Search

Characterization of 13 isolates of luminescent bacteria from specimens of Cleidopus gloriamaris has revealed differences between these organisms and previously described species of Photobacterium. Further studies on the biochemical and serological attributes of these and other luminescent bacteria are essential, before ecological studies on the fish-microbe symbiosis are possible.

The Leiognathid (pony) fish, which occur in the shallow coastal waters of the tropical and subtropical Indo-Pacific region, are capable of emitting a bright light from their ventral surface. In these small fish, as in several other (but not all) bioluminescent fish, the source of light is symbiotic luminous bacteria, maintained within a special organ (Harvey, 1952, 1958 ; Buchner, 1965). The evidence that bacteria are involved as symbionts has come from microscopic observations to gether with the fact that cultures of luminous bacteria have been obtained from the organs (Harms, 1928 ; Haneda, 1940, 1950) . In the present experiments addi tionai proof of the bacterial origin of the light is presented, together with evidence that the symbiotic bacteria are distinct from many of the free living luminescent bacteria which may be isolated directly from sea water in the same area. In different grOU@)5of fish there are very different and sometimes highly elabo rate types of organs and modes for display of the bacterial light. In pony fish the system involves several special and unusual elements. The organ itself, which surrounds the esopilagus like a donut, and communicates with it †? zA±a paired ducts (Haneda, 1940, 1950) is literally packed with bacteria. Upon dissection it is always fotlnd to be emitting light, brightly and continuously, irrespective of the tinle of day or other environmental factors. The light reaches the outside (ventral) surface via indirect and somewhat sophisticated optics. The gut tract makes a loop into the wall of the swim (air) bladder at the site of the light organ. Tile organ thus faces directly into the swim bladder, and is provided with an eyelid-like flap which can control the amount of light shining into the air-filled bladder. The swim bladder is internally reflecting, being lined with guanine crystals—the same material which is responsible for tue silvery skin of many fish (Denton, 1970, 1971). The ventral portion of the swim bladder is only partially reflective (“half silvered―), and to it attach specialized translucent ienticular muscle cells. The optical arrangement thus takes the light from a small source and causes it to be evenly diffused over a larger area, namely most of the ventral part of the body. As shown below, it was possible to estimate the total number of viable bacteria within an organ, and to compare this with the weight of the organ. Such counts indicate that a large percentage, if not all, of the bacteria which are crammed into the ducts of the organ are both viable and bioluminescent. The isolated symbiotic bacteria were compared with free living bacteria isolated directly from sea water in the same area where the fish were collected. Although tilebacteriafrom the two sourceshad similarcolonialmorphology, and were the same in certain other special respects, tile two appeared to possess distinctly differ ent types of luciferases, corresponding to types previously known from different strains (Hastings, Weber, Friedland, Eberhard, Mitchell and Gunsalus, 1969), hut not previouslycorrelatedwith symbioticand freelivinglifestyles.

Cells of a species of luminescent bacteria were treated with a gas stream containing products formed by the photochemical oxidation of cis -2-butene and NO. Luminescence and viability decreased with the time of irradiation of reactants. The rate of luminescence decrease was dependent on the ratio of the initial concentrations of cis -2-butene and NO with a ratio of 2 giving the most rapid loss. Known photochemical oxidation products, ozone, NO 2 , formaldehyde, acetaldehyde, and PAN were examined individually. Aldehydes did not appear to contribute to the decrease in luminescence but ozone and PAN did. Although NO 2 alone up to 0.5 ppm produced no decrease, it may contribute to the luminescence loss in the total irradiation mixture. The total oxidant concentration produced upon irradiation gave luminescence decreases which were matched by comparable concentrations of pure ozone. However, this does not imply that luminescence losses caused by photochemical oxidants are due only to ozone. A possible mechanism of the toxic effect and the interpretation of this bioassay in relation to other organisms are briefly discussed.

WE have postulated previously that the catalytic decomposition by catalase of hydrogen peroxide to molecular oxygen and water is accompanied by changes in the valency of catalase iron1. This supposition is strongly supported by spectroscopic and manometric experiments carried out on azide-catalase. As additional evidence in support of this view, we brought forward the results of experiments carried out in Barcroft differential manometers, showing that under certain conditions even the activity of free catalase is inhibited when oxygen is completely removed from the surrounding medium and replaced by pure nitrogen. Our manometric experiments were repeated by Weiss and Weil-Malherbe2 who, using Warburg manometers, failed to obtain this inhibition. Johnson and van Schouvenburg3 also failed to confirm our results using luminescent bacteria as indicators of decomposition of hydrogen peroxide by catalase in complete absence of oxygen. However, we repeated our experiments from time to time, using different enzyme preparations, and invariably confirmed our previous results. The failure by other workers to obtain similar inhibition was explained by us4, 5 as being due probably either to some difficulty in complete elimination of oxygen from Warburg manometers or to some defects in their enzyme preparations.