Rhamnose biosynthesis is not impaired by the deletion of putative rfbC genes, slr0985 and slr1933, in Synechocystis sp. PCC 6803

Abstract

Cyanobacterial extracellular polymeric substances (EPS) mainly composed of heteropolysaccharides can be attached to the cell wall as capsular polysaccharides (CPS) or released to the environment as released polysaccharides (RPS). These polymers have an unusually high diversified monosaccharidic composition, making them attractive for biotechnological/biomedical applications. However, their production is still poorly understood, hindering their optimization for industrial needs. This work aimed to better understand the biosynthesis of the 6-deoxy sugars, fucose and rhamnose, in the model cyanobacterium Synechocystis sp. PCC 6803. To that end, genes encoding proteins putatively involved in the biosynthesis of GDP-L-fucose [sll1213 (fucS)] and dTDP-L-rhamnose [slr0985 (rfbC1) and slr1933 (rfbC2)] were deleted. As previously observed, ΔfucS had significant growth impairment, and its RPS did not contain any fucose or rhamnose. Here, we also showed that both deoxyhexoses' pathways are completely impaired in ΔfucS. In contrast, both ΔrfbC1 and ΔrfbC1ΔrfbC2, although producing significantly less RPS and more CPS than the wild type, did not show major differences regarding the RPS monosaccharidic composition. These results strongly suggest that their gene products are not essential for rhamnose biosynthesis. Transcriptional analysis revealed that one of the gmd genes (slr1072) putatively encoding a GDP-mannose 4,6-dehydratase was upregulated in all the knockout strains and that the three EPS-related genes in the same operon as rfbC1 (slr0982, slr0983, and slr1610) were upregulated in both ΔrfbC strains. Altogether, our results reveal that rhamnose biosynthesis in Synechocystis depends on FucS but not on the putative RfbC enzymes, underlining the need to further elucidate the mechanisms involved in the biosynthesis of this deoxyhexose.IMPORTANCEThis study contributes to the overall knowledge of deoxyhexoses' biosynthesis in Synechocystis sp. PCC 6803. Here, we demonstrated that the ΔfucS strain not only produces EPS without fucose and rhamnose, but that both pathways are completely impaired. Furthermore, we also showed that the deletion of both putative rfbC genes does not affect rhamnose biosynthesis despite having an impact on carbohydrate production/export, shifting RPS to CPS production. Altogether, our results suggest that the rfbC genes are not correctly annotated and highlight the intricacies and/or potential crosstalk between the two deoxyhexose pathways, yet to be completely unraveled in Synechocystis. The understanding of the cyanobacterial EPS assembly and export is crucial for the optimization of their production and tailoring for industrial/commercial applications.

Keywords: Synechocystis; cyanobacteria; deoxyhexoses; extracellular polymeric substances (EPS); fucose; rhamnose.

Fig 1

Biosynthetic pathways and genomic context of genes involved in the synthesis of fucose and rhamnose in Synechocystis sp. PCC 6803. (A) Schematic representation of the putative biosynthetic pathways of guanosine diphosphate-L-fucose (GDP-Fuc) and deoxythymidine diphosphate-L-rhamnose (dTDP-Rha). The red line indicates feedback inhibition. (B) Locus of the genes encoding enzymes putatively involved in the biosynthetic pathways of GDP-Fuc and dTDP-Rha, with the gene knockouts in this study highlighted in red, blue, and green. Both the locus tag and the gene name/symbol are provided when available. Unk./Hyp.—genes encoding unknown or hypothetical proteins.

Fig 2

Aggregation/sedimentation and cell wall ultrastructure of Synechocystis sp. PCC 6803 wild type (WT) and ΔfucS, ΔrfbC1, and ΔrfbC1ΔrfbC2 strains. (A) Micrographs of the strains at low cell densities highlighting the clumping phenotype of ΔfucS, scale bar: 20 µm. Insets highlight differences in sedimentation. (B) Sedimentation index (%) at 0, 24, and 48 h. Data represent means ± SD (n = 3). (C) Ultrastructure of the cell wall with the S-layer highlighted (arrow) and inset with the putative detached S-layer in ΔfucS also highlighted (asterisk). Scale bars: 200 nm. Cells were grown in BG11 medium at 30°C under a 12 h light (25 μE m⁻² S⁻¹)/12 h dark regimen, with orbital shaking at 150 rpm.

Fig 3

Growth, total carbohydrates, released polysaccharides, and capsular polysaccharides of Synechocystis sp. PCC 6803 wild type (WT) and ΔfucS, ΔrfbC1, and ΔrfbC1ΔrfbC2 strains. (A) Growth (optical density at 730 nm) and micrograms of chlorophyll a per milliliter of culture and production of (B) total carbohydrates, (C) released polysaccharides, and (D) capsular polysaccharides expressed as micrograms of carbohydrates per microgram of chlorophyll a. Data represent means ± SD (n ≥ 3), and individual measurements are shown. Statistical analysis performed using one-way analysis of variance, followed by Dunnett’s multiple comparisons, is shown for the last timepoint. Significant differences are identified: *(P ≤ 0.05), **(P ≤ 0.01), ***(P ≤ 0.001), and ****(P ≤ 0.0001).

Fig 4

Detection of rhamnose on the cell surface of Synechocystis sp. PCC 6803 wild type (WT) and ΔfucS, ΔrfbC1, and ΔrfbC1ΔrfbC2 strains using a rhamnose-binding protein tagged with GFP (gp17-GFP). (A) Micrographs depicting the red autofluorescence signal and the green signal from the gp17-GFP. The right panels show the merging of the other two micrographs. Scale bar: 10 µm. (B) Quantification of fluorescence associated with gp17-GFP bound to rhamnose at the surface of the cells measured in 96-well microplates. Data represent means ± SD (n = 3), and individual measurements are shown. Statistical analysis consists of one-way analysis of variance, followed by Dunnett’s multiple comparisons. Significant differences are identified: **(P ≤ 0.01) and ****(P ≤ 0.0001).

Fig 5

Analysis of the relative normalized expression of genes putatively related to the biosynthesis of deoxyhexoses and EPS in Synechocystis sp. PCC 6803 wild type (WT) and ΔfucS, ΔrfbC1, and ΔrfbC1ΔrfbC2 strains. RNA was extracted from cells of the different strains collected at three timepoints: 0, 7, and 21 days. RT-qPCR analysis of (A) gmd (slr1072), rfbC1 (slr0985), and rfbC2 (slr1933) expressions in Synechocystis sp. PCC 6803 wild type and ΔfucS, (B) rfbB (slr0982), rfbF (slr0983), slr1610 (putative methyltransferase), gmd (slr1072), fucS (sll1213), and slr1933 (rfbC2) expressions in wild type and ΔrfbC1, and (C) rfbB (slr0982), rfbF (slr0983), slr1610 (putative methyltransferase), gmd (slr1072), and fucS (sll1213) expressions in wild type and ΔrfbC1ΔrfbC2. The normalized fold expression of the target genes relative to wild type is represented at each timepoint. Data from three biological and three technical replicates were normalized against two reference genes (sll1212 and sll1395), error bars represent the standard deviations, and individual measurements are shown. Statistical analysis was performed using t-test, and significant differences are identified: *(P ≤ 0.05), **(P ≤ 0.01), ***(P ≤ 0.001), and ****(P ≤ 0.0001).