Preliminary Report on Air Sampling of
Particle-Associated Microcystins and BMAA
Pilot Study in Lee County, Florida: Fall 2018 – Winter 2019

Michael L. Parsons, Ph.D.

Jordan Peccia, Ph.D.

Professor of Marine Science

Thomas E. Golden, Jr. Professor of

Director, Coastal Watershed Institute

Environmental Engineering

Florida Gulf Coast University

Yale University

10501 FGCU Blvd

9 Hillhouse Avenue

Fort Myers, FL 33965

New Haven, CT 06520

mparsons@fgcu.edu

jordan.peccia@yale.edu

(239) 590-7526

Wyatt Arnold, Ph.D.
Post-doctoral Scientist
Environmental Engineering
Yale University
9 Hillhouse Avenue
New Haven, CT 06520
wyatt.arnold@yale.edu

 Project Description. Extensive Microcystis blooms occurred in the tidal Caloosahatchee River and
in adjoining Cape Coral canals in southwest Florida in late summer/early fall 2018. The Florida
Department of Environmental Protection conducted monthly sampling in the area during the
bloom, and reported the presence of Microcystis aeruginosa and microcystin concentrations as
high as 46 µg L-1 in their September 17, 2018 sampling event (https://floridadep.gov/dear/algalbloom/content/algal-bloom-sampling-results). Exposure to concentrations >20 µg L-1 in
recreational waters are considered to be “high risk” for acute health effects by the World Health
Organization
(https://www.epa.gov/nutrient-policy-data/guidelines-and-recommendations).
Microcystis (and most cyanobacteria) are also known to produce β-N-methylamino-L-alanine
(BMAA), which is thought to be a factor in some neurodegenerative disorders (e.g., ALS).
Knowing that microcystins and BMAA were present in the waters affected by the cyanobacteria
bloom, people were concerned that they were being exposed to these compounds via inhalation of
aerosols or particles (i.e., “algal dust”). Discussions with Lori Backer (CDC) and a thorough
review of the literature revealed that evidence for an airborne vector for microcystin and BMAA
was sparse and often inconclusive (e.g., Codd et al. 1999; Cox et al. 2009; Backer et al. 2010;
Chernov et al. 2017). To that end, we initiated a pilot study to determine 1) if microcystins and
BMAA were present in airborne aerosols or particles; and 2) the sizes of particles these compounds
were associated with. The latter is particularly important as the smaller the particle, the deeper in
the lungs the particle can get and the more likely any associated toxins (or other potentially
bioactive compounds) can get into the bloodstream (i.e., if particles can enter the alveoli). As
people are still being exposed to Microcystis in some areas of Cape Coral and along the
Caloosahatchee River, there is an urgent need to determine if airborne particles (or aerosols) are
an exposure vector that needs to be addressed and accounted for when assessing risks to human
health in the region.
Project Implementation. We used Anderson
Impactor air samplers for this pilot study (Figure
1). These samplers offer many advantages to
other methods, and provide additional data to
allow us to test our primary question: is inhalation
a viable vector for microcystin and BMAA
exposure in humans? Firstly, the samplers sizefractionate the air-borne particles to allow us to
assess if/and how many particles can enter the
respiratory tract, and to what degree (i.e., head to
bronchi to alveoli). Secondly, the vacuum pumps
that suck the air into the samplers are calibrated,
which allowed us to calculate the volume of air
samplers and the concentration of the particles
(and microcystins and BMAA) in the air.

Figure 1. Schematic representation of the
impactor sampler and the Section of the
respiratory tract represented by each stage.
Image provided by PharmaTutor.

 Two deployments were conducted for this study. The first involved the deployment of two air
samplers. One sampler was deployed next to an ongoing Microcystis bloom, and another at a
control site upwind of the bloom) for 20 days (September 21 – October 11, 2018). The “bloom
site” was a residential property on a dead end canal in Cape Coral (26.59861 N; 81.93139 W;
Figures 2 & 3). A Microcystis bloom was evident at the site at the initiation of sampling (but past
the peak in biomass earlier in the month; Figure 4), but waned over the course of the sampling
period (Figure 5). The “control site” was the FGCU Vester Field Station in Bonita Springs, FL
(26.33055 N; 81.8375 W; Figure 6). No cyanobacteria
blooms were evident in the vicinity of the field station
during the course of the sampling period.
We also collected water samples before and after each
deployment at each site (50 mL Falcon® centrifuge tubes;
four total each sampling trip) to determine toxin
concentrations in the water during and after deployment.
The filters (one for each stage in each sampler) and water
samples were tested for microcystins and BMAA using
Enzo Life Sciences ELISA microcystin test kits
(http://www.enzolifesciences
.com/ALX-850-319/
microcystins-adda-specific-elisa-kit/) and ABRAXIS
ELISA BMAA test kits (https://www.abraxiskits.com/
uploads/products/docfiles/387_BMAA%
20R010213.pdf).

Figure 2. Location of Cape Coral
“bloom site” at a residential property
on a dead end canal.

Wind data were also collected from the Page Field Airport in Fort Myers from weather.gov (station
KFMY) to determine the predominant wind directions and speeds during sampler deployment.

Figure 3. Deployment of air
sampler at Cape Coral
residence. (Adam Catasus,
graduate student, FGCU).

Figure 4. Microcystis bloom
present at the Cape Coral study
site at the beginning of the air
sampler deployment on
September 21, 2018.

Figure 5. The Microcystis
bloom waned over the course of
the sampling period at the Cape
Coral study site (photograph
taken on October 11, 2018).

 The second deployment took place from December 21,
2018 through January 18, 2019 (28 days) at the FGCU
Buckingham campus (Figure 7). This area was chosen
because 1) it is a secure location (part of FGCU); and 2)
it is at least 2 km away from any “large” bodies of water
(i.e., those that show up on Google Earth). The purpose
of this second deployment was to test new filters
(Whatman filter paper 41) to reduce potential signal
noise in the analysis. We also obtained two more air
samplers, so four samplers in total were placed at the
Buckingham property on a picnic table at an outside
pavilion (Figure 8). The use of four samplers provided
more material to examine for microcystins and BMAA
(and possibly qPCR to detect and quantify Microcystis),
as well as the ability to test replicates (and
reproducibility).

Figure 6. Location of the Vester Field
Station “control site” in Bonita
Springs, Florida.

Figure 8. A view of the area, samplers, and
pavilion at the Buckingham campus.
Figure 7. Location of the Buckingham campus.
The red scale bar = 2 km. The white circle
shows the location of the Buckingham campus.
The yellow star shows the location of the
pavilion the air samplers were set up under.

Findings. First deployment. The ELISA test kits meet all QA/QC standards needed for this study.
Calibration curves worked well, positive controls were within range, and negative controls were
non-detect. The microcystin standard provided by the manufacturer produced a concentration
value of 0.6 (µg L-1), within the acceptable range of 0.75  0.185 µg L-1. The blank produced a
(false) microcystin value of 0.05 µg L-1, below the established limit of detection for the analysis
(0.15 µg L-1).
Analysis of the filter samples from the control and bloom sites indicated that microcystins and
BMAA were detected at both sites, and concentrations were not statistically different between the
two sites (Figure 7). Both compounds were present in all size fractions tested at both locations
(Table 1 & Figure 8), and were not consistently higher in any size fraction between the sites. Also
note that the compounds were detected on the filters with the smallest pore sizes (< 1.0 µm), which

 indicates these compounds can reach the alveoli, and therefore have access to the respiratory
pathway into the bloodstream.
Microcystins and BMAA were found in the water samples at both locations (Table 2), albeit at
lower concentrations at the control site versus the bloom site, particularly in regard to microcystin.
Microcystin levels were 5,627 µg L-1 at the Cape Coral site at the initiation of air sampling on
September 21, 2018 (approximately 200x higher than the 20 µg L-1 “high risk” threshold for
recreational waters established by the World Health Organization (Table 2). As the bloom waned
over the 20-day period, however, microcystin concentrations fell to 1.2 µg L-1, and were through
below this threshold value. While the Vester Field Station water sample collected at the initiation
of sampling was not tested, the microcystin concentrations in water at the termination of the
sampling were found to be very low (0.1 µg L-1).

Figure 7. Comparison of total microcystin and
BMAA concentrations at Vester (Control) and
Cape Coral (Bloom).
Table 1. Microcystin concentrations measured in
the air impactor stages from the “Bloom” site (Cape
Coral) and “Control” site (Vester Field Station).
Note that no filter was used in Stage 0 of the
“Control” sampler.
Cape Coral (Bloom)
Sample (respiratory
system equivalent)
Stage 7
(head)
Stage 6
(head)
Stage 5
(pharynx)
Stage 4
(trachea)
Stage 3
(secondary bronchi)
Stage 2
(terminal bronchi)
Stage 1
(alveoli)
Stage 0
(alveoli)
Stage F
(alveoli)

microcystin
concentration (ng/m3)
0.1937
0.2095
0.1560
0.1209
0.1083
0.1236
0.0635
0.0572
0.0826

Vester Field Station
(Control) Sample
Stage 7
(head)
Stage 6
(head)
Stage 5
(pharynx)
Stage 4
(trachea)
Stage 3
(secondary bronchi)
Stage 2
(terminal bronchi)
Stage 1
(alveoli)
Stage 0
(alveoli)
Stage F
(alveoli)

microcystin
concentration (ng/m3)
0.1656
0.1631
0.1663
0.1639
0.1300
0.1349
0.0531
not used
0.1112

Figure 8. Difference in toxin concentration
between Vester (Control) and Cape Coral
(Bloom) size fractions (filter pore diameters
in µm).

 The home owner who graciously allowed us to set
up the air sampler at their residence had saved the
HVAC filters and an air mask that were used
during the height of the cyanobacteria bloom in
late August – early September, 2018. The
microcystin concentrations were higher on these
filters, but also represent a composite sample (i.e.,
equal to the sum of most of the filter stages (0 – 5)
of the air samplers. While the different nature of
sampling between the air samplers and HVAC
filters (and dust mask) prevent direct comparison
of the microcystin and BMAA concentrations, the
fact that both compounds were detected indicates
the compounds were associated with particles
inside the residence.

Table 2. Concentrations of microcystin and
BMAA from water samples at the two study
sites at the beginning and end of deployment.

Analysis of the Page Field airport wind data
indicated that the predominant wind direction over
the course of the study was East South East (ESE;
99.65°). Therefore, it is not likely that the
detection of microcystins and BMAA at the
control site (Vester) was due to wind transport
from the bloom at Cape Coral.
Second deployment. The filters from two of the air
samplers (one Yale unit; one FGCU unit) were
analyzed for toxins. Microcystin extracts were run
in triplicate; BMAA in duplicate (more extract
volume was needed for the BMAA analysis). The
standard curves for both analyses were >0.99,
indicating that toxin quantification data are highly
accurate. For both compounds, concentrations
were far lower than seen in the fall; approximately
10 times lower for microcystin and 5 times lower
for BMAA. Interestingly, however, toxins were
still detected, and at all size fractions.

Figure 9. Microcystin and BMAA
concentrations from the air samples (FC and
FM; Out; red) and the HVAC (H1 and H2)
and dust mask (Mask) filters from Cape
Coral residence.

Interpretation of Results. Although these results
are preliminary, the findings presented herein are
significant. The most important finding was that
microcystins and BMAA were found in
association with particles that can travel through the respiratory tract to the alveoli (Figure 8 and
Table 1), demonstrating that 1) microcystins and BMAA can be airborne; and 2) inhalation is a
viable vector for microcystin and BMAA exposure in humans. Secondly, both compounds were

 detected at both locations in the fall, albeit at low concentrations. The second deployment at
Buckingham also detected both compounds at all size fractions, but at much lower concentrations
(10x and 5x lower for microcystin and BMAA, respectively). These findings suggest that we
measured “elevated” concentrations of microcystins and BMAA in the fall, and that lower
background concentrations (i.e., present in the everyday environment) were documented in the
winter (Buckingham) deployment. Although concentrations were “elevated” in the fall,
microcystin concentrations were very low (<0.21 ng m -3; not known if BMAA concentrations are
considered low or not due to lack of health-related data). Additionally, the Vester Field Station
(located south of Cape Coral) was upwind of Cape Coral during the course of the fall study,
indicating that the Cape Coral cyanobacteria bloom was not likely the source of the compounds at
filtered at Vester. Additionally, the fact that microcystin and BMAA levels were similar at both
locations in the fall indicates there was no expected “dilution effect” if Cape Coral was a source
of the particles (i.e., we would expect microcystin and BMAA concentrations to be higher near the
source).
Our fall sampling effort occurred during the waning periods of this bloom in late September/early
October, as depicted by the low microcystin concentrations measured at the end of the sampling
(1.2 µg L-1). As the Microcystis bloom was much larger in August and early September, it is likely
that toxin-laden particles would have been higher during the peak of the bloom. In a study of
freshwater lakes in California, however, Backer et al. (2010) reported that airborne microcystins
were not correlated with Microcystis cell concentrations, indicating that the production of airborne
microcystins appears to be more complicated than being a simple function of cell concentrations.
Conclusions. This pilot study demonstrated that microcystins and BMAA are associated with
airborne particles capable of being inhaled deep into the lungs (the alveoli), thereby representing
a possible vector of exposure in humans. This work expands upon other studies that have
demonstrated that microcystins and BMAA can aerosolize and be inhaled (Backer et al. 2010; Cox
et al. 2009) by providing more data on the size fractionation of the particles with which
microcystins and BMAA were associated. By doing so, the argument for an inhalation-based
vector of exposure is further strengthened, meriting further examination and assessment.
The other significant findings of this pilot study was that 1) microcystin and BMAA concentrations
were similar in the fall at the “Bloom” and “Control” sites, suggesting that the source of these
compounds may be more ubiquitous to the region (i.e., anywhere near water); and 2) the lower
concentrations measured at Buckingham in the winter suggests that levels may be lower inland
away from water bodies (and the potential cyanobacterial source of the compounds) And/or in the
winter/dry season (when cyanobacteria biomass is lower). Further research is needed to assess
whether these concentrations represent any chronic health risks, as has been reported in other
studies; e.g., Zhang et al. (2015) reporting a 3% increase in non-alcoholic liver disease for every
10% increase in cyanobacteria bloom coverage in their study of freshwater ponds and lakes and
nearby communities across the United States.
This study was a pilot study; future deployments will expand temporal and spatial coverage and
improve statistical/analytical power. Inclusion of other research expertise (e.g., epidemiology)
will also allow us to put results in context with human health exposure risk assessment.

 References
Backer, L.C., McNeel, S.V., Barber, T., Kirkpatrick, B., Williams, C., Irvin, M., ... & LePrell, R.
2010. Recreational exposure to microcystins during algal blooms in two California lakes. Toxicon,
55(5), 909-921.
Codd, G., Bell, S., Kaya, K., Ward, C., Beattie, K., & Metcalf, J.S. 1999. Cyanobacterial toxins,
exposure routes and human health, European Journal of Phycology, 34:4, 405-415, DOI:
10.1080/09670269910001736462.
Cox, P.A., Richer, R., Metcalf, J.S., Banack, S.A., Codd, G.A., & Bradley, W.G. 2009).
Cyanobacteria and BMAA exposure from desert dust: a possible link to sporadic ALS among Gulf
War veterans. Amyotrophic Lateral Sclerosis, 10(sup2), 109-117.
Chernoff, N., Hill, D.J., Diggs, D.L., Faison, B.D., Francis, B.M., Lang, J.R., Larue, M.M., Le, T.T., Loftin, K.A., Lugo, J.N., Schmid, J.E., & Winnik, W.M. 2017. A critical review of the
postulated role of the non-essential amino acid, β-N-methylamino-l-alanine (BMAA) in
neurodegenerative disease in humans: Journal of Toxicology and Environmental Health - Part B
(Critical Reviews), p. 1-47, doi:10.1080/10937404.2017.1297592.