Department of Biological Sciences, University of Warwick, Coventry CV4 7AL, UK1
Max-Planck-Institut für Marine Mikrobiologie, Bremen, Germany2
Unité de Physiologie Microbienne (CNRS URA 2172), Institut Pasteur, Paris, France3
The H. Steinitz Marine Biology Laboratory, InterUniversity Institute of Eilat, Eilat, Israel4
Author for correspondence: David J. Scanlan. Tel: +44 24 76 528363. Fax: +44 24 76 523701. e-mail: dscanlan{at}bio.warwick.ac.uk
![]() |
ABSTRACT |
---|
![]() ![]() ![]() ![]() ![]() ![]() ![]() |
---|
Keywords: cyanobacteria, Red Sea, oligonucleotide probes, 16S rRNA, TSA
Abbreviations: DAPI, 4,6-diamidino-2-phenylindole; HL, high light; HRP, horseradish peroxidase; LL, low light; TETON, 4-(1,4,7,10-tetra-oxadecyl)-1-naphthol; TSA, tyramide signal amplification
The GenBank accession numbers for the sequences reported in this paper are AF311217 (RCC278, EQPAC1), AF311218 (RCC277, NATL1MIT), AF311219 (RCC280, NATL2B), AF311220 (RCC264, TAK9803-2), AF311291 (WH7803), AF311292 (WH8018) and AF311293 (WH8103).
![]() |
INTRODUCTION |
---|
![]() ![]() ![]() ![]() ![]() ![]() ![]() |
---|
Whilst the construction of 16S rRNA gene clone libraries, dot-blot hybridization and denaturing gradient gel electrophoresis (DGGE) are valuable techniques to analyse natural communities and to provide a picture of microbial diversity, the processes of DNA extraction and PCR amplification may introduce biases (von Wintzingerode et al., 1997 ). Furthermore, the abundance of a particular genotype can only be expressed as a relative value. Whole-cell hybridization using 16S rRNA-targeted oligonucleotides avoids these potential biases, allows the actual abundance of the genotype to be evaluated and also permits visualization of the spatial relationships of the cells in situ (Amann et al., 1997
, 1995
). For the majority of non-photosynthetic organisms, oligonucleotides monolabelled with a fluorescent dye are employed. However, for the hybridization of cyanobacteria, the signal from such probes has often been found to be insufficient to be visualized against the strong background of autofluorescence when viewed by epifluorescence microscopy (Schönhuber et al., 1999
). Fluorescently labelled probes have been used successfully for hybridization of Synechococcus and Prochlorococcus cells (Binder & Liu, 1998
; Worden et al., 2000
) when used in combination with flow cytometry and using a general bacterial probe EUB338 which targets one of the more accessible regions of the 16S rRNA (Fuchs et al., 1998
). However, for hybridization of Prochlorococcus cells, it was necessary to use peptide nucleic acid (PNA) probes since DNA probes gave poor results (Worden et al., 2000
). Another method for improving the signal of hybridized cells is to use horseradish peroxidase (HRP)-labelled probes combined with a fluorescent detection system based on tyramide signal amplification (TSA) (Schönhuber et al., 1997
, 1999
). Alternatively, hybridization with biotin-labelled probes and reaction with a streptavidinHRP conjugate prior to detection using TSA, a more indirect hybridization protocol successfully employed for bacteria (Lebaron et al., 1997
), should also overcome problems caused by autofluorescence. Here we report the use of a direct in situ hybridization method for members of the genus Prochlorococcus, initially optimized on cells of cultured isolates and then applied to the analysis of natural Prochlorococcus communities sampled from the North Atlantic and the Red Sea. Using a suite of 16S rRNA-targeted oligonucleotides we extend previous work on the distribution and niche occupancy of specific Prochlorococcus genotypes and demonstrate that closely related high-light-adapted genotypes can show remarkably different depth distributions in these two oceanic regions.
![]() |
METHODS |
---|
![]() ![]() ![]() ![]() ![]() ![]() ![]() |
---|
DNA sequencing and analysis.
The 16S rRNA gene sequences of the Prochlorococcus sp. strains EQPAC1, TAK9803-2, NATL1MIT and NATL2B were obtained after PCR amplification of two overlapping fragments using two separate pairs of primers: 27F (5'-AGAGTTTGATCMTGGCTCAG-3') and OXY1313R (5'-CTTCACGTAGGCGAGTTGCAGC-3'), and OXY107F (5'-GGACGGGTGAGTAACGCGTGR-3') and 1518R (5'-AAGGAGGTGATCCANCCRCA-3'). Bi-directional sequencing was performed directly on the PCR products using an ABI 373A automated sequencer (Applied Biosystems). For the Synechococcus sp. strains WH7803, WH8018 and WH8103, 16S rRNA genes were amplified by the PCR using the primers 27F and 1518R and inserted into the pCRII TA vector (Invitrogen). For the non-axenic Synechococcus sp. strain WH8018, clones containing cyanobacterial sequences were screened by PCR amplification with the primers OXY107F and OXY1313R. Clones were bi-directionally sequenced as described for the Prochlorococcus 16S rRNA genes. Sequence alignment and phylogenetic analysis was carried out using the ARB sequence database tools (http://www.mikro.biologie.tu-muenchen.de/pub/ARB/). Initial analysis was carried out using the maximum-likelihood method, the maximum-parsimony method and evolutionary distance (Jukes and Cantor model for DNA) with almost full-length sequences, using 1364 positions after omitting positions where there were gaps, incomplete data or sequence ambiguity. Shorter sequences were then added in to each of the three resulting trees by the maximum-parsimony method. A consensus tree (based on the maximum-likelihood tree) was drawn to represent the combined results of the three trees and multifurcations were created when the branching order was not supported by all three methods.
Probes.
The ARB sequence database tools were used for probe design and probe matching. The probe sequences and their specificities are shown in Table 1. The bacterial probe EUB338 (5'-GCTGCCTCCCGTAGGAGT-3') (Amann et al., 1990
) and the oligonucleotide NON338 (5'-ACTCCTACGGGAGGCAGC-3') (Wallner et al., 1993
), which served respectively as positive and negative controls, were monolabelled with the fluorochromes CY3 and BODIPY-FL (Interactiva).
|
Cell fixation of cultured isolates.
Prochlorococcus and Synechococcus cells (approx. 2 ml of liquid cultures) were pelleted by centrifugation (10000 g, 5 min) and resuspended in 200 µl PBS (130 mM sodium chloride, 10 mM sodium phosphate buffer, pH 7·2). Fixation was performed either by mixing immediately with 1 vol. ethanol (96%, v/v) or by adding paraformaldehyde (4%, w/v, in PBS) to a final concentration of 3% (w/v) and incubation for 1 h on ice. After fixation with paraformaldehyde, the cells were washed once with PBS, resuspended in the same buffer and then mixed with 1 vol. ethanol (96%, v/v). The fixed samples were stored at -20 °C.
Sampling and fixation of natural samples.
Water samples from the North Atlantic Ocean were collected from discrete depths (3, 20, 30, 40, 60, 80 and 100 m) on 14 July (36·82° N, 19·24° W; Station 25) and 15 July (36·63° N, 19·24° W; Station 26) 1996 with a rosette of 25-litre Go-Flo bottles during a Natural Environment Research Council, Plankton Reactivity in the Marine Environment (PRIME) cruise, aboard the RRS Discovery. Aliquots (1 ml) were immediately fixed by adding glutaraldehyde to a final concentration of 0·1% (v/v) prior to freezing in liquid nitrogen.
Samples from the Red Sea were also taken from discrete depths (10, 30, 50, 80, 100 and 150 m) on 18 October 1999 at Station A (29·28° N, 34·55° E) (Lindell & Post, 1995 ) using a hydrographic cable with 30-litre Niskin bottles. Samples were fixed in two different ways: (i) 100 ml seawater (for depths 10, 30, 50 and 80 m) or 150 ml seawater (for depths 100 and 150 m) were vacuum-filtered onto polycarbonate membranes (GTTP 4700, 0·2 µm pore size, Millipore) using a filter tower with a diameter of 40 mm and the cells were fixed by overlaying the membranes for 1 h with 5 ml paraformaldehyde (1%, w/v). The fixative was then drawn through the membrane and washed with 5 ml PBS. (ii) 50 ml seawater was fixed with glutaraldehyde (0·1%, v/v, final concentration) for 10 min at 4 °C. The paraformaldehyde-fixed filters and the 50 ml glutaraldehyde-fixed liquid samples were then stored at -20 °C.
Flow cytometric data for Prochlorococcus, Synechococcus and heterotrophic bacteria cell numbers were obtained as described previously (Jacquet et al., 1998 ).
In situ hybridization
Preparation of cells of cultured Prochlorococcus and Synechococcus strains.
Fixed cells of cultured Prochlorococcus and Synechococcus strains were spotted into wells on microscope slides (Paul Marienfeld KG, Germany), dried at 45 °C for 5 min and then dehydrated in an ethanol series, 50%, 80% and 96% (v/v), for 3 min each. The slides were then air-dried and the cells permeabilized by adding 30 µl lysozyme solution [1 or 5 mg ml-1of lysozyme (109000 U mg-1; Fluka) dissolved in 100 mM Tris/HCl (pH 8·0), 50 mM EDTA] and incubating for 30 min at 37 °C. After rinsing the slides with distilled water, the cells were dehydrated as described above.
Preparation of natural samples.
For in situ hybridization of the North Atlantic samples, aliquots (400 µl) of the glutaraldehyde-fixed samples were filtered onto polycarbonate filters (GTTP 2500; 0·2 µm pore size; Millipore) using a filter tower with a diameter of 5 mm. The filters were then washed once with 1 ml PBS and divided into quarters. For the Red Sea glutaraldehyde-fixed samples, each sample (50 ml) was filtered onto a single polycarbonate filter, but using a filter tower with a diameter of 20 mm. After filtration, the filters were washed with 2 ml PBS. Further treatment was as described for cultured strains except that the first ethanol dehydration step was omitted.
Hybridization with fluorescently labelled probes.
Hybridizations with fluorochrome-labelled probes were performed as described previously (Manz et al., 1992 ). The fluorochrome-labelled probes EUB338 and NON338 were hybridized in buffer containing 20% (v/v) formamide.
Hybridization with HRP-labelled probes.
The hybridization procedure was based on those described previously (Amann et al., 1992 ; Schönhuber et al., 1999
). Each well on the microscope slide or filter portion was overlaid with 10 µl hybridization buffer [0·9 M NaCl, 20 mM Tris/HCl (pH 7·2), 0·01% (w/v) SDS and 2% blocking reagent (Boehringer Mannheim)] which contained 50 ng of an HRP-labelled probe and different concentrations of formamide depending on the probe used. The formamide concentrations used in hybridizations with the HRP-labelled probes were as follows: 405PRO, 405SYN and 1445NATL, 60%; 645HLI, 181HLI, 181LL and CYA664, 55%; 645HLII, 645LL and NON338, 50%. Probe NON338 was always used as a negative control in hybridizations with cultured Prochlorococcus and Synechococcus strains and with environmental samples. Slides were incubated in humid chambers at 35 °C for 2 h and then washed at the same temperature for 30 min in 50 ml wash buffer [20 mM Tris/HCl (pH 7·2), 5 mM EDTA, 0·01% (w/v), SDS] containing NaCl at molarities equivalent to the formamide concentrations present in the hybridization buffer (0·019 M, 0·013 M and no NaCl for formamide concentrations of 50%, 55% and 60%, v/v, respectively). Hybridization of HRP-labelled probes was detected initially by incubation with the non-fluorescent substrate TETON [4-(1,4,7,10-tetraoxadecyl)-1-naphthol; Boehringer Mannheim] as described previously (Schönhuber et al., 1999
). For detection with the fluorescent substrate fluoresceintyramide (Schönhuber et al., 1997
, 1999
), slides were first equilibrated in TNT buffer [0·1 M Tris/HCl (pH 7·5), 0·15 M NaCl, 0·05 % (v/v), Tween 20] for 15 min at room temperature. After drying around each well/filter portion, 10 µl fluoresceintyramide substrate solution [1:1 ratio of amplification diluent (a constituent of the TSA kit) and 40% (w/v), dextran sulphate with 1/50 vol. fluoresceintyramide] was added to each well/filter and then incubated at room temperature in the dark for 1060 min. The wells or filters were then carefully rinsed with distilled water and washed in TNT buffer at 55 °C for 10 min. After a final rinse in distilled water, the slides were dried and mounted in a 4:1 mixture respectively of Citifluor (Agar Scientific) and Vectashield (Vector Laboratories). Filters were stained with 4',6-diamidino-2-phenylindole (DAPI) prior to mounting (see Counting on filters section). Slides were examined using either a Zeiss Axiophot2 microscope (Carl Zeiss) equipped for epifluorescence with a high-pressure mercury bulb (100 W) and specific filter sets 01 (excitation 365/12 nm, emission LP 397 nm) and 09 (excitation BP450490, emission LP 520 nm) (Carl Zeiss) and Chroma HQ 41007 (Chroma Tech.), or an Olympus BX-60 microscope (Olympus Optical) equipped with a high-pressure mercury bulb (100 W) and specific filter sets U-MWU (excitation BP330385 nm, BA420 nm) and U-MNB (excitation BP470490 nm, BA515 nm) (Olympus Optical). Epifluorescence micrographs and phase-contrast images were captured onto Kodak film P1600X. The epifluorescence micrographs of cells hybridized with a positive or negative probe were taken at the same exposure in order to provide a fair comparison.
Counting on filters.
To obtain total bacterial counts filters were counterstained with DAPI following hybridization. Filters were stained for 7 min with a solution (about 5 µl) of DAPI at a concentration of 1 µg ml-1 in water and then rinsed briefly by dipping in sterile distilled water, followed by 96% (v/v) ethanol. Filters were air-dried and placed onto glass slides in the mounting medium described above. For the North Atlantic samples, 20 randomly chosen fields of view were counted using the DAPI filter set to enumerate total cells and the fluorescein filter set for the probe-specific cell counts. For each microscope field, probe-specific counts were expressed as a percentage of the total cells visualized by DAPI and the mean abundance and standard deviation calculated from these values. The total numbers of cells counted per filter ranged from 360 to 1280. For the Red Sea samples, for which greater volumes of sample were available, 10 randomly chosen fields of view were counted in the same way as for the North Atlantic samples and the total number of cells counted per filter was at least 1400. Filters for which no signals were observed after inspecting at least 20 random fields of view were deemed negative and were not counted.
![]() |
RESULTS |
---|
![]() ![]() ![]() ![]() ![]() ![]() ![]() |
---|
|
|
|
In order to improve the positive:negative probe signal ratio, HRP-labelled oligonucleotides were used in hybridizations with Prochlorococcus cells. With this approach the hybridization signal is amplified by the activity of the probe-conferred enzyme, leading to a deposit of multiple substrate molecules in the cell.
Whole-cell hybridization using HRP-labelled oligonucleotides
Probe CYA664, which is complementary to 16S rRNA gene sequences of all known Prochlorococcus and marine Synechococcus strains and to some chloroplasts (Schönhuber et al., 1999 ), was initially employed as an alternative to probe EUB338 to test the utility of HRP-labelled oligonucleotides for whole-cell hybridizations with cultured strains. As for EUB338, its target region in the 16S rRNA of E. coli is readily accessible (Fuchs et al., 1998
). Hybridization was detected using either the non-fluorescent substrate TETON or the fluorescent substrate fluoresceintyramide. With the former, revealing hybridization by production of a blue precipitate, no positive results were obtained. In contrast, a 10 min incubation with the fluoresceintyramide substrate yielded extremely bright green single-cell hybridization signals with all strains examined (see for example Fig. 2e
, left panel).
|
Evaluation of probe specificities
The optimal hybridization conditions were determined for all of the genotype-specific probes listed in Table 1 by hybridizing each of them to cells of all of the cultured Prochlorococcus and Synechococcus strains (see section above) at different formamide concentrations (30, 40, 50, 55 and 60%). Although the target 16S rRNA sequence among the strains examined often differed by a single mismatch only, specific hybridization for each probe was clearly observed at formamide concentrations of 50, 55 or 60% (see Methods for probe-specific formamide concentrations). Hybridizations were performed at the optimal formamide concentrations at least twice and the specificity of each probe was consistently observed. The clear difference between a positive and negative hybridization signal is shown in Fig. 2(a
e
), where three Prochlorococcus strains representative of HLI, HLII and LL genotypes respectively and two Synechococcus strains were each hybridized with a positive (left panel) and negative probe (right panel). The autofluorescence of the Prochlorococcus cells hybridized with the negative probe was very low (Fig. 2a
c
, right panel) in contrast to Synechococcus cells, which showed significant yellow/orange autofluorescence (Fig. 2d
and e
, right panel). Despite the strong autofluorescence of the Synechococcus cells, the difference between a negative signal (yellow/orange) and a positive signal (green) could clearly be observed. Each epifluorescence photomicrograph has been presented beside the corresponding phase-contrast image to demonstrate the efficient labelling of all cells. The signal intensities observed for each probe were very similar, in spite of potentially more difficult accessibility of the 645 target site compared to that of the 181 region of the 16S rRNA (Fuchs et al., 1998
). Figs 2(a)
and 2(c)
(right panels) demonstrate the specificity of the 181LL and 645HLI probes respectively, which both distinguish between Prochlorococcus sp. PCC9511 and NATL1MIT by only a single mismatch in the respective rRNA target sites (Table 1
). Furthermore, probe 181LL did not hybridize to Synechococcus sp. strain WH8103 despite there also being only a single mismatch between the 16S rRNA sequence and the probe (Fig. 2e
). The 16S rRNA sequences of some marine Synechococcus strains are targeted by probe 181LL (see Table 1
) and give a positive hybridization signal as illustrated for strain WH8018 (Fig. 2d
).
Application of probes to identify Prochlorococcus sp. genotypes in the North Atlantic and the Red Sea
The seawater samples available to us from the North Atlantic had originally been fixed with glutaraldehyde for flow cytometric analysis. Nevertheless, preliminary hybridizations with selected samples from different depths of Station 26 in the North Atlantic yielded very bright signals when using probes CYA664 and 405PRO. For the genotype-specific probes the signals were lower and the order of brightness was subjectively assessed as: 181LL>645HLI>645LL. Since hybridization of probe 181HLI to the filters resulted in a high background that masked positive signals, this probe was not used further. No positive signals were observed by hybridization of probe 645HLII to all of the samples from Stations 25 and 26. Some larger cells were observed after hybridization to probe CYA664 and these cells were believed to be representatives of Synechococcus. The number of these cells was very low, consistent with flow cytometric data for the days sampled at either side of Stations 25 and 26, which indicated that Prochlorococcus sp. accounted for approximately 8097 % of the total picophytoplankton (West & Scanlan, 1999 ).
The first results of enumerating hybridizing cells as the fraction (mean±SD) of total bacterial counts by DAPI staining from different depths (3, 20 and 40 m) of Station 25 in the North Atlantic Ocean are shown in Table 2. The mean fraction of cells collected from 20 and 40 m and hybridizing with the two general probes CYA664 and 405PRO were very similar. This suggests, in agreement with previous data (West & Scanlan, 1999
), that the cyanobacterial community in these water columns is almost entirely composed of members of the genus Prochlorococcus, which are all recognized by probe 405PRO. For the 3 m sample, there was also good agreement between the mean fraction of cells hybridizing with probe 645HLI compared to probes CYA664 and 405PRO, suggesting that the genotypes detected were almost entirely assignable to the HLI cluster. However, for the deeper samples (20 and 40 m) the sum of cells detected by the 645HLI and 645LL genotype-specific probes only accounted for about 50% of the total mean fraction of Prochlorococcus sp. determined with probes CYA664 and 405PRO, one or both of the two genotypes evidently being underestimated. Indeed, the signal intensities observed in samples from deeper depths with both genotype-specific probes were lower than those obtained for the 3 m sample, which no doubt contributed to the lower counts. These initial hybridizations on filters were carried out using a fluoresceintyramide incubation time of 30 min. In an attempt to improve the signals with the genotype-specific probes, subsequent substrate incubation times were extended to 1 h. Additionally, probe 181LL was selected for hybridizations in preference to probe 645LL, since it gave higher signals (see above). With these changes, the mean fraction of total Prochlorococcus sp. detected by CYA664 and the sum of genotype-specific fractions were in much better agreement. This is shown in Table 2
, both for a sample from 80 m at Station 25 and for all samples of Station 26. The typical signals observed by hybridization with probes 645HLI and 181LL, following the improved protocol, are shown in Fig. 3(a)
and Fig. 3(b)
. For the 80 m sample of Station 25, probe 181LL detected 70% of the mean fraction of cells identified by the CYA664 probe. The results for the samples from Station 26 showed an even better agreement between the enumerations made with the two genotype-specific probes, which together accounted for 8795% of the mean fraction determined by hybridizations with probe CYA664. A comparison of the hybridization results and enumerations made by DAPI staining indicate that at both stations HLI-adapted Prochlorococcus sp. dominated the surface water, representing approximately 25% of the total bacterial counts, whereas at 80 m, the LL-adapted genotypes were dominant and the proportion of both genotypes together decreased to about 14% of total counts. Additionally, at a depth of 40 m at Station 25, both HLI- and LL-adapted Prochlorococcus sp. genotypes were detected.
|
|
The mean fraction of Prochlorococcus, or Prochlorococcus plus Synechococcus, as a percentage of total bacteria as assessed by in situ hybridization was then compared to that estimated by flow cytometry (Table 4). The two independent methods gave similar results although the estimates by in situ hybridization were often slightly lower than those obtained by flow cytometry, except for the 100 m sample, where the converse was observed.
|
![]() |
DISCUSSION |
---|
![]() ![]() ![]() ![]() ![]() ![]() ![]() |
---|
Although initially it seemed possible that the permeability of the cells was inadequate, the success of the HRP-labelled probes combined with the TSA detection system showed this not to be the case. In contrast, the alternative detection system offered by the non-fluorescent, precipitable substrate TETON, although successful in hybridizations with some cyanobacteria (Schönhuber et al., 1999 ), was not suitable for use in Prochlorococcus. It is possible that, as a consequence of a low rRNA content together with the small cells (less than 0·8 µm in diameter) typical for members of this genus, the number of probe-conferred HRP molecules is too low to allow the formation of sufficient oxidized substrate for the production of a visible precipitate (Schönhuber et al., 1999
). For a given tyramide incubation time, the fluorescent signals from cells of cultured Prochlorococcus marinus and Prochlorococcus sp. isolates were significantly brighter than those obtained in situ, particularly in deep-water samples. An explanation for the lower hybridization signals observed in cells of Prochlorococcus sp. from natural samples may be a low rRNA content per cell due to slow growth. Indeed, cell cycle analysis of natural Prochlorococcus sp. populations in the equatorial Pacific Ocean has revealed that deep populations grow more slowly than surface populations (Vaulot et al., 1995
). However, since surface populations may exhibit growth rates equivalent to those observed in cultured isolates (about one doubling per day) (Moore & Chisholm, 1999
), there must be additional factors which may contribute to the reduced hybridization signals observed in natural samples. It is possible that there are structural or physiological differences between cultured isolates and natural populations that, in the latter, decrease the permeability of the cells or perhaps reduce the accessibility of the 16S rRNA target sites.
For the initial enumerations made at Station 25, the mean fraction of cells detected by the genotype-specific probes 645HL and 645LL was too low to account for the total number of Prochlorococcus sp. determined either with CYA664 or 405PRO. However, the replacement of probe 645LL by probe 181LL, and the extension of the fluoresceintyramide incubation time, enhanced the signals so that the mean fraction of cells hybridizing with the genotype-specific probes was raised to the level determined with the CYA664 probe. Thus it is most likely that in the earlier enumerations a proportion of the hybridizing cells gave signals that fell below the detection threshold, rather than the discrepancies in the analyses being due to the existence of genotypes other than those that can be recognized by the three genotype-specific probes employed. The dominance of Prochlorococcus sp., the observed transition of HLI- to LL-adapted genotypes from surface to deep waters respectively, and the absence of HLII genotypes from these particular sampling stations in the North Atlantic Ocean, are entirely consistent with previous molecular data (West & Scanlan, 1999 ). The degree of variation of the data as observed by the mean±SD values was relatively high due to the low numbers of cells and their uneven distribution on the filters. Nevertheless, there was good agreement between the mean fractions of total Prochlorococcus as a percentage of total bacteria as assessed by probes 405PRO and CYA664 compared to the sum of the two Prochlorococcus genotypes.
In situ hybridization of Prochlorococcus cells sampled from the Red Sea revealed a remarkably different community structure, both in the genotypes present and in their vertical distribution. Despite the stratification of the water column (Fig. 4b), HLII genotypes were dominant at all depths and even outnumbered LL genotypes by around 4 times at 100 m, the only depth at which LL genotypes were observed in significant numbers. The dominance of HLII genotypes in this oceanic region is further supported by the retrieval of environmental sequences of the ntcA gene (encoding a cyanobacterial-specific transcriptional activator involved in global nitrogen control) since they were found to cluster together with the ntcA genes of HLII cultured Prochlorococcus strains (Lindell, 2000
). The ability of HLII genotypes to colonize the euphotic zone from the surface to a depth below the 1% light level is remarkable and differs from the Prochlorococcus community structure in the eastern North Atlantic Ocean, where the HL- and LL-adapted populations were clearly partitioned (West & Scanlan, 1999
) (see Table 2
). The data for the Red Sea are consistent with flow cytometric observations on samples from the same water column in November 1998 that contained dimly fluorescent (HL-adapted) Prochlorococcus cells from surface water down to 120 m. This depth distribution partly overlapped with that of strongly fluorescent (LL-adapted) cells that were observed from 80 m down to 175 m (A. F. Post, unpublished results). Photophysiological studies of HL-adapted Prochlorococcus strains have not indicated that the HLII genotype strains have greater photoacclimative abilities than HLI genotype strains (Moore & Chisholm, 1999
). However, these studies also revealed that at light levels corresponding to a depth of 118 m, some HL- and LL-adapted strains had similar growth rates, which may help explain the coexistence of HLII and LL Prochlorococcus at 100 m in the Red Sea. It is possible that in deeper layers the HLII cells are present but are growing slowly. Therefore it would be very interesting to carry out an in situ hybridization study in parallel with flow cytometric cell cycle analysis (see Vaulot et al., 1995
) in order to estimate the growth rates of cells at different depths.
Despite the use of a tyramide incubation time of 1 h, hybridization of the genotype-specific probes to the Red Sea samples gave a lower estimation of Prochlorococcus cell numbers compared to the group-specific probes. This may be explained by an underestimation of HLII genotypes due to the lower signal conferred by this probe, or could also be due to the presence of genotypes not targeted by our probes. Since the HLII probe targets a region of the 16S rRNA that may be less accessible to oligonucleotides (Fuchs et al., 1998 ), it is worth testing the use of unlabelled helper oligonucleotides which are designed to bind to regions adjacent to that of the target probe and were shown to improve the hybridization signals in E. coli (Fuchs et al., 2000
). Future retrieval and sequencing of further Prochlorococcus environmental 16S rDNA clones will be required in order to verify that all of the genotypes present are indeed targeted by the probes described here. The statistical variation observed within the Red Sea data was less than that of the North Atlantic samples due to a higher density of cells on the filters, which allowed a greater number of cells to be counted. Despite this variation, the fact that the mean fractions of Prochlorococcus assessed by in situ hybridization agreed so well with Prochlorococcus abundance determined by flow cytometry (Table 4
) lends strong credence to the use of this in situ hybridization protocol in quantitative analyses.
The original examination of 16S rRNA sequences obtained for a number of Prochlorococcus strains isolated from surface and deep waters from different oceanic regions demonstrated a clear correlation with their depth of isolation (Scanlan et al., 1996 ) whilst closely related strains were found in different geographical locations such as the Atlantic and Pacific Oceans (Urbach et al., 1998
). These findings are extended in this study, which showed that LL genotypes comparable to the NATL1MIT/NATL2B isolates seem to be ubiquitous, having been detected only at depth, but in both the Red Sea and the North Atlantic. However, for the HL genotypes the picture is more complicated. The results presented here, and other preliminary data which show a dominance of HLI in the Mediterranean Sea and HLII in the Sargasso (N. West & D. Scanlan, unpublished data), indicate that the relative abundance of the two types of HL populations and their vertical distribution may differ significantly with geographical location. This may reflect other more subtle features of the water column itself such as salinity, temperature, prevailing oceanic circulation or the source of the Prochlorococcus seed population.
Our results demonstrate that an in situ hybridization method can be used to rapidly assess the community structure of Prochlorococcus sp. in the natural environment. Moreover, using this technique, estimation of Prochlorococcus abundance, as a proportion of total bacteria, can be directly correlated with flow cytometric data for Prochlorococcus and bacterial cell numbers, and the two independent methods have been shown to compare favourably. The fixation of samples is simple and the analysis does not require time-consuming DNA extraction procedures and is free of the potential biases of the PCR. The counting of cells using epifluorescence microscopy is the most labour-intensive part of the method and this step may be improved in the future by using a laser scanning cytometer for the counting of cells on a filter (Parthuisot et al., 2000 ) or by optimizing the hybridization method in liquid for analysis by flow cytometry. Both of these methods would allow the rapid counting of large numbers of cells, either at the genus level or of particular genotypes, thus providing very precise estimates of population dynamics. Moreover, by combining in situ hybridization analyses using 16S rRNA-targeted oligonucleotides with single-cell immunofluorescence assays for protein markers of nutrient stress (Scanlan et al., 1997
), it should be possible to determine the specific physiological state of individual Prochlorococcus sp. genotypes. Such a multiple-probing approach will be essential to elucidate the nutritional adaptation and functional importance of the various ecotypes of Prochlorococcus sp. within natural communities.
![]() |
ACKNOWLEDGEMENTS |
---|
We thank Thérèse Coursin (Institut Pasteur) for making available the 16S rRNA sequence of Prochlorococcus sp. strain PCC 9511 prior to publication, Mike Zubkov for providing the PRIME natural samples and for helpful discussions, Dominique Marie for Red Sea flow cytometric data, and Fred Partensky, Jean Blanchot, Florence Legall and Solange Legall for providing Prochlorococcus sp. cultures.
![]() |
REFERENCES |
---|
![]() ![]() ![]() ![]() ![]() ![]() ![]() |
---|
Amann, R. I., Zarda, B., Stahl, D. A. & Schleifer, K.-H. (1992). Identification of individual prokaryotic cells with enzyme-labeled, rRNA-targeted oligonucleotide probes. Appl Environ Microbiol 58, 3007-3011.[Abstract]
Amann, R. I., Ludwig, W. & Schleifer, K. H. (1995). Phylogenetic identification and in situ detection of individual microbial cells without cultivation. Microbiol Rev 59, 143-169.[Abstract]
Amann, R., Glöckner, F. O. & Neef, A. (1997). Modern methods in subsurface microbiology: in situ identification of microorganisms with nucleic acid probes. FEMS Microbiol Rev 20, 191-200.
Binder, B. J. & Liu, Y. C. (1998). Growth rate regulation of rRNA content of a marine Synechococcus (cyanobacterium) strain. Appl Environ Microbiol 64, 3346-3351.
Brosius, J., Dull, T., Sleeter, D. D. & Noller, H. F. (1981). Gene organization and primary structure of a ribosomal RNA operon from Escherichia coli. J Mol Biol 148, 107-127.[Medline]
Capone, D. G., Zehr, J. P., Paerl, H. W., Bergman, B. & Carpenter, E. J. (1997). Trichodesmium, a globally significant marine cyanobacterium. Science 276, 1221-1229.
Chisholm, S. W., Frankel, S. L., Goericke, R., Olson, R. J., Palenik, B., Waterbury, J. B., Westjohnsrud, L. & Zettler, E. R. (1992). Prochlorococcus marinus nov. gen. nov. sp.: an oxyphototrophic marine prokaryote containing divinyl chlorophyll a and b. Arch Microbiol 157, 297-300.
Ferris, M. J. & Palenik, B. (1998). Niche adaptation in ocean cyanobacteria. Nature 396, 226-228.
Fuchs, B. M., Wallner, G., Beisker, W., Schwippl, I., Ludwig, W. & Amann, R. (1998). Flow cytometric analysis of the in situ accessibility of Escherichia coli 16S rRNA for fluorescently labeled oligonucleotide probes. Appl Environ Microbiol 64, 4973-4982.
Fuchs, B. M., Glöckner, F. O., Wulf, J. & Amann, R. (2000). Unlabeled helper oligonucleotides increase the in situ accessibility to 16S rRNA of fluorescently labeled oligonucleotide probes. Appl Environ Microbiol 66, 3603-3607.
Glöckner, F. O., Amann, R., Alfreider, A., Pernthaler, J., Psenner, R., Trebesius, K. & Schleifer, K. H. (1996). An in-situ hybridization protocol for detection and identification of planktonic bacteria. Syst Appl Microbiol 19, 403-406.
Jacquet, S., Lennon, J. F., Marie, D. & Vaulot, D. (1998). Picoplankton population dynamics in coastal waters of the northwestern Mediterranean Sea. Limnol Oceanogr 43, 1916-1931.
Lebaron, P., Catala, P., Fajon, C., Joux, F., Baudart, J. & Bernard, L. (1997). A new sensitive, whole cell hybridization technique for detection of bacteria involving a biotinylated oligonucleotide probe targeting rRNA and tyramide signal amplification. Appl Environ Microbiol 63, 3274-3278.[Abstract]
Lindell, D. (2000). Assessing the nitrogen status of marine prokaryotic phytoplankton using molecular methods. PhD thesis, Hebrew University of Jerusalem.
Lindell, D. & Post, A. F. (1995). Ultraphytoplankton succession is triggered by deep winter mixing in the Gulf of Aqaba (Eilat), Red Sea. Limnol Oceanogr 40, 1130-1141.
Manz, W., Amann, R., Ludwig, W., Wagner, M. & Schleifer, K.-H. (1992). Phylogenetic oligodeoxynucleotide probes for the major subclasses of proteobacteria: problems and solutions. Syst Appl Microbiol 15, 593-600.
Moore, L. R. & Chisholm, S. W. (1999). Photophysiology of the marine cyanobacterium Prochlorococcus: ecotypic differences among cultured isolates. Limnol Oceanogr 44, 628-638.
Moore, L. R., Rocap, G. & Chisholm, S. W. (1998). Physiology and molecular phylogeny of coexisting Prochlorococcus ecotypes. Nature 393, 464-467.[Medline]
Neilan, B. A., Jacobs, D., Del Dot, T., Blackall, L. L., Hawkins, P. R., Cox, P. T. & Goodman, A. E. (1997). rRNA sequences and evolutionary relationships among toxic and nontoxic cyanobacteria of the genus Microcystis. Int J Syst Bacteriol 47, 693-697.
Otsuka, S., Suda, S., Li, R. H., Watanabe, M., Oyaizu, H., Matsumoto, S. & Watanabe, M. M. (1998). 16S rDNA sequences and phylogenetic analyses of Microcystis strains with and without phycoerythrin. FEMS Microbiol Lett 164, 119-124.
Partensky, F., Hess, W. R. & Vaulot, D. (1999). Prochlorococcus, a marine photosynthetic prokaryote of global significance. Microbiol Mol Biol Rev 63, 106-127.
Parthuisot, N., Catala, P., Lemarchand, K., Baudart, J. & Lebaron, P. (2000). Evaluation of ChemChrome V6 for bacterial viability assessment in waters. J Appl Microbiol 89, 370-380.[Medline]
Rippka, R., Coursin, T., Hess, W. & 7 other authors (2000). Prochlorococcus marinus Chisholm et al., 1992 subsp. pastoris subsp. nov. strain PCC 9511, the first axenic chlorophyll a2/b2-containing cyanobacterium (Oxyphotobacteria). Int J Syst Evol Microbiol 50, 18331847.
Scanlan, D. J., Hess, W. R., Partensky, F., Newman, J. & Vaulot, D. (1996). High degree of genetic variation in Prochlorococcus (Prochlorophyta) revealed by RFLP analysis. Eur J Phycol 31, 1-9.
Scanlan, D. J., Silman, N. J., Donald, K. M., Wilson, W. H., Carr, N. G., Joint, I. & Mann, N. H. (1997). An immunological approach to detect phosphate stress in populations and single cells of photosynthetic picoplankton. Appl Environ Microbiol 63, 2411-2420.[Abstract]
Schönhuber, W., Fuchs, B., Juretschko, S. & Amann, R. (1997). Improved sensitivity of whole-cell hybridization by the combination of horseradish peroxidase-labeled oligonucleotides and tyramide signal amplification. Appl Environ Microbiol 63, 3268-3273.[Abstract]
Schönhuber, W., Zarda, B., Eix, S., Rippka, R., Herdman, M., Ludwig, W. & Amann, R. (1999). In situ identification of cyanobacteria with horseradish peroxidase-labeled, rRNA-targeted oligonucleotide probes. Appl Environ Microbiol 65, 1259-1267.
Urbach, E. & Chisholm, S. W. (1998). Genetic diversity in Prochlorococcus populations flow cytometrically sorted from the Sargasso Sea and Gulf Stream. Limnol Oceanogr 43, 1615-1630.
Urbach, E., Scanlan, D. J., Distel, D. L., Waterbury, J. B. & Chisholm, S. W. (1998). Rapid diversification of marine picophytoplankton with dissimilar light-harvesting structures inferred from sequences of Prochlorococcus and Synechococcus (Cyanobacteria). J Mol Evol 46, 188-201.[Medline]
Urdea, M. S., Warner, B. D., Running, J. A., Stempien, M., Clyne, J. & Horn, T. (1988). A comparison of non-radioisotopic hybridization assay methods using fluorescent, chemiluminescent, and enzyme labeled synthetic oligodeoxyribonucleotide probes. Nucleic Acids Res 16, 4937-4956.[Abstract]
Vaulot, D., Marie, D., Olson, R. J. & Chisholm, S. W. (1995). Growth of Prochlorococcus, a photosynthetic prokaryote, in the equatorial Pacific Ocean. Science 268, 1480-1482.
Wallner, G., Amann, R. & Beisker, W. (1993). Optimizing fluorescent in situ hybridization with ribosomal RNA-targeted oligonucleotide probes for flow cytometric identification of microorganisms. Cytometry 14, 136-143.[Medline]
West, N. J. & Scanlan, D. J. (1999). Niche-partitioning of Prochlorococcus populations in a stratified water column in the eastern North Atlantic Ocean. Appl Environ Microbiol 65, 2585-2591.
von Wintzingerode, F., Göbel, U. B. & Stackebrandt, E. (1997). Determination of microbial diversity in environmental samples: pitfalls of PCR-based rRNA analysis. FEMS Microbiol Rev 21, 213-229.[Medline]
Worden, A. Z., Chisholm, S. W. & Binder, B. J. (2000). In situ hybridization of Prochlorococcus and Synechococcus (marine cyanobacteria) spp. with rRNA-targeted peptide nucleic acid probes. Appl Environ Microbiol 66, 284-289.
Wyman, M., Gregory, R. P. F. & Carr, N. G. (1985). Novel role for phycoerythrin in a marine cyanobacterium, Synechococcus strain DC2. Science 230, 818-820.
Received 20 November 2000;
revised 14 March 2001;
accepted 28 March 2001.