar code. The bands for one variety can be compared to bands for other varieties to detect similarities and differences. The more similarities there are, the more related the two varieties are, and the parents of the variety (or sibling varieties sharing the same parents) can be determined.
Although breeders generally have a very good idea of the origins and probable genetic advantages of the varieties or lines they develop, fingerprinting adds greater certainty to their work and helps them to work more rapidly. Closely related lines frequently share the same characteristics; thus, if one line has a favorable performance under certain conditions, lines closely related to it probably will, too. Also, in hybrid breeding, lines that are unrelated generally create better performing hybrids than lines that are related to each other or are very similar genetically. DNA fingerprinting can help breeders decide which varieties to cross with which.
Another problem that can confound plant breeding is that varieties found in several parts of the world (or even the same country or province) can have the same name but may not even be related. Fingerprinting can determine if varieties with the same name are truly genetically identical. This information helps breeders and also helps genebank curators decide which seed to conserve.
August, 2004