Acute Toxicity by Oral Co-Exposure to Palytoxin and Okadaic Acid in Mice

The frequent occurrence of marine dinoflagellates producing palytoxin (PLTX) or okadaic acid (OA) raises concern for the possible co-presence of these toxins in seafood, leading to additive or synergistic adverse effects in consumers. Thus, the acute oral toxicity of PLTX and OA association was evaluated in mice: groups of eight female CD-1 mice were administered by gavage with combined doses of PLTX (30, 90 or 270 μg/kg) and OA (370 μg/kg), or with each individual toxin, recording signs up to 24 h (five mice) and 14 days (three mice). Lethal effects occurred only after PLTX (90 or 270 μg/kg) exposure, alone or combined with OA, also during the 14-day recovery. PLTX induced scratching, piloerection, abdominal swelling, muscle spasms, paralysis and dyspnea, which increased in frequency or duration when co-administered with OA. The latter induced only diarrhea. At 24 h, PLTX (90 or 270 μg/kg) and OA caused wall redness in the small intestine or pale fluid accumulation in its lumen, respectively. These effects co-occurred in mice co-exposed to PLTX (90 or 270 μg/kg) and OA, and were associated with slight ulcers and inflammation at forestomach. PLTX (270 μg/kg alone or 90 μg/kg associated with OA) also decreased the liver/body weight ratio, reducing hepatocyte glycogen (270 μg/kg, alone or combined with OA). No alterations were recorded in surviving mice after 14 days. Overall, the study suggests additive effects of PLTX and OA that should be considered for their risk assessment as seafood contaminants.


Introduction
Marine dinoflagellates include different harmful species producing a variety of nonproteinaceous toxins that can affect human health [1]. A serious sanitary problem is associated with consumption of shellfish and/or other edible marine organisms accumulating these toxins, which can cause diverse adverse effects linked to the chemical structure of the ingested compounds. An additional concern is raised by the simultaneous exposure to multiple algal toxins that can co-occur as marine foodstuff contaminants. In particular, the combined action of structurally different toxins could lead to additive or synergistic toxic effects due to their distinct mechanisms of action and/or to an increased absorption of some toxins consequent to intestinal damage induced by other harmful compounds [2,3]. Thus, as recommended by the European Food Safety Authority (EFSA), the hazard from combined exposure to multiple algal toxins should be characterized to update the risk assessment of these compounds as seafood contaminants [4].
In the Mediterranean Sea, a frequent algal toxin accumulated in edible bivalves is okadaic acid (OA) [5][6][7][8][9][10][11], a lipophilic polyether produced mainly by Dinophysis dinoflagellates causing a human gastrointestinal syndrome named diarrhetic shellfish poisoning (DSP) [12]. OA is a potent inhibitor of protein phosphatases acting on phosphoserine or phosphothreonine residues in cell proteins [13,14]. Inhibition of these enzymes by OA leads to a hyperphosphorylated level of different proteins, which in turn influences a series of cellular mechanisms, including those at the basis of diarrhea induction or tumor promotion [15]. In particular, diarrhea as an acute effect of OA is ascribed mainly to an increased

Results
The study was carried out in two experimental phases (experiment 1 and 2). Experiment 1 was a pilot study, in which mice were administered with the 30 µg/kg PLTX (low dose) combined with 370 µg/kg OA, or with the same doses of each individual toxin. In experiment 2, the doses of PLTX were increased at 90 µg/kg (mid dose) and 270 µg/kg (high dose), while the dose of OA remained unchanged (370 µg/kg). Since experiments 1 and 2 were parts of a single study, the obtained results are cumulatively described for a comprehensive overview.

Mortality
Within 24 h after administration, no lethality was recorded in mice treated with OA (370 µg/kg) or with the low PLTX dose (30 µg/kg) as single toxins, whereas the mid-or the high dose of PLTX alone (90 and 270 µg/kg) induced the death of 2/8 mice in each group. Similarly, after the toxin co-administration, only the mid-or the high PLTX dose combined with OA resulted in lethal effects (2/8 mice in each group) ( Table 1).  During the 14-day recovery period, lethal events occurred in mice administered with the mid-dose of PLTX as a single toxin (1/3 mice, day 8) and in those co-administered with the high PLTX dose and OA (1/3 mice, day 5) ( Table 1).

Signs of Toxicity
Administration of OA (370 µg/kg) induced only transitory diarrhea, visible within 2 h from its administration (4/16 mice). Within 24 h, no signs of toxicity were noted in mice administered with the low PLTX dose (30 µg/kg). Increasing the dose at 90 µg/kg, PLTX induced recurrent episodes of scratching in one mouse, which recovered within a few hours, whereas 2/8 mice that died within 24 h showed piloerection, sedation and muscular spasms, associated with abdominal dilation in one mouse. These signs were accompanied by tremors, jumping, paralysis (mainly of the hind limbs) and dyspnea in 1-3/8 mice administered with the high dose of PLTX (270 µg/kg) as an individual toxin. Combined administration of PLTX and OA induced signs of toxicity also in mice receiving the low PLTX dose (transitory diarrhea: 2/8 mice; recurrent episodes of scratching within 2 h: 3/8 mice; piloerection: 1/8 mice). These signs of toxicity, associated with more severe ones, such as tremors, loss of righting reflex, paralysis dyspnea and muscular spasms, occurred in mice administered with the mid-or the high PLTX dose combined with OA, some of them being longer as compared to those recorded in mice administered only with PLTX. In particular, the mid-PLTX dose and OA induced transitory diarrhea (3/8 mice), piloerection and abdominal dilation (1/8 mice), as well as sedation, muscular spasms, loss of righting reflex, dyspnea and/or paralysis (mainly of the hind limbs) in 2/8 mice that died. These signs, accompanied by tremors, also occurred in spontaneously dead mice and  During the recovery period, signs of toxicity were recorded only in two spontaneously dead mice: progressive sedation and dilated abdomen occurred in 1/3 mice treated with the mid-PLTX dose and in 1/3 mice co-administered with the high PLTX dose combined with OA (data not shown).

Body Weight
As compared to controls, no significant body weight changes were recorded in mice within 24 h from the administration of OA (370 µg/kg) or the low PLTX dose (30 µg/kg) as single toxins. On the other hand, a slight but significant body weight reduction (about 5%) was recorded in mice exposed to the mid-or the high PLTX dose (90 or 270 µg/kg), both as a single toxin or combined with OA ( Figure 1).
During the 14-day recovery, no significant changes in body weight were recorded between controls and mice administered with OA or the low PLTX dose as single toxins, whereas a significant body weight reduction occurred in those administered with the mid-PLTX dose (10-18% up to day 8, when one mouse died) or with the high PLTX dose (8-9%, up to day 3). The combined toxins administration resulted in a significant body weight reduction only in mice co-exposed to OA and mid-PLTX dose (8%, at day 3) or to OA and the high PLTX dose (10-15% up to day 5, when one mouse died) ( Figure 1).

Food Consumption
Food consumption was calculated from the ratio between the diet eaten by the group of mice and the total body weight. At 24 h from administration, mice treated with the midor the high PLTX dose (90 or 270 μg/kg) showed a reduced food intake (17% in each group), as compared to controls. Similarly, mice co-administered with these PLTX doses combined with OA showed reduced food consumption (16% and 21%, respectively) (Figure 2).
A reduced food intake was also recorded during the 14-day recovery in mice treated with the mid-PLTX dose as a single toxin (9-19% reduction, up to day 8) and in those coadministered with OA and the high PLTX dose (7-17% reduction, up to day 5). In the latter, food consumption up to day 5 was lower than that of mice administered with the same dose of PLTX or OA as individual toxins (7-13% or 10-12%, respectively) ( Figure 2).

Food Consumption
Food consumption was calculated from the ratio between the diet eaten by the group of mice and the total body weight. At 24 h from administration, mice treated with the midor the high PLTX dose (90 or 270 µg/kg) showed a reduced food intake (17% in each group), as compared to controls. Similarly, mice co-administered with these PLTX doses combined with OA showed reduced food consumption (16% and 21%, respectively) ( Figure 2). Food consumption by mice recorded one day after treatment (day 1) and during the recovery period up to day 14. Data represent the daily food consumption by each group of mice (g/kg body weight/day).

Gross Pathology and Relative Organs Weight
At 24 h, no macroscopic signs were recorded in the main organs of mice treated with the low PLTX dose (30 μg/kg) as an individual toxin, whereas exposure to 90 or 270 μg/kg PLTX induced redness of the intestinal wall (1/5 mice at each dose). Administration of OA as an individual toxin (370 μg/kg) induced only a pale fluid accumulation in the proximal tract of the small intestinal lumen (6/10 mice). This finding was also recorded in 4/5 mice A reduced food intake was also recorded during the 14-day recovery in mice treated with the mid-PLTX dose as a single toxin (9-19% reduction, up to day 8) and in those co-administered with OA and the high PLTX dose (7-17% reduction, up to day 5). In the latter, food consumption up to day 5 was lower than that of mice administered with the same dose of PLTX or OA as individual toxins (7-13% or 10-12%, respectively) ( Figure 2).

Gross Pathology and Relative Organs Weight
At 24 h, no macroscopic signs were recorded in the main organs of mice treated with the low PLTX dose (30 µg/kg) as an individual toxin, whereas exposure to 90 or 270 µg/kg PLTX induced redness of the intestinal wall (1/5 mice at each dose). Administration of OA as an individual toxin (370 µg/kg) induced only a pale fluid accumulation in the proximal tract of the small intestinal lumen (6/10 mice). This finding was also recorded in 4/5 mice co-administered with OA and low PLTX dose (30 µg/kg), accompanied by intestinal wall redness (3/5 mice). These macroscopic changes co-occurred in all the mice administered with the mid-or high PLTX dose combined with OA, with digestive tract redness extended to the gastric wall. The frequency of intestinal and gastric wall redness was significantly higher in mice co-exposed to the mid-or high PLTX dose and OA, as compared to that recorded in mice treated with the corresponding doses of PLTX as a single toxin (Table 3). During the 14-day recovery, macroscopic alterations (gastrointestinal wall redness, pale fluid accumulation in the small intestinal lumen and swollen abdominal cavity) were noted only in spontaneously dead mice (1/3 mice treated with 90 µg/kg PLTX alone, day 8; 1/3 mice treated with OA combined with 270 µg/kg PLTX, day 5). No macroscopic alterations were noted in surviving mice sacrificed at the end of the withdrawal period (data not shown).
Concomitantly to necropsy, the relative weight of the main organs (organ weight/body weight ratio) was determined. As compared to controls, at 24 h from administration, only a reduced liver/body weight ratio was recorded in mice given the high PLTX dose as an individual toxin (24%) and in those co-administered with OA and the mid-or the high PLTX dose (25% in each group). The relative liver weight of mice co-administered with OA and the mid-PLTX dose was also significantly lower than that of mice treated with the same doses of each individual toxin (25% or 17%, respectively). Moreover, the relative liver weight of mice co-administered with OA and the high PLTX dose was significantly lower (25%) than that of mice administered with only OA (Figure 3a).
No significant changes in relative organ weight were recorded at the end of the recovery period (Figure 3b). administered with OA and the mid-PLTX dose was also significantly lower than that of mice treated with the same doses of each individual toxin (25% or 17%, respectively). Moreover, the relative liver weight of mice co-administered with OA and the high PLTX dose was significantly lower (25%) than that of mice administered with only OA ( Figure  3a).
No significant changes in relative organ weight were recorded at the end of the recovery period (Figure 3b).

Blood Chemistry
The blood volume sampled from the majority of mice administered with the low PLTX dose, alone or associated with OA, was not sufficient for the blood chemistry analyses, and the relevant data are not reported.
Within 24 h from administration, no significant changes in the blood chemistry parameters occurred in mice administered with each toxin alone, as compared to controls. Only a not significant increase in AST serum level was recorded in mice administered with OA or the mid-and the high PLTX dose as individual toxins (30%, 41% or 84%, respectively). A not significant increase in AST was also recorded in mice co-administered with OA and the mid-or the high PLTX dose (21% and 55%, respectively). In addition, mice given the high PLTX dose combined with OA showed a significant reduction (42%) of AP serum level (Figure 4a). After 14-day recovery, only a significant increase in AST serum level (81%) was recorded in mice co-administered with OA and the high PLTX dose, as compared to controls.

Blood Chemistry
The blood volume sampled from the majority of mice administered with the low PLTX dose, alone or associated with OA, was not sufficient for the blood chemistry analyses, and the relevant data are not reported.
Within 24 h from administration, no significant changes in the blood chemistry parameters occurred in mice administered with each toxin alone, as compared to controls. Only a not significant increase in AST serum level was recorded in mice administered with OA or the mid-and the high PLTX dose as individual toxins (30%, 41% or 84%, respectively). A not significant increase in AST was also recorded in mice co-administered with OA and the mid-or the high PLTX dose (21% and 55%, respectively). In addition, mice given the high PLTX dose combined with OA showed a significant reduction (42%) of AP serum level (Figure 4a). The AST serum concentration in these mice was also higher than that of mice administered with the corresponding dose of OA or PLTX as individual toxins (62% or 72%, respectively), even at marginal significance level (0.05 < p < 0.10) (Figure 4b).

Light Microscopy
Histological analysis showed tissue changes in the liver and/or forestomach of mice sacrificed 24 h after the administration of PLTX alone or combined with OA. At hepatic level, the fine vacuolated and granulated cytoplasm observed in hepatocytes of controls was reduced in mice (5/5) administered with the high PLTX dose (270 μg/kg), singly or  After 14-day recovery, only a significant increase in AST serum level (81%) was recorded in mice co-administered with OA and the high PLTX dose, as compared to controls. The AST serum concentration in these mice was also higher than that of mice administered with the corresponding dose of OA or PLTX as individual toxins (62% or 72%, respectively), even at marginal significance level (0.05 < p < 0.10) (Figure 4b).

Light Microscopy
Histological analysis showed tissue changes in the liver and/or forestomach of mice sacrificed 24 h after the administration of PLTX alone or combined with OA. At hepatic level, the fine vacuolated and granulated cytoplasm observed in hepatocytes of controls was reduced in mice (5/5) administered with the high PLTX dose (270 µg/kg), singly or combined with OA. This finding is compatible with a reduced glycogen content provoked by PLTX administration (Figure 5).

Light Microscopy
Histological analysis showed tissue changes in the liver and/or forestomach of mice sacrificed 24 h after the administration of PLTX alone or combined with OA. At hepatic level, the fine vacuolated and granulated cytoplasm observed in hepatocytes of controls was reduced in mice (5/5) administered with the high PLTX dose (270 μg/kg), singly or combined with OA. This finding is compatible with a reduced glycogen content provoked by PLTX administration (Figure 5). At gastric level, some mice co-exposed to the mid-(90 µg/kg) or the high PLTX dose (270 µg/kg) and OA showed slight to mild ulcers in the non-glandular part, accompanied by inflammatory cell infiltration (2/5 mice, in each group). An inflammatory reaction typical of gastritis with edema and polymorphonuclear cells infiltrate was particularly evident in the forestomach of 2/5 mice co-administered with OA and the high PLTX dose ( Figure 6).
No tissues changes were recorded in mice sacrificed after 14-day recovery (data not shown).
At gastric level, some mice co-exposed to the mid-(90 μg/kg) or the high PLTX dose (270 μg/kg) and OA showed slight to mild ulcers in the non-glandular part, accompanied by inflammatory cell infiltration (2/5 mice, in each group). An inflammatory reaction typical of gastritis with edema and polymorphonuclear cells infiltrate was particularly evident in the forestomach of 2/5 mice co-administered with OA and the high PLTX dose ( Figure 6).
No tissues changes were recorded in mice sacrificed after 14-day recovery (data not shown).

Discussion
In the last decades, harmful algal blooms have increased throughout the world's seawaters with negative impacts to human health and local economies. In the Mediterranean Sea, blooms of Dinophysis dinoflagellates producing OA and/or its analogues frequently occurred since the nineteen-eighties with frequent episodes of DSP [5][6][7][8][9][10][11]. In the last decades, blooms of Ostreopsis (mainly Ostreopsis cf. ovata) were also recorded along the Mediterranean coasts [39][40][41][42][43][44]. Concomitantly to these events, PLTX and its analogues had been detected in different edible marine organisms, also at concentrations exceeding the maximum admitted level proposed by EFSA [64][65][66][67][68][69][70]. In these situations, OA and PLTX could co-occur in seafood, raising concerns for possible additive or synergistic effects in consumers. Thus, to characterize the hazard by oral co-exposure to OA and PLTX, a comparative acute oral toxicity study was carried out in mice: the effects of OA (370 µg/kg) and PLTX (30,90 or 270 µg/kg) co-administration were compared to those of each individual toxin.
Within 24 h, the mid-or the high dose of PLTX (90 or 270 µg/kg) induced lethal effects, even individually or combined with OA (370 µg/kg), with the same incidence (2/8 mice in each group). Lethality data indicate no additive or synergistic effects between PLTX and OA and put in evidence the high lethal potency of PLTX. Its individual lethal dose (90 µg/kg) was lower than 300 µg/kg, previously recorded as NOEL (no observed effect level) in mice under the same experimental conditions [59], as well as those lower than 100 µg/kg, previously recorded as a non-lethal dose for mice, even prolonging the time of observation to 96 h [60]. This finding highlights the variable inter-individual sensitivity to PLTX, already noted in mice as variable lethality after repeated oral administration [61], and suggested by epidemiological data in humans [48,[50][51][52][53]. The different inter-individual sensitivity to PLTX could be related, at least in part, to genetic factors, such as a variable inter-individual expression of Na + /K + -ATPase subunits' isoforms, as suggested by an in vitro toxicogenetic study on human monocytes [71].
As compared to controls, administration of the mid-or the high PLTX dose resulted in a reduced food intake (17% in each group), comparable to that of mice exposed to the same PLTX doses combined with OA (16% and 21% reduction, respectively). These data cannot be statistically analyzed, being derived from food consumption by each whole group of animals and not from the mean consumption by each mouse, but they support the adverse effects of PLTX when combined with OA. In parallel, administration of these PLTX doses resulted in a slight but significant body weight loss (about 5%), which was not influenced by OA co-administration. In addition, at these doses, PLTX caused a series of clinical signs (scratching, piloerection, sedation, muscular spasms, tremors, jumping, paralysis of the hind limbs and/or dyspnea) suggesting the involvement of the neuromuscular system, some of them previously noted after single oral administration of PLTX or 42-hydroxy-PLTX in mice [58][59][60]72]. On the other hand, OA induced only visible transitory diarrhea (4/16 mice), comparable in incidence with that recorded in mice co-exposed to OA and each dose of PLTX (2-3/8 mice). Mice co-administered with the two toxins showed the combined signs of toxicity induced by each individual toxin. Although they appeared longer and/or more frequent as compared to those recorded after the individual toxin administration, the significance of these differences is not statistically evident due to the limited number of mice. Two signs (scratching and piloerection) occurred even in mice receiving the low PLTX dose (30 µg/kg) combined with OA, but were not noted after the administration of the same doses of each individual toxin. Thus, globally, these findings suggest an additive effect between PLTX and OA, supported also by necropsy and histology findings.
Necropsy showed distinct macroscopic changes in the gastrointestinal tract of mice administered with OA or the mid-and high PLTX dose: pale fluid accumulation in the proximal tract of the small intestine occurred in mice exposed to OA, whereas redness of the small intestinal wall was noted in mice administered with PLTX. These alterations cooccurred in mice co-exposed to OA and even to the low PLTX dose, indicating an additive effect. Moreover, mice receiving OA with the mid-or the high PLTX dose showed redness of the digestive tract extended to the gastric wall, where histological analysis revealed the presence of forestomach lesions (slight focal ulcers and inflammatory cells infiltrate; 2/5 mice in each group). An inflammatory reaction of the forestomach submucosa typical of gastritis, with edema and polymorphonuclear cells infiltrate, was more evident in mice co-exposed to the high PLTX dose and OA. Gastric lesions (weak erosions and slight fluid accumulation) were previously recorded in mice orally administered with 200 µg/kg PLTX or its analogue ostreocin-d [73], while acute forestomach inflammation was noted after higher doses (≥424 µg/kg) of PLTX or 42-hydroxy-PLTX oral administration [59,72]. Gastric alterations were previously recorded in mice after single oral OA administration. An acute inflammation of the forestomach squamous mucosa and erosions of the fundic mucosa were recorded after the administration of 2-10 µg OA/mouse [26], whereas light erosion of surface forestomach epithelial cells was noted at 150 µg/kg [30]. In addition, bloody content in the gastric lumen and/or wounds were noted after the administration of OA doses ≥750 µg/kg [23], whereas acute forestomach submucosal inflammation with epithelial vacuolar degeneration and/or reactive hyperplasia of keratinized epithelium were recorded at 1000 µg/kg [21]. Furthermore, our study indicates the stomach as a target of PLTX and OA association, with an additive effect between the two toxins. While the individual toxins did not induce any gastric alteration visible by light microscopy, the mid-or the high PLTX dose combined with OA induced ulceration and inflammation in the forestomach, probably due to the combined local irritant action. Even though an adverse effect at the forestomach is relevant for rodents rather than for humans, which are devoid of this anatomical structure, the combined irritant effects of PLTX and OA at the digestive tract should be considered to update the risk assessment of these toxins as seafood contaminants.
Administration of the high dose of PLTX as a single toxin or its co-administration even at the mid-or the high dose with OA also affected the liver, as shown by the decreased liver/body weight ratio. As compared to controls, the high PLTX dose induced 24% reduction of liver/body weight ratio, comparable to that recorded after the mid-or the high PLTX dose combined with OA (25% in each group). In addition, the liver/body weight ratio in mice co-exposed to the mid-PLTX dose and OA was significantly lower (17%) than that of mice exposed to the same PLTX dose as the individual toxin, indicating its additive effect with OA. On the other hand, histological analysis suggests a decreased glycogen content in hepatocytes of almost all the animals given the highest PLTX dose, both as a single toxin or combined with OA. Furthermore, these findings were not accompanied by increased serum levels of enzyme indices of liver injury. Only a reduction of alkaline phosphatase (42%) was recorded in mice co-exposed to OA and the high PLTX dose, but the enzyme levels remained within the physiological range of mice [74]. A decreased glycogen content in liver cells was previously noted in mice after single oral administration of PLTX doses ≥600 µg/kg [59] or OA doses ≥500 µg/kg [23]. OA had been previously shown to induce ultrastructural alterations in liver cells at doses ≥500 µg/kg [23], necrotic foci and lipid vacuoles at 700 µg/kg [33], and cytoplasmic vacuolation at 2000 µg/kg [21]. Although our findings do not show a hepatotoxic effect for OA, they confirm the liver as a target of PLTX at a dose as low as 270 µg/kg, which is lowered to 90 µg/kg if associated with OA.
Adverse effects due to PLTX administration, even with lethal outcomes, occurred during the 14-day recovery in mice administered with the mid-dose of PLTX or co-administered with the high PLTX dose and OA: 1/3 mice died at days 8 and 5, respectively. Signs of toxicity (progressive sedation and abdominal dilation) and macroscopic alterations (gastrointestinal wall redness and pale fluid accumulation in the small intestine) were noted only in these spontaneously dead mice. Other findings were recorded only in these groups of mice (reduced body weight and food consumption; 81% increase in AST serum level only in those co-administered with the high PLTX dose and OA). Furthermore, no liver or gastric alterations were observed in surviving mice at the end of the withdrawal period, suggesting a complete recovery of the toxic effects.

Dose Selection and Experimental Design
Selection of the toxins' doses. The dose of OA (370 µg/kg) was selected on the basis of a preliminary study as a non-lethal dose inducing fluid accumulation in the small intestine and mild visible diarrhea. At this dose, OA co-administration with PLTX would avoid or reduce the excretion of the latter through feces, which could limit its effects. In our experimental conditions, the single oral administration of this OA dose to CD-1 female mice induced mild transitory diarrhea within few hours and/or fluid accumulation in the small intestine notable within 24 h by necropsy. PLTX was administered at three doses (30,90 and 270 µg/kg), selected on the basis of previous findings after its acute and repeated oral toxin administration to CD-1 female mice. In particular, in our experimental conditions: (i) after acute oral PLTX administration, the NOEL (no observed effect level) was 300 µg/kg [59]; (ii) after daily oral administration for 7 days, the NOAEL (no observed adverse effect level) was 3 µg/kg/day, whereas toxic and even lethal effects occurred after four daily oral administrations at 30 µg/kg/day [61]. Hypothesizing an additive or synergistic effect, a pilot experiment (experiment 1) was carried out co-administering 370 µg/kg OA and 30 µg/kg PLTX. Then, in a subsequent study (experiment 2), PLTX was administered at 3-fold increased doses (90 and 270 µg/kg), the dose of OA remaining unchanged (370 µg/kg).
Experimental design. At day 0 (D0), groups of 8 mice were administered by gavage with a single dose of OA (370 µg/kg) combined with PLTX (30, 90 or 270 µg/kg), or with the corresponding doses of each toxin alone. Control mice were administered with the vehicle (10 ml/kg PBS, containing 1.8% ethanol, v/v). After administration, mice were monitored up to 24 h, when subgroups of 5 mice were sacrificed. Subgroups of 3 mice were maintained for a 14-day withdrawal period, recording clinical signs, body weight and food consumption, daily in the morning. Food consumption was calculated from the amount of diet eaten by all the mice of each group divided by the total body weight. At the scheduled times of sacrifice (24 h after the treatment: day 1, D1; 14 days after the treatment: day 14, D14), mice were anesthetized by intraperitoneal injection of tiletamine/zolazepam (20 mg/kg; Zoletil ® ; Virbac; Milan, Italy) and xylazine (5 mg/kg; Virbac; Milan, Italy), and bled to death through the abdominal aorta to collect blood for chemistry analyses (see Section 4.4). Then, mice were necropsied, and liver, heart, lungs, kidneys, spleen and brain were removed and weighed. Samples of these organs and other tissues (see Section 4.5) were fixed in neutral buffered formalin for the histological analysis. Similarly, animals that died spontaneously were immediately weighed, and the blood was collected for chemistry analyses; the main organs and tissues were weighed and/or fixed for the histological analysis. The groups of treatment, the number of mice and the scheduled sacrifice are reported in Table 4.  0  8  5  3  2  0  370  8  5  3  3  30  0  8  5  3  4  30  370  8  5  3   Experiment 2   1  0  0  8  5  3  2  0  370  8  5  3  3  90  0  8  5  3  4  270  0  8  5  3  5  90  370  8  5  3  6 270 370 8 5 3

Blood Chemistry Analyses
Blood was allowed to clot for 15 min at room temperature and centrifuged at 2000× g for 10 min at 4 • C. Serum was separated and stored at −80 • C until the analyses were carried out. Using an automatized analyzer (AU400 Olympus with Beckman Coulter reagents), the following parameters were determined: alkaline phosphatase (AP), alanine aminotransferase (ALT), aspartate aminotransferase (AST), glutamate dehydrogenase (GDH), total cholesterol, creatinine, glucose, total proteins, triglycerides, urea, albumin, globulins, albumin/globulins ratio, sodium ions, potassium ions, chloride ions, calcium ions and inorganic phosphorus (Pi).

Histological Analysis
Heart, liver, lungs, kidneys, spleen, stomach, duodenum, jejunum, colon, rectum, pancreas, thymus, cerebrum, cerebellum, spinal cord, uterus, ovaries and skeletal muscle (soleus) were fixed in neutral buffered 10% formalin, embedded in paraffin and cut in sections of 5 µm. Sections were deparaffinized, rehydrated and stained with hematoxylin and eosin, following standard techniques for histological analyses by light microscopy. Pictures were obtained with a Nikon eclipse i 50 microscope equipped with a DS-Vi1 digital camera and NIS-Elements Microscope Imaging version 3.2 Software (Nikon Instruments; Tokyo, Japan).

Statistical Analysis
Data are expressed as mean ± standard error (S.E.). Significant differences between groups were calculated by one-way analysis of variance, followed by the Dunnett's test for multiple comparisons of unpaired data, accepting p values lower than 0.05 as significant. Frequency of clinical signs and of macroscopic changes at gross pathology is expressed as number of mice showing the sign(s) or alteration(s)/number of treated mice, and significant differences between groups were calculated by Fisher's exact test, accepting p values lower than 0.05 as significant.

Conclusions
The acute toxicity study in mice demonstrates that single oral co-administration of PLTX (30,90 or 270 µg/kg) and OA (370 µg/kg) resulted in a LOAEL (lowest observed adverse effect level) of 30 µg/kg PLTX combined with 370 µg/kg OA. Moreover, it highlights a high oral lethality of PLTX, which is not influenced by OA co-administration. The other severe clinical signs of toxicity by PLTX and OA association are ascribed mainly to PLTX and are increased in frequency and/or duration by OA coadministration, which suggests an additive effect between these toxins. These aspects should be considered to update the risk assessment of PLTX and OA as seafood contaminants.