Representational difference analysis: critical appraisal and method development for the identification of unique DNA sequences from prokaryotes

Infectivity and Screening of Anti-piperaquine Genes in Mice Infected with Piperaquine-Sensitive and Piperaquine-Resistant Plasmodium berghei

BACKGROUND

Piperaquine (PQ) is one of the major components of artemisinin-based combination therapy for malaria. However, the mechanism of PQ resistance has remained unclear.

METHODS

In this study, we infected mice with PQ-resistant Plasmodium berghei ANKA strain line (PbPQR) or PQ-sensitive P. berghei ANKA strain line (PbPQS) and their survival rates, parasitemia, and spleen sizes were compared. In addition, we constructed genomic DNA subtractive library of spleens from the infected mice, and screened the potential PQ-resistant related genes from genomic DNA of PbPQR line using the representational difference analysis (RDA) method. Clones of the subtractive library were screened by PCR, and related genes were sequenced and analyzed using BLAST software of NCBI.

RESULTS

Compared to PbPQS-infected mice, PbPQR-infected mice survived significantly longer, and had significantly lowered parasitemia rate and significantly increased splenomegaly. Among the total of 502 clones picked, 494 were sequenced and 96 unique PCR fragments were obtained; in which 24 DNA fragments were homologous to chromosomes related to immune function of mice. ORF Finder blasting showed that at the protein level, 26 encoded proteins were homologous to 18 hypothetical PbANKA proteins and 13 encoded proteins were homologous to "ferlin-like protein" family of PbANKA. In addition, there were more immune-related DNA molecules, ubiquitous PbANKA homology at the ORF fragment level, and enriched ferlin-like protein families identified from PbPQR-infected mice than those from PbPQS-infected mice.

CONCLUSION

These findings suggest that PbPQR may induce stronger protective immune response than that of PbPQS in infected mice.

Genomic screening in dioecious "yerba mate" tree (Ilex paraguariensis A. St. Hill., Aquifoliaceae) through representational difference analysis

The "yerba mate" tree, Ilex paraguariensis, is a functionally dioecious crop species with economic relevance in several South American countries. We report a genomic screening accomplished through representational difference analysis (RDA) in male and female I. paraguariensis trees. The aim of the present paper was to investigate the occurrence of sex-related genomic differences in order to develop an early gender detection molecular method that could help reducing energy inputs during the "yerba mate" processing and that could be suitable for breeding programs. An intra-experiment redundancy was detected via SSCP analysis and sequence characterization. Taking together both reciprocal RDA assays, fragments isolated can be discriminated into three main categories. The first category of fragments shows spurious affinities with available deposited sequences and could be considered as specific to I. paraguariensis. The second category comprises sequences identified as organellar or ribosomal plant DNA. Sequences grouped in the third category involve clones akin to conserved domains of retrotransposons (RNaseH, integrases and/or chromodomains) from at least two distinct lineages of Ty3/Gypsy retrotransposons and one from Ty1/Copia retroelements, which in addition are associated to sex determination regions of the Solanaceae, Caricaceae and Salicaceae. A contig sequence was assembled that codes for an integrase core domain and a chromodomain. The phylogenetic analysis of the so-called IPRE (for I. paraguariensis retroelement) integrase domain indicates that it belongs to the Del lineage of the Chromoviridae. This is the first report of mobile elements isolated and detected from the "yerba mate" tree. Although RDA derived fragments, so far tested, have been retrieved from both sexes with similar sequences, association to sex related regions cannot be completely discarded. Implications of present results are further discussed.

Genetic differences between Bradyrhizobium japonicum variant strains contrasting in N(2)-fixation efficiency revealed by representational difference analysis

Two variant strains of Bradyrhizobium japonicum, derived from SEMIA 566, adapted to the stressful environmental conditions of the Brazilian Cerrados and characterized by contrasting capacities for N(2) fixation, were compared by representational difference analysis (RDA). Twenty-four gene sequences that are unique to the highly effective strain S 370 were identified, eight showing high similarity to known genes, nine encoding putative proteins and seven representing conserved hypothetical or hypothetical proteins; they were classified in eight functional categories. Among those genes, some were highlighted for their known or potential functions in plant-microbe interactions. The nodulation outer protein P (nopP), related to the type-III secretion system (TTSS) and a major determinant of nodulation of some tropical legumes, was detected in the genome of strain S 370. Three coding sequences (CDS) identified by RDA were expressed in proteomics experiments with B. japonicum strain USDA 110 (ChvE and NopP). The use of the sequences identified by RDA in the highly effective strain S 370 might represent an important tool to speed up strain selection programs, accelerating pre-screening procedures. Additionally, the conserved hypothetical and hypothetical proteins identified in strain S 370 might encode important but still unknown proteins related to the symbiosis that deserve further study.

Identification of strain-specific DNA of Actinobacillus actinomycetemcomitans by representational difference analysis

A genomic subtraction method, the representational difference analysis (RDA), was tested for its use in identifying strain-specific DNA in Actinobacillus actinomycetemcomitans. Subtraction of strain D7S with strain HK1651 yielded D7S-specific 2.3-kilobase (kb) islet-A and 5.3-kb islet-B. Islet-A contains a 1.5-kb region that is homologous to a region found in the A. actinomycetemcomitans plasmid pVT745. Islet-B contains a 2.1-kb homolog of vgr, a component of a DNA repeat element rhs. The distribution of these islets among A. actinomycetemcomitans strains was further examined by polymerase chain reaction. Islet-A was found in nine serotype a and two serotype b strains but was missing from 34 strains. Islet-B was found in one serotype a strain, four serotype d strains and two serotype e strains, but was missing from 34 strains.

Hybridization monitor: A method for identifying differences between complex genomes

We have developed a method to identify and amplify differential fragments between two complex genomes. This technique, named hybridization-monitored genome differential analysis (HMDA), incorporates a monitor system into a PCR-based solid subtraction hybridization that tracks the entire hybridization process. This is achieved by monitoring the subtraction progress using PCR analysis of the conserved sequence of 18S rDNA in the tester sample after each round of subtraction. Homologous fragments can then be eliminated when bound to the driver DNA immobilized on a solid membrane. The hybridization continues until the conserved DNA sequence of 18S rDNA can no longer be detected, and most of the unbound DNA fragments left in the liquid were mainly the tester-specific fragments, thus greatly decreasing the complexity of DNA template of PCR amplification, increasing the amplification efficiency of differences accordingly, and ensuring high positive efficiency and coverage across the tester genome. We have applied the technique in a comparison between the genomes of Saccharomyces cerevisiae and Schizosaccharomyces pombe, which are two completely sequenced organisms. Results indicated that 95% of the subtracted clones have been confirmed to be different to the driver analyzed using the BLASTN homology alignment. With this technique, 240-fold enrichment of differences is obtained, and the coverage of the difference is up to 79%. These results indicate that HMDA can efficiently identify sequences that differ between two complex genomes.

Novel DNA sequences from natural strains of the nitrogen-fixing symbiotic bacterium Sinorhizobium meliloti

Variation in genome size and content is common among bacterial strains. Identifying these naturally occurring differences can accelerate our understanding of bacterial attributes, such as ecological specialization and genome evolution. In this study, we used representational difference analysis to identify potentially novel sequences not present in the sequenced laboratory strain Rm1021 of the nitrogen-fixing bacterium Sinorhizobium meliloti. Using strain Rm1021 as the driver and the type strain of S. meliloti ATCC 9930, which has a genome size approximately 370 kilobases bigger than that of strain Rm1021, as the tester, we identified several groups of sequences in the ATCC 9930 genome not present in strain Rm1021. Among the 85 novel DNA fragments examined, 55 showed no obvious homologs anywhere in the public databases. Of the remaining 30 sequences, 24 contained homologs to the Rm1021 genome as well as unique segments not found in Rm1021, 3 contained sequences homologous to those published for another S. meliloti strain but absent in Rm1021, 2 contained sequences homologous to other symbiotic nitrogen-fixing bacteria (Rhizobium etli and Bradyrhizobium japonicum), and 1 contained a sequence homologous to a gene in a non-nitrogen-fixing species, Pseudomonas sp. NK87. Using PCR, we assayed the distribution of 12 of the above 85 novel sequences in a collection of 59 natural S. meliloti strains. The distribution varied widely among the 12 novel DNA fragments, from 1.7% to 72.9%. No apparent correlation was found between the distribution of these novel DNA sequences and their genotypes obtained using multilocus enzyme electrophoresis. Our results suggest potentially high rates of gene gain and loss in S. meliloti genomes.