Comparison of the seasonal variations of Synechococcus assemblage structures in estuarine waters and coastal waters of Hong Kong

Abstract

Seasonal variation in the phylogenetic composition of Synechococcus assemblages in estuarine and coastal waters of Hong Kong was examined through pyrosequencing of the rpoC1 gene. Sixteen samples were collected in 2009 from two stations representing estuarine and ocean-influenced coastal waters, respectively. Synechococcus abundance in coastal waters gradually increased from 3.6 × 10(3) cells ml(-1) in March, reaching a peak value of 5.7 × 10(5) cells ml(-1) in July, and then gradually decreased to 9.3 × 10(3) cells ml(-1) in December. The changes in Synechococcus abundance in estuarine waters followed a pattern similar to that in coastal waters, whereas its composition shifted from being dominated by phycoerythrin-rich (PE-type) strains in winter to phycocyanin-only (PC-type) strains in summer owing to the increase in freshwater discharge from the Pearl River and higher water temperature. The high abundance of PC-type Synechococcus was composed of subcluster 5.2 marine Synechococcus, freshwater Synechococcus (F-PC), and Cyanobium. The Synechococcus assemblage in the coastal waters, on the other hand, was dominated by marine PE-type Synechococcus, with subcluster 5.1 clades II and VI as the major lineages from April to September, when the summer monsoon prevailed. Besides these two clades, clade III cooccurred with clade V at relatively high abundance in summer. During winter, the Synechococcus assemblage compositions at the two sites were similar and were dominated by subcluster 5.1 clades II and IX and an undescribed clade (represented by Synechococcus sp. strain miyav). Clade IX Synechococcus was a relatively ubiquitous PE-type Synechococcus found at both sites, and our study demonstrates that some strains of the clade have the ability to deal with large variation of salinity in subtropical estuarine environments. Our study suggests that changes in seawater temperature and salinity caused by the seasonal variation of monsoonal forcing are two major determinants of the community composition and abundance of Synechococcus assemblages in Hong Kong waters.

FIG 1

(A) Sampling stations. NM3 is located in the western waters of Hong Kong, which are influenced by freshwater discharge from the Pearl River, whereas PM7 is a typical coastal station located in the eastern waters of Hong Kong. (Adapted from reference with permission of the publisher [copyright 2013 Society for Applied Microbiology and John Wiley & Sons Ltd.].) (B) NMDS plot showing the relationship of hydrographical and trophic conditions at sampling stations.

FIG 2

Abundances and relative abundances of Synechococcus lineages in Hong Kong waters. (A and B) Abundances of PC-type and PE-type Synechococcus in Hong Kong waters measured by flow cytometry analysis (replotted from Liu et al. [21]). (C and D) Calculated abundances of Synechococcus lineages in each month (abundance of each lineage = relative abundance × total Synechococcus abundance measured by flow cytometry). (E and F) Relative abundances of Synechococcus lineages in samples collected in each month. *, S5.1-Undesc is the undescribed clade represented by Synechococcus sp. strain miyav.

FIG 3

Heat map displaying the relative abundances of the top 50 most abundant OTUs across the samples. The data were transformed by square root transformation. The OTU name, lineage, and most similar strain in the NCBI database are shown on the left. OTU14 and OTU42 (red), affiliated with clade IX Synechococcus sp. RS9901, have a niche similar to that of PC-type Synechococcus. *, undescribed clade represented by Synechococcus sp. strain miyav; **, unclassified OTU (the OTU's representative sequence was less than 90% identical to the reference sequences in this study).

FIG 4

Diversity and richness indices of Synechococcus assemblages in Hong Kong waters.

FIG 5

NMDS plot showing the relationship of the Synechococcus communities based on the relative abundance of each Synechococcus lineage. The samples formed three groups, PM7(Summer), NM3(Wet), and Dry. The AMOVA test showed that the compositions of the three groups were significantly different from each other. The sizes of the bubbles indicate the abundance of Synechococcus in each sample (the abundance of each sample was square root transformed).

FIG 6

Four lineages that contribute most to the variation among group PM7(Summer), group NM3(Wet), and group Dry using SIMPER analysis. The bars show the average relative abundance of each lineage in each group; the numbers above the bars are the contributions of each lineage to the dissimilarity between groups. The three groups were determined according to an NMDS plot. *, undescribed clade (Undesc).

FIG 7

Correlation triplot based on an RDA depicting the relationship between the environmental factors and Synechococcus communities. The relative abundance of each lineage was normalized by square root transformation, and environmental data were Z-score transformed. *, factors significantly correlated with variation in the Synechococcus community (P < 0.05).

FIG 8

Growth curves of Synechococcus strains in f/2 media with different salinities. MW03 and WH 5701 are PC-type Synechococcus. WH 8012, WH 7803, PS01, and MW02 are PE-type Synechococcus.