You are on page 1of 7

Chemico-Biological Interactions 246 (2016) 45e51

Contents lists available at ScienceDirect

Chemico-Biological Interactions
journal homepage: www.elsevier.com/locate/chembioint

The effects of water sample treatment, preparation, and storage prior


to cyanotoxin analysis for cylindrospermopsin, microcystin and
saxitoxin
Lisa Kamp, Jennifer L. Church, Justin Carpino, Erin Faltin-Mara, Fernando Rubio*
Abraxis LLC, 124 Railroad Drive, Warminster, PA 18974, USA

a r t i c l e i n f o

a b s t r a c t

Article history:
Received 12 June 2015
Received in revised form
18 December 2015
Accepted 21 December 2015
Available online 29 December 2015

Cyanobacterial harmful algal blooms occur in freshwater lakes, ponds, rivers, and reservoirs, and in
brackish waters throughout the world. The wide variety of cyanotoxins and their congeners can lead to
frequent exposure of humans through consumption of meat, sh, seafood, blue-green algal products and
water, accidental ingestion of contaminated water and cyanobacterial scum during recreational activities,
and inhalation of cyanobacterial aerosols. Cyanotoxins can also occur in the drinking water supply. In
order to monitor human exposure, sensitive analytical methods such as enzyme linked immunosorbent
assay and liquid chromatography-mass spectrometry are often used. Regardless of the analytical method
of choice, some problems regularly occur during sample collection, treatment, storage, and preparation
which cause toxin loss and therefore underestimation of the true concentration. To evaluate the potential
inuence of sample treatment, storage and preparation materials on surface and drinking water samples,
the effects of different types of materials on toxin recovery were compared. Collection and storage
materials included glass and various types of plastics. It was found that microcystin congeners LA and LF
adsorbed to polystyrene, polypropylene, high density polyethylene and polycarbonate storage containers, leading to low recoveries (<70%), cylindrospermopsin and saxitoxin did not adsorb to the containers tested. Therefore, this study shows that glass or polyethylene terephthalate glycol containers are
the materials of choice for collection and storage of samples containing the cyanotoxins cylindrospermopsin, microcystins, and saxitoxin. This study also demonstrated that after 15 min chlorine
decreased the concentration of microcystin LR to <40%, microcystin LA and saxitoxin to <15%, therefore
quenching of drinking water samples immediately upon sample collection is critical for accurate analysis.
In addition, the effect of various drinking water treatment chemicals on toxin recovery and the behavior
of those chemicals in the enzyme linked immunosorbent assays were also studied and are summarized.
2016 Elsevier Ireland Ltd. All rights reserved.

Keywords:
ELISA
Microcystins
Cylindrospermopsin
Saxitoxin
Adsorption to storage vessels
ADDA

1. Introduction
Cyanobacterial harmful algal blooms (CyanoHABs) occur in
freshwater lakes, ponds, rivers, and reservoirs, and in brackish
waters throughout the world. Organisms responsible include an

Abbreviations: ADDA, (2S, 3S, 8S, 9S, 4E, 6E)-3-amino-9-methoxy-2, 6, 8trimethyl-10-phenyl-4, 6-decadienoic acid); CyanoHABs, cyanobacteria harmful
algal blooms; ELISA, enzyme linked immunosorbent assay; gpg, grains per gallon;
HDPE, high density polyethylene; LC-MS, liquid chromatography-mass spectrometry; LOD, limit of detection; MC, microcystin; PETG, polyethylene terephthalate
glycol; PC, polycarbonate; PP, polypropylene; PS, polystyrene; RT, room temperature; SOP, standard operating procedure.
* Corresponding author.
E-mail address: frubio@abraxiskits.com (F. Rubio).
http://dx.doi.org/10.1016/j.cbi.2015.12.016
0009-2797/ 2016 Elsevier Ireland Ltd. All rights reserved.

estimated 40 species, primarily belonging to the genera Anabaena,


Aphanizomenon, Cylindrospermopsis, Lyngbya, Microcystis, Nostoc,
and Oscillatoria (Planktothrix) [1]. Cyanobacterial toxins (cyanotoxins) include cytotoxins and biotoxins responsible for acute lethal, acute chronic, and sub-chronic poisoning of both wild and
domestic animals as well as humans. Cyanotoxins include the
neurotoxins anatoxin-a, anatoxin-a(s), and saxitoxin, and the
hepatotoxins microcystins, nodularins and cylindrospermopsin [2].
Microcystins (MCs) are toxic cyclic peptides produced by certain
cyanobacterial genera such as Anabaena, Microcystis, Nostoc and
Planktothrix [3]. Differences in the coding of the non-ribosomal
peptide synthetase/polyketide synthase complex responsible for
microcystin production have resulted in more than 100 microcystin
variants being reported to date [4]. Cylindrospermopsin is a

46

L. Kamp et al. / Chemico-Biological Interactions 246 (2016) 45e51

polycyclic uracil derivative containing guanadino and sulfate


groups. It is also zwitterionic, making it highly water soluble.
Cylindrospermopsin is toxic to liver and kidney tissue [5] and is
thought to inhibit protein synthesis and to covalently modify DNA
and/or RNA [6]. Saxitoxin is a water soluble tricyclic neurotoxin that
acts as a sodium channel blocker, preventing normal cellular
function, leading to paralysis [7].
The wide variety of cyanotoxins and their congeners can lead to
frequent exposure of humans through the consumption of meat,
sh, seafood, blue-green algal products and untreated drinking
water; accidental ingestion of contaminated water and cyanobacterial scum during recreational activities; and inhalation of
cyanobacterial aerosols [8e12]. Cyanotoxins have even been found
in the treated drinking water supply (City of Toledo, Ohio, USA,
August 2014, [34]). In order to monitor human exposure, sensitive
analytical methods such as enzyme linked immunosorbent assays
(ELISA) and liquid chromatography-mass spectrometry (LC/MS) are
often used. Regardless of the analytical method of choice, some
problems regularly occur during sample collection, treatment,
storage, and preparation which cause toxin loss and therefore underestimation of the true concentration [8,13e20].
The aim of the current study was to assess the pitfalls of
microcystins, cylindrospermopsin, and saxitoxin sample treatment,
preparation and storage of drinking and surface water samples. For
Microcystins, drinking and surface water samples were fortied
with representative congeners having various degrees of hydrophobicity (MC-LR, -LA, -LF, Fig. 1). Similarly, cylindrospermopsin
and saxitoxin were also added to surface and drinking water
samples. The samples were subsequently quantied using
commercially available ELISA kits. To determine the potential inuence of sample treatment, storage and preparation materials, the
effects of different types of materials on toxin recovery were
assessed. Collection and storage materials included glass, high
density polyethylene (HDPE), polyethylene terephthalate glycol
(PETG), polycarbonate (PC), polypropylene (PP), and polystyrene
(PS). In addition, the effect of various drinking water treatment
chemicals on toxin recovery and the behavior of those chemicals in
the ELISAs were also studied.
2. Materials and methods
2.1. Materials
Unless otherwise stated, materials were obtained or purchased
as follows: Abraxis, Warminster, PA, USA [Microcystins-ADDA
polyclonal indirect ELISA, Microcystins-(ADDA)-DM monoclonal
direct ELISA] [21,22], Cylindrospermopsin, and Saxitoxin ELISA kits;
MC-LR, MC-LA, MC-LF, cylindrospermopsin, and saxitoxin toxins,
10 mg/L solutions]; J.G. Finneran, Vineland, NJ (glass vials); Thermo
Fisher Scientic, Waltham, MA (Nalgene PETG, PP, HDPE bottles);
VWR, Bridgeport, NJ (PC bottles); Sarstedt, Newton, NC (PS tubes).
Chemicals were of reagent grade: Alfa Aesar, Ward Hill, MA
(aluminum sulfate, sodium silicouoride, sodium hexametaphosphate); Amresco, Solon, OH (sodium chlorite); EMD, Gibbstown, NJ
(charcoal); Fisher Scientic, Fair Lawn, NJ (sodium carbonate);
Fluka, St. Louis, MO (potasium permanganate); JT Baker, Center
Valley, PA (calcium oxide); Sigma, St. Louis, MO (ascorbic acid, sodium thiosulfate); Environmental Express, Charleston, SC (glass
ber lters, 1.2 mm).
2.2. Storage container recovery and sample holding time
Drinking water, with a residual chlorine concentration of
0.7 ppm (Warminster, PA), and surface water (Pine Run Reservoir,
Doylestown, PA), 90 mg/L total organic carbon, were collected in

Fig. 1. Structures of various common microcystin congeners. The variable amino acids
(labeled circle) largely determine the relative hydrophobicity of the whole molecule
and are also reected in the name and abbreviation of the respective microcystincongener.

the Fall of 2014. Drinking water samples were treated with 100 mg/
L of sodium thiosulfate as described in the Ohio EPA- ADDA by
ELISA Analytical Methodology SOP [23]. Samples were aliquoted
into glass bottles and fortied individually with 1 mg/L of MC-LR,
MC-LA or MC-LF; 0.5 mg/L cylindrospermopsin; or 0.1 mg/L saxitoxin. In samples fortied with saxitoxin, the Abraxis saxitoxin
sample preservation solution was added at a ratio of 1:10 prior to
toxin addition. Solutions were further aliquoted into pre-rinsed (3
times with fortied sample) glass, HDPE, PC, PETG, PP, or PS storage
containers, then tested immediately (within 2 h of spiking and
aliquoting into containers) as well as after storage at 4  C for 24, 48,
and 96/120 h.
2.3. Sodium thiosulfate quenching study
Samples were collected in glass containers. Immediately after
collection, drinking water samples were fortied with MC-LR or
MC-LA at 1.0 mg/L, cylindrospermopsin at 0.5 mg/L, or saxitoxin
(after sample preservation) at 0.1 mg/L. Samples were then split, one
set treated with 100 mg/L sodium thiosulfate, and both sets stored
and tested for the time periods specied above.
2.4. Drinking water chemicals assay tolerance
Chemicals commonly used in the water treatment process

L. Kamp et al. / Chemico-Biological Interactions 246 (2016) 45e51

(personal communication with Ohio EPA) were evaluated to


determine the tolerance levels in the ELISA. The chemicals were
prepared in deionized water, with the exception of samples for
saxitoxin analysis which were also treated with the sample preservation solution as specied above. All samples were prepared in
glass bottles and across a range of concentrations, including concentrations at and above those concentrations commonly used in
water treatment. Water treatment chemical solutions were prepared as grain per gallon (gpg) or parts per million (ppm)
depending on the unit of measure which is commonly used for
chemical application in water treatment plants in the USA. Solutions containing reagents which were insoluble at the desired
testing levels were ltered using glass ber lters prior to analysis.
All water treatment chemical tolerances were evaluated without
quenching or pH adjustment unless indicated: aluminum sulfate
(with and without pH adjustment) at 0, 1, 2, 3, 4, 5, 6, and 10 grains
per gallon (gpg); sodium carbonate (soda ash) at 0, 0.25, 2.5, and 25
gpg; potasium permanganate (with and without quenching) at 0,
0.5, 2, 7, 10 ppm; sodium chlorite (with and without quenching) at
0, 0.1, 1, 10 ppm; sodium silicouoride at 0, 0.1, 1, 10 ppm; sodium
hexametaphosphate at 0, 2.5, 25, 250 ppm; carbon at 0, 2, 15,
40 ppm; calcium oxide (with and without pH adjustment) at 0, 0.8,
8, 80 gpg.
2.5. pH study
Drinking water, quenched with sodium thiosulfate, and surface
water samples in glass bottles were adjusted to a pH range between
3 and 11 using ascorbic acid (100 mg/mL) or 1 N sodium hydroxide
as required. Toxins were then fortied as listed above and analyzed
in the ELISAs.
2.6. Calculations and statistical data analysis
ELISAs were conducted and data were processed according to
the manufacturer's instructions. Briey, absorbance values were
measured at 450 nm using a Molecular Devices plate reader and
concentrations were interpolated by comparing the absorbance
values obtained for the samples with the calibration data achieved
via 4-parameter regression analysis. Means SDs were calculated
from at least two replicate analyses. Recovery data were compared
to a theoretical 100% recovery in glass, with acceptable criteria for
samples as 30% of theoretical recovery. Data are presented as
means SD of two individual experiments (n 2), if not stated
otherwise.
3. Results and discussion
3.1. Effect of storage container on toxin recovery
Reliable sample treatment, preparation, and storage prior to
analysis of cyanotoxins in drinking and surface waters is an ongoing
issue in the safety and risk assessment of human and animal
exposure and also in conjunction with human intoxications with
fatal outcomes [24]. Cylindrospermopsin and saxitoxin did not
adsorb to any of the tested vessel types (Supplementary Materials).
The susceptibility of microcystins to adsorptive loss to sample
containers varies signicantly depending on congener, container
and water type (Fig. 2). A signicant decrease in recovery relative to
the theoretical recovery range of 70e130% was detected over time
for MC-LR in deionized water stored in PS, PP, and HDPE; MC-LA in
PP and HDPE; and MC-LF in PS, PP, HDPE, and PC (Fig. 2a). In
drinking water, decreased recoveries were obtained for MC-LR in
HDPE; MC-LA in HDPE; and MC-LF in PS, PP, HDPE, and PC (Fig. 2b).
Lake water recoveries for MC-LR and MC-LA in all storage

47

containers were above 70%, however the more hydrophobic


congener MC-LF exhibited lower than 70% recoveries in PP and PC
(Fig. 2c). Similar absorptive behavior was observed in a collaborative study on Microcystins in human blood serum [18,19]. We have
seen a report of absorption of arginine containing microcystins
(MC-RR) by GF/C lters [20], however in our work we have not
observed this adsorptive behavior. This may be due to differences in
the GF/C lters used, such as the manufacturing process or lter
housing. In light of the differences seen in the performance with
GF/C lters, any commercial lter should be thoroughly validated
before use.
When comparing the structures of the three MC congeners
employed in this experiment, it becomes quickly apparent that
these congeners differ only in their variable L-amino acid moieties
(Fig. 1). It thus follows that the chemical behavior (hydrophobicity
and surface adhesion capabilities) of the entire MC molecule would
be governed by the biochemical properties of the two amino acids.
The biochemical properties of the amino acids in question (arginine, alanine, leucine, and phenylalanine) have distinct differences
in hydrophobicity, whether these were determined using theoretical calculations, octanol/water partition coefcients, aqueous twophase systems, or experimental assays [25,26]. Based on the latter
analyses, a grading of increasing hydrophobicity can be established
as arginine << alanine, leucine < phenylalanine. The presence of
planar aromatic rings of phenylalanine adds an additional factor, i.e.
adhesion capabilities [27]. Consequently, the combination of these
four amino acids within the two variable L-amino acid residues
would predict the following grading of increasing hydrophobicity:
MC-LR < MC-LA < MC-LF, and increasing adhesion capability: MCMC-LA, MCeLR << MC-LF. The latter theoretical considerations are
corroborated by experimentally determined 1-octanol/water
partition coefcients (log P) for MC-LR, and -LF of 2.16, and 3.56,
respectively [28,29] and more importantly by HPLC [30], and also
conrmed in the collaborative work with microcystins in human
serum [18,19].
Surface waters tend to contain many inorganic and organic
molecules that could potentially saturate sample container binding
sites, making the surface of those vessels less likely to adsorb the
more hydrophilic microcystins. In view of the chemical and structural considerations, it is not surprising that the less hydrophobic
MC-LR in surface waters can be recovered more reliably from all
storage vessels, even after prolonged storage times (Fig. 2c). The
data obtained using quenched drinking water indicate that the MCLR and MC-LA congeners adsorb to HDPE containers and MC-LF
adsorbs to all of the containers tested, with the exception of glass
and PETG (Fig. 2b). Deionized water treated with sodium thiosulfate (Fig. 2a) exhibited recoveries of <70% with many more of the
congeners and storage vessel types than the other two water
matrices tested, most likely due to the very low concentration of
organic and inorganic molecules present. Cylindrospermopsin and
saxitoxin are more hydrophilic, therefore they do not tend to
adsorb to the tested container surfaces. To prevent adsorptive
microcystins loss, it is recommended that only glass or PETG
sample containers be used during sample collection and storage.
Any of the tested storage vessels can be used for cylindrospermopsin and saxitoxin sample collection and storage.
3.2. Cyanotoxin recovery from chlorinated water with and without
sodium thiosulfate quenching
Dissolved chlorine is an oxidizing agent used for disinfection in
nished drinking water. Sodium thiosulfate is one of the chemicals
used for dechlorination of water, also known as quenching [31].
Chlorine causes the breakdown of the cyanotoxin molecules
[32,33]. The data in Fig. 3a and b illustrates the breakdown of

48

L. Kamp et al. / Chemico-Biological Interactions 246 (2016) 45e51

Fig. 2. Effect of type of storage container on microcystin congener recovery. Waters fortied with 1 mg/L Microcystin-LR, LA, or LF (from left to right, respectively) and stored in
polyethylene terephthalate glycol (PETG), polystyrene (PS), polypropylene (PP), high density polyethylene (HDPE), and polycarbonate (PC) containers for 0, 24, 48, and 120 h Fig. 2a
deionized water, 2b. drinking water, and 2c. surface water. All container results reported as % recovery compared to recovery in glass.

microcystins LR and LA, saxitoxin and cylindrospermopsin with


exposure to chlorine over time. The quenching of drinking water
samples with sodium thiosulfate can preserve the toxin content of
samples for at least 5 days, prolonging the holding time before
analysis (Fig. 4). The breakdown of microcystins LR and LA as well
as saxitoxin occurs rapidly: after only 15 min of contact time,
approximately 50% of the initial spiked MC-LR concentration remains, less than 40% of the MC-LA and just 15% of saxitoxin.

Cylindrospermopsin is also degraded, but its degradation is


signicantly slower, with approximately 60% remaining after 24 h.
The data obtained illustrates that drinking water samples must be
quenched immediately upon collection, especially when conducting technology comparisons, such as those between ELISA,
where analysis can be run immediately and onsite, and LC/MS,
which usually requires samples to be shipped overnight and analysis may take as long as several days to complete. The data obtained

Fig. 3. Effect of chlorine on cyanotoxin degradation over time. Effect of dissolved chorine at 0.7 ppm (without quenching) on the degradation of Microcystin-LR, -LA (1 mg/L) and
saxitoxin (0.1 mg/L) (Fig. 3a); and cylindrospermopsin (0.5 mg/L) (Fig. 3b) over time. Error bars represent SD from the mean of two individual experiments (n 2).

L. Kamp et al. / Chemico-Biological Interactions 246 (2016) 45e51

49

Fig. 4. Effect of sodium thiosulfate quenching on cyanotoxin degradation over time.


MC-LR recovery
MC-LA recovery
Saxitoxin recovery
Cylindrospermopsin
recovery. Recovery of Microcystin-LR, -LA (1 mg/L), saxitoxin (0.1 mg/L) and cylindrospermopsin (0.5 mg/L) over time after sodium thiosulfate quenching of dissolved chorine
(0.7 ppm). Error bars represent SD from the mean of two individual experiments (n 2).

indicates that drinking water, after quenching, when strored


refrigerated in glass or PETG bottles, can be stored for at least 5 days
without negatively affecting results. The data also indicates that
surface waters, if stored refrigerated in glass or PETG bottles, can be
held up to 5 days without signicant degradation (Supplementary
Materials and Fig. 5).

3.3. Effect of sample pH on cyanotoxin recovery


As the pH of surface water samples can vary, a pH tolerance
study was conducted to obtain the pH parameters within which the
ELISAs can be conducted. The pH tolerance is dened as the
maximum or minimum sample pH which does not produce a false
positive result (above the assay LOD) in an unspiked sample or a
recovery which is <70% or >130% of the theoretical recovery obtained from a spiked sample at neutral pH. Toxin recovery (%) results on fortied samples (shown in Fig. 5), indicate that the
microcystins ELISA can tolerate pH 4e11, the cylindrospermopsin
ELISA pH 3e11, and the saxitoxin ELISA pH 3e11. All unspiked
samples at the various pH units tested gave results below the
respective ELISA LOD.

3.4. ELISA drinking water chemicals tolerance study


The ELISA results obtained after testing numerous chemicals
used in the drinking water treatment process demonstrate the
assays to be robust at those levels usually used during the treatment process (Table 1). Assay tolerance is dened as the maximum
concentration of a compound which a sample can contain without
producing a false positive result (above the assay LOD) in an
unspiked sample or a recovery which is <70% or >130% of the
theoretical recovery obtained from an untreated spiked sample.
The following chemicals up to the stated concentrations exhibited
no interference in the MC-ADDA ELISA: sodium thiosulfate (1 g/L),
aluminum sulfate (with pH adjustment to within assay tolerance
levels 100 gpg, without pH adjustment 4 gpg), sodium carbonate
(25 gpg), potassium permanganate (raw 2 ppm, quenched 10 ppm),
sodium chlorite (0.75 ppm raw, quenched 10 ppm), sodium silicouoride (10 ppm), sodium hexametaphosphate (250 ppm), calcium oxide (with pH adjustment to within assay tolerance levels
2000 gpg, without pH adjustment 4 gpg).
The following chemicals up to the stated concentration exhibited no interference in the MC (ADDA) DM ELISA: sodium thiosulfate
(1 g/L), aluminum sulfate (9 gpg), sodium carbonate (25 gpg),

Fig. 5. pH sample tolerance on cyanotoxin ELISAs. Recovery (%) of cyanotoxin ELISAs on surface water samples at various pH levels. Microcystin-LR was fortied with 1 mg/L,
cylindrospermopsin with 0.5 mg/L, and saxitoxin with 0.1 mg/L. Error bars represent SD from the mean of two individual experiments (n 2).

50

L. Kamp et al. / Chemico-Biological Interactions 246 (2016) 45e51

Table 1
Microcystins-ADDA, microcystins (ADDA)-DM, saxitoxin, and cylindrospermopsin ELISAs water treatment chemicals assay tolerances.
Water treatment
chemical

Microcystins-ADDA

Microcystins (ADDA)-DM

Saxitoxin

Cylindrospermopsin

Aluminum Sulfate

100 gpg with pH adjustment to


within assay pH tolerancea
25 gpg

9 gpg

10 gpg

10 gpg

25 gpg

25 gpg

25 gpg

10 ppm with quenching using


0.1 mg sodium thiosulfate per 1 mL
sampleb
8 gpg

10 ppm
10 ppm

10 ppm with quenching using


0.1 mg sodium thiosulfate per 1 mL
sampleb
20 gpg with pH adjustment using
ascorbic acid to within assay pH
tolerancea
10 ppm
10 ppm

250 ppm

250 ppm

Sodium Carbonate
(Soda Ash)
Potassium
Permanganate
Calcium Oxide (Lime)

Sodium Silicouoride
Sodium Chlorite

10 ppm with quenching using 1 mg 10 ppm with quenching using
sodium thiosulfate per 1 mL sampleb 1 mg sodium thiosulfate per 1 mL
sampleb
2000 gpg with pH adjustment
4 gpg
using ascorbic acid to within assay
pH tolerancea
10 ppm
10 ppm
10 ppm with quenching using
5 ppm with quenching using
0.1 mg sodium thiosulfate per 1 mL 0.1 mg sodium thiosulfate per
sampleb
1 mL sampleb
250 ppm
250 ppm

Sodium
Hexametaphosphate
2 ppm with ltering at time of
Carbonc
samplingc
a
b
c

2 ppm with ltering at time of


samplingc

40 ppm with ltering at time of 2 ppm with ltering at time of


samplingc
samplingc

Natural pH of solution outside of the tolerance level of the assay, therefore chemical tolerance levels determined after adjustment to within the pH tolerance of the assay.
Strong oxidizer which may affect toxin, therefore chemical tolerance level of assay determined after quenching solutions with sodium thiosulfate.
Carbon is commonly used to remove toxins from water samples, therefore tolerance level due to effect of carbon on toxin, not on assay performance.

potassium permanganate (raw 0.01 ppm, quenched using 1 mg


sodium thiosulfate per mL of sample, 10 ppm), sodium chlorite (raw
1 ppm, quenched using 0.1 mg of sodium thiosulfate per mL of
sample, 5 ppm), sodium silicouoride (10 ppm), sodium hexametaphosphate (250 ppm), calcium oxide (4 gpg).
The cylindrospermopsin and saxitoxin ELISAs exhibited common tolerance levels for sodium thiosulfate (1 g/L), aluminum
sulfate (10 gpg), sodium carbonate (25 gpg), potassium permanganate (raw 0.5 ppm, quenched 10 ppm), sodium chlorite (10 ppm),
sodium silicouoride (10 ppm), and sodium hexametaphosphate
(250 ppm). Tolerances varied however with calcium oxide (8 gpg)
for the saxitoxin ELISA and 20 gpg with pH adjustment to within
assay tolerance using ascorbic acid for the cylindrospermopsin
ELISA.
4. Conclusion
Appropriate treatment, preparation, and storage of samples
prior to analysis for cyanotoxins in drinking and surface waters is
an ongoing issue not only in the safety and risk assessment of human and animal exposure but also in conjunction with human intoxications with fatal outcomes. It was found that microcystin
congeners LA and LF adsorb to polystyrene, polypropylene, high
density polyethylene and polycarbonate storage containers, leading
to low recoveries (<70%), cylindrospermopsin and saxitoxin did not
adsorb to the containers tested. Therefore, this study shows that of
the containers tested, glass or polyethylene terephthalate glycol are
the materials of choice for collection and storage of samples containing the cyanotoxins cylindrospermopsin, microcystins, and
saxitoxin. Results indicate that the microcystins ELISA can tolerate
pH 4e11, the cylindrospermopsin ELISA pH 3e11, and the saxitoxin
ELISA pH 3e11. This study also demonstrates that exposure to
chlorine for 15 min decreases the concentration of microcystin LR
to <40%, microcystin LA and saxitoxin to <15%, while cylindrospermopsin degraded slowly with approximately 60% remaining after 24 h. Therefore quenching of drinking water samples
immediately upon sample collection is critical for accurate analysis.
As the present study shows, the ELISAs tested were tolerant of the
chemicals commonly used during the drinking water treatment
process and chemicals commonly found in surface waters,

demonstrating that ELISA-based methods are suitable tools for the


detection and monitoring of cyanotoxins in drinking and surface
waters. Regardless of the technology used, when performing cyanotoxin analysis, special care should be given to the sample
collection, treatment, preparation, and storage prior to analysis.
Appendix A. Supplementary data
Supplementary data related to this article can be found at http://
dx.doi.org/10.1016/j.cbi.2015.12.016.
References
[1] E.J. Carpenter, W.W. Carmichael, Taxonomy of Cyanobacteria, in:
G.M. Hallegraeff, et al. (Eds.), Manual on Harmful Marine Microalgae, IOC
Manuals and Guides No. 33 UNESCO, 1995, pp. 373e380.
[2] W.W. Carmichael, The Cyanotoxins, in: J. Callow (Ed.), Advances in Botanical
Research 27, Academic Press, London, 1997, pp. 211e256.
[3] S. Merel, D. Walker, R. Chicana, S. Snyder, E. Baures, O. Thomas, State of
knowledge and concerns on cyanobacterial blooms and cyanotoxins, Environ.
Int. 59 (2013) 303e327.
[4] J. Puddick, M.R. Prinsep, S.A. Wood, S.A.F. Kaufononga, S.C. Cary, D.P. Hamilton,
High levels of structural diversity observed in microcystins from Microcystis
CAWBG11 and characterization of six new microcystin congeners, Mar. Drugs
12 (2014) 5372e5395.
[5] P.R. Hawkins, M.T. Runnegar, A.R. Jackson, I.R. Falconer, Severe hepatotoxicity
caused by the tropical cyanobacterium (blue-green alga) Cylindrospermopsis
raciborskii (Woloszynska) Seenaya and Subba Raju isolated from a domestic
water supply reservoir, Appl. Environ. Microbiol. 50 (1985) 1292e1295.
[6] G.R. Shaw, A.A. Seawright, M.R. Moore, P.K.S. Lam, Cylindrospermopsin, a
cyanobacterial alkaloid: evaluation of its toxicologic activity, Ther. Drug
Monit. 22 (2000) 89e92.
[7] M.E. Van Apeldoorn, H.P. van Egmond, G.J.A. Speijers, G.J.I. Bakker, Toxins of
cyanobacteria, Mol. Nutr. Food Res. 51 (2007) 7e60.
[8] T.A. Msagati, B.A. Siame, D.D. Shushu, Evaluation of methods for the isolation,
detection and quantication of cyanobacterial hepatotoxins, Aquat. Toxicol.
78 (2006) 382e397.
[9] S. Wood, D. Dietrich, Quantitative assessment of aerosolized cyanobacterial
toxins at two New Zealand lakes, J. Environ. Monit. 13 (2011) 1617e1624.
[10] A.H. Heussner, L. Mazija, J. Fastner, D.R. Dietrich, Toxin content and cytotoxicity of algal dietary supplements, Toxicol. Appl. Pharmacol. 265 (2012)
263e271.
[11] V. Mulvenna, K. Dale, B. Priestly, U. Mueller, A. Humpage, G. Shaw, G. Allinson,
I. Falconer, Health Risk Assessment for Cyanobacterial Toxins in Seafood, Int. J.
Environ. Res. Public Health 9 (2012) 807e820.
[12] Y.S. Cheng, Y. Zhou, C.M. Irvin, B. Kirkpatrick, L. Backer, Characterization of
aerosol containing microcystin, Mar. Drugs 5 (2007) 136e150.
[13] J.S. Metcalf, P. Hyenstrand, K.A. Beattie, G.A. Codd, Effects of physicochemical
variables and cyanobacterial extracts on the immunoassay of microcystin-LR

L. Kamp et al. / Chemico-Biological Interactions 246 (2016) 45e51


by two ELISA kits, J. Appl. Microbiol. 89 (2000) 532e538.
[14] P. Hyenstrand, J.S. Metcalf, K.A. Beattie, G.A. Codd, Effects of adsorption to
plastics and solvent conditions in the analysis of the cyanobacterial toxin
microcystin-LR by high performance liquid chromatography, Water Res. 35
(2001) 3508e3511.
[15] P. Hyenstrand, J.S. Metcalf, K.A. Beattie, G.A. Codd, Losses of the cyanobacterial
toxin microcystin-LR from aqueous solution by adsorption during laboratory
manipulations, Toxicon 39 (2001) (2001) 589e594.
[16] J.S. Metcalf, G.A. Codd, Analysis of cyanobacterial toxins by immunological
methods, Chem. Res. Toxicol. 16 (2003) 103e112.
[17] I.R. Falconer, A.R. Humpage, Health risk assessment of cyanobacterial (bluegreen algal) toxins in drinking water, Int. J. Environ. Res. Public Health 2
(2005) 43e50.
[18] A.H. Heussner, S. Alatner, L. Kamp, F. Rubio, D.R. Dietrich, Pitfalls in microcystin extraction and recovery from human blood serum, Chem. Biolo.
Interact. 223 (2014) 87e94.
[19] A.H. Heussner, I. Winter, S. Alatner, L. Kamp, F. Rubio, D.R. Dietrich, Comparison of two ELISA-based methods for the detection of microcystins in
blood serum, Chem. -Biol. Interact. 223 (2014) 10e17.
[20] S. Rogers, J. Puddick, S.A. Wood, D.R. Dietrich, D.P. Hamilton, M.R. Prinsep, The
effect of Cyanobacterial Biomass Enrichment by Centrifugation and GF/C
Filtration on Subsequent Microcystin Measurement, Toxins 7 (2015)
821e834.
[21] USEPA, Technology Brief: Immunoassay Test Kits for Microcystins, 2012 (EPA/
600/S/12/511).
[22] W.J. Fischer, I. Garthwaite, C.O. Miles, K.M. Ross, J.B. Aggen, A.R. Chamberlin,
N.R. Towers, D.R. Dietrich, Congener-independent immunoassay for microcystins and nodularins, Environ. Sci. Technol. 35 (2001) 4849e4856.
[23] Ohio EPA Total (Extracellular and Intracellular) Microcystins - ADDA by ELISA
Analytical Methodology Version 2.0, January 2015. Last accessed on 5/13/
2015,
http://epa.ohio.gov/Portals/28/documents/HABs/
MicrocystinMethodology.pdf.
[24] S.M. Azevedo, W.W. Carmichael, E.M. Jochmsen, K.L. Rinehart, S. Lau,

51

G.K. Eaglesham, Toxicology 181e182 (2002) 441e446.


[25] H. van de Waterbeemd, H. Karajiannis, N. El Taya, Lipophilicity of amino acids,
Amino Acids 7 (1994) 129e145.
[26] N. Gulyaeva, A. Zaslavsky, P. Lechner, A. Chait, B. Zaslavsky, pH dependence of
the relative hydrophobicity and lipophilicity of amino acids and peptides
measured by aqueous two-phase and octanol-buffer partitioning, J. Pept. Res.
61 (2003) 71e79.
rk, F. Hanke, C.A. Palma, P. Samori, M. Cecchini, M. Persson, Adsorption of
[27] J. Bjo
aromatic and anti-aromatic systems on graphene through pp stacking,
J. Phys. Chem. Lett. 1 (2010) 3407e3412.
[28] P. de Maagd, A. Hendriks, W. Seinen, D. Sijm, pH-dependent hydrophobicity of
the cyanobacteria toxin microcystin-LR, Water Res. 33 (1999) 677e680.
[29] C.J. Ward, G.A. Codd, Comparative toxicity of four microcystins of different
hydrophobicities to the protozoan, Tetrahymena pyriformis, J. Appl. Microbiol. 86 (1994) 874e882.
[30] M. Craig, T. McCready, H. Luu, M. Smillie, P. Dubord, C. Holmes, Identication
and characterization of hydrophobic microcystins in Canadian freshwater
cyanobacteria, Toxicon 31 (1993) 1541e1549.
[31] APHA, AWWA, WEF, Standard Methods for Examination of Water and
Wasterwater, twenty-second ed., American Public Health Association, Washington, 2012.
[32] B.C. Nicholson, G.R. Shaw, J. Morrall, P.J. Senogles, T.A. Woods, J. Papageorgiou,
C. Kapralos, W. Wickramasinghe, B.C. Davis, G.K Eagkesham, Chlorination for
degrading saxitoxins (paralytic shellsh poisons) in water, Environ. Technol.
24 (2003) 1341e1348.
[33] E. Rodriguez, G.D. Onstad, T.P.J. Kull, J.S. Metcalf, J.L. Acero, U. von Gunten,
Oxidative elimination of cyanotoxins: Comparison of ozone, chlorine, chlorine
dioxide and permanganate, Water Res. 41 (2007) 3381e3393.
[34] T. Dungjen, D. Patch, Toledo-area Water Advisory Expected to Continue
Through Sunday as Leaders Await Tests; Water Stations to Remain Open.
Microcystin Found in Samples; Boiling Not Recommended, Toledo Blade,
Toledo, http://www.toledoblade.com/local/2014/08/02/City-of-Toledo-issuesdo-no-drink-water-advisery.html#PcVfL8Fx8bmFHv13.99.