A Computer-Simulated Restriction Fragment Length Polymorphism Analysis of Bacterial Small-Subunit rRNA Genes: Efficacy of Selected Tetrameric Restriction Enzymes for Studies of Microbial Diversity in Nature

Craig L. Moyer¹, James M. Tiedje², Fred C. Dobbs³ and David M. Karl¹

¹Department of Oceanography, School of Ocean and Earth Science and Technology, University of Hawaii, Honolulu, HI 96822

²Center for Microbial Ecology, Department of Crop and Soil Sciences and of Microbiology, Michigan State University, East Lansing, Michigan 48824

³Department of Oceanography, Old Dominion University, Norfolk, Virginia 23529

Abstract

An assessment of 10 tetrameric restriction enzymes (TREs) was conducted bv using a computer-simulated restriction fragment length polymorphism (RFLP) analysis for over 100 proximally and distally related bacterial small-subunit (SSU) rRNA gene sequences. Screening SSU rDNA clone libraries with TREs has become an effective strategy because of logistic simplicity, commercial availability, and economy. However, the rationale for selecting the type and number of TREs has not been svstematicaliy evaluated. Our objective was to identify the optimal combination of TREs for RFLP screening of cloned SSU rRNA genes from undefined bacterial clone libraries. After computer-simulated TRE digestion. the resultant fragments were categorized on the basis of the frequency of different restriction fragment size classes. Three groups of distribution patterns for the TREs were determined and further examined via graphical exploratory data anaivsis. The RFLP size-frequency distribution data for each group of enzymes were then used to infer phylogenetic relationships via the neighbor-joining method. The resulting bootstrap values and the correct placement of node bifurcations were used as additional criteria to evaluate the efficacy of the selected TREs. These RFLP data were compared with known phylogenetic relationships based on SSU rRNA sequence analysis as defined by the Ribosomal Database Project. A heuristic approach testing random combinations of TREs showed that three or more TRE combinations detected >99% of the operational taxonomic units (OTUs) within the model data set. OTUs that remained undetected after three TRE treatments had a median sequence similarity of 96.1%. Of the 10 restriction enzymes examined, HhaI, RsaI, and BstUI (group 3) were the most efficacious at detecting and differentiating bacterial SSU rRNA genes on the basis of their ability to correctly classify OTUs. Group 3 TREs are therefore recommended for screening in studies using bacterial SSU rRNA genes as descriptors of in situ microbial diversity.