HYMENOBACTER ROSEOSALIVARIUS GEN. NOV., SP. NOV. FROM CONTINENTAL ANTARCTIC SOILS AND SANDSTONE - BACTERIA OF THE CYTOPHAGA FLAVOBACTERIUM/BACTEROIDES LINE OF PHYLOGENETIC DESCENT/

Aseptically collected sandstone and soil samples from the antarctic Dr y Valleys were inoculated into oligotrophic media and incubated under low light intensities. A total of 41 Gram-negative isolates were obtai ned with reddish colonies spreading on agar. A sandstone isolate and f our soil strains were characterized further. They were nearly identica l in morphological, physiological, biochemical and chemotaxonomic prop erties. They produced large amounts of extracellular polymer and utili zed for growth: glucose, saccharose, mannitol, sorbitol, L-aspartate, malate and acetate, but not D-ribose, adonitol, DL-alanine, glutamate, glycolate, lactate or succinate. All strains hydrolyzed gelatin, star ch, casein, xylan, Tweens 80 or 60 and dead or living yeast cells, but not cellulose or pectin. Nitrate was not reduced, ethanol was not oxi dized and acid was not produced from maltose, mannitol or dulcitol. Am monia was not produced from peptone. They were strictly aerobic. Major fatty acids were n 16:1 d 9, n 16:1 d 11, n 17:1 d 11, and i 15:0. Th e strains contained the quinone MK-7 and phosphatidylethanolamine as t he main phospholipid. The base ratio ranged from 55 to 61 mol% G+C. A 16S rRNA sequence analysis of strains AA-688 and AA-718 showed these t o be identical and to represent a special phylogenetic group within th e Cytophaga / Flavobacterium / Bacteroides major line of descent. Thre e soil strains labeled ''Taxeobacrer'' Txc1, Txg1, and Txo1 (REICHENBA CH, 1992) belonged to the same group but had lower sequence similariti es (<95%). Some of their characteristics were different from chose of the antarctic strains: the utilization of C-compounds, hydrolysis of p olymers, temperature tolerances, major fatty acids and base ratios. Tx c1 and Txg1 may later have to be considered as members of this group, possibly on the species level, while Txo1 could represent a different but related genus. It is concluded that the five antarctic strains rep resent a new genus and species for which the name Hymenobacter roseosa livarius is proposed. The type strain is AA-718(T) (DSM 11622(T)).