Introduction

Coastal areas, open oceans, and brackish systems have been affected worldwide by harmful algal blooms (HABs) over the last decades (Anderson et al. 2012; Hallegraeff 2010). Microalgal bloom events are characterized by the cellular proliferation and, occasionally, high biomass accumulation of toxic or noxious algal species that are harmful to aquatic fauna and food web dynamics (Neves and Rodrigues 2020). HABs occurrence has increased recently in various coastal regions around the globe which is likely due to natural dispersion of species by currents and winds, organisms transport via human activities (e.g., ballast water and aquaculture), increase in water column temperature, coastal eutrophication, and the better detection of HABs and their toxins in the environment (e.g., Glibert and Burkholder 2006; Gobler 2020; Granéli et al. 2011; Griffith and Gobler 2020; Wells et al. 2015; Xiao et al. 2019; Zohdi and Abbaspour 2019).

Many algal species synthesize toxins that are responsible for poisoning incidents worldwide. Most of these toxins are secondary metabolites produced by algal cells, but many other bioactive metabolites are simple molecules (e.g., polyunsaturated aldehydes in diatoms) derived from their primary metabolism (Ianora et al. 2011). Moreover, dissolved toxins released in the water by algal cells may surpass intracellular toxin levels in the late stages of algal growth (MacKenzie et al. 2004), which may be a risk factor for fauna survival (Castrec et al. 2018; Costa 2016). Generally, HABs are associated with fauna contamination by biotoxins that may eventually lead to mortality of marine species (Branch et al. 2013; Landsberg 2002; Shumway 1990). Among marine shellfish, bivalves are recognized as common vectors of toxins to their consumers regarding their intoxication susceptibility through filter-feeding on toxic algae (Basti et al. 2018; Deeds et al. 2008; Lopes et al. 2019; Neves et al. 2019), although, non-bivalve molluscs may also act as toxin vectors (Costa et al. 2005; Mafra Jr et al. 2019; Neves et al. 2015a; Shumway 1995). Deposit-feeders may accumulate toxins by the ingestion of biodeposits (e.g., pseudofeces, feces) containing toxins or directly by microalga consumption (e.g., Dowsett et al. 2011; Neves et al. 2015a; Persson et al. 2006, 2008; Persson and Smith 2009). In addition, carnivorous molluscs may accumulate toxins via trophic transfer by predation on contaminated prey (e.g., Bricelj et al. 2012; Carreto et al. 1996; Ito et al. 2004; Wekell et al. 1996).

The phylum Mollusca is the second largest in the animal kingdom, including living species and fossil records, and extremely diverse in terms of animal forms, feeding modes, and behaviors (Kershaw 1983). The four most abundant classes of living molluscs are the Polyplacophora (e.g., chitons), Bivalvia (e.g., bivalves), Gastropoda (e.g. snails), and Cephalopoda (e.g., squids and octopuses). Molluscs provide essential ecosystem services such as habitat structure, water purification through filtration, serve as food item for a wide range of predators—including humans, and are economically valuable (reviewed in Gutiérrez et al. 2003; Neves et al. 2020). The average annual growth rate of global food production in mariculture systems has increased 5.1% from 2010 to 2017 (FAO - Food and Agriculture Organization of the United Nations 2019), in which the world production of molluscs in marine waters comprised US$33.9 billion in 2018 (FAO - Food and Agriculture Organization of the United Nations 2016). Molluscan shellfish may be environmentally exposed to harmful algal blooms; thus, the edible mollusc may accumulate toxins and become unsafe for human consumption. The consumption of contaminated shellfish is a global public health concern since phycotoxins may lead to dangerous human poisonings (Neves and Rodrigues 2020). Besides human health risks, phycotoxin toxicity may cause physiological injuries in molluscan shellfish altering global fisheries production and the quality of marine ecosystems (Ajani et al. 2017; Khora and Jal 2018; Shumway 1990). In this way, this review aimed to present an overview of the main impacts of harmful algal blooms on molluscan shellfish by evaluating the exposure of cultivated molluscs in marine environments to microalgal and cyanobacteria species that form blooms and/or synthesizes toxins. Moreover, it was assessed the multiple responses of molluscs to HABs, the trophic transfer of phycotoxins, the dynamics of phycotoxins in molluscs, and the risks for human health. Finally, future perspectives were provided considering the economic importance of shellfish production and the impacts of contaminated molluscs on human health. The present review is within the context of the Sustainable Development Goal 14 (i.e., Life Below Water) of the United Nations by increasing the research efforts in marine sciences in order to improve the ocean health and environmental and human safety.

Material and methods

An extensive literature search limited to English was conducted using indexed databases (ISI Web of Science, Scopus, and Google Scholar) and reference lists of the retrieved papers searching for the effects of harmful alga and/or phycotoxins on marine molluscs, by applying a combination of the keywords “molluscs,” “mollusks,” “Mollusca,” “phycotoxins;” and pairwise combinations of “harmful alga,” “algal bloom,” and “toxic algae.” Publications included in the analysis were those that reported exposure of marine molluscs to harmful alga species and/or phycotoxins. All types of scientific publications were included except books that were not freely available online.

Of the 2420 results obtained from the literature search, 150 scientific publications were selected after thorough inspections for subject adherence and thus retrieved for further analyses (see Supplementary Material Table S1 to access all the reference list and publication details). Our analytical database consisted in data compiled from scientific studies published up to 2015. Considering that the first record on the effects of harmful alga and/or phycotoxins on marine molluscs dated back to 1972, our extensive literature search encompassed a time span of 44 years, ceasing intentionally in 2015 to coincide with the last scientific summary for policy makers (Kudela et al. 2015), provided by the SCOR-IOC Scientific Steering Committee of the Global Ecology and Oceanography of Harmful Algal Blooms (GEOHAB; UNESCO). The responses of molluscan species described in the literature were evaluated, compiled, and discussed here in three distinct sub-sections: (1) feeding and behavioral responses, (2) metabolic and physiological reactions, and (3) fitness consequences. A Spearman correlation was performed to test for statistical significance between marine production of molluscs (in million tonne) and the relative number of available studies by molluscan category. Statistical analysis was performed using the software Statistica 8.0 (StatSoft).

Moreover, data from the available literature reporting values of phycotoxin concentrations in molluscs and their consumers were used to calculate the biomagnification factor (BMF), which is the ratio between the toxin concentration measured in aquatic consumers and their prey (sensu Gray 2002).

Results and discussion

Data overview

The number of published studies is highly asymmetric among molluscan groups, with 90% of studies focused on bivalves, 5.3% on cephalopods, and 4.7% on gastropods. The highest research effort on bivalves is likely related to the greater importance of this molluscan group as a food item, its higher commercial value, and the amount produced in mariculture systems. In addition, filter-feeder bivalves are traditionally monitored for human safety consumption; while carnivorous, herbivorous, and detritivorous molluscs were lately recognized as phycotoxin sources (e.g., Bricelj et al. 2012; Dowsett et al. 2011; Ito et al. 2004; Neves et al. 2015a; Persson et al. 2008; Persson and Smith 2009).

Production of farmed molluscs in 2013 (more than 21 million tonnes) represented 14% of the total seafood and fish produced worldwide (FAO - Food and Agriculture Organization of the United Nations 2014; Ritchie and Roser 2019). The amount of marine shellfish production varies according to each molluscan groups (FAO - Food and Agriculture Organization of the United Nations 2012): (1) clams, cockles, and arkshells ~5 million tonnes; (2) oysters ~4 to 5 million tonnes; (3) mussels ~2 million tonnes; (4) scallops and pectens ~2 million tonnes; (5) other marine molluscs, including cephalopods ~0.5 to 1 million tonne, and (6) gastropods ~0.5 million tonne (Fig. 1a). Aquaculture production of bivalves in marine waters accounts for more than 90% of the whole production of farmed molluscs. High production in mariculture systems demands monitoring of physical, chemical, and biological conditions and animal quality, which seems to reflect the higher number of studies focusing on harmful alga effects on bivalves (Fig. 1b). Among the molluscan categories, the impacts of harmful algae were similarly evaluated by scientific studies on clams, cockles, and arkshells (26.6% of retrieved papers), oysters (26.2%), and mussels (26.6%), followed by scallops and pectens (14.2%), and finally cephalopods (3.4%), and gastropods (3%) (Fig. 1b). A significant positive correlation was found between the marine production of molluscs (in million tonne) and the number of available studies by molluscan category (Spearman correlation, ρ= 0.84, p <0.05).

Fig. 1
figure 1

a World aquaculture production of molluscs in marine waters (data compiled from FAO - Food and Agriculture Organization of the United Nations 2012). b Pencent proportion of retrieved scientific studies (n=150) among the six molluscan categories: clams, cockles, and arkshells; oysters; mussels; scallops and pectens; cephalopods; and gastropods

Full size image

The number of species addressed by retrieved scientific publications consisted of 56 bivalves, seven gastropods, and seven cephalopods. All species names are presented in Supplementary Material Table S1. Most of the information available, in the percentage of scientific studies, focused in only six bivalve species: the oysters Crassostrea gigas (19%) and C. virginica (12%), the mussels Mytilus edulis (14%) and M. galloprovincialis (9%), and the clams Mercenaria mercenaria and Ruditapes philippinarum (~7%). These six bivalve species have been listed by FAO as the major farmed molluscs in marine waters (FAO - Food and Agriculture Organization of the United Nations 2006-2020), which problably prompted more scientific efforts on these commercially valuable species.

Scientific studies compiled in this review addressed 36 different taxa of harmful microalgae (listed in Table 1), encompassing a great variety of toxins that may lead to shellfish poisoning syndromes in humans. Most of the studies evaluated the impacts of dinoflagellates exposure on molluscan shellfish, in which Dinophyceae class comprised 89% of all microalgal and cyanobacteria species. However, molluscan responses to marine cyanobacteria and diatoms have also been addressed in the retrieved studies. Dinoflagellates can produce a great variety of toxic compounds that are responsible for six of the seven human poisoning syndromes induced by phycotoxins (reviewed in James et al. 2010). A higher proportion of studies targeted the genus Alexandrium (37%), probably because of its wide geographic distribution and the high recurrence of its blooms (Van Dolah 2000).

Table 1 Harmful algal or cyanobacteria species or higher taxonomic level addressed in the 150 compiled scientific studies, its geographic distribution (from Algae Base - Guiry and Guiry 2016), and the main toxins produced by each species
Full size table

The main responses of shellfish to marine harmful algae

Molluscs show a great variety of responses when exposed to marine harmful algae or directly to phycotoxins. In addition, bioactive extracellular compounds—noxious exudates released by harmful cells—can induce cytotoxic, allelopathic, and hemolytic effects upon marine bivalves (e.g., Basti et al. 2014a; Borcier et al. 2017; Ford et al. 2008; Smolowitz and Shumway 1997). A detailed compilation of the main routes of exposure to harmful algae or their toxic compounds and related responses of marine molluscan shellfish is graphically presented in Fig. 2.

Fig. 2
figure 2

Theoretical diagram of the main routes of exposure to harmful algae (e.g., exudates contact, algal cell ingestion, toxin accumulation) and related responses of marine molluscan shellfish based on the compiled literature. *Impacts attributed to bioactive extracellular compounds

Full size image

Feeding and behavioral responses

The filtration and ingestion rates of most bivalve species, both juveniles and adults, decrease when they are exposed to harmful algal cells, either in suspension or deposited in the substrate (e.g., Haubois et al. 2007; May et al. 2010; Shumway et al. 2006; Springer et al. 2002). Exposure to bioactive extracellular compounds also inhibits clearance and ingestion rates of bivalves (Borcier et al. 2017; Matsuyama et al. 1997). However, an exception is the oyster Ostrea edulis that has increased its filtration rate when exposed to the toxic dinoflagellate Alexandrium (=Protogonyaulax) tamarense (Shumway and Cucci 1987). Moreover, molluscs that are periodically exposed to harmful algal blooms may have evolved mechanisms to exploit toxic algae as food with no harmful effects (Bricelj et al. 2005; Shumway and Cucci 1987; Twarog and Yamaguchi 1975).

Decreases in filtration and ingestion rates may be a consequence of active avoidance by the recognition of harmful cells. For example, selective feeding on non-toxic algal cells was described for the Manila clam Ruditapes philippinarum exposed to Heterocapsa circularisquama (Basti et al. 2011a, 2016). Reduction in filtration and ingestion rates can also be a consequence of behavioral and/or physiological changes in shell-valve activity and shell closures. Toxic alga exposure can induce short and incomplete shell-valve closure reactions (Basti et al. 2009), partial or complete closure of shells or valves (Estrada et al. 2007; Neves et al. 2015b; Romero-Geraldo et al. 2014), siphon retraction (Basti and Segawa 2010; Shumway and Cucci 1987), and increase in the reaction time and in the number of stimuli necessary for shell-valve closure response (Neves et al. 2019). In addition, molluscs exposed to Alexandrium species that synthesize paralytic shellfish toxins (PSTs) showed higher rates of shell-valve clapping (Argopecten purpuratus; Hégaret et al. 2012), an increase in the duration of shell-valve aperture and closure (Crassostrea gigas; Tran et al. 2010, 2015), and a negative correlation between shell-valve activity and toxic alga concentration (C. gigas; Haberkorn et al. 2011). Molluscs can also exhibit shorter duration of escape response, burrowing and swimming activity, byssal thread production, muscle paralysis, changes in cardiac activity, and neurophysiological responses when exposed to toxic algae (e.g., Basti et al. 2016; Escobedo-Lozano et al. 2012; reviewed in Gainey Jr and Shumway 1988; Hégaret et al. 2007a, 2012; Neves et al. 2015a; Shumway and Cucci 1987; reviewed in Shumway and Gainey Jr 1992). Behavioral responses in bivalve molluscs can indicate in vivo sensitivity to harmful algae and their toxins (Bricelj et al. 1996), making them more susceptible to predation and parasites infestation.

Algal cells can be avoided through pre-ingestive mechanisms (e.g., pseudo-feces) (e.g., Basti and Segawa 2010; Escobedo-Lozano et al. 2012; Estrada et al. 2007; Mafra Jr et al. 2009; Romero-Geraldo et al. 2014; Wikfors and Smolowitz 1995) or eliminated as viable cells or cysts in biodeposits of bivalves and gastropods (e.g., Hégaret et al. 2007a; May et al. 2010; Neves et al. 2015a, 2015b; Rolland et al. 2012; Shumway et al. 2006; Springer et al. 2002). The elimination of intact cells by molluscan species decreases the assimilation and digestion of toxic cells and the potential for toxins accumulation in their tissues during HABs.

Metabolic and physiological reactions

The elimination of accumulated toxins is an important protective strategy adopted by invertebrates. Toxin elimination occurs mainly through toxin release via excretion and detoxification mechanisms in some molluscan species (e.g., Bogan et al. 2007a; Choi et al. 2003; Contreras et al. 2012; Dowsett et al. 2011; Guéguen et al. 2008; Sephton et al. 2007; Suzuki et al. 2003). Detoxification in bivalves is composed by an initial fast depuration with elimination of unassimilated toxins and toxic cells, followed by a slow reduction in toxins that have been assimilated and incorporated in molluscan tissues (Cabado et al. 2020; Duan et al. 2021; Silvert et al. 1998). Metabolic and physiological reactions in molluscs as a consequence of digestive detoxification and/or use of toxins in metabolic processes may transform toxic compounds into products with different toxicities (Hégaret et al. 2009a). Some molluscan species biotransform toxic compounds into less toxic products; for example, the epimerization in shellfish tissues causes an 11-fold decrease in saxitoxin toxicity (e.g., Choi et al. 2003; FAO - Food and Agriculture Organization of the United Nations 2004; Fast et al. 2006; Gárate-Lizárraga et al. 2004). However, PSTs toxicity may also increase through an increase in the proportion of potent carbamate derivates (e.g., saxitoxin and decarbamoylsaxitoxin) compared to the less toxic N-sulfocarbamate forms (Doucette et al. 2006). Variations in toxin biotransformation within different tissues have also been described for bivalves. Digestive glands, followed by gills, have demonstrated higher toxin transformation activity while mantle, adductor muscle, and siphon tissues have shown a lower potential for toxin transformation (Fast et al. 2006).

Toxicity level varies widely among molluscan species depending on toxin retention, depuration rates, toxin nature, and density of microalga cells. Considering detoxification in bivalves, fast detoxifiers are those species that reduce toxins at rates between 6 and 17% per day (e.g., Mercenaria mercenaria, Mytilus edulis) (reviewed in Bricelj and Shumway 1998). Slow detoxifiers exhibit rates of toxin elimination smaller than 1% per day (e.g., Placopecten magellanicus, Saxidomus nuttalli), and these species could take several months or even years to reach safe toxin concentrations for consumption (reviewed in Bricelj and Shumway 1998). A similar time span for reduction in the content of paralytic toxins and brevetoxins has been experimentally demonstrated in molluscan shellfish during depuration phase: 23 days of depuration for the purple clam Hiatula rostrata (after 18 days of exposure to Alexandrium minutum; Chen and Chou 2001), 13 days of depuration for the noble scallop Chlamys nobilis and the green mussel Perna viridis (after 3 days of exposure to A. tamarense, Choi et al. 2003), 8 days of depuration for the Pacific oyster Crassostrea gigas (after 13 days of exposure to A. minutum, Guéguen et al. 2008), and 15 days of depuration for the eastern oyster Crassostrea virginica and the hard clam Mercenaria mercenaria (after 8 days of exposure to Karenia brevis, Griffith et al. 2013). Retention time of domoic acid varied from few hours to days in mussels and from several months to years in scallops, considering the time needed for species to reach toxin concentrations below regulatory levels (Fernández et al. 2004). Detoxification rates were not evaluated in cephalopods, and only one study evaluated toxin concentration in snails (Littorina littorea) during depuration phase after exposure to A. minutum, but no significant decrease in saxitoxin content was noticed during 6 days of depuration (Neves et al. 2015b).

Several methods have been tested to accelerate detoxification time of commercially important bivalves contaminated with phycotoxins as an effort to reduce the duration of harvest interdiction (e.g., Cabado et al. 2020; Guéguen et al. 2008; Lassus et al. 2005; Shumway et al. 1995). Osmotic stress, electric shocks, decrease in pH, and chlorination have not been effective (Shumway et al. 1995). Recently, thermal procedures were tested in naturally PSTs-contaminated mussels, clams, and scallops and showed promising results (Cabado et al. 2020). Biotransformation of phycotoxins using enzymes is still under investigation, and assays with ozonization of seawater led to contradictory results (Lassus et al. 2005). The role of silt in depuration rates of molluscs has not been clearly established (Guéguen et al. 2008). The use of silt was suggested to accelerate detoxification process as it seems to have an effect on feeding physiology increasing the filtration rate (Kiørboe and Møhlenberg 1981). Active feeding during detoxification of molluscs seems to be important to accelerate the rates of gut evacuation and overall metabolism of toxins. Molluscs fed non-toxic algae during depuration phase were able to detoxify PSTs faster than molluscs held in filtered seawater (without food) (e.g., Bricelj and Shumway 1998; Guéguen et al. 2008; Lassus et al. 2005, 2007b). In addition, as the density of the non-toxic alga increases, the efficiency of mussel detoxification rate seems to enhance proportionately (Blanco et al. 1997). Alternative methods to reduce phycotoxin accumulation in shellfish molluscs have been also experimentally tested, in which mussels exposed to a phytochemical extract of cinnamaldehyde showed lower concentrations of DSTs in the digestive gland (Duan et al. 2021).

Mathematical models have been developed to predict the time duration of the phycotoxin detoxification process from economically important molluscan species (Blanco 2006; Blanco et al. 2005). The evaluation of the feasibility of detoxification models in terms of cost-effectiveness is necessary to meet the demand of shellfish industry and farmers for the implementation of an industrial-scale detoxification system that allows detoxification of contaminated shellfish in shorter time duration. The reduction in shellfish toxicity through toxin elimination and biotransformation into less toxic compounds has ecological and economic importance. When toxin content reaches levels below regulatory limits, shellfish may be harvested again, decreasing the risk of negative ecological effects and human health risks caused by the consumption of contaminated molluscs, as well as the economic losses for the producer. For example, a conservative estimate of the average annual economic impact resulting from HABs in the USA was close to US$ 75 million from 1987 to 2000 (Hoagland and Scatasta 2006).

Biological processes for toxin depuration and other cellular protection systems in marine molluscs are expensive in terms of energy requirements for organisms (Sokolova et al. 2011). Other physiological responses were also shown by molluscan shellfish after exposure to harmful algae and phycotoxins (e.g., changes in ammonia excretion and oxygen consumption and changes in cardiac activity, muscle paralysis, byssal thread production, and neurophysiological responses) (Fig. 2). Therefore, increased rates of molluscan metabolic activity to depurate phycotoxins, commonly related to stressful conditions, may be detrimental for individual fitness and the maintenance of populations exposed to HABs.

Fitness consequences

Exposure to toxic algae may directly damage DNA structure or modify gene expression in shellfish. Physical damages have been described as morphological alteration in cytoplasmic and nuclear membrane, chromatin condensation, DNA fragmentation and release of fragments into the cytoplasm (Estrada et al. 2014), primary DNA lesions in digestive glands (Dizer et al. 2001), qualitative and quantitative impacts on DNA (Mat et al. 2013), and increase in DNA damage in gill tissues (Flórez-Barrós et al. 2011). Harmful algae may also change expression levels of genes involved in antioxidant defense, cell detoxification, immune system, stress response, and cell cycle regulation in bivalves (e.g., Fabioux et al. 2015; García-Lagunas et al. 2013; Mat et al. 2013; Medhioub et al. 2013; Rolland et al. 2014; Tran et al. 2015). For example, an increase in the transcription levels has been described for genes encoding for antioxidant enzymes (superoxide dismutase and catalase), mitochondrial enzymes (cytochome oxidase I and III, cytochrome b, ATP synthase-6, and ND1), ion channels (K+ and Ca2+), heat-shock proteins, bi-phase transcriptional response of C-type lectin and toll-like receptor genes, and expression of differential genes in molluscs immune cells (e.g., Astuya et al. 2015; Núñez-Acuña et al. 2013; Romero-Geraldo et al. 2014).

Populations from toxin-affected areas may undergo genetic adaptation to toxins with natural selection against individual’s sensitive to toxins (Twarog 1974), as demonstrated for the softshell clam Mya arenaria (Bricelj et al. 2005, 2010; Connell et al. 2007). Populations that have been previously exposed to toxic algae seem to be less affected compared to naive populations. For example, individuals of M. arenaria from a naive population showed higher mortality rates (> 30%) compared to individuals from a population previously exposed to Alexandrium tamarense (2–8% of mortality; MacQuerrie and Bricelj 2008). Resistant populations are less sensitive to phycotoxicity and accumulate higher toxin amounts; thus, resistant individuals are potential toxin vectors representing an extra risk for human and ecosystem health.

Immune response in molluscs is modulated by stressful conditions and consists in a response to stress by specialized cells—the hemocytes (Barcia and Ramos-Martínez 2011). Hemocyte responses vary as a function of toxin type and concentration (Dizer et al. 2001; Flórez-Barrós et al. 2011), exposure time (Astuya et al. 2015; Jones et al. 1995), microalga species (Hégaret et al. 2011), and the frequency of population exposure to harmful alga (Haberkorn et al. 2010; Hégaret and Wikfors 2005a). Even the exposure to bioactive extracellular compounds may induce deleterious effects on bivalve hemocytes, as demonstrated for Alexandrium species (Ford et al. 2008). The compilation of studies retrieved in the present review (see in the Supplementary Material Table S1) indicates that harmful algae usually induce an increase in total hemocyte counts (i.e., number) and production of reactive oxygen species (ROS), followed by a decrease in hemocyte viability, relative size, and intracellular complexity (e.g., Castrec et al. 2018; Galimany et al. 2008a; Hégaret and Wikfors 2005a, 2005b, Hégaret et al. 2010). Phagocytic activity has either decreased or increased after exposure to harmful algae depending on microalga and molluscan species (e.g., Galimany et al. 2008a; Hégaret et al. 2009b; Lassudrie et al. 2020), suggesting a species-specific response for this hemocyte parameter. Shifts in the ratio of hyalinocyte to granulocyte have also been reported and were associated with increases in granulocyte percentage (Hégaret and Wikfors 2005b). Granulocytes are the main cells in phagocytosis and encapsulation mechanisms; thus, shifts in hemocyte cellular types suggest changes in immune function. In summary, contradictory hemocyte responses were reported for molluscs exposed to harmful algae, and can be either immunostimulant or immunosuppressant. Some immunosuppressive effects evidenced by molluscs include decreases in hemocyte viability, phagocytosis, ROS production, and cell adhesion (e.g., Hégaret et al. 2009b, 2011). Immunosuppressive responses leave the organisms more susceptible to pathogens, and, therefore, harmful algae exposure may trigger bacteria proliferation (Galimany et al. 2008b), infection, and prevalence of protozoan and trematodes (e.g., Perkinsus marinus and Bucephalus sp.) in molluscan tissues (Hégaret et al. 2012; Lassudrie et al. 2015a).

Marine molluscs and other invertebrates are infected by several parasites, most often trematodes, that cause shifts in host resistance to environmental factors (reviewed in Berger and Kharazova 1997; Costello et al. 2021; Couch and Fournie 2021). More severe effects of phycotoxins are expected on parasitized molluscs. For example, toxic algae exposure of Manila clams (Ruditapes philippinarum) that were highly parasitized by protozoans (Perkinsus olseni) induced severe immune modulation (Hégaret et al. 2007b) and disrupted the pro/-anti-oxidant response of clams (Lassudrie et al. 2014). In addition, exposure to harmful algae may change the relationship of bivalves (i.e., hosts) with their parasite (e.g., P. olseni) and pathogen (e.g., herpesvirus, OSHv-1) reducing the prevalence and intensity of infection (da Silva et al. 2008; Lassudrie et al. 2015b). Negative effects of harmful algae and their exudates on parasitic protozoan viability, survival, and morphology were addressed by in vitro assays (da Silva et al. 2008; Hégaret et al. 2011). Harmful algae may act directly on the parasite providing paradoxally a short-duration relief for the host (Soudant et al. 2013). For example, the harmful alga Prorocentrum minimum and its exudates were found to induce mortality and morphological changes in the cells of the protozoan Perkinsus olseni, suggesting an antagonistic suppression of parasite transmission and proliferation in natural environment (Hégaret et al. 2011).

Beyond pathogens and environmental pollutants, harmful algae are agents of pathologies in molluscan tissues inducing (1) inflammatory responses, (2) tissue degeneration, (3) cell and tissue death, (4) alterations in cellular number, size and type, and (5) shifts in fluid accumulation within tissues (Table 2). In the view of studies retrieved in the present review (Table 2), the most frequent histopathologies induced by harmful microalgae are hemocyte infiltration and aggregation (inflammatory response), formation of ceroidosis and epithelial desquamation (tissue degeneration), and necrosis (cell and tissue death). Bioactive extracellular microalgal compounds may also induce tissue damages. For example, bioactive compounds released by Alexandrium minutum cells induced lesions on mantle and muscle tissues of the king scallop Pecten maximus (Borcier et al. 2017). Histopathological lesions primarily occur in tissues that interact with harmful cells, among them gills and mantle, are the most affected tissues in bivalves while digestive tissues are the most in gastropods (Supplementary Material Table S1). The intensity and prevalence of histopathologies depend on the density of toxic algae, duration of toxin exposure, and water temperature (Basti et al. 2015b). Damaged tissues of molluscs exposed to harmful algae may recover depending on the level in which those tissues were affected. Tissues of the blue mussel (Mytilus edulis) affected by the exposure to the toxic dinoflagellate Alexandrium fundyense have recovered fast (i.e., within 6 days) from inflammatory responses and tissue damage (i.e., lipofuchsin granules) (Galimany et al. 2008c). However, advanced histopathologies can cause non-reversible changes that compromise vital functions of molluscs and lead to the death of individuals (e.g., Akberali and Trueman 1985; Cuevas et al. 2015; Yee-Duarte et al. 2018).

Table 2 Types of histopathology in marine molluscan shellfish induced by harmful alga
Full size table

Reproductive capacity may be negatively affected by toxic algae exposure with changes in production and morphology of spermatozoa, reduction in viability, motility and swimming velocity of spermatozoa, reduction in egg viability and fertilization, and increase in ROS production by oocytes (e.g., Basti et al. 2013; Le Goïc et al. 2014; Rolton et al. 2015). However, all the effects on reproductive cells were evaluated in only three oyster species (Pinctada fucata martensii, Crassostrea virginica, and C. virginica) exposed to dinoflagellates producers of saxitoxins and brevetoxins (Basti et al. 2013; Le Goïc et al. 2014; Rolton et al. 2015).

Larva and juvenile life stages are usually more affected by alga toxicity than adult individuals, and larval survival decreases with increased toxicity (Rolton et al. 2014; Talmage and Gobler 2012). Besides that, flagellated cells of the genus Pfiesteria have also reduced larval survival by attacking and consuming molluscan larvae (Shumway et al. 2006; Springer et al. 2002). Negative effects on larval growth, development, and lipid synthesis have been attributed to harmful algae exposure (e.g., Basti et al. 2011b; Mu and Li 2013; Rolton et al. 2014; Talmage and Gobler 2012; Wikfors and Smolowitz 1995). In addition, negative effects on embryo cleavage, inhibition of embryonic and newly hatched development, increased embryo abnormalities, damages to feeding and gut apparatus during embryonic development, reduction or interruption of embryo hatching, delay in metamorphosis of larval stages, and decrease in larval activity were shown in a few species of oysters and clams (e.g., Basti et al. 2013, 2014b, 2015a; Mu and Li 2013; Rolton et al. 2014, 2015).

Mass mortalities of marine invertebrates during HABs are attributed to the toxins produced by harmful species (Faimali et al. 2012). Lethal effects in adult bivalves have been related to toxic dinoflagellates exposure (e.g., Carella et al. 2015; MacQuerrie and Bricelj 2008; Shumway and Cucci 1987; Shumway et al. 2006; Smolowitz and Shumway 1997). However, mortality of marine bivalves may also be induced by other secondary metabolites and/or unknown toxins produced by harmful algae, other than the characterized toxins. High mortalities of Japanese and noble scallops, respectively Patinopecten yessoensis and Mimachlamys (=Chlamys) nobilis, experimentally exposed to a strain of Dinophysis caudata that synthesizes pectenotoxin-2 (PTX-2), have been attributed to unknown toxins and/or secondary metabolites (Basti et al. 2014a). The implication of unknown toxins and/or secondary metabolites was suggested since mortality of scallops due to PTX-2 has never been observed, neither in the wild nor under laboratory conditions (Basti et al. 2014a).

Trophic transfer of phycotoxins

Contaminated molluscs act directly as toxin vectors to their consumers: carnivorous gastropods (Chen and Chou 1998; Wekell et al. 1996), cephalopods (Costa et al. 2009; Lopes et al. 2014), lobsters (Jiang et al. 2006), crabs (Oikawa et al. 2004), marine birds (Shumway et al. 2003), sea otters (Kvitek et al. 1991; Miller et al. 2010), and humans (García et al. 2004; Laurent et al. 2012). Moreover, molluscs may act indirectly as a source of phycotoxins following the elimination of toxic cells in their biodeposits, which makes the toxins available for deposit-feeders and coprophagous organisms (e.g., Hégaret et al. 2007c; Mafra Jr et al. 2009; Neves et al. 2015b, 2019). Toxins are transferred through the marine food web from the toxic algae to molluscs and then towards higher trophic levels, which is the typical pathway of phycotoxin bioaccumulation (reviewed in Basti et al. 2018; reviewed in Turner et al. 2021). Transfer of phycotoxins via food web interactions can have serious and deleterious effects on marine organisms (reviewed in Landsberg 2002), and lead to different human poisonings and even to death (García et al. 2004; reviewed in Neves and Rodrigues 2020).

Biomagnification is defined as an increase in the concentration of toxins in successively higher trophic levels (reviewed in Gray 2002). On the contrary, biodilution is defined as a decrease in toxin concentrations through the trophic web. For example, microcystins tend to biodilute rather than to biomagnify through trophic transfer in aquatic food webs (Ibelings et al. 2005; Kozlowsky-Suzuki et al. 2012; Sotton et al. 2014). The biomagnification factor indicates if trophic transfer is dominated by the process of biomagnification (BMF mean and 95% CI >1) or biodilution (BMF mean and 95% CI <1) (Kozlowsky-Suzuki et al. 2012). Only two studies, from the 150 scientific publications retrieved in this review, provided simultaneous measurements of toxin concentration in molluscs and their consumers: the scallop Chlamys nobilis and its predator the spiny lobster (Jiang et al. 2006), and the clam Donax sp. and its predator Octopus vulgaris (Lopes et al. 2014). Thus, data considered in the present review for BMF calculation was based only on concentrations of PSTs. We have found a tendency for toxin reduction, with BMF smaller than one (0.10 ±0.12), that suggests PST biodilution through trophic transfer. However, further studies including different species of molluscs and their potential predators, as well as diverse toxins are needed to properly evaluate BMFs of marine phycotoxins.

Phycotoxin dynamics in molluscs and risks for human health

Most information on toxin accumulation in marine molluscs comes from routine monitoring programs for toxins in shellfish (Supplementary Material Table S1). Toxin content in molluscan shellfish may be related to its uptake, metabolism, depuration (Hégaret et al. 2009a), and differential sensitivity to toxins (Twarog 1974; Twarog et al. 1972). Toxin accumulation in molluscs varies widely among species (reviewed in Bricelj and Shumway 1998), even when considering species from the same taxonomic class (e.g., FAO/IOC/WHO- Food and Agriculture Organization of the United Nations/ Intergovernmental Oceanographic Commission of UNESCO/ World Health Organization 2004; Mello et al. 2010; Miller et al. 2010; Suzuki and Mitsuya 2001). Phycotoxin accumulation in molluscs may also be affected by environmental factors (e.g., temperature, exposure time, and salinity) and diverse toxin profiles in harmful alga species from distinct geographic regions (Lassus et al. 2007a; Molinet et al. 2010). Therefore, the equilibrium between the rates of uptake, metabolism, and elimination of toxins depends on several factors related to the toxic microalga, molluscan species, and the environmental conditions that influence both the HAB species and the molluscan shellfish (reviewed in Basti et al. 2018). In summary, the amount of accumulated toxins varies according to mollusc feeding mode and ingestion rates, size of individuals, species resistance and susceptibility to phycotoxins, density of toxic algae, and phycotoxin group (e.g., saxitoxin, domoic acid, okadaic acid).

Marine bivalves, cephalopods, and gastropods accumulate phycotoxins, occasionally exceeding maximum levels for safety consumption, and represent a risk for human health. Bivalves often accumulate higher toxin amounts than cephalopods and gastropods (Table 3). Mussels tend to accumulate high levels of toxins faster than other bivalves and usually eliminate them more quickly (e.g., Lassus et al. 1989; Mafra Jr et al. 2015; reviewed in Shumway et al. 1995; Wohlgeschaffen et al. 1992). The toxin dynamics in mussels make them interesting candidates to be used as the most appropriate model, among molluscs, to allow early detection of HABs. In contrast, toxin concentration in gastropods is usually below regulatory levels for safe consumption (e.g., Dowsett et al. 2011; Kim et al. 2012; Neves et al. 2015a, 2015b), suggesting that gastropods accumulate less toxins than bivalves. Moreover, toxin accumulation in molluscs is not homogeneous; toxin content varies among tissues; and toxins migrate to different tissues during biotransformation. Data compilation from retrieved scientific studies indicates that in overall digestive glands concentrate more toxins (~81% total toxin) than other tissues-gonads (~12% total toxin), adductor muscles (~5% total toxin), and brachial hearts (~28% total toxin) (see Table 3). Higher toxin concentration in digestive glands seems to be directly related to their role in food digestion and absorption.

Table 3 Concentration of phycotoxins in the tissues of marine molluscan shellfish. Minimum and maximum values are presented when data were available in the literature
Full size table

Routine monitoring programs for toxins in shellfish aim to reduce human risks by detecting phycotoxin concentration levels in molluscs which may lead to harvesting restrictions or closure to prevent human contamination. The European Commission (EC) regulates the production and marketing of live bivalves to ensure that contaminated shellfish are not placed on the market (European Commission, 2004 - EC 853/2004; European Commission, 2013 - EC 786/2013). Molluscs represent 11% of European Union (EU) seafood consumption, and EU production supplies 65% of its domestic mollusc demand (O’Mahony 2018). The National Shellfish Sanitation Program (NSSP), recognized by the U.S. Food and Drug Administration (FDA), is responsible for the sanitary control of shellfish production and sale for human consumption in the USA (FDA - U.S. Food and Drug Administration 2017). The current regulatory limits for marine phycotoxins are presented in Table 4.

Table 4 Regulatory limits applied to marine toxins established by U.S. Food and Drug Administration (FDA - U.S. Food and Drug Administration 2017) and European Union/European Food Safety Authority (EC 853/2004 and 786/2013) and related human poisoning syndromes
Full size table

Human health effects induced by shellfish poisonings include several general symptoms (e.g., nausea, diarrhea, vomiting, headache, fever) and symptoms specifically induced by the phycotoxin group, for example, tingling of fingers and toes and muscle paralysis (PSP), confusion and short-term memory loss (ASP), and cardiovascular symptoms (CFP) (reviewed in Nicolas et al. 2017).

Conclusion and perspectives

Harmful algae impact on marine molluscan shellfish is complex to trace and predict, since they rely on molluscan species, harmful algae species, toxic compounds, and synergetic interactions of these and other barely known factors. Moreover, the few detected effects of HABs on gastropods and cephalopods indicate that more attention should be devoted to these molluscan groups for both ecological and human health considerations. Further studies including different species of molluscs and their potential prey and predators are also needed to evaluate the trophic transfer of marine biotoxins through food webs.

Considering the perspective of climate change which may favor an increase in the occurrence and intensity of HABs, the intensification of efforts to expand the knowledge about HABs impacts on marine molluscs is crucial to mitigate the damages on economy and human health. More information about the consequences of harmful algae exposure on reproductive features of marine molluscan shellfish is necessary in order to mitigate the impacts on cultivated shellfish and predict HABs impacts on farmed populations. Moreover, further investigations on the feasibility of detoxification in terms of cost-effectiveness are also needed to allow detoxification of contaminated shellfish in shorter periods and reduce the adverse effects for human health and excessive economic losses.

An increase in the research effort in the field of marine sciences is one of the targets of the Sustainable Development Goal 14 (United Nations Development Programme) in order to improve ocean health and, consequently, environmental and human safety. Moreover, the conservation and sustainable use of the oceans, seas, and marine resources can contribute to reduce the frequency and distribution of marine HABs and their impacts on public health, fishery, and marine ecosystems worldwide.