Based on animal toxicological studies, guidelines for exposure to microcystins via ingestion have been developed for the action level (red mode) – situation 1. The guideline values are extrapolated from animal experiments and make various assumptions about exposure, and also include uncertainty factors. Uncertainty factors are used to account for safety margins, errors in extrapolation from animal experiments to human risk, and other limitations associated with experiments or limited data.
In this document tolerable daily intakes (TDIs) for microcystins are calculated based on data from two separate animal toxicological studies: a 13-week mouse study (Fawell et al, 1999) conducted with purified microcystin-LR via gavage, and a 44-day pig study (Falconer et al, 1994; see Box A2.1) carried out with cyanobacterial bloom material in the drinking water, containing nine microcystin congeners but no microcystin-LR. A TDI is defined as an estimate of the intake of a substance over a lifetime that is considered acceptable without appreciable health risk. The TDIs were used to calculate maximum allowable values (MAVs) of microcystins in recreational water under the following assumptions: the proportion of microcystin intake from recreational water is 100 per cent (ie, there are no other microcystin sources or exposures), and there is an average consumption of 100 mL of water per day (see Box A2.1).
Usually only a small proportion (less than 20 per cent) of the total microcystin load in a water body is extracellular (outside the cells) when cells are healthy (Orr and Jones 1998; Park et al, 1998). Therefore cells can only provide an approximate measure of microcystin concentrations present in a sample. The microcystin MAVs can be translated to an equivalent worst-case cell density of Microcystis sp. based on toxin quota data (Wood, Rhodes et al, 2008; Wood unpublished data; Box A2.1). This cell density is then converted into an equivalent biovolume of total cyanobacterial material (see Box A2.1) to gauge the potential hazard of other cyanobacteria, irrespective of whether toxin status is known.
The rationale for using cell counts, which are converted to biovolume estimates, rather than toxin concentrations to prompt management actions is that for most practical purposes cell counting is still primarily used by water managers to detect algae/cyanobacterial-related water-quality problems. This is because the testing is widely available and provides relatively rapid and cost-effective information. Cyanotoxin testing is offered commercially by a few organisations within New Zealand (see Appendix 8), although cost may be prohibitive for large sample numbers. Biovolumes must, however, be regarded as an indicator or ‘surrogate’ for a potential toxin hazard. These should be used to prompt actions, such as toxin monitoring, that are outlined in the alert levels framework (Section 3).
Box A2.1: Derivation of a guideline for microcystin and cyanobacterial exposure during recreational activities over a lifetime
Tolerable daily intakes for recreational exposure to microcystins were calculated (Table A2.1) using data from a 13-week mouse study (Fawell et al, 1999) and a 44-day pig study (Falconer et al, 1994) and the following equation:
TDI = NOAEL or LOAEL (1)
Table A2.1: Summary of data and uncertainty factors used to calculate TDIs for microcystins
|Study||Falconer et al, 1994||Fawell et al,1999|
|Duration||44 days||13 weeks|
|Material/toxin||Cyanobacterial bloom material, containing nine microcystin congeners but no microcystin-LR, via drinking water||Purified microcystin-LR via gavage|
|LOAEL to NOAEL||2||_|
|Sum of uncertainty factors||1000d||500d|
|TDI (μg/kg bw/day)||0.088||0.08|
a LOAEL = lowest observed adverse effect level – the lowest dose at which adverse health effects are observed.
b NOAEL = no observed adverse effect level – the highest dose at which no adverse health effects are observed.
c As measured by PP2A assay (worst case scenario).
d These safety factors do not incorporate an additional safety factor for tumour promotion. The risk scenario of a swimmer, kayaker, sailor, etc, whether adult or child, being repeatedly but discontinuously exposed during a short visit, should only incorporate liver damage as the endpoint. Incorporation of tumour promotion, and thus an additional safety factor, would suggest continuous exposure (eg, via drinking water or food).
The TDIs are used to calculate maximum allowable values (MAVs) for microcystins during recreational exposure.
MAVs based on TDIs, derived from Falconer et al, 1994 (Table A2.1)
Child = 0.088 μg/kg/day x 15 kg ×10 = 13.2 μg/L total microcystins (2)
Adult = 0.088 μg/kg/day ×70 kg ×10 = 61.6 μg/L total microcystins (3)
- 0.088 μg/kg body weight per day is the TDI (Table A2.1)
- 15 is the average weight of a child in kg (equation 2) and 70 is the average weight of an adult in kg (equation 3)
- 10 is the conversion from the amount of water accidentally swallowed per day (approximately 100 mL) to litres.
MAVs based on TDIs derived from Fawell et al, 1999 (Table A2.1)
Child = 0.08 μg/kg/day x 15 kg ×10 = 12 μg/L total microcystins (4)
Adult = 0.08 μg/kg/day ×70 kg ×10 = 56 μg/L total microcystins (5)
- 0.08 μg/kg body weight per day is the TDI (Table A2.1)
- 15 is the average weight of a child in kg (equation 4) and 70 is the average weight of an adult in kg (equation 5)
- 10 is the conversion from the amount of water accidentally swallowed per day (approximately 100 mL) to litres.
The child exposure guideline derived for microcystins (measured as total microcystins and expressed as microcystin-LR toxicity equivalents),1 from the Fawell et al, (1999) study, provided the lowest MAV (12 μg/L) and is used as the action level (red mode) – situation 1 guideline.
To derive a cell number that is equivalent to this toxin hazard, a microcystin cell quota of 6.3 × 10–7 μg total microcystins/cell is assumed. This data was obtained from five Microcystis sp. isolates from Lake Horowhenua. Microcystis sp. material analysed from this lake produces the highest values recorded in New Zealand (Wood, Stirling et al, 2006). Toxin quotas can vary under culture conditions. Wood and Dietrich (unpublished data) recently measured similar microcystin toxin quotas in environmental samples collected from Lake Rotoura (Kaikoura, South Island). Maximum microcystin cell quotas of 0.9 × 10–7 μg total microcystins/cell were recorded in their study.
Table A2.2: Toxin quotas of Microcystis sp. isolated from Lake Horowhenua
|Strain no.||Genus||Microcystins/cell (pg/cell)|
|Average toxin quota|| |
Source: Wood, Rhodes et al, 2008; Wood, unpublished data
Therefore, the equivalent concentrations of toxic cells of Microcystis sp.that are tolerable for a small child and an adult during recreational activities are:
Child =12 μg/L ×10–3 L/mL = 19,000 cells/mL (6)
6.3 ×10–7 μg /cell
Adult =56 μg/L ×10–3 L/mL = 90,000cells/mL (7)
6.3 ×10–7 μg /cell
- 12 μg/L is the MAV guideline (equation 4) for cyanobacterial exposure in children (equation 6), and 56 μg/L is the MAV guideline (equation 5) for cyanobacterial exposure in adults (equation 7)
- 10–3 is the conversion from litres to millilitres
- 6.3 × 10–7 is the toxin cell quota for total microcystins/cell.
The approximate biovolume equivalent to 19,000 cells/mL of Microcystis sp. is1.8mm3/L (see Appendix 4).
Of the New Zealand planktonic species, Microcystis spp. are the only confirmed microcystin producers. The production of microcystins by Anabaena spp. and Planktothrix spp. is suspected due to the detection of microcystins in environmental samples dominated by these species and/or via the detection of genus-specific genes involved in microcystin production in environmental samples (Wood, Stirling et al, 2006; Wood, unpublished data).
It is recommended that the biovolume of greater than 1.8 mm3/L be applied as an ‘equivalent’ guideline for populations of known potentially toxic cyanobacteria (see Tables 1 and 2) other than Microcystis sp. The rationale is that the hazard from toxicity is unlikely to exceed the worst case for an equivalent biovolume of highly toxic Microcystis spp. containing microcystins. This should allow protection from significant risk while further health risk assessments are made.
It is generally accepted that there is insufficient toxicological data available to calculate quantitative guidelines for cyanotoxins other than microcystins (eg, anatoxins and saxitoxins). In New Zealand there is a lack of information on toxin quotas from other toxin-producing species. There is substantial data on the anatoxin-a producer Aphanizomenon issatschenkoi (Wood, Rasmussen et al, 2007; Selwood et al, 2007). This data was used to calculate the concentration of anatoxin-a that would be present if the 1.8 mm3/L biovolume threshold was applied to a bloom of Aph. issatschenkoi. Although there is little data available on the toxicity of anatoxin-a, the value of 8.9 µg/L is only slightly higher than the PMAV of 6 µg/L developed for the Drinking-water Standards for New Zealand 2005 (Ministry of Health, 2005b) seems acceptable for a recreational-use water body if the assumptions used in the calculation of the microcystin TDI are adopted.
Box A2.2: Calculation of anatoxin-a for a sample containing 1.8 mm3/L of Aphanizomenon issatschenkoi
Biovolume of Aphanizomenon issatschenkoi (see Appendix 4):
= 89 mm3
Concentration of cells to obtain a biovolume of 1.8 mm3/L:
= (1.8 mm3/L x 10-3 L/mL)/(89 mm3/cell x 10-9 mm3/mm3)
= 20,200 cells/mL
Amount of toxin for this cell concentration:
= 22,200 cell/mL x 0.4 pg/mL
= 8.9 ug/L
A second guideline (ie, action level (red mode – situation 2) is required for circumstances where high cell densities, or scums, of ‘non-toxic’ cyanobacteria are present; that is, where the cyanobacterial population has been tested and shown not to contain known toxins (anatoxins, cylindrospermopsins, microcystins, nodularins or saxitoxins). Where the microcystin-related biovolume guideline is exceeded and no known cyanotoxins are present, it is appropriate to issue warnings if either the total biovolume of all cyanobacterial material exceeds 10 mm3/L or scums are consistently present (ie, scums are seen at some time each day at the recreational site).
This guideline recommendation is based on the work of Stewart et al, (2006), where it was shown that there was an increase in the likelihood of symptom reporting in bathers above a cyanobacterial cell surface area equivalent to this approximate biovolume. The potential symptoms reported above this cell surface area are primarily mild respiratory complaints. The biovolume represents a conversion from the surface area units given by Stewart et al, (2006), where total surface area of 12 mm2/mL is given as being equivalent to a total biovolume of approximately 12.5 mm3/L. This value is rounded here to a more conservative value of 10 mm3/L (two significant figures) to account for the uncertainties associated with sampling cyanobacterial populations in typical water bodies and with estimating cell densities from cell counting, which are subsequently used to derive biovolumes.
The action level (red mode) – situation 3 guideline accounts for protection from health hazards associated with the occurrence of cyanobacteria at high levels in general, demonstrated in particular by the consistent presence of scums (ie, where scums occur daily at a number of sites in a water body). This is consistent with the WHO Level 3 guideline for the occurrence of scums (WHO, 2003).
The benthic alert-level framework is based on three tiers of alert levels, with cyanobacterial abundance and attachment the triggers for changes in threshold levels. Methods for conducting a site survey are given in Section 4.2.3. The percentage coverage thresholds are based on preliminary observations. For example, the surveillance level (green mode), with less than 20 per cent coverage, is common in many rivers in New Zealand and does not necessarily indicate that a proliferation event is likely. If these mats do detach they are likely to be quickly washed away. When abundance is over 50 per cent (action level – red mode), mats commonly detach from the substrate and are more likely to accumulate along shorelines or catch in vegetation. Once mats become easily accessible the health risks are higher.
The framework is based on preliminary research and observations, and it is anticipated that these will require further refining as knowledge and monitoring tools improve. Note that the alert-level framework is designed to manage risks to recreational users. The levels given in the framework are not relevant for addressing risks to dogs that actively seek out and consume cyanobacterial mats. Raising public awareness (eg, through information/warning signs, see Appendix 9), media releases (see Appendix 10) and information pamphlets and signs (see Appendices 9 and 11), are recommended to reduce human exposure and dog poisonings.
Toxin testing should be used to help further define the health risk at sampling sites. However, quantitatively measuring toxin levels (even if samples are collected quantitatively) in benthic cyanobacteria can be problematic. This is due to the requirement for short sample turnaround times: in most commercially analysed samples, toxin concentrations are reported in micrograms of toxin per kilogram of wet weight, and it is difficult to standardise the volume of liquid and inorganic material within a mat. There is also a lack of toxicological data available, making it difficult to determine appropriate threshold values for anatoxins. For these reasons, no toxin concentrations have been recommended within the framework. It is anticipated that future research will enable the inclusion of toxin thresholds.