Abstract
The majority of marine cnidarians contain fluorescent and non-fluorescent pigments, but little is known about the role of these compounds in the life histories of these animals. To address this problem, two hypotheses were tested. The first hypothesis was that the fluorescent colors found in many cnidarians are due to the presence of proteins similar to the green fluorescent protein (GFP) originally isolated from the hydrozoan medusa Aequorea victoria . This was tested using chemical extractions and purification of the fluorescent compounds from cnidarian tissue (Chapter 1), by cloning cDNAs responsible for green, cyan and red fluorescence from five species representing three different families within class Anthozoa (Chapter 2), and by sequencing two nuclear fluorescent protein (FP) genes, including introns and upstream and downstream regions (Chapter 4). The fluorescent substances were conclusively shown to be proteins related to GFP. The basic organization of the FP genes was that of a standard nuclear gene with CAT and TATA transcription promoters, a poly-A terminator sequence, and standard invertebrate intron-exon splice sites. Intron length, intron number, and the 5' and 3 ' untranslated sequences varied between the reconstructed FP genes, indicating that FPs have a potential for use as phylogenetic markers. The second hypothesis was that the fluorescent proteins have a photoprotective role in cnidarian biology. In order to address this, several different types of experiments were carried out. First, an analysis of the mutability of the proteins cloned in Chapter 2 was performed (Chapter 3). This demonstrated that the various FPs are not maximally fluorescent, as the photoprotection hypothesis would predict. It was also shown that the red FP can be turned orange, yellow, or green with single amino acid substitutions. This has interesting implications concerning the origin and maintenance of red fluorescence that are discussed at length in the text. Second, a series of in vivo and in vitro experiments were performed, looking for changes in fluorescence in several species of scleractinian corals in relation to light intensity (Chapter 5). The conclusion of this work is that the fluorescent proteins from the scleractinian species studied do not appear to be photoprotective, although many factors confound the measurement of in vivo fluorescence. The measurable amount of fluorescence can change in an animal over time, but the changes are not well correlated with light intensity.