Axenic Biofilm Formation and Aggregation by Synechocystis sp. Strain PCC 6803 Are Induced by Changes in Nutrient Concentration and Require Cell Surface Structures

Microbes can exist as suspensions of individual cells in liquids and also commonly form multicellular communities attached to surfaces. Surface-attached communities, called biofilms, can confer antibiotic resistance to pathogenic bacteria during infections and establish food webs for global nutrient cycling in the environment. Phototrophic biofilm formation is one of the earliest phenotypes visible in the fossil record, dating back over 3 billion years. Despite the importance and ubiquity of phototrophic biofilms, most of what we know about the molecular mechanisms, genetic regulation, and environmental signals of biofilm formation comes from studies of heterotrophic bacteria. We aim to help bridge this knowledge gap by developing new assays for Synechocystis, a phototrophic cyanobacterium used to study oxygenic photosynthesis and biofuel production. With the aid of these new assays, we contribute to the development of Synechocystis as a model organism for the study of axenic phototrophic biofilm formation.

ABSTRACT

Phototrophic biofilms are key to nutrient cycling in natural environments and bioremediation technologies, but few studies describe biofilm formation by pure (axenic) cultures of a phototrophic microbe. The cyanobacterium Synechocystis sp. strain PCC 6803 (here Synechocystis) is a model microorganism for the study of oxygenic photosynthesis and biofuel production. We report here that wild-type (WT) Synechocystis caused extensive biofilm formation in a 2,000-liter outdoor nonaxenic photobioreactor under conditions attributed to nutrient limitation. We developed a biofilm assay and found that axenic Synechocystis forms biofilms of cells and extracellular material but only when cells are induced by an environmental signal, such as a reduction in the concentration of growth medium BG11. Mutants lacking cell surface structures, namely type IV pili and the S-layer, do not form biofilms. To further characterize the molecular mechanisms of cell-cell binding by Synechocystis, we also developed a rapid (8-h) axenic aggregation assay. Mutants lacking type IV pili were unable to aggregate, but mutants lacking a homolog to Wza, a protein required for type 1 exopolysaccharide export in Escherichia coli, had a superbinding phenotype. In WT cultures, 1.2× BG11 medium induced aggregation to the same degree as 0.8× BG11 medium. Overall, our data support that Wza-dependent exopolysaccharide is essential to maintain stable, uniform suspensions of WT Synechocystis cells in unmodified growth medium and that this mechanism is counteracted in a pilus-dependent manner under altered BG11 concentrations.

IMPORTANCE Microbes can exist as suspensions of individual cells in liquids and also commonly form multicellular communities attached to surfaces. Surface-attached communities, called biofilms, can confer antibiotic resistance to pathogenic bacteria during infections and establish food webs for global nutrient cycling in the environment. Phototrophic biofilm formation is one of the earliest phenotypes visible in the fossil record, dating back over 3 billion years. Despite the importance and ubiquity of phototrophic biofilms, most of what we know about the molecular mechanisms, genetic regulation, and environmental signals of biofilm formation comes from studies of heterotrophic bacteria. We aim to help bridge this knowledge gap by developing new assays for Synechocystis, a phototrophic cyanobacterium used to study oxygenic photosynthesis and biofuel production. With the aid of these new assays, we contribute to the development of Synechocystis as a model organism for the study of axenic phototrophic biofilm formation.

INTRODUCTION

Synechocystis is a model microorganism for photosynthesis (1, 2) and is frequently chosen for metabolic engineering to enhance biofuel production (reviewed in references 3 and 4). Knowledge of the environmental signals and molecular mechanisms of axenic biofilm formation by microbial phototrophs would inform rational engineering of customized cellular adhesion for biofuel applications, as well as address knowledge gaps in the ecological roles of phototrophs in establishing mixed-species biofilms in diverse environments. A few recent studies of axenic phototroph biofilm formation have been reported (reviewed below). However, compared to the wealth of axenic heterotrophic biofilm studies, progress for axenic phototrophs has been limited because the available biofilm assays are relatively slow, and model organisms for axenic phototrophic biofilm formation have not been established. As a result, axenic biofilm formation by phototrophs remains poorly understood relative to the prominence of phototrophic biofilms in the earliest fossil record (5–7, 114) and to their importance in global nutrient cycling (8–11), astrobiology and space exploration (12–14), and biofuel production (15–17).

We hypothesize that microbial phototrophs, specifically Synechocystis species, are able to form axenic biofilms and use cell surface structures such as exopolysaccharides (EPS), the S-layer, and pili to attach and adhere to surfaces, similar to other biofilm-forming heterotrophic bacteria (reviewed in reference 18 and below). To test these hypotheses, we surveyed the literature and used BLAST searches to identify Synechocystis genes and predicted homologs known to be important for heterotrophic biofilm formation. These were targeted for deletion in Synechocystis.

At neutral pH, bacteria have a net negative charge conferred on them by EPS (described below). A general model of the role of a cell’s surface charge in binding of cells to a substratum has been described by the Derjaguin-Landau-Verwey-Overbeek (DLVO) theory (reviewed in references 19 and 20). This model predicts that particles (cells) have a more stable (nonsedimenting) colloid suspension with increasing electrostatic charge of the particles. The extended DLVO model (XDLVO) adds cell surface hydrophobicity to predictions of cell interactions. Proteins, lipids, S-layer glycoproteins, membrane vesicles, and lipopolysaccharides have all been shown to increase the hydrophobicity of a cell’s surface in a variety of bacteria (21–24), which increases the tendency of cells to aggregate in aqueous (hydrophilic) environments. Many cyanobacteria synthesize EPS (reviewed in references 25 and 26).

Below we highlight three known or hypothesized components of Synechocystis EPS: Wza-dependent EPS, the released polysaccharide (RPS) colanic acid, and cellulose. Synechocystis EPS have been biochemically characterized using acid hydrolysis and chromatography of the resulting sugar monomers (27, 28). Selection of naturally occurring mutants by anion-exchange chromatography found that the increasing negative charge of mutants correlated with either increased uronic acids or increased total amount of EPS. Additionally, these strains had decreased rates of cell sedimentation, suggesting that the net charge conferred by EPS influences cell-cell repulsion (29) in Synechocystis.

Bacterial EPS synthesis and export systems are best characterized in Escherichia coli. This includes the type I capsular polysaccharide, which requires proteins, including Wzx, Wzy, Wza, and Wzc, for the later stages of EPS synthesis and export (30). Wza and Wzc proteins act as a gating mechanism/polymerase and as an outer membrane porin of type I capsule in E. coli, respectively (31). In many heterotrophic bacteria, mutants lacking Wza do not synthesize capsule (32) and are also impaired for biofilm formation (33–35). The Synechocystis protein Sll1581 is a putative homolog to Wza (28% identity; 42% similarity), consistent with its localization to the outer membrane of Synechocystis (36). In a previous study, mutants lacking Sll1581 had EPS levels less than 25% that of wild-type (WT) level (37). Compared to WT cells, this mutant showed spontaneous auto-sedimentation without aggregation in supernatants, as measured by a 3-week standing flask assay of liquid cultures.

A second putative Synechocystis homolog, Sll0923, has 21% identity and 39% similarity to E. coli protein Wzc. Synechocystis mutants lacking the Wzc protein had the same nonsedimenting phenotype as the WT in standing flasks of supernatant and correspondingly only a 50% reduction in EPS levels compared to the WT level. We did not find putative homologs to Wzx and Wzy in Synechocystis but predict that the activities of these proteins in Wza-dependent EPS may be provided by ABC transporters, similar to type 2 and type 3 capsule in E. coli, as suggested previously (28).

Colanic acid, a type of released polysaccharide (RPS), is synthesized and exported using the same proteins as type I capsule in E. coli, but it is under different genetic regulation. Colanic acid is produced by E. coli strains that do not synthesize Wzi, a protein specific to anchoring type 1 capsule to the cell surface (30, 38). We did not find a homolog to Wzi in Synechocystis using a BLASTP search. This is consistent with previously reported detection of RPS in WT Synechocystis supernatants and reduced levels of this RPS in supernatants of wzc mutant cultures (29, 37). The role of Synechocystis RPS in cell binding, if any, is not known.

Cellulose is a component of EPS associated with the cell surface, promoting adhesion in many heterotrophic bacteria (39). Cellulose has also been detected in the EPS of several cyanobacterial species (40, 41). Instead of cellulose synthases like BcsA that are common in heterotrophic bacteria, cyanobacteria use CesA, the cellulose synthase protein conserved in higher plants (42, 43). Synechococcus PCC 7002 and Thermosynechococcus vulcanus RKN (40, 44) were shown to have cellulose-dependent aggregation. Synechocystis contains one cellulose synthase motif, DDD35QXXRW, in Sll1377, which has homology to the N-terminal region (48% query coverage) of CesA in Thermosynechococcus vulcanus RKN (BLASTP search [45] using default parameters). In a study of 12 diverse cyanobacterial species, cellulose was not detected in Synechocystis or Synechococcus elongatus PCC 7942 in nonaggregated cultures (43). From these findings, it is inconclusive whether Synechocystis synthesizes cellulose during aggregation.

A second line of evidence in the literature ties cellulose-dependent aggregation to environmental signals via highly conserved role of secondary messenger c-di-GMP (cyclic di-GMP) (46, 47). Nutrient limitation such as carbohydrate starvation is frequently correlated with increased c-di-GMP levels in heterotrophs (48); carbohydrate starvation results in both nutrient and energy limitation in heterotrophs. The cyanobacterium Thermosynechococcus vulcanus RKN undergoes cellulose-dependent aggregation in response to blue light (40, 44), an energy-limited condition, in a process requiring formation of c-di-GMP by the protein SesA (Tlr0924). Two studies show that c-di-GMP levels are correlated with aggregation in Synechocystis under different energy-limiting conditions although cellulose measurements were not reported (49, 50).

In addition to the role of EPS in biofilm formation and aggregation, we also investigated the roles of the S-layer and pili. The Synechocystis S-layer protein (Sll1951) is glycosylated and forms a surface layer with hexagonal symmetry, which can be imaged via transmission electron microscopy (TEM) as a honeycomb-like surface texture on WT cells (51, 52). S-layer mutants in heterotrophs have a range of adhesion phenotypes, ranging from superbinding to completely biofilm deficient, depending on species (reviewed in reference 21). S-layer mutants have enhanced biofilm formation compared to that of the WT (a superbinding phenotype) in Bacillus cereus (53), Caulobacter crescentus (54), and various Clostridium difficile 630 strains (55). In contrast, S-layer mutants of Streptococcus gordonii are deficient in aggregation (56).

There are six major classes of pili (and/or homologous structures called fimbriae and curli) which pathogens such as E. coli, Pseudomonas aeruginosa, Salmonella enterica, and Neisseria species use for attachment, adhesion, and biofilm formation during infection (reviewed in reference 57). Additionally, biofilm initiation depends on motility in many bacterial species (reviewed in reference 58), including gliding motility conferred by type IV pili. Synechocystis has thousands of type IV pili arranged peritrichously and extending several micrometers beyond the cell surface (29, 59). These type IV pili are glycosylated along their entire length. Mutations causing altered glycosylation of PilA, the pilin structural subunit, cause defects in gliding motility in Synechocystis (60). PilC is a predicted cytoplasmic chaperone protein for export of PilA monomer for pilin assembly in diverse bacteria. Consistent with this prediction, Synechocystis pilC deletion mutants (slr0162-slr0163) are apiliate (bald) and amotile (61). Interestingly, knockout of Pcc7942_2069, a putative homolog of type II secretion protein E and type IV pilus assembly protein PilB, causes biofilm formation in Synechococcus elongatus PCC 7942 (here S. elongatus) (62). This mutation is proposed to block an unidentified molecule that inhibits secretion of biofilm enhancing proteins in WT S. elongatus (63). A subsequent study found certain piliated S. elongatus mutants also underwent a degree of sedimentation, adhesion, and biofilm formation compared to levels in the WT, indicating that loss of type IV pili is not a prerequisite for biofilm formation (64).

In this study, we document extensive biofilm formation in a large outdoor photobioreactor used to grow WT Synechocystis. We adapted the crystal violet assay commonly used for biofilm study of heterotrophic bacteria in order to screen conditions leading to biofilm formation by axenic WT Synechocystis cultures. We also developed rapid aggregation and flocculation assays to further characterize environmental signals and cell surface biochemistry of cell binding. We engineered targeted genetic mutations of genes sll1581 (wza), slr0923 (wzc), sll1951 (gene coding for an S-layer hexamer), and slr0162-slr0163 (pilC) required for cell surface structures. We then screened these mutants for biofilm formation and aggregation phenotypes. Finally, we compared biofilm formation, aggregation, and motility phenotypes of WT and mutant strains with measurements of outer membrane proteins and cellulose content of extracellular matrices. We summarized our findings and interpret their significance to biotechnology and microbial ecology.

(This article was submitted to an online preprint archive [65].)

RESULTS AND DISCUSSION

Biofouling in outdoor PBR is correlated with use of hard water to prepare BG11 medium.

We grew wild-type Synechocystis cultures in a 2,000-liter outdoor photobioreactor (PBR). As shown in Fig. 1, when cultures were inoculated into growth medium BG11 prepared from sanitized (but not axenic) hard (tap) water, we observed consistent and florid growth of macroscopic green biofilms (n = 3; PBR cleaned and sanitized between replicates). This unwanted biofilm formation (biofouling) occurred in stages characteristic of heterotroph biofilms that began with isolated colonies and progressed to thick confluent growth over approximately 1 week of mid-log culture growth (growth curve data not shown). Visible biofilm grew only in illuminated areas, suggesting that this process was driven by obligate phototrophs such as Synechocystis (66, 67) and not by other microbes present in this nonaxenic environment. In subsequent rooftop PBR cultures in BG11 medium prepared with softened tap water, visible macroscopic biofilm was absent (data not shown). Our data are consistent with the hypothesis that Synechocystis formed biofilms only when nutrient levels were reduced due to precipitation with calcium in hard water.

FIG 1.

Biofouling of a nonaxenic rooftop photobioreactor during growth of WT Synechocystis. (A) During lag phase, no biofilm growth was evident. (B to G) Images taken every 24 h during rapid growth (approximate doubling every 24 h over a period of 5 days). Biofouling such as in the representative images shown was correlated with using hard tap water to prepare BG11 medium; no biofouling was evident when softened tap water was used. Scale bar, ~2 cm. The glass PBR tubes are ~20 cm in diameter.

Synechocystis forms axenic biofilms when the concentration of BG11 medium is altered.

To study axenic Synechocystis biofilm formation under controlled laboratory conditions, we adapted the crystal violet assay typically used to study heterotrophic biofilm formation (34, 68) by introducing nutrient evaporation and dilution steps that induce Synechocystis biofilm formation. We used this modified crystal violet assay to screen Synechocystis biofilm formation in a range of growth conditions including 0 to 120 rpm shaking, 4 to 50 μmol/(m²/s) photons (photosynthetically active radiation [PAR]), with biofilm growth measured daily for up to 5 days (data not shown). In the experiment shown in Fig. 2, maximum biofilm formation, as determined by crystal violet staining, occurred under conditions of 32 μmol/(m²/s) photons (PAR), 30°C, and 72-rpm shaking, at 72 h (treatment condition). It is noteworthy that unlike many heterotrophic bacteria, Synechocystis did not form visible biofilms during the crystal violet assay unless induced by these changes in nutrient concentration (P value of 0.02). Simply diluting cultures as for subculturing, without the preceding evaporation step, did not induce biofilm formation (control condition; see Materials and Methods). Scanning laser confocal microscopy used for the experiment shown in Fig. 2B revealed that isolated microcolonies were approximately 200 to 300 μm wide, 1 to 2 cells tall, and uniformly distributed on the submerged portion of the coverslips. Fig. 2C shows that stained material bound to coverslips included cells and extracellular material.

FIG 2.

Axenic biofilm formation by WT Synechocystis requires shift in nutrient concentrations. (A) “Attached” data series shows axenic biofilm formation grown from 3-ml cultures on glass coverslips in 12-well tissue culture plates. A standard crystal violet assay had only background (growth medium only) levels of crystal violet staining (“Control” condition), as measured by crystal violet absorbance at the OD₆₀₀. Modifying the assay to include evaporation followed by dilution of growth medium induced biofilm formation (“Treatment” condition). Crystal violet was eluted from cellular material bound to glass coverslips after 72 h of biofilm growth at 32 μmol/(m²/s) photons with shaking at 72 rpm. Planktonic cells (“Suspended” data series) were also measured at 72 h as measured by absorbance at OD₇₃₀. Both data series are shown on the same y axis. Error bars correspond to one standard deviation from sample mean. (B) Scanning laser confocal microscopy of auto-fluorescent cells showing biofilm structure approximately 1 to 2 cells tall, as measured along the z axis and indicated with green trace lines (not to scale) showing height of microcolonies at white cross sections. (C) Phase-contrast microscopy showed that material stained with crystal violet corresponded with attached cells and extracellular material (×1,000 magnification). Abs., absorbance.

Synechocystis requires type IV pili and the S-layer to form biofilms.

We used allelic exchange of the Km^r-sacB markers with genes essential for type IV pili (pilC [slr0162-slr0163]), EPS (sll1581 and sll0923), and the S-layer (sll1951) to engineer mutant strains. Three independent isolates of each fully segregated clone were confirmed by PCR and sequencing. Strains and plasmids used in this study are listed in Table 1; primers are listed in Table 2. We assessed the biofilm phenotypes of mutant strains using the modified crystal violet assay described above. For each condition, we measured four biological replicates (unique cultures). At least four biofilm coupons served as technical replicates for each biological replicate. Figure 3 shows that levels of crystal violet staining from the pilC mutants (strain number SD519) (optical density at 600 nm [OD₆₀₀] of 0.23 ± 0.02; P = 0.02) and S-layer mutants (e.g., SD523) (OD₆₀₀ of 0.11 ± 0.03, P value = 0.01) were significantly lower than the WT level (SD100) (OD₆₀₀ of 0.78 ± 0.23). We conclude that type IV pili and the S-layer are essential for biofilm formation by Synechocystis. Our Wza deletion mutants (SD517) appeared to have a growth defect (OD₇₃₀ of 0.39 ± 0.05) compared to growth of the WT (OD₇₃₀ of 0.64 ± 0.11; P = 0.01). Growth and/or energy is required for biofilm formation in certain other bacteria (58, 69). We could not determine from these data whether growth or Wza-dependent EPS is required for biofilm formation by Synechocystis.

TABLE 1.

Strains and plasmids used in this study

Strain or plasmid	Relevant genotypea	Note (reference)	Source
Strains
SD100	WT Kazusa	Wild type, amotile (81)	Gift from Willem Vermaas lab
SD500	WT Paris	Wild type, motile (81)	Pasteur Cyanobacterial Collection
SD515	Paris sll1581::Kmr-sacB	wza mutant	This study
SD517	Kazusa sll1581::Kmr-sacB	wza mutant	This study
SD519	Kazusa slr0162-slr0163::Kmr-sacB	pilC mutant	This study
SD520	Paris slr0162-slr0163::Kmr-sacB	pilC mutant	This study
SD522	Paris sll1951::Kmr-sacB	S-layer mutant	This study
SD523	Kazusa sll1951r::Kmr-sacB	S-layer mutant	This study
SD546	Kazusa sll0923::Kmr-sacB	wzc mutant	This study
SD577	Kazusa slr0162-slr0163::Kmr-sacB pΨ552	pilC mutant with extrachromosomal pilC	This study
Plasmids
pGEM 3Z	pUC18 derivative	Cloning vector	Promega Corporation
pΨ540	sll1581 up::Kmr-sacB::sll1581 down	Suicide vector for constructing SD515 and SD517	This study
pΨ541	slr0162-slr0163 up::Kmr-sacB::slr0162-slr0163 down	Suicide vector for constructing SD519 and SD520	This study
pΨ543	sll1951 up::Kmr-sacB::sll1951 down	Suicide vector for constructing SD522 and SD523	This study
pΨ546	sll0923 up::Kmr-sacB::sll0923 down	Suicide vector for constructing SD546	This study
pΨ552	slr0162-slr0163	Expression vector for pilC	This study
pΨ568	RSF1010 derivative	Expression vector	Gift from Soo-Young Wanda
pPSBA2KS	pSL1180 derivative (Pharmacia Biotech)	Source of Kmr-sacB markers (106)	Gift from Willem Vermaas lab

^a“Up” and “down” refer to regions flanking the gene of interest targeted for replacement (upstream and downstream, respectively; see Materials and Methods).

TABLE 2.

Primers used in this study^a

Primer name	Primer sequence	Description	Construct
Up wza F NheI (Eag)	GGATTGGCTAGCATTCATAGCATTCGGCCGATG	5′ primer for upstream region of wza	pΨ540
Up wza R SphI	CCTGGATACGACGCATGCCATCATCTAAG	3′ primer for upstream region of wza	pΨ540
Dn wza F BamHI	GCTTGCAGGGCATAATTTTTGGATCCACAAGATTCC	5′ primer for downstream region of wza	pΨ540
Dn wza R NheI (stop)	CCAACTTTAAGAAACAATGATGTAGTCTTAGCTAGCACCAGTGGTTAAACT	3′ primer for downstream region of wza	pΨ540
Up Wzc F SacI	GCACTGTCCCTATGCGAGCTCAAGCCAGTA	5′ primer for upstream region of wzc	pΨ546
Up wzc R BamHI	GCTTGTAAGGCAGGGGATCCACCAACT	3′ primer for upstream region of wzc	pΨ546
Dn wzc F BamHI	CGCTTGATCACGGATCCATATCTAGAAATAAAAACCAGTTCAGA	5′ primer for downstream region of wzc	pΨ546
Dn wzc R Sph R	GGTTCATCACTCAAAAGCATGCTGGCATCC	3′ primer for downstream region of wzc	pΨ546
Up s-layer F BamHI	GGGCAGTAAGCGACGGGATCCAGCTCGTTTAAG	5′ primer for upstream region of sll1951	pΨ543
Up s-layer R NheI	GGATTTAATCTCTAAATCTGCTAGCTAAAGTTACGG	3′ primer for upstream region of sll1951	pΨ543
Dn s-layer F NheI	GCACTTTTCAGACACTTGCTAGCGGCCGGGGAAAA	5′ primer for downstream region of sll1951	pΨ543
Dn s-layer R SphI	GGTTGGTCTTACTATAGCATGCAGGTGGTAACGGA	3′ primer for downstream region of sll1951	pΨ543
Up pilC F BamHI	CCAATGCTCTGCGGGGATCCTTACGGGAAGATCCG	5′ primer for upstream region of pilC	pΨ541
Up pilC R NheI	GGTCAGATGATTAGGGGGCTAGCACCGAAAAACTTATG	3′ primer for upstream region of pilC	pΨ541
Dn pilC F NheI (Eag)	GCCGCTAGCGTTGTGAAGAGAGTACGGCCGCAC	5′ primer for downstream region of pilC	pΨ541
Dn pilC R SphI	GCGGCATTCCCAAGTAAAGCATGCGCTCTTTAA	3′ primer for downstream region of pilC	pΨ541
pilC compl F NdeI	CCCCCTAATCATATGACCCCAAACTATTAAGC	5′ primer for pilC	pΨ552
pilC compl R SalI	GTCACAAGCAATCAGTCGACAGCAGAGC	3′ primer for pilC	pΨ552

^a“Up”/“upstream region” and “down”/“downstream region” refer to regions flanking the gene of interest targeted for replacement (see Materials and Methods).

FIG 3.

The S-layer and type IV pili are required for biofilm formation by Synechocystis. The “attached” data series shows crystal violet binding measured at the OD₆₀₀. The “suspended” data series shows planktonic growth measured at OD₇₃₀. Wza, SD517; Slyr, SD523; PilC, SD519; WT, SD100. Both data series are shown on the same y axis. Error bar corresponds to one standard deviation from the sample mean.

Synechocystis aggregation requires cellular energy production.

Our crystal violet data show that pili and the S-layer are necessary but not sufficient for biofilm formation: presence of these surface structures did not cause WT cells to form biofilms unless some additional unknown factor was induced, such as by changes in nutrient concentrations. We wanted to improve our understanding of the environmental signals and molecular mechanisms of cell-cell binding in Synechocystis. Aggregation is related to biofilm formation in that it also involves cell-cell binding and results in multicellular structures. Both aggregates and biofilms are relevant to many of the same ecological processes and biotechnology applications (70, 71). Additionally, the amount of cellular material in the Synechocystis biofilms we grew on coverslips was insufficient for convenient biochemical analyses, likely due to the small culture volume (3 ml per biofilm coupon). Therefore, we developed a rapid aggregation assay to further characterize cell-cell binding by Synechocystis. Note that unlike biofilms, which are attached to the coverslip surface, aggregates comprise only cell-cell binding and are not attached to a substratum.

Figure 4 shows that WT Synechocystis cultures aggregated an average of 56% ± 6% total biomass within 8 h when shifted to reduced-strength medium (0.8× BG11 medium) compared to negative-control cultures resuspended in supernatant (P < 0.01). This result is similar to our biofilm data (Fig. 2A). Our data are consistent with reports of a marine cyanobacterium, Synechococcus sp. strain WH8102, producing aggregates when either the phosphorus or nitrogen concentration was lowered (72, 73). Similarly, lowering iron or phosphorus nutrient concentration induced aggregation and synthesis of extracellular material by the marine cyanobacterium Trichodesmium erythraeum IMS101 (74).

FIG 4.

Aggregation of WT Synechocystis requires cellular energy production and is not affected by removal of soluble microbial products. Spnt, negative control resuspended in supernatant; 0.8×, cultures resuspended in 0.8× BG11 medium; DCMU, cultures resuspended in 0.8× BG11 medium supplemented with DCMU; Dark, cultures resuspended in 0.8× BG11 medium and then incubated in the dark; D→L, cultures resuspended in 0.8× BG11 medium incubated in the dark for 8 h and then shifted to the light; Sp+1.2×, cultures incubated in supernatant spiked with nutrient stocks to final concentration of approximately 1.2× BG11 medium; 1.2×, cultures resuspended in fresh 1.2× BG11 medium. Each condition was assessed with at least four biological replicates (n = 4). Error bar corresponds to one standard deviation of the sample mean.

We also investigated the role of cellular energy production in aggregation. Compared to positive controls, cultures induced for aggregation by a shift to 0.8× BG11 medium remained suspended when incubated in the dark (5.6% ± 6.0% aggregation; P < 0.01), or when 5 μmol of DCMU [3-(3,4-dichlorophenyl)-1,1-dimethylurea] was added to 0.8× BG11 cultures incubated in the light (0.9% ± 1.6%; P < 0.01). Dark conditions and DCMU both prevent photoautotrophic growth (66, 75, 76). When cultures in 0.8× BG11 medium were shifted to illuminated conditions after being incubated for 8 h in the dark, they eventually aggregated to the same degree as without dark incubation (46.7% ± 9.8%). We conclude that the aggregation phenotype requires cellular energy production; i.e., it is not caused solely by a change in the chemical or physical environment, such as addition of cationic coagulants for alga dewatering (77, 78), or by a change in ionic strength of medium directly affecting the hydrophobicity and adhesiveness of cells, as described in XDLVO theory (79).

Soluble microbial products do not influence aggregation.

During our aggregation assay, conditioned supernatant was exchanged for fresh BG11 medium. In this step, the extracellular environment was modified by removal of soluble microbial products (SMP) (80). SMP of different bacteria include the proposed secreted inhibitors and enhancers of biofilm formation by S. elongatus (63) and RPS (released polysaccharides) of E. coli, which could influence aggregation. The pH, salinity, and osmolarity of the culture were also altered and could be signals for aggregation, such as by inducing a stress response. Nutrient stress, osmotic stress, and salt stress each cause different stress responses in Synechocystis (81–85). One or more of these environmental signals may induce aggregation in wild-type strains subjected to a change in nutrient concentrations.

To distinguish between SMP and nutrient signals in inducing aggregation, we spiked 100-ml cultures resuspended in supernatant with microliter volumes of concentrated BG11 stock solutions to a final concentration of approximately 1.2× BG11 medium (assuming supernatants of mid-log cultures are approximately 1.0× BG11 medium) in the presence of SMP. As a control, we also tested aggregation when nutrient concentration was increased by a shift to fresh BG11 medium at a 1.2× concentration. As shown in Fig. 4, we found no significant difference in amounts of aggregation in 1.2× BG11 medium in supernatant compared to that in fresh 1.2× BG11 medium. Furthermore, we show that the amount of aggregation is the same whether nutrient strength is increased or decreased, regardless of the presence of SMP (1-factor analysis of variance [ANOVA]). One possible interpretation is that the environmental signal driving aggregation is not related to nutrient status but, rather, to osmotic or salt status, which could occur if salt or osmolyte levels are either too high or too low. We conclude that removal of SMP had no effect on cell-cell binding under the conditions tested. Further testing is needed to identify the precise role(s) of changing BG11 medium concentration in inducing aggregation.

We show in Fig. 5 and in Video S1 in the supplemental material the phenotype of a simple and rapid flocculation assay, where cell-cell binding results in larger, buoyant flocs rather than smaller sinking aggregates. We note that the IUPAC (International Union of Pure and Applied Chemistry) does not distinguish between flocculation and aggregation, both of which refer to the formation of multiparticle clusters due to destabilization of a colloid suspension (86). We use these different terms here as a convenient way to distinguish between two different assays.

FIG 5.

Synechocystis flocculation is a rapid, robust phenotype. (A) Unaggregated WT cells in supernatant (left) and aggregated cells in 1.2× BG11 medium (right). (B) Unflocculated cells in supernatants (left). Flocculation was induced by transferring culture to room temperature autoclaved supernatants spiked with BG11 stocks to a final approximate concentration of 1.2× BG11 medium (right). Cell-cell binding was much faster and resulted in higher percentages of total bound biomass in WT cultures than in the aggregation assay (representative image shown).

The flocculation assay is faster than the aggregation assay (2 h versus 8 h; see Video S1 for time lapse of flocculation). Additionally, the flocculation assay effects a higher percentage of biomass than the aggregation assay (Fig. 5). The flocculation assay also may be invaluable for characterization of buoyant flocs, which are the cause of cyanobacterial surface blooms, or water blooms (blooms). Blooms and buoyant flocs have been specifically and extensively studied due to their role in harmful algal blooms (HABs), which cause malodorous and/or toxic effects, with negative impact on water recreation and ecology (reviewed in references 87 to 89). Some phototrophs use gas vesicles to regulate their buoyancy; Synechocystis does not harbor genes known for gas vacuole formation (90). One study described buoyant floc formation by Synechocystis when 5 mmol of CaCl₂ (20 times the levels of 1× BG11 medium) was added; this study also reported no flocculation in BG11, even when concentrations were increased up to 50× (91). Our Synechocystis flocculation assay uses a nutrient strength of 1.2× BG11 medium, resuspended in autoclaved room temperature supernatants.

Type IV pili but not the S-layer are required for aggregation.

Figure 6 shows aggregation phenotypes between WT and mutant strains in 0.8× BG11 medium. As with the WT, uninduced mutant strains maintain a stable colloid suspension without aggregating (data not shown). Aggregation by the S-layer mutant (SD523; 58.4% ± 19.8%) was not significantly different from that of the WT (60.9% ± 8.2%). This result is not consistent with our data from the modified crystal violet assay, which showed that the S-layer mutant was inhibited for biofilm formation compared to that of the WT (Fig. 3). As the S-layer mutant was deficient in biofilm formation but not aggregation, it is possible that the S-layer is important for initial attachment and has greater influence in cell-glass binding than in cell-cell binding. Previous studies have shown that increasing substratum surface roughness decreases the surface interaction energy, facilitating binding (92, 93). Similarly, it may be that Synechocystis binding to smooth glass coverslips during the modified crystal violet assay is less favorable than binding to a rough cell surface during the aggregation assay. Additionally, there is precedence for the influence of the pH and ionic strength of the growth medium on the binding phenotypes of S-layer mutants, including studies of Lactobacillus, Clostridium, and Geobacillus, three Gram-positive genera (55, 79). If this variation in hydrophobicity also affects Synechocystis S-layer mutants, it may play a larger role in binding to glass than binding to other cells.

FIG 6.

Cell surface structures differentially modulate Synechocystis aggregation. WT and mutant strains analyzed by an aggregation assay (shift to 0.8× BG11 medium). WT, SD100; Slyr, SD523; Wzc, SD546; Wza, SD517; pilC, SD519; Comp, SD577. Error bar corresponds to one standard deviation from the sample mean.

The pilC deletion mutants (SD519) were severely deficient in aggregation (13.0% ± 9.5%), compared to the WT level (60.9% ± 8.2%; P = 0.01). This is consistent with the pilC mutants also being deficient in biofilm formation. Introducing the pilC gene and putative promoter region back into the pilC mutants via an expression plasmid restored WT levels of aggregation (SD577; 57.9% ± 12.7%). Synechocystis pili extend several micrometers beyond the cell surface in all directions (27, 59). The role of pili in WT aggregation thus may partly be due to increasing the effective volume occupied by each piliated cell, thereby promoting cell-cell contact through crowding. The filamentous shape of pili also facilitates binding by reducing the effective radius at the point of surface contact with pilus tips, as predicted by DLVO theory; this is also consistent with pili playing a larger role than the S-layer in aggregation (Fig. 5). Since we showed that SMP do not influence aggregation (Fig. 4), our data preclude PilC-mediated secretion of released factors analogous to the biofilm enhancers and inhibitors secreted by S. elongatus (62–64).

Mutants lacking wza-dependent EPS have a superbinding phenotype.

Mutants lacking wzc (SD546) had WT levels of aggregation (59.2% ± 5.8%), whereas mutants lacking wza had significantly higher levels of aggregation (80.8% ± 8.7% versus 60.9% ± 8.2%; P = 0.01). This is consistent with previously reported results of EPS affecting cell-cell repulsion (29, 37). In one study, a 3-week standing flask assay of cultures in unaltered supernatants showed that mutants lacking Wza fell out of suspension (sedimented without aggregation) more readily than WT or Wzc mutants. Wza mutants were also shown to have less EPS and a smaller zeta potential (approximately –21 mV) than WT (approximately –35 mV). Zeta potential is a proxy for quantifying the electrostatic charge of cell surfaces (94). The authors of the study concluded that Wza-dependent EPS production promotes dispersal of planktonic WT cells via electrostatic repulsion, consistent with the superbinding phenotype of the wza deletion mutants reported here. This conclusion is also consistent with a third study reporting constitutive aggregation and binding to glass of Synechocystis mutants lacking EPS (28). (These mutations in wzt [kpsT] and wzm [kpsM] are predicted to function as ABC transporters in the same pathway as wza and wzc.)

No differential expression of cellulose or outer membrane proteins detected during aggregation.

As with biofilm formation, type IV pili are necessary but not sufficient for aggregation by WT Synechocystis: an additional factor, such as induction by change in concentration of BG11 medium, is required for aggregation to occur. We hypothesized that WT Synechocystis was producing adhesive molecules on its surface in response to changes in growth medium, causing aggregation. To detect these putative adhesive molecules, we isolated outer membrane protein fractions of WT and mutant strains under treatment and control aggregation conditions. Samples were split before loading on gels to include both boiled and nonboiled preparations of proteins in order to detect any heat-labile proteins. As shown in Fig. S1, we did not detect any putative proteins that were differentially expressed in the treatment (aggregated) cultures versus those of the negative control, as determined by altered band patterns of Coomassie-stained gels analyzed by SDS-PAGE. This result may indicate that aggregation is not mediated by synthesis of an outer membrane protein adhesin, but mass spectrometry would be more definitive.

Aggregated cells of Thermosynechococcus vulcanus RKN were visibly dispersed by treatment with cellulase enzyme, consistent with the role of cellulose in causing aggregation in this species (40). We used the published cellulase assay to test for a role of cellulose in Synechocystis aggregation. We saw no significant change in degree of aggregation in cellulase-treated cultures compared to that in a negative control (data not shown). As shown in Fig. S2, we also tested for presence of cellulose in the purified extracellular matrices of aggregated and control cultures using the glucose oxidase assay, which quantifies the amount of glucose released by digestion with cellulase enzyme, as shown previously with T. vulcanus RKN samples (40). While our results do indicate that cellulase-liberated glucose is present in our samples (P < 0.03), there was no significant variation between aggregated and unaggregated samples. A mutational analysis of the biofilm and aggregation phenotypes of a Synechocystis mutant lacking the gene sll1377, which contains the putative cellulose synthase domain, would be more definitive.

Based on this and previous studies, our current understanding is that Synechocystis aggregation is bioenergy dependent, is induced directly or indirectly by changes in nutrient concentrations but not SMP removal in a process that requires type IV pili but not cellulose, and is inhibited by Wza-dependent cell-bound EPS. Additionally, we show in Fig. S3 that mutants lacking the S-layer and Wza phenocopied the WT for gliding motility and phototaxis. This indicates that the influence of these mutations on biofilm formation and aggregation is not due to epistatic effects on motility pathways that may be indirectly required for cell binding. Additionally, our WT Kazusa strain (Table 1, SD100) with amotile type IV pili is competent for aggregation and biofilm formation, indicating that motility is not required for cell binding in Synechocystis.

Although the precise molecular mechanisms of Synechocystis cell binding remain to be determined, our data are consistent with the relative importance of these factors as reported in other bacteria: cellular energy/growth was essential for any aggregation to occur, whereas the smaller effects of the S-layer and Wza-dependent EPS on aggregation and biofilm formation suggest that their role is due to electrostatic and/or hydrophobic contributions to the cell surface. Likewise, our data indicate that pili have a larger impact than the S-layer or Wza-dependent EPS, which could be attributed to their additional roles in increasing effective culture density and/or reducing the effective radius of the point of surface contact, which is predicted by DLVO theory to increase binding.

Conclusion.

Our results further the development of Synechocystis as a potential model organism for studies of axenic phototrophic biofilm formation and aggregation. We report convenient new axenic assays relevant to ecological and biotechnological studies of phototrophic cell binding under controlled laboratory conditions, namely, a modified crystal violet assay for biofilm formation, in addition to aggregation and flocculation assays. These new assays enable much more rapid analysis (<72 h versus weeks) of WT Synechocystis cell binding phenotypes than those published previously. This is due to using changes in nutrient concentration to induce binding of exponentially growing Synechocystis cultures rather than growing cultures in blue light (49) or light-activated heterotrophic growth (LAHG) (50, 113), which necessitates using slow-growing cultures.

We demonstrate the utility of these assays in performing mutational analysis to identify cell surface structures influencing cell-cell binding, namely type IV pili, Wza-dependent exopolysaccharide, and the S-layer. These findings include the report of a nonbiofouling strain of Synechocystis, the pilC deletion mutant SD519, which would be an advantageous genotype for more efficient cultivation in planktonic PBRs or open ponds. Additionally, we used these assays to determine that change in the nutrient concentration of Synechocystis cultures is an immediately useful environmental signal for rapid, economical harvest of 60% biomass of WT Synechocystis cells (or 80% of wza mutant cells) within 8 h. These assays also comprise new tools for performing studies on the molecular biology of axenic cell binding by a phototrophic bacterium, whereas historically, axenic biofilm assays have been conducted primarily with heterotrophic bacteria.

Overall, our data and previously published studies are consistent with a model for WT Synechocystis cell-cell interactions regulated by two different mechanisms depending on growth conditions. Under optimal growth conditions, the negatively charged Wza-dependent EPS keeps cells distributed in a stable colloid suspension by electrostatic repulsion as predicted by XDLVO theory; this could benefit the cell by limiting self-shading that would otherwise be caused by sedimentation (37) or transient contact in cell suspensions (95). Under altered nutrient conditions, blue light, or LAHG, this cell-cell repulsion is overcome through an unknown mechanism that, based on studies in other bacteria, likely includes synthesis and export of adhesive molecules to the cell surface. Recently, the facultative phototroph Rhodopseudomonas palustris was shown to form EPS-dependent axenic biofilms in a standard crystal violet assay without requiring an additional induction signal (96). Characterizing additional phyla among obligate and facultative phototrophs would enable testing of whether this relationship between metabolism, EPS, and biofilm formation is conserved.

In oligotrophic natural environments such as lakes and pelagic zones of open oceans, cyanobacterial cell density is much lower than that typically used for lab cultures (97), reducing the number and size of aggregates detected in these natural environments (73). However, migration of cyanobacteria through the water column is a normal part of their seasonal adaptation, forming blooms on lake surfaces in spring and benthic layers in the winter (87, 88, 97). Aggregation under nutrient-limited conditions contributes benefits to phototrophs (reviewed in reference 89), including relocation of cells to nearby microniches that may not be as nutrient limited.

Cyanobacterial aggregates (particulate organic carbon) also have significant ecological implications as they have recently been identified as important contributors to carbon flux to lower ocean depths, which has a major impact on oceanic food webs (98, 99). Global warming may disrupt these natural cycles in a number of ways, including increased temperatures and ocean acidification; overall, climate change is predicted to increase the growth of cyanobacterial blooms, including those species known to be toxic (100, 101). Additional studies will be helpful in developing strategies to mitigate these negative effects on ocean food webs and in optimizing cyanobacteria for production of sustainable food, fuel, and other valuable commodities (102, 103).

MATERIALS AND METHODS

Culture growth conditions.

Synechocystis cultures were grown at 30°C in 100 ml of BG11 medium (104) in 500-ml Erlenmeyer flasks until they reached mid-log phase (OD₇₃₀ of 0.6 to 0.8). Cultures were illuminated continuously with 50 μmol/(m²/s) photons (PAR) and mixed at 120 rpm on a platform shaker. Cultures were bubbled with a supply of sterile, humidified air at 0.8 mm/min as measured by a flow meter (Cole Palmer). Air was sterilized by passing through a 0.22-μm-pore-size filter and prehumidified by bubbling through a side-arm flask of distilled H₂O (dH₂O). The flasks were incubated in growth chambers maintained at a constant 30% humidity. The cultures were diluted to an OD₇₃₀ of approximately 0.15 and allowed to double at least twice before assessment of cell-binding phenotypes. Medium was supplemented with 50 μg/ml kanamycin sulfate for growth of mutant strains carrying a kanamycin resistance marker. Strain SD577, carrying plasmid pΨ552, was propagated with medium containing 30 μg/ml each of streptomycin and spectinomycin.

Modified crystal violet assay.

A standard crystal violet assay was adapted to Synechocystis from methods described previously (microtiter dish biofilm formation assay [68] and a plastic binding assay [69]). Cultures were diluted in 100 ml of BG11 medium at a starting OD₇₃₀ of approximately 0.15 and grown again to log phase as follows: for negative-control cultures (uninduced), growth conditions were as described above; for treatment cultures (induced), the side-arm flask for humidification was removed, resulting in evaporation of a culture flask to about an 84-ml volume over 24 h, equivalent to a nutrient strength of approximately 1.20× BG11 medium. Therefore, returning this culture to 1.0× BG11 medium at the start of a biofilm assay (described below) introduces a shift from higher to lower nutrient conditions.

Cells from control and treatment cultures were harvested by centrifugation at 6,000 × g for 5 min and resuspended in 1.0× BG11 medium to an OD₇₃₀ of 0.05. Three milliliters of culture was added to each well of a 12-well plate (Corning Costar, catalog no. 07–200–82; Fisher Scientific) that contained a 22-mm-thick glass coverslip as a biofilm substratum (12–540–B; Fisher Scientific). Glass coverslips were trimmed previously to fit vertically into the plate wells using a diamond scribe (Ted Pella, Inc.). Plates with inserted coverslips were placed in cross-linker (Spectroline Spectrolinker XL-1500). Plate lids were removed and also placed face up in cross-linker. Materials were sterilized by UV radiation (254 nm) for 400 s at 1,500 μW/cm² (600 mJ/cm²). After inoculation of wells, tissue plate edges were sealed to plate lids with Parafilm and cellophane tape to minimize evaporation. Sealed plates were incubated for 72 h on a platform shaker at 72 rpm under 32 μmol/(m²/s) photons (PAR), in a chamber with 30% humidity. BG11 medium (1.0×) without inoculum was used as a blank. Coverslips were removed and rinsed for 10 s per side with a strong stream of BG11 medium from a squeeze bottle, and excess solution was wicked off by standing the coverslip edgewise on absorbent paper for 5 s. Coverslips were then stained by insertion in wells containing 4 ml of 0.01% (wt/vol) aqueous crystal violet solution for 5 min in a separate tissue culture plate. Unbound stain was rinsed and wicked away, as above. Coverslips were dried in ambient air overnight in the dark and used for qualitative assessment (imaging of macroscopic staining patterns). The final culture density at the OD₇₃₀ of each well was also measured to correlate planktonic culture growth with biofilm growth. For quantifying biofilms, dried coverslips were placed in small weighing boats, and crystal violet stain was dissolved in 1 ml of dimethyl sulfoxide (DMSO) on a platform with shaking for 20 min in the dark or until coverslip stain was removed (up to 45 min). Crystal violet absorbance was measured at 600 nm as a proxy for the amount of cellular material bound to coverslips.

Confocal microscopy imaging of biofilms.

Biofilms were grown on glass coverslips and rinsed (without staining) as described in the modified crystal violet assay (above) and placed on ice to cool. A total of 100 μl of 1.6% low-melt agarose in sterile isotonic solution (1% NaCl [wt/vol] in dH₂O) was immediately applied to the biofilm facing up on the chilled coverslip. Coverslips with agarose facing up were then placed in a small petri dish, attached to the dish with dental wax, and immersed in isotonic solution. Biofilms were imaged using a Leica TCS SP5 II with a 10× or 63× differential interference contrast (DIC) dipping lens. Sample fluorescence was excited by an argon laser and collected through a Texas red filter. Biofilm heights were assessed manually by measuring samples of interest using the z axis.

Plasmid and strain construction.

DNA manipulation was carried out using standard procedures (105). Suicide plasmids for replacing Synechocystis genes with a Km^r-sacB cassette were constructed by four-part ligation into the commercial vector pGEM 3Z (Promega), a pUC18 derivative. The Km^r-sacB cassette from pPSBA2ks (106) contains markers for kanamycin resistance and sucrose sensitivity. PCR fragments of approximately 500 bp located upstream and downstream of each gene target were amplified from the Synechocystis genome using primers listed in Table 2 to target locations of double-homologous recombination flanking the gene of interest for each suicide vector. Flanking regions, pGEM 3Z, and the Km^r-sacB cassette were stitched together by restriction digest and ligation as follows. Briefly, NheI- and EagI-digested restriction sites were generated between the two flanking sequences to accommodate the digested Km^r-sacB fragment, and BamHI and SphI (New England Biolabs) sites allowed insertion of these three fragments into the pGEM 3Z multicloning site. For example, we created the plasmid pΨ541 (for replacing pilC with Km^r-sacB) by ligating digested PCR products amplified from upstream and downstream regions of pilC with digested Km^r-sacB and pGEM 3Z.

For replacing wzc, BamHI and XbaI were used to insert the Km^r-sacB cassette, and SacI and SphI were used to insert the fragments to create pΨ546. Ligation reactions were transformed into competent E. coli (5-alpha; New England Biolabs), and transformants were screened by antibiotic selection. Clones were identified by DNA sequencing.

Design of Km^r-sacB suicide plasmids incorporated the genomic context of each gene to avoid introducing polar effects of neighboring genes as follows. For sll1581, the last 100 bp were left intact in order not to delete putative upstream promoter-containing region of neighboring sll1582. Similarly, the putative promoter regions of sll1581 and ssr2843 overlap; therefore, this region was not included in the replacement region in order to preserve native expression of ssr2843 in sll1581 deletion mutants. For the remaining mutant strains, the gene and upstream region (predicted promoter) were targeted for Km^r-sacB replacement.

Transformation of naturally competent or electrocompetent Synechocystis cultures.

Mutants of Synechocystis were generated as previously described (107). Briefly, cells from log-phase culture were harvested as above and resuspended to a 200-μl volume equivalent of an OD₇₃₀ of 2.5. Four micrograms of suicide vector was added to Synechocystis cells and incubated for 6 h in BG11 medium without antibiotic, with intermittent shaking. The transformation reaction product was plated onto a Nuclepore Track-etch membrane (Whatman) on a BG11 agar plate. This was then incubated for 24 to 72 h at 30°C with 50 μmol/(m²/s) photons (PAR) until a green lawn appeared. Following this incubation, membranes were transferred to a BG11 plate containing 50 μg/ml kanamycin sulfate and incubated for approximately 2 weeks until the lawn disappeared, and then single colonies grew. To ensure complete segregation of the transformants, colonies were repropagated three to five times on BG11 agar plates supplemented with 50 μg/ml kanamycin sulfate. Kanamycin-resistant colonies were screened via PCR with primers internal to genes targeted for deletion to determine complete segregation of chromosomes, indicating loss of the WT gene sequence.

Since natural competence in Synechocystis requires type IV pili (59), we prepared electro-competent cultures of our apiliate pilC mutants in order to transform them with plasmid pΨ552 for expressing PilC, as described previously (108, 109). We harvested cells from 50 ml of log-phase cultures as described above. Cells were resuspended in 500 μl of sterile 10% glycerol solution. Samples of cells (60 μl) were mixed with up to 10 μg of purified plasmid (300 to 3,000 ng/μl DNA in dH₂O). Cells and DNA were added to 0.1-cm electroporation cuvettes and then pulsed with 12 kV/cm, 25 μF, and 400 Ω. Cells were resuspended in 900 μl BG11 medium, transferred to test tubes with 2 ml of additional BG11 medium and incubated as described previously until the OD₇₃₀ doubled. To select for transformants, cells were harvested by centrifugation and resuspended in 500 μl of supernatant for plating a range of volumes on selective medium (BG11 medium supplemented with 30 μg/ml each of streptomycin and spectinomycin).

Isolation and analysis of outer membrane protein fractions.

Cells were lysed and fractionated using standard methods (110). Fifteen-milliliter volumes of cultures were harvested by centrifugation at 6,000 × g for 5 min; cells were stored at 80°C. Cells were resuspended in 1.2 ml of 50 mM ammonium bicarbonate buffer solution with HALT protease inhibitor (ThermoFisher) on ice. Six hundred microliters of sample was added to 2-ml cryovials with 400 μl of 0.1-mm zirconium beads. Cells were lysed by bead beating (Mini BeadBeater; BioSpec) for 7 cycles at maximum speed (one cycle of 30 s of beating followed by 2 min of incubation on ice). Whole-cell lysates were fractionated using differential centrifugation as follows. Lysates were transferred to new tubes. Unlysed cells were harvested at 1,600 × g for 5 min. Supernatants were transferred, and the total membrane fraction was harvested at 16,000 × g for 1 h. Total membrane pellet was resuspended in 500 μl of 20 mM Tris, pH 8, and 500 μl of 0.8% Sarkosyl on ice and incubated at 4°C with inversion for 90 min. The outer membrane fraction was pelleted at 16,000 × g for 8 h to 12 h at 4°C. Supernatant was removed, and a final pellet of enriched outer membrane fraction was resuspended in 50 μl of 20 mM Tris buffer, pH 8. Samples were quantified via bicinchoninic acid (BCA) assay (Sigma-Aldrich). Volumes equivalent to equal protein (20 μg per well) were added to SDS loading buffer, heated (or not heated) at 100°C for 10 min, and loaded onto gradient acrylamide gels for PAGE. Protein bands were visualized via Coomassie staining (105).

Aggregation assay.

Cells from 100 ml of mid-log culture (OD₇₃₀ of 0.6 to 0.8) were harvested by centrifugation as described above, resuspended in either supernatant (negative control) or 0.8× BG11 medium (treatment condition), and decanted into 100-ml glass graduated cylinders. The starting OD₇₃₀ was measured, and the standing cultures were incubated at 30°C with illumination as described above. The final OD₇₃₀ after 8 h was measured by sampling 1 ml of culture from the 50-ml mark on the cylinder. Aggregation was reported as normalized percent change in OD as follows: [(final OD – starting OD)/starting OD] ×100. Negative percent aggregation indicated that the culture density increased over time due to cell growth while minimal aggregation occurred.

Flocculation assay.

Synechocystis cultures were grown to mid-log phase (OD₇₃₀ between 0.6 and 0.8), and cells were harvested as described above. Supernatants were decanted into large flasks (5 liters or more), capped with foil, and placed in secondary containment pans. Supernatants were autoclaved for 5 min on the gravity cycle with no drying (total time of the autoclave cycle should be no more than about 15 min). Supernatants were cooled to 30°C with an ice bath using a digital thermometer to monitor temperature. BG11 stock solutions were added to cooled supernatants to increase medium strength to a final approximate concentration of 1.2× BG11, assuming that supernatant contributes approximately 1.0× BG11 medium. Harvested cells were resuspended in the prepared supernatants. Cultures were then decanted into graduated cylinders and incubated and measured as described above.

Cellulase treatment of aggregated cultures.

Cellulase digestion of aggregates was performed as described previously (44). Three milliliters of aggregated cultures was transferred to 15-mm glass test tubes, and 100× stock solutions of cellulase were prepared in dH₂O and added to final concentrations of 0.60 U/ml cellulase (product number C0615; Sigma-Aldrich). An equal volume of water was added to a negative-control culture. Cultures were mixed and incubated without shaking at 30°C in the light. After incubation, cultures were gently resuspended, and 1-ml samples were transferred to semi-micro cuvettes and allowed to settle for 1 h. Dispersal of aggregates was determined by increase in absorbance at OD₇₃₀ between treated and control conditions.

Isolation of ECM for biochemical characterization.

Extracellular matrix (ECM) of cells, including the S-layer and exopolysaccharide, was removed and purified by a combination of mechanical and chemical separation from intact cells as described previously (28, 111). Cultures (100 ml) at an OD₇₃₀ of approximately 0.6 to 0.8 were centrifuged for 20 min at 6,000 × g. Cells were resuspended in 10 ml of supernatant with 60 μl of formaldehyde and incubated at 4°C for 1 h. Four milliliters of 1 N NaOH was added, mixed gently, and incubated again at 4°C for 3 h to disrupt ionic bonding between the ECM and the cell. Cells were centrifuged for 20 min at 20,000 × g to physically separate ECM from intact cells. Supernatants were passed through 0.2-μm-pore-size filters to remove trace cells. A negative control of BG11 medium was also sterile filtered to detect background levels of cellulose from the filtration membrane and/or dialysis membrane (see below). Samples and negative controls were lyophilized, resuspended in 2 ml of dH₂O, and dialyzed into three 4-liter volumes of dH₂O using 4,000-molecular-weight-cutoff (MWCO) reconstituted cellulose membranes (Tube-O-Dialyzer, 786–616; G Biosciences). Samples were then lyophilized a second time, resuspended in 100 μl of dH₂O, and stored in 20-μl aliquots at −80°C until analysis.

Digestion and quantitation of cellulose from ECM using glucose oxidase assay.

Cellulose from ECM was digested and quantified as described previously (40), with modifications. ECM samples and a cellulose positive control (C8002; Sigma-Aldrich) were digested with cellulase enzyme (C0615; Sigma-Aldrich) to liberate glucose, which was detected by a glucose oxidase assay. Reaction mixtures were prepared in 100 μl with a final concentration of 4% (wt/vol) cellulose polymer (positive control) or 0.89× ECM. Cellulase enzyme was prepared in 40 mM sodium acetate buffer and added to reaction mixtures for a final cellulase concentration of 0.6 U/ml. Reaction mixtures were digested at 37°C for 72 h. Digested samples were centrifuged at 13,200 × g for 1 min, and supernatants were transferred to new tubes and stored at −80°C.

To measure glucose released by digestion with cellulase enzyme, a BioAssay Systems EnzyChrom glucose oxidase assay kit (EGOX-100) was used according to the manufacturer’s instructions as described previously for cyanobacterial samples (44), with modification. Working reagent was prepared to include glucose oxidase enzyme and exclude addition of 2 M d-glucose solution. Glucose oxidase enzyme solution (G0543-10KU; Sigma) was diluted in dH₂O to a final reaction volume concentration of 8.3 U/ml of glucose oxidase. Working reagent was combined with 20 μl of digested ECM (1×) or a 1:200 dilution of digested cellulose polymer positive control in a 96-well plate. Plates were covered with lids and vortexed gently, centrifuged briefly, and then incubated at room temperature for 20 min. Fluorescence was measured at an excitation of 530 nm and emission of 585 nm with a SpectraMax M5 plate reader (Molecular Devices) using Softmax Pro software, version 6.2.2.

Phototaxis and motility assay.

The phototaxis assay was adapted from Burriesci and Bhaya (112). Log-phase cultures were diluted to an OD₇₃₀ of about 0.25, and 10-μl volumes were spotted on swarm agar (BG11 medium prepared with 0.5% Difco BactoAgar) in grid-lined square petri dishes, such that inocula were oriented directly over grid lines. Plates were sealed in parafilm and incubated at 30°C and 30% ambient humidity under directional illumination [a dark box with a light source of 30 μmol/(m²/s) photons (PAR) at one end]. Strains were graded as motile and phototactic using qualitative assessment of growth if they had blurred edges and had elongated away from the grid line, compared to the negative-control strain, which grows in a disc with crisp edges centered on top of the grid line (see Fig. S3 in the supplemental material).