raising the temperature for a while in pressurized aquaria (Cottin et al. 2008).
However, the increase takes hours to achieve, so it can only be part of the physiology
of surviving rapid temperature shifts (Fig. 15.5). Aspects of the high-temperature
enzymology of bacteria associated with Alvinella are better studied. Symbiotic
bacteria have not been found internally in Alvinella spp. However, copious colonies,
sometimes described as a “fleece”, of filamentous bacteria live on dorsal and lateral
epidermal bumps along the edges of each segment. The bumps are glandular with a
mucoid secretion evident in surface cells and externally, into which the long (up to 0.6
mm) bacterial colonies are twisted (Desbruyères et al. 1985). Thus, the worm’s
surface is specially elaborated to accommodate these episymbionts. Other
chemoautotrophic bacteria live in the intersegmental grooves and line the inner
surfaces of the tube. Most endosymbionts at vents are gammaproteobacteria, but Cary
et al. (1997) have shown that most of the external filamentous forms on Alvinella are
episilonproteobacteria (as are those of Rimicaris). There are two dominant bacterial
“phylotypes” (i.e. with significantly variant SSU RNA) (Campbell et al. 2001). One
of those has been cultured, named Nautilia profundicola, and its complete genome
sequenced (Campbell et al. 2008) to characterize its metabolism in detail. It is,
indeed, a “chemolithotroph”, but not a sulfide oxidizer. Rather, according to Campbell
et al. (2008), the “cells are strictly anaerobic. Chemolithoautotrophic growth occurs
with molecular hydrogen or formate as the electron donor and elemental sulfur as the
electron acceptor, producing hydrogen sulfide”. In the Alvinella consortium there are
also sulfate or sulfite reducing epsilonproteobacteria that oxidize formate and acetate
in laboratory cultures, and others that are sulfur-oxidizing autotrophs (Campbell et al.
2001). So, the Alvinella-associated bacterial community is phylogenetically restricted,
but still trophically complex.
(^) The worms have a feeding apparatus with grooved and ciliated tentacles,
comparable to that with which ampharetids, a related polychaete family, gather
particulate food. In addition, alvinellids have a peculiar lip structure which may gather
bacteria from surfaces in bulk. However, it has not been possible to observe feeding
on the bacterial associates directly. Bacteria on the inner tube wall are almost certainly
grazed. Those on the dorsum may be grazed, since the worm is very flexible, or they
may supply inocula to the colonies on the tube walls. Not all of the behavior that we
would like to see is sufficiently accessible. The anterior gut contents are indeed
filamentous bacteria, and other bacteria are compressed in the foregut into mucous-
bound masses (Desbruyères et al. 1985). Solid feces in the posterior gut are mostly
elemental sulfur, but also contain glucosamine, an ammoniated sugar found in
bacterial cell walls (Saulnier-Michel et al. 1990). Thus, the diet is certainly the
bacterial associates. The sulfur suggests a significant role for bacteria using sulfate as
an oxidant.
(^) Lee et al. (2008) isolated and identified genes for two metabolic enzymes in DNA