spectrum (1 – T, where T is the transmittance of the sam- ple) and the fluorescence (Fl) excitation spectrum, as described previously (Cogdell et al. 1981; Noguchi et al. 1990). The efficiency of Car ? Bchla ET, ascribed as the ratio of Fl/(T – 1), was averaged over the 0-0 and 0-1 bands of the carotenoid molecules where the bacterio- chlorin molecules do not absorb after the spectra were normalized to the Qx-transition of the Bchla molecules situated at ca. 590 nm. It was assumed that the ET from the Qx transition to the Qy transition of Bchls was 100%. The room-temperature fluorescence excitation and fractional absorption spectra were measured with the same spectral bandwidth of 2 nm. The FL emission was measured at the peak position with the slits wide open (20 nm) to ensure that all the fluorescence was detected.
Carotenoid isolation, purification and structure determination
Pigments were removed from cells by two extractions with acetone/methanol (7:2, v/v) and each extraction was accompanied by sonication of the mixture. Then the sol- vent was removed by evaporation. The pigment extracts were analyzed by HPLC equipped with a lBondapak C18 column (8 · 100 mm, RCM type; Waters, USA) and eluted with methanol (1.8 ml/min). Anion exchange chro- matography (DEAE-Toyopearl 650M, Tosoh, Japan) was used to separate the bacteriochlorins and polar lipids from the carotenoids (Takaichi et al. 2001). Absorption spectra of the carotenoids were recorded with a photodiode-array detector (MCPD-3600; Otsuka Electronics, Japan) attached to the HPLC system. The molar absorption coefficient at maximum wavelength for each carotenoid in the eluent of methanol was assumed to be the same. Relative molecular masses were determined by field-desorption mass spec- trometry using a double-focusing gas chromatograph/mass spectrometer equipped with a field-desorption apparatus (M-2500, Hitachi, Japan). The 1H-NMR (500 MHz) spec- tra in CDCl3 at 24C were measured with a UNITY INOVA-500 (Varian, USA) system. The peak assignments of NMR spectra were made on the basis of 1H–1H COSY and NOESY analysis. The circular diochroism (CD) spectrum was measured by a J-820 spectropolarimeter (JASCO, Japan) in diethyl ether/i-pentane/ethanol (5:5:2, by vol.) at 20C (Takaichi et al. 2001).
Results and discussion
In June 1999, during the course of a survey of the previ- ously unexplored South Andros Black Hole system marked
temperature anomaly was recorded at 17.8 m depth, where the water temperature increased sharply from 29 to 36C before decreasing again after a further 1-m increase in depth (Fig. 2). This depth marks the boundary between the oxic brackish upper water mass and the denser anoxic saline lower layer. Since the South Andros Black Hole has no known direct connection to the sea, except through rock fractures and local porosity, water exchange is severely restricted. Hence, the physico-chemical gradients that develop are unusually stable and the boundaries sharp, confirming the absence of mixing (Schwabe and Herbert 2004). In this respect the South Andros Black Hole can be considered analagous to meromictic lake see references (van Gemerden and Mas 1995; Overmann et al. 1991) and references therein. Coincident with the sharp increase in water temperature a 1 m thick gelatinous layer (plate) of anoxygenic phototropic purple sulfur bacteria was observed (Fig. 2, hatched horizontal bar). The dominant purple sulfur bacteria of this warm, saline, and sulfide rich layer have been isolated and identified as members of the genera Allochromatium and Thiocapsa. The development of dense populations of anoxygenic phototropic bacteria is not uncommon at the pycnocline in stratified lakes (Over- mann et al. 1996a, b). However, the thickness of these plates is typically of the order of a few tens-of-centimeters and they are more usually located higher in the water column (van Gemerden and Mas 1995). To our knowledge, this study presents the first evidence of the ability of such bacterial populations to significantly increase the ambient
Water depth (m)
30 32 34 Temperature (°C)
Fig. 2 Day-time (solid circles) and night-time (open circles) tem- perature profiles of the water column and spatial location (hatched horizontal bar) of the phototropic purple sulfur bacteria