Genomic Exploration of Produce Degradation 9
amplified a 439-bp DNA fragment in all E. carotovora ssp. atroseptica isolates
tested, but not in isolates of the other E. carotovora subspecies or in atypical isolates
(Yahiaoui-Zaidi et al., 2003).
Building on their previous work (Ward and De Boer, 1994), De Boer and Ward
(1995) developed a primer set, ECA1f-ECA1r, in which only E. carotovora subsp.
atroseptica DNA was amplified. However, the genetic nature of these primers is
unknown, even compared with the most currently available GenBank database (Table
1.1). Targeting the E. chrysanthemi group, Nassar et al. (1996) developed a primer
set based on the sequence of E. chrysanthemi pelADE gene. A 420-bp amplified
fragment was obtained for all 78 E. chrysanthemi strains tested. No amplified fragment
was observed with the other Erwinia species and organisms of different genera.
Smid et al. (1995) developed three PCR primers. A combination of two, ERW-
FOR-ATRREV, is specific to E. carotovora subsp. atroseptica and another combination
of two, ERWFOR-CHRREV, is specific to E. chrysanthemi. Database searching (Table
1.1) shows that ERWFOR and ATRREV share no similarity to any known bacterial
DNA sequences. In contrast, primer CHRREV falls into the locus of a protease gene
from E. chrysanthemi. As the soft rot erwinia genome sequence database grows, the
genetic nature of all PCR primers together with their amplicons can be identified.
The PCR-sequencing format for pathogen detection can eliminate the nonspecific
amplification problem that has been one of the major concerns in PCR application
for diagnostics. Similar to sequences from the rrn operon, SNPs from other genomic
loci, if any, could be a valuable addition to soft rot erwinia genomics.
1.3 DNA GENOMICS
1.3.1 DNA-DNA HOMOLOGY
Although nucleotide sequences were not particularly identified as the target, DNA-
DNA homology study is one type of genome comparison. Instead of looking for
any particular locus or loci, all of the bacterial genetic contents, including coding
and noncoding regions, are compared simultaneously. As a result, differences are
described at the whole-genome level. The higher the DNA-DNA homology, the more
closely related the bacterial strains. DNA-DNA homology is by definition a critical
parameter for a bacterial species. DNA-relatedness studies of soft-rot organisms can
be traced back to the early 1970s, when Brenner et al. (1973) showed that strains
of Pectobacterium carotovorum and Pectobacterium chrysanthemi belonged to dis-
tinct DNA homology groups. Supported by DNA homology data, Gardan et al.
(2003) recently proposed that three subspecies of P. carotovorum should be elevated
to the species level.
Because a DNA homology test is laborious and time-consuming, it is not suitable
for routine use. In a study with fluorescent pseudomonas, Cho and Tiedje (2001)
developed a random genome fragment microarray in an effort to overcome the
disadvantages of whole-genome DNA-DNA hybridization. Sixty to 96 genome frag-
ments of approximately 1 kb from each of four fluorescent Pseudomonas species
were spotted on microarrays. Genomes from 12 well-characterized fluorescent
Pseudomonas strains were labeled with Cy dyes and hybridized to the arrays. Cluster