1. Introduction
Picocyanobacteria are known to be a diverse and widely distributed group of cyanobacteria that occupy a wide range of ecological niches including freshwater, brackish and marine microhabitats [1] [2] . In general, cyanobacterial populations can be described using photosynthetic accessory pigments and size-structure analysis [1] [3] to provide detailed descriptions of these populations. The picocyanobacteria can produce secondary metabolites including microcystin (MC) and its variants and B-Methylamino-L-alanine (BMAA) [4] [5] . Recent genomic [6] and 16s metabarcoding [7] analysis has confirmed that commonly found picocyanobacteria (Order Synechococcales) can produce anatoxin-a. As described elsewhere [7] , the presence of (pico)cyanobacteria can be heavily influenced by salinity.
The purpose of this project was to verify methods necessary to collect (floatation and serial filtration) and analyze (light microscopy, fluorometry and 16s metabarcoding) cyanobacterial populations in salinity microhabitats with a focus on picocyanobacteria. The 16s metabarcoding portion of this project was utilized to determine where this method could provide “value-added” information to our interpretations of cyanobacterial population dynamics. By confirming the presence/absence of cyanotoxins, specifically anatoxin-a, at varying concentrations from these populations we hoped to develop an initial profile of the potential for exposure.
2. Materials and Methods
Field Sampling
Samples of whole lake water (WLW) and net (NET) water samples were collected from five different sites representing a range of saline environments on a series of dates between May and October from the shoreline (maximum depth 1m) (Figure 1), including Upper Chilmark (CHP-Up) (lat./long. 41.3419907, 70.7188816), Chilmark-2 (CHP2) (lat./long. 41.3501158, 70.6936283), Edgartown Great Pond-9 (EGP-9) (lat./long. 41.3741377, 70.5608743), Tisbury Great Pond-7 (TGP-7) (lat./long. 41.3502175, 70.6516254) and Tashmoo Pond (TSH-Sen) (lat./long. 41.4507261, 70.6240343). Samples were placed in 500 mL brown amber bottles, placed on ice and transported to the laboratory for further analysis. Size fractionation using floatation, serial filtration with gravity and serial filtration with pressure were used for sample processing and collection [7] . Samples were stored at −4˚C prior to cyanotoxin and fluorometric analysis. Samples for 16s meta barcoding analysis were collected by passing a known volume of either <5 µm, WLW or NET water samples through a Sterivex filter (0.2µm pore size) using a 50-mL syringe and stored at −80˚C prior to analysis.
Quantification of (pico)cyanobacterial biomass
All water samples were prepared, preserved, stored and analyzed as previously described [7] . Prior to fluorometric analysis, the hand-held fluorometry units were calibrated [8] for both phycocyanin (PC) and phycoerythrin (PE), accessory
![]()
![]()
Figure 1. Location of sample collection sites, Martha’s Vineyard, MA.
pigments primarily associated with cyanobacterial (phycocyanin) and picocyanobacterial (phycoerythrin) biomass. The whole lake water sample (WLW) was fractionated (<50 µm, <10 µm, <5 µm, <0.2 µm) according to the project QAPP and samples frozen for a single freeze-thaw extraction (SFT) prior to fluorometric analysis using the handheld FluoroquikTM fluorometer (excitation wavelengths 595 nm for phycocyanin and 545 nm for phycoerythrin) using a quantitation limit for phycocyanin (PC) at 3 µg·L−1 and phycoerythrin (PE) at 0.1 µg·L−1. The 16s meta-barcoding results were used as a quality assurance/quality control (QA/QC) measure as previously described [7] . The number of reads was used as a measure of (pico)cyanobacterial abundance.
Quantification of (pico)cyanobacterial cyanotoxin
The samples for cyanotoxin analysis were collected, prepared and stored following the protocol that was used for fluorometric analysis with two additional freeze-thaw cycles Samples were vortexed on high speed for 3 minutes followed by sonication for 3 minutes to complete the triple freeze-thaw extraction process as previously described [7] Leland N.J. et al. 2023). Water samples were prepared for enzyme-linked immunosorbent assay (ELISA) anatoxin-a (ATX) analysis using vacuum centrifugation (i.e. 10X) as previously described [7] . Replicate (n = 2) anatoxin-a concentrations are reported, where the detection limit for the ELISA analysis [9] was 0.15 µg·L−1.
Salinity Interference with anatoxin-a ELISA analysis
Potential interference when sample concentration would be required was evaluated using a concentration series (Figure 2), where a stock sample of seawater at 30 ppt was prepared. The samples were placed in microcentrifuge tubes and vacuum centrifuged to the desired concentration factors of 5X, 10X and 20X
![]()
Figure 2. Validation of anatoxin-a concentrations using ELISA analysis along a salinity gradient.
for final blank salinity values that approximated 150 ppt, 300 ppt, and 600 ppt. The data confirm that salinity interference will occur for samples (i.e. dissolved cyanotoxins) at 30 ppt that are subsequently concentrated (5X, 10X and 20X) prior to ATX analysis. To refine of this analysis, a stock sample of seawater at 23.7 ppt was prepared, samples placed in microcentrifuge tubes and vacuum centrifuged to the desired concentrations factors of 1.2X, 2X, 4X, 5X, 6X, 7X, 8X, 10X, 12X, and 15X for final blank salinity values of 31 ppt, 50 ppt, 101 ppt, 117 ppt, 134 ppt, 162 ppt, 207 ppt, 250 ppt, 298 ppt and 365 ppt. The data confirm that salinity interference will occur for ELISA analysis of anatoxin-a at varying salinities. Specifically, we found that interference may occur for anatoxin-a at salinities greater than 150 ppt (Figure 2).
16s metabarcoding
The DNA was extracted from a 0.22-micron filter for each size fraction of water samples collected for this study. The total DNA for each sample fraction was amplified with 16s primer 515 – 926 R [10] . Amplicons were then sequenced on the Novaseq platform (250 bp paired-end reads). Sequences were trimmed, denoised, and assigned taxonomy with Qiime2 [11] . Each ASV (unique sequences found among all samples) was assigned a taxonomy with the sklearn algorithm using a feature classifier generated from the Silva reference database (v132, 99 OTU). To compare the total number of sequences that were assigned to either Nostocales or Synechococcales, we collapsed the taxonomic assignments down to the level of species.
Statistical data analysis
Statistical analysis of cyanotoxin and fluorometric data was performed using Sigma Plot Version 14 software. Parametric analysis (Pearson’s correlation coefficients and linear regression analysis) were used to describe relationships between variables. For linear regression analysis autocorrelation (Durbin-Watson = 2.0), leverage (Studentized deleted residuals: SDR > 2) collinearity (VIF > 3), and influence (Cooks distance: Cd > 4/n and Difference in Fits: DFFits = 2 × sq.rt. [(p + 1)/(n − p − 1)] where n = number of observations, p = number of variables (including the constant) were examined.
(Pico)cyanobacterial detection and biomass quantification
This project used photosynthetic accessory pigments phycocyanin (PC) (Table 1) and phycoerythrin (PE) (Table 2) to quantify cyanobacterial biomass where PC is cyanobacterial specific and PE is considered a “signature” pigment for picocyanobacteria and potentially other cyanobacterial genus [12] [13] . With the exception of PC in Tashmoo Pond, all samples were above the limit of detection for both PC and PE. Analysis of cyanobacterial accessory pigments PC and PE have been used in marine environments to aid in the classification to species level [2] [14] . The range of PC concentrations (Table 1) for our samples was similar to those previously observed for Dolichospermum spp., from freshwater mesotrophic ponds [3] [7] that were confirmed using light microscopy. Dolichospermum sp. has been shown to tolerate salinity concentrations similar to those observed in Chilmark Pond [15] [16] . The picocyanobacteria (<5 µm) being reported here could be described as PC-rich or PE-rich based on habitat, where PC-rich are found in near-shore waters and PE-rich are more representative of open marine systems [1] [2] .
![]()
Table 1. Phycocyanin concentrations (µg·L−1) and salinity in water samples from Martha’s Vineyard sites.
![]()
Table 2. Phycoerythrin concentrations (µg·L−1) and salinity in water samples from Martha’s Vineyard.
We verified the shifts in population structure and composition using two different analytical techniques including fluorometry and 16s metabarcoding. The fluorometric analysis using phycocyanin (PC) confirmed a positive correlation (r = 0.738, p < 0.001) between salinity and cyanobacterial community structure as shown in Figure 3, where picocyanobacteria were measured as a proportion of the whole lake water (<5 µm/WLW%). A similar analysis using phycoerythrin (PE) had a weak negative correlation (r = −0.09, p = 0.683). The principal component analysis (Figure 4) identified the degree to which variables influenced the structure and composition of the cyanobacterial populations with salinity and two principle components (PC1 = 39% and PC2 = 23%). These results are similar to those described for Lake Ellesmere/Te Waihora [17] where salinities ranged from 5 - 13.5 ppt and halotolerant picocyanobacteria dominated. Both approaches (fluorometry and metabarcoding) verified unique populations at the sites that were tested.
The 16s metabarcoding confirmed the particulate material collected from the picocyanobacteria sample (<5 µm) from Upper Chilmark Pond (CHPUp) as belonging to the Order Synechococcales, specifically Cyanobium PCC-6307. Analysis of samples collected from Chilmark-2 (CHP-2), Edgartown Great Pond (EGP-9), Tisbury Great Pond (TGP-7), and Tashmoo Pond (TSH-Sen) confirmed varying abundance of Cyanobium PCC-6307 and Synechococcus 9902 (Figure 5). These findings are similar to the previous reports on the diversity of picocyanobacteria in this coastal area [2] confirming the influence of salinity microhabitats on the structure and composition of cyanobacterial populations. Absolute abundance of the top seven genus (Figure 5) confirmed the presence of Dolichospermum spp. and Sphaerospermopsis spp. in Upper Chilmark Pond. While halotolerant Dolichospermum spp. is known to produce anatoxin-a, Sphaerospermopsis spp. has been shown to produce guanitoxin, formerly known as anatoxin-a(s). [18]
(Pico)cyanobacterial populations and production of anatoxin-a
Anatoxin-a was detected in all of the water samples collected throughout the entire sampling season (Table 3). The concentrations are similar to those previously reported for anatoxin-a concentrations in picocyanobacteria [6] [7] using ELISA analysis. The similarity in anatoxin-a concentrations may be influenced by the high proportion of picocyanobacteria within the samples (Figure 3). In Chilmark Pond (CHP-Up) which contains the halotolerant BFC Dolichospermum sp., and picocyanobacteria, we confirmed a positive correlation between cyanobacterial biomass, as PC, and anatoxin-a concentrations (Figure 6) where r = 0.438, p = 0.0536 and a positive correlation using PE as a measure of cyanobacterial biomass (r = 0.348, p = 0.133).
![]()
Figure 3. Correlation between salinity (ppt) and cyanobacterial community structure as described using phycocyanin from Martha’s Vineyard sites. (r = 0.738, p < 0.001).
![]()
Figure 4. Principle component analysis of net (BFC), whole lake water (WLW) and <5 µm (<5) using 16s metabarcoding samples from Martha’s Vineyard sites. Variable weights of influence after accounting for salinity, PC1 = 39%, PC2 = 23%.
![]()
Figure 5. Relative abundance of cyanobacterial species in samples from Martha’s Vineyard sites as determined by 16s metabarcoding.
![]()
Figure 6. Correlation analysis of (pico)cyanobacterial biomass and anatoxin-a in Upper Chilmark Pond (r = 0.438, p = 0.0536).
![]()
Table 3. Anatoxin-a concentrations (µg·L−1) in water samples from Martha’s Vineyard sites.
3. Conclusions
Our investigation has demonstrated that serial filtration using gravity and pressure is an effective field and laboratory method to isolate picocyanobacteria. It was demonstrated that salinity had an influence on the cyanobacterial community structure and composition, where the halotolerant bloom forming cyanobacterium Dolichospermum spp. (Order Nostocales) was found in salinity microhabitats up to 6 ppt (Upper Chilmark Pond and Chilmark Pond 2) in addition to picocyanobacteria (Order Synechococcales). Salinity microhabitats between 6 - 30 ppt, including Edgartown Great Pond-9 and Tisbury Great Pond-7 supported colonial and single-celled picocyanobacteria, while Tashmoo Pond exclusively supported picocyanobacteria (Order Synechococcales). The picocyanobacteria Cyanobium 6307 was exclusive to Upper Chilmark Pond, while mixtures of Cyanobium 6307 and Synechococcus 9902 were found at higher salinities in Chilmark-2, Edgartown Great Pond-9, Tisbury Great Pond-7, and Tashmoo Pond. These findings were verified using both fluorometric and 16s metabarcoding techniques.
A range of anatoxin-a concentrations were documented from sites of varying salinity, ranging from 1 ppt - 31 ppt. We were able to identify the picocyanobacteria to the species level using 16s metabarcoding techniques, where Cyanobium 6307 was present (exclusively) in Upper Chilmark Pond. Varying proportions of Cyanobium 6307 and Synechococcus 9902 were found in Chilmark Pond, Edgartown Great Pond-9, Tisbury Great Pond-7 and Tashmoo Pond. We were able to affirm a positive linear correlation between cyanobacterial biomass (PC) and anatoxin-a in Upper Chilmark Pond.
Acknowledgements
This research was supported by contributions of field collection and processing by the water quality staff of The Martha’s Vineyard Commission (Sheri Caseau and Eleah Caseau). Financial support was provided by the Martha’s Vineyard Commission. Use of laboratory space and equipment was provided by UNH-Center for Freshwater Biology and Ecotoxicology. The project 16s metabarcoding was supported by a grant from the United States Geological Survey #G20AP00034.