3.1. Cylindrospermopsin
Cylindrospermopsin (CYN) occurs in fresh and brackish waters worldwide, due to the presence of the cyanobacterial genera
Cylindrospermopsis,
Aphanizomenon,
Anabaena,
Raphidiopis,
Lyngbya and
Umezakia [
14].
Studies of bioaccumulation of cylindrospermopsin in gastropod snails, bivalves [
15], crustaceans [
16], amphibian tadpoles and fish [
17] demonstrated that this toxin is concentrated into tissues from free solution and from toxic
Cylindrospermopsis cells. The highest accumulation was seen in mussels with a whole-body concentration of almost 3 mg/kg dry weight, with the maximum tissue concentration found in haemolymph. Cylindrospermopsin appears in muscle tissue as well as viscera, increasing the possibility of consumption in these seafoods.
Human poisoning from CYN has been previously recorded. In Palm Island in 1979, for example, 150 people received hospital treatment for an unusual hepatoenteritis after drinking water from a reservoir that was treated with copper to remove a
Cylindrospermopsis algal bloom [
18]. The absence of toxin exposure information, however, makes this case unusable for the purposes of deriving a TDI. There have, however, been several published accounts of the oral toxicity of cylindrospermopsin in animals, with the majority of studies using a single dose [
19,
20,
21]. Repeat oral dosing after a two week interval showed unexpectedly enhanced toxicity, indicating residual damage to the animals from the first dose [
22].
A study by Humpage and Falconer [
23], following the protocols set out by the OECD for subchronic oral toxicity assessment in rodents, exposed male Swiss Albino mice to cylindrospermopsin through drinking water and through gavage (dosing by mouth) [
9]. The first trial used a cylindrospermopsin-containing extract from cultured
Cylindrospermopsis raciborskii, supplied in drinking water for 10 weeks. The dose of cylindrospermopsin ranged from 0 to 657 μg/kg/day at 4 levels. The animals were examined clinically during the trial and showed no ill effects other than a small dose-related decrease in body weight compared to controls after 10 weeks. Liver and kidney weights were significantly higher with increasing dose. Several biochemical indicators of liver function showed dose-related changes. For instance, serum total bilirubin and albumin increased, while serum bile acids decreased. Liver enzyme changes in the serum showed a different pattern to those seen with acute liver poisoning or hepatitis, as only a small increase in serum alanine aminotransferase and a larger increase in alkaline phosphatase were observed. There was also a decrease in aspartate aminotransferase. The most substantial change observed was in the urine protein/creatinine concentration, which decreased sharply with dose. This was interpreted as reflecting decreased protein synthesis in the kidney through inhibition by the toxin. Histopathological examination of all internal organs showed changes only in the liver and kidney. Dose-related hepatocyte damage and renal proximal tubule necrosis were observed [
23].
When it was apparent from these results that lower oral doses were required to find the No Observed Adverse Effect Level, Humpage and Falconer [
23] carried out a second trial in which mice were dosed orally by gavage over 11 weeks with 0, 30, 60, 120 and 240 μg/kg/day of purified cylindrospermopsin. The same trends in serum parameters were seen, but with no statistically significant changes. Organ weights showed most sensitivity to these low doses with significant increases in body weight, and as a percentage of body weight, in liver, kidney, adrenal glands and testis. Minor histopathological damage was seen in the liver at the two upper dose levels, and in kidney proximal tubules at the highest dose. Urine protein/creatinine decreased progressively with dose, reaching significance at 120 μg/kg/day of oral cylindrospermopsin.
At very low dose levels of toxins compensatory changes occur in metabolism to restore homeostasis. The increases in organ weight can be expected to compensate for reductions in function as seen in the liver and kidneys, and compensation for stresses resulting from the toxin, for example in the adrenal glands. It therefore becomes subjective to decide where the No Observed Adverse Effect Level occurs, depending on which effect is considered adverse. From the urine protein data it is clear that the NOAEL is below 120 μg/kg/day. However statistically significant change in kidney weight occurred at 60 μg/kg/day.Thus to adopt the conservative viewpoint that the most sensitive response should be considered as the indicator of adverse effect, the dose of 30 μg/kg/day was accepted as the NOAEL from these trials [
23]. Recent studies have corroborated this value and shown that both males and females are affected to a similar degree [
24].
Thus using 30 μg/kg/day as the NOAEL, the TDI for cylindrospermopsin can be calculated.
TDI(μg/kg/day) = 30 ÷ Uncertainty factors (4)
The uncertainty factors are 10 for intraspecies variability, 10 for interspecies variability and an additional factor of 2 given that there is recent evidence that CYN has teratogenic [
25] and reproductive effects [
26], and there is preliminary evidence that it may be carcinogenic [
20,
22,
27]. In these circumstances a reasonable additional uncertainty factor of 2 is applicable.
Step 1: TDI = 30 ÷ 200 = 0.15 μg/kg/day
Step 2: Acceptable limit per day = 0.15 μg/kg/day × bodyweight (kg) × 1.0 (allocation factor)
Table 2.
Acceptable daily limit for cylindrospermopsin with age and bodyweight.
Table 2.
Acceptable daily limit for cylindrospermopsin with age and bodyweight.
Age group (years) | Average bodyweight (kg) | Acceptable daily limit (μg/day) |
---|
≥17 | 74 | 11 |
2–16 | 38 | 5.7 |
Step 3: Obtain high-level consumption data for consumers aged 17 years and above and 2–16 years old—see
Table 1.
Step 4: Derive health guideline level (μg/kg) for cylindrospermopsin in whole seafood sample—see
Table 3. Acceptable limit (μg/day) ÷ consumption (kg/day) (that is Step 2 ÷ Step 3).
Table 3.
Health guideline values for cylindrospermopsin toxin in seafood.
Table 3.
Health guideline values for cylindrospermopsin toxin in seafood.
Health Guideline Value (μg/kg of whole organism sample) |
---|
Age group (years) | Fish | Prawns | Mussels/Molluscs |
---|
≥17 | 29 | 29 | 62 |
2–16 | 18 | 24 | 39 |
Preliminary
in vitro evidence suggests that deoxyCYN has similar potency to CYN, and so this analogue should be included in monitoring programs and toxicity assessments [
28,
29].
3.2. Microcystins
Microcystins have been the most thoroughly investigated cyanobacterial toxin group, and is still the major toxin group under investigation. The majority of human and animal microcystin-related poisonings worldwide have been associated with the presence of the cyanobacterial species Microcystis aeruginosa and M. flos-aquae. Microcystins may also be produced by species of the planktonic genera Anabaena, Planktothrix (Oscillatoria), Nostoc, and Anabaenopsis.
The microcystins are a family of cyclic peptide toxins, containing seven peptide-linked amino acids, in which acids in L-configuration occupy two positions in the ring. A range of L-amino acids may take these positions, with consequences for toxicity over a range approaching ten-fold. To standardize guideline values, microcystins toxicity is expressed as toxicity equivalent to microcystin-LR (leucine, arginine) [
6].
The most significant recorded human poisoning event due to microcystins occurred in Brazil in 1996 at the Caruaru Dialysis Clinic [
30]. Cyanobacterial toxins contaminated the clinic’s water source, so that intravenous exposure to microcystins and cylindrospermopsins during routine renal dialysis treatment led to acute liver failure in 100 patients and resulted in 76 deaths [
30]. An Australian study has revealed toxic liver damage (an increase in the activity of the hepatic enzyme–glutamyl-transferase) coincided with a bloom of
M. aeruginosa in 1981 in a drinking water supply in Armidale NSW [
31,
32]. At lower doses there is also evidence that microcystin’s effects on cell regulation may increase the growth rate of existing tumours (tumour promotion) [
33]. This evidence has been provided by both experimental and epidemiological studies [
34] with microcystins and nodularins implicated in tumour promotion in both the liver [
35,
36] and colon [
37]. However, reliable dose-response data for studies recording human exposure to microcystins is lacking, therefore animal studies must be relied upon to derive a TDI.
The only animal study that has met the OECD criteria for subchronic oral toxicity assessment in rodents is that of Fawell
et al. (1994) [
38]. Their 13-week oral gavage study of mice exposure met the criteria for experimental design, duration of exposure and used both sexes of animal. Fawell
et al. (1994) concluded that the NOAEL for microcystin-LR was 40 μg/kg/day. This is supported by an oral toxicity study carried out in pigs, which resulted in a Lowest Observable Adverse Effect Level (LOAEL) of 100 μg/kg/day of microcystin-LR equivalents [
6]. The International Agency for Research on Cancer has recently classified microcystin-LR as a “possible human carcinogen” (Class 2B) [
39]. The cited mechanism of action is protein phosphatase inhibition and so the assumption of a threshold dose below that no adverse effect occurs still applies. It is appropriate, therefore, to derive a TDI from the NOAEL.
Thus using 40 μg/kg/day as the NOAEL, the TDI for microcystin-LR (and equivalent toxins) can be calculated:
TDI (μg/kg/day) = 40 μg/kg/day ÷ uncertainty factors (5)
The uncertainty factors are 10 for intraspecies variability, 10 for interspecies variability, and an additional factor of 2 for limitations in data, including evidence of tumour promotion, suspicion of carcinogenesis [
39], conflicting data in teratogenesis, and recent evidence of reproductive toxicity.
Step 1: TDI = 40μg/kg/day ÷ 200 = 0.2μg/kg/day.
Step 2: Acceptable limit per day = 0.2μg/kg/day × bodyweight( kg) × 1.0 (allocation factor)
Table 4.
Acceptable daily limit for microcystins with age and bodyweight.
Table 4.
Acceptable daily limit for microcystins with age and bodyweight.
Age group (years) | Average bodyweight (kg) | Acceptable daily limit (μg/day) |
---|
≥17 | 74 | 14.8 |
2–16 | 38 | 7.6 |
Step 3: Apply high level consumption data from
Table 1.
Step 4: Derive health guideline level for seafood for microcystin-LR and similar toxins.
Table 5.
Derived health guideline values for microcystins in seafood.
Table 5.
Derived health guideline values for microcystins in seafood.
Health Guideline Value (μg/kg of whole organism sample) |
---|
Age group (years) | Fish | Prawns | Mussels/Molluscs |
---|
≥17 | 39 | 39 | 83 |
2–16 | 24 | 32 | 51 |
3.3. Nodularin
Nodularin is a hepatotoxin produced by
Nodularia spumigena.
N. spumigena is primarily regarded as a brackish water species and forms blooms in estuarine lakes in Australia and New Zealand and in the Baltic Sea in Europe [
40]. In addition to these saline environments, there have also been frequent blooms in the freshwater lakes of the lower River Murray in South Australia [
41]. As a brackish water species
N. spumigena is the most common toxic cyanobacterial species in the Gippsland Lakes.
Nodularin is structurally very similar to microcystin and has a similar mode of toxicity showing the same hepatotoxic effects through the inhibition of protein phosphatases [
42]. Some have suggested it may be more carcinogenic than microcystin [
43]. No human poisonings have been recorded as a result of ingestion of
N. spumigena [
31] however it is “at least as hepatotoxic as microcystins for intraperitoneal exposure in experimental animals and, given its identical mode of action, can be regarded as presenting at least the same risk to human health as microcystins if ingested in drinking water” [
40]. Due to the structural similarities between microcystins and nodularin and the lack of animal studies looking at the health effects associated with exposure to nodularin it is acceptable that the guideline value for microcystins be applied to nodularin. Several published risk assessments have used a similar approach [
5,
44]. For calculations, refer to the ‘Microcystins’ section above.
3.4. Saxitoxins
There have been no recorded cases of human poisonings as a result of ingestion of saxitoxins produced by cyanobacteria [
6]. There are, however, documented cases where saxitoxins arising from dinoflagellates have led to neurotoxic effects as well as death in humans [
45]. The established health guideline value for saxitoxins produced by dinoflagellates is 0.8 mg/kg (STX toxicity equivalents) in bivalve mussels (shellfish). This value is used by Food Standards Australia New Zealand (FSANZ Food Standard 1.4.1) and the Victorian Shellfish Quality Assurance Program [
46].
Cyanobacteria produce different analogues of saxitoxin as compared to microalgae. In the U.S.
Aphanizomenon produces saxitoxins, however the only known cyanobacterial producer in Australia is
Anabaena circinalis [
47]. Current evidence suggests that this species produces mainly the less toxic C-toxin analogues, along with lesser amounts of the more toxic analogues commonly found in marine microalgae. However, it is known that acidic or alkaline conditions and heat can chemically convert the C-toxins to the more toxic variants, and that similar bioconversions can occur within shellfish [
48]. There is currently no information on the degree of inter-conversion that occurs from these low toxicity variants to the more toxic ones during cooking or digestion in the stomach, although it is known that the more toxic variants are stable at normal cooking temperatures [
49].
The health guideline value of 0.8 mg/kg is a long-standing limit that has been used for marine saxitoxins. There has been a long history of success (nearly 50 years) associated with this level, with no evidence of human illnesses from commercially harvested products [
45]. Weckell
et al. attempted to trace the origin of the guideline value and noted that it originated from a U.S. Marine Biotoxins Program [
7]. The limit was established in the 1930s ‘based on bioassays measuring toxic activity in mice’, but the exact details of its derivation are uncertain [
7]. Despite this, the saxitoxin guideline value has been used by all major regulatory agencies around the world for many years and it does appear to be protective of public health [
7].
However, in 2009 the European Food Safety Authority (EFSA) was asked by the European Commission to review the existing ML for saxitoxin. Following a review of published literature EFSA established an ARfD for saxitoxin of 0.5 μg saxitoxin equivalents/kg bw based on human data [
49]. Together with an estimate of acute dietary intake for shellfish, EFSA revised the ML down to 75 μg STX equivalents/kg shellfish meat. As the existing saxitoxin ML of 0.8 mg/kg in the Food Standards Code has a long history of effective protection for human health, this substantial downward revision is under further consideration.