A modular plasmid toolkit applied in marine bacteria reveals functional insights during bacteria-stimulated metamorphosis

Abstract

A conspicuous roadblock to studying marine bacteria for fundamental research and biotechnology is a lack of modular synthetic biology tools for their genetic manipulation. Here, we applied, and generated new parts for, a modular plasmid toolkit to study marine bacteria in the context of symbioses and host-microbe interactions. To demonstrate the utility of this plasmid system, we genetically manipulated the marine bacterium Pseudoalteromonas luteoviolacea, which stimulates the metamorphosis of the model tubeworm, Hydroides elegans. Using these tools, we quantified constitutive and native promoter expression, developed reporter strains that enable the imaging of host-bacteria interactions, and used CRISPR interference (CRISPRi) to knock down a secondary metabolite and a host-associated gene. We demonstrate the broader utility of this modular system for testing the genetic tractability of marine bacteria that are known to be associated with diverse host-microbe symbioses. These efforts resulted in the successful conjugation of 12 marine strains from the Alphaproteobacteria and Gammaproteobacteria classes. Altogether, the present study demonstrates how synthetic biology strategies enable the investigation of marine microbes and marine host-microbe symbioses with potential implications for environmental restoration and biotechnology. IMPORTANCE Marine Proteobacteria are attractive targets for genetic engineering due to their ability to produce a diversity of bioactive metabolites and their involvement in host-microbe symbioses. Modular cloning toolkits have become a standard for engineering model microbes, such as Escherichia coli, because they enable innumerable mix-and-match DNA assembly and engineering options. However, such modular tools have not yet been applied to most marine bacterial species. In this work, we adapt a modular plasmid toolkit for use in a set of 12 marine bacteria from the Gammaproteobacteria and Alphaproteobacteria classes. We demonstrate the utility of this genetic toolkit by engineering a marine Pseudoalteromonas bacterium to study their association with its host animal Hydroides elegans. This work provides a proof of concept that modular genetic tools can be applied to diverse marine bacteria to address basic science questions and for biotechnology innovations.

Keywords: CRISPRi; Hydroides; golden gate; marine; metamorphosis; modular; symbiosis; tubeworm; violacein.

Fig 1

Schematic overview of the modular plasmid system and quantitative promoter measurements. (A) Schematic representation of the modular golden gate assembly plasmid parts with flanking BsaI cut sites (dashed lines). Overlapping 4 bp overhangs are color coordinated. The modular broad-host-range (BHR) backbone (pBTK402) contains inverted BsaI cut sites and an RFP dropout. (B) Golden Gate Assembly is performed in a one-tube reaction by digesting the backbone and insert part plasmids with BsaI and ligating with T4 ligase. (C) A modular stage-1 plasmid is complete when all overlapping inserts are successfully assembled in order. (D and E) Luciferase assays of P. luteoviolacea strains expressing plasmids with different promoters during exponential, stationary, or biofilm growth driving a Nanoluciferase (NLuc) gene where (D) shows CP25-NLuc-T7, PA3-NLuc-T7, Ptac-NLuc-T7 and (E) compares native MACs macS and macB promoters. Luminescence, as relative luminescence units (RLUs), is normalized to optical density at 600 nm (OD₆₀₀) and plotted on a log base 10 scale. The dashed line indicates P. luteoviolacea cells expressing a non-luminescent plasmid as represented by the dotted line (Y = 524 RLU/OD₆₀₀). Plotted is the mean of three biological replicates. Error bars indicate standard deviations.

Fig 2

CRISPRi knockdown of secondary metabolite production in P. luteoviolacea. (A) Schematic representation of modular CRISPRi parts adapted to include dCas9-bla and Ptac sgRNA parts, pMMK601, and pMMK602, respectively. Part plasmids are combined, and a BsmBI Golden Gate Assembly was performed. (B) Schematic representation of the violacein gene cluster vioABCD in P. luteoviolacea and the violacein molecular structure. The CRISPRi system was assembled with an sgRNA targeting the vioA gene (pMMK603) and employed to knock down violacein production in P. luteoviolacea. (C) P. luteoviolacea with gfp (pMMK815) or vioA (pMMK816) sgRNA plasmids grown on marine agar plates. (D) Quantification of violacein production (measured at 580 nm) between P. luteoviolacea containing gfp or vioA sgRNA plasmids. Asterisks indicate significant differences (*P = 0.02, Dunn’s multiple comparisons test). Bars represent the mean of eight total replicates and error bars indicate standard deviations.

Fig 3

Functional knockdown of MACs and visualization of P. luteoviolacea during the tubeworm-microbe interaction. (A) Schematic depicting P. luteoviolacea and the production of MACs, which induce tubeworm metamorphosis. CRISPRi single-guide RNA (sgRNA) targeting the macB MACs baseplate gene prevents MACs from assembling, rendering the bacterium unable to induce metamorphosis. Cells that produce intact MACs are able to induce tubeworm metamorphosis. A strong fluorescent reporter strain (BHR-CP25-gfp) enabled visualization of live tubeworm-bacteria interactions. (B) Bar graph representing biofilm metamorphosis assays with P. luteoviolacea carrying a CRISPRi plasmid targeting macB or gfp and Hydroides tubeworms. A P. luteoviolacea ∆macB strain with a sgRNA targeting gfp and a treatment without bacteria (no bacteria) were included as controls. Biofilm concentrations were made with cells at OD₆₀₀ 0.1. Bars plotted show the average of 12 replicates, performed across three independent experiments. Each well contained 20–40 worms. Error bars indicate standard deviations. Statistical significance between treatments was determined by Dunn’s multiple comparisons test (N = 12). (C and D) Merged fluorescence and DIC micrographs of Hydroides elegans juveniles imaged 24 h after the competent larvae were exposed to inductive biofilms of P. luteoviolacea containing plasmids with (C) CP25-gfp or (D) CP25-NLuc. Strains containing NLuc plasmids were used as a negative control to account for autofluorescence. Yellow arrows show accumulation of fluorescent bacteria in the Hydroides juvenile pharynx. Scale bar is 100 µm.

Fig 4

Marine Proteobacteria are amenable to plasmid uptake and stable replication of toolkit plasmids. (A) Maximum likelihood phylogeny built using the whole genomes of 12 strains selected for manipulation and successfully conjugated in this study (52, 53). All strains used in this study are known for their interaction with a range of marine biota and the icons depicting their associated host are shown in the vertical box. Gammaproteobacteria strains are highlighted in purple and Alphaproteobacteria strains are shown in gold. Scale bar is 0.4 and bootstraps were generated using the rapid-bootstrapping method (54). The tree was rooted at the midpoint with FigTree (v1.4.4). (B) Fluorescence and DIC overlay micrographs of overnight cultures containing constitutively expressed RSF1010 ori fluorescence vector (CP25-gfp-T7). Scale bar is 5 µm. Stars denote environmental strains that serve as the first reported conjugation for that genera.