INTRODUCTION

On a global scale, benthic dinoflagellates have been much less studied than their planktonic counterparts, perhaps because they tend to form much less dramatic blooms, and their role in marine food webs is more cryptic. Nevertheless, over the past two decades, in particular, many benthic dinoflagellate species have attracted increasing attention as the causative agents of benthic harmful algal blooms (BHABs) and related syndromes, such as ciguatera fish poisoning (CFP) and diarrhetic shellfish poisoning (DSP) in human consumers of seafood (^{Heredia-Tapia et al., 2002}). Other BHAB species have recently emerged as the cause of acute respiratory distress in humans due to the release of toxic aerosols created by wind and wave action in coastal ecosystems (^{Blanfuné et al., 2015}). Although the mooted arguments for a general global increase in the number and frequency of HAB events were not sustained in the recent analysis of global databases (^{Hallegraeff et al., 2021}), the evidence indicates that the impacts on human populations and marine ecosystems have increased substantially. A similar pattern is apparent concerning BHAB events - almost all of which are generated by benthic dinoflagellates - but the associated databases are much less comprehensive than for planktonic blooms.

A recent review on toxigenic marine benthic dinoflagellates in Latin America (^{Durán-Riveroll et al., 2019a}) is essentially a status report focusing on key BHAB species, mainly of the genera Amphidinium Clap. et J. Lachm., Coolia Meunier, Fukuyoa Gómez, Lopes et Lin, Gambierdiscus Adachi et Fukuyo, Ostreopsis Schmidt and Prorocentrum Ehrenb. The geographical coverage in that review comprises the eastern Pacific, Caribbean Sea, Gulf of Mexico, Atlantic coastal waters of Latin America, and part of the southeastern coast of the USA, and thus includes all the relevant coasts of Mexico. A national review on HABs in Mexico (^{Durán-Riveroll et al., 2019b}) in the context of coastal contamination and climate change considers mainly planktonic species but also includes information on benthic dinoflagellates in the Mexican Caribbean. The last International Conference on Harmful Algae held online in October 2021 in La Paz, Mexico (^{ICHA, 2021}), featured recent HAB studies. Special sessions opened avenues for future research directions on BHAB dinoflagellate community studies, e.g., published abstracts on the section Ciguatera and Benthic HABs (^{ICHA, 2021}). Some BHAB research from Mexico has been summarized in short papers on diversity (^{Durán-Riveroll et al., 2022}) and microbiome associations with bacteria (^{Martínez-Mercado et al., 2022}) in the conference proceedings (^{Band-Schmidt & Rodríguez-Gómez, 2022}).

In fact, the only dinoflagellate species confirmed to form high-magnitude benthic blooms with potentially harmful consequences (i.e., BHABs) in Mexico are Prorocentrum rhathymum A. R. Loebl., Sherley et R. J. Schmidt (^{Gárate-Lizárraga & Martínez-López, 1997}; ^{Gárate-Lizárraga & Muñetón-Gómez, 2008}), P. lima (Ehrenb.) F. Stein (^{Heredia-Tapia et al., 2002}), Blixaea quinquecornis (T. H. Abé) Gottschling (^{Gárate-Lizárraga et al., 2006}; ^{Gárate-Lizárraga & Muñetón-Gómez, 2008}; ^{Okolodkov et al., 2016}), and Amphidinium cf. carterae (^{Gárate-Lizárraga, 2012}, ²⁰²⁰; ^{Gárate-Lizárraga et al., 2019}).

Linkages of other known toxigenic benthic dinoflagellates (e.g., Gambierdiscus and Fukuyoa species) to toxin syndromes such as CFP in Mexico are plausible but remain circumstantial. ^{Parrilla-Cerrillo et al. (1993)} reported an outbreak (200 cases) of CFP in Baja California Sur in 1984 due to consumption of snapper (Lutjanus) species. From 1992-1995, more fish poisoning events were registered at different locations in Baja California Sur, suggesting the presence of ciguatoxins (CTXs) in carnivorous sea bass (Serranidae) and snapper (Lutjanidae) fish (^{Ochoa et al., 1997}). In May 1993, another event of apparent ciguatoxicity in Mexico occurred, and it was based on the consumption of fish caught at Alijos Rocks, 300 miles offshore of Bahía Magdalena in Baja California Sur (^{Lechuga-Devéze & Sierra-Beltrán, 1995}); although the toxin identification remains uncertain, the symptoms were characteristic of CFP in the affected fishermen. From 1993-1996, human poisoning events with similar symptoms also occurred at two locations in the Bahía de La Paz area including Isla El Pardito, after consumption of liver of the same fish families. In these cases, the presence of CTX1 was indicated by chromatography analysis but not structurally confirmed (^{Núñez-Vázquez et al., 1998}). Based on data from the Secretariat of Health of Mexico and scientific articles, during a 29-year period (1984-2013), 464 human poisoning cases caused by consumption of carnivorous fish were recorded from the genera Lutjanus, the groupers Epinephelus and Mycteroperca in the Pacific, and Lutjanus and the barracuda Sphyraena in the Caribbean (^{Núñez-Vázquez et al., 2000}). Most reports were from the states of Baja California Sur (52%), Quintana Roo (35%) and Yucatán (10%) and of Mexican tourists poisoned by eating fish in Cuba (3%). Although collectively ascribed to benthic dinoflagellates considered the precursors of ciguatera from the genera Amphidinium Clap. et J. Lachm., Coolia Meunier, Fukuyoa, Gambierdiscus, Ostreopsis Schmidt, and Prorocentrum, there is no conclusive evidence of particular dinoflagellate species linked to these specific events.

This review updates the current knowledge of benthic dinoflagellates in Mexican waters based on integrated field and laboratory studies interpreted in an ecological context. Laboratory experimental studies on cultured benthic dinoflagellate isolates and from live freshly collected field specimens from Mexico have been initiated only since around the turn of the 21st century. The focus herein is on unresolved taxonomic and nomenclatural issues, species diversity, biogeographical distribution and toxigenicity, with special emphasis on allelopathic species interactions. For the first time, this comprehensive approach allows a better understanding of the role of dinoflagellates in benthic communities and their functional interactions during BHAB events in marine ecosystems in Mexico. This review provides a more detailed integration of BHAB events and associated species, but with an exclusive focus on Mexico, and adds a missing dimension of knowledge on non-harmful dinoflagellate species.

DISCUSSION

Morphological diversity. More than 190 species of benthic dinoflagellates belonging to 45 genera had been described by 2014, primarily based on morphological criteria (^{Hoppenrath et al., 2014}), but the list is being constantly revised and extended by new morphological observations and molecular taxonomic data. Morphological traits of unarmored benthic dinoflagellate cells can be difficult to discriminate among closely related species due to extreme phenotypic plasticity in shape and surface microstructure. This plasticity results from “the ability of an organism to change morphology in response to stimuli or inputs from the environment” (^{West-Eberhard, 2008}), and in some cases is presumably adaptive. In comparison, for armored, but dinoflagellate cells, the overall shape tends to be more constrained, but identification is complicated by phenotypic plasticity in thecal plate tabulation and variation in superficial features such as pore patterns and surface microstructure. When taxonomists lack experience with a particular taxon, alternative designations can be given to the same species because some benthic taxa have thecal tabulations difficult to interpret (^{Hoppenrath et al., 2014}).

Most published information on the morphological diversity of marine benthic dinoflagellates in Mexico has been based on light microscopy (LM) or scanning electron microscopy (SEM). Numerous morphological studies on field or cultured material are derived from specimens collected from the coastal waters of Quintana Roo, Yucatán, Campeche, Veracruz, Baja California Sur, Baja California, Chiapas, Oaxaca, Guerrero, Colima (Revillagigedo Archipelago), Michoacán and Nayarit (Isla Isabel) (Fig. 1, Table 1). The inter-seasonal or annual dynamics of epibenthic/epiphytic dinoflagellate assemblages and their relationships with some physicochemical variables have been studied in some coastal localities (^{Okolodkov et al., 2007}, ²⁰¹⁴; ^{Aguilar-Trujillo et al., 2014}, ²⁰¹⁷; ^{Martínez-Cruz et al., 2015}; ^{Irola-Sansores et al., 2018}). To our knowledge, no studies on the direct influence of physicochemical factors on dinoflagellate morphology have been conducted in Mexico. Individual species or entire epibenthic assemblages were considered in the above cited studies, and in some cases, the epiphytic assemblages were comprehensively described. In comparison, however, sand-dwelling species inhabiting Mexican waters have been neglected and deserve more attention.

Figure 1 Location of the origin of marine benthic dinoflagellates and related studies reported in Mexico (1942-2022). Mexican states, including islands where field sampling was performed, and benthic dinoflagellates were found, are shaded light gray. Only geographic names mentioned in the text are given. 

Dinoflagellate reviews have highlighted the importance of life-history transitions in HAB dynamics (e.g., ^{Bravo & Figueroa, 2013}), but reports of confirmed sexual cysts of benthic dinoflagellate species remain scarce on a global scale. There are publications on resting (sexual) cysts of planktonic dinoflagellate species collected from the benthic environment in the Mexican Pacific (cited below) that are outside the scope of this review. Studies of dinoflagellate cysts (presumably, hypnozygotes) of planktonic species of the Gonyaulacales, Peridiniales and Gymnodiniales dominate the literature from Mexico (^{Martínez-Hernández & Hernández-Campos, 1991}; ^{Peña-Manjarrez et al., 2001}, ²⁰⁰⁵, ²⁰⁰⁹; ^{Morquecho & Lechuga-Devéze, 2003}, ²⁰⁰⁴; ^{Kielt, 2006}; ^{Pospelova et al., 2008}; ^{Vásquez-Bedoya et al., 2008}; ^{Morquecho et al., 2009}; ^{Limoges et al., 2010}, ²⁰¹³; ^{Helenes et al., 2020}; ^{Gu et al., 2021}), whereas the resting cysts and life history transitions of benthic species are poorly known. As noteworthy exceptions, from the benthic community, cysts of Amphidinium cf. carterae Hulburt have been recorded in shallow coastal lagoons (^{Gárate-Lizárraga, 2012}, ²⁰²⁰; ^{Gárate-Lizárraga et al., 2019}) and those of A. thermaeum Dolapsakis et Economou-Amilli in laboratory cultures isolated from Bahía de La Paz (^{Herrera-Herrera, 2022}).

The distinctive morphological features of vegetative cells of dinoflagellates may be reflected in the overall morphology and/or thecal paratabulation (armored species) of their respective cyst forms, but this is rarely the case. In Mexican waters, many undescribed cyst forms may belong to known species, but the life-history relationships remain undefined.

In general, vegetative cells of benthic dinoflagellates often tend to be more dorso-ventrally compressed compared to planktonic forms, probably due to their natural habitats on surfaces of macroalgal thalli, on seagrass leaves, or between sand grains. Presumably, this is an adaptive strategy to optimize substrate attachment and deal with turbulence and hydrodynamic flow in more stationary habitats. In any case, diagnostic morphological features more or less easily visible in planktonic species are often less obvious in benthic species. For example, in planktonic species within the Peridiniales and Gonyaulacales, these features are usually the sulcal platelets that are rarely suitable for identifying benthic dinoflagellate species.

Figure 2 Marine benthic dinoflagellates from Mexican coastal waters: A - Amphidinium theodorei (Bahía de la Paz); B - Bysmatrum caponii (Veracruz Reef System (VRS), Veracruz); C - Cabra cf. aremonica (northern Yucatán); D - Coolia malayensis (Veracruz); E - Durinskia sp. (northern Yucatán); F - Gambierdiscus caribaeus (northern Yucatán); G - Ostreopsis heptagona (VRS, Veracruz); H - Ostreopsis lenticularis (Isla Pérez, Arrecife Alacranes); I - Ostreopsis cf. ovata (northern Yucatán); J - Plagiodinium belizeanum (northern Yucatán); K - Prorocentrum hoffmannianum (northern Yucatán); L - Prorocentrum lima (Isla Pérez, Arrecife Alacranes); M - Prorocentrum rhathymum (northern Yucatán); N - Prorocentrum microcephala (northern Yucatán); O - Prorocentrum stenosoma (northern Yucatán). Images A and D were taken with a JEOL JSM-6360LV SEM (ICMyL-UNAM); B was taken with a JEOL JCM-5310LV (Facultad de Ciencias - UNAM); C and E-O were taken with a JEOL JSM 7600F (CINVESTAV-IPN, Unidad Mérida). Scale bar: 5 μm in A, 10 μm in B-O. 

Prorocentrum is the richest genus in benthic samples in terms of the number of species among all benthic dinoflagellate genera: of more than 80 Prorocentrum species, at least 29 are benthic (^{Hoppenrath et al., 2014}). In this genus, the platelets of the periflagellar area are of great importance in distinguishing among species, whereas for the identification of planktonic species, the cell shape (in fact, the valve outline) is usually sufficient. In Mexican coastal waters, the most common Prorocentrum species, preliminarily identified as P. lima and P. hoffmannianum M.A. Faust, turned out to be species complexes that are not fully resolved from morphological criteria on a global basis (^{Cembella et al., 2021}).

The validity of Prorocentrum rhathymum has caused endless discussion among researchers ever since the publication of the original description of P. mexicanum by ^{Osorio-Tafall (1942)} (^{Cortés-Altamirano & Sierra-Beltrán, 2003}; ^{Litaker et al., 2011}; ^{Gómez et al., 2017}; for further references, see ^{Cembella et al., 2021}). ^{Cortés-Altamirano & Sierra-Beltrán (2003)} later separated these species. Prorocentrum emarginatum Fukuyo and P. sculptile M. A. Faust are both relatively common but never dominant in epiphytic dinoflagellate assemblages of the Gulf of Mexico; these species are difficult to distinguish, and it remains unclear whether or not they are synonymous.

After the publication of new molecular and toxigenicity data on Gambierdiscus, it became apparent that the earlier reports of the most notorious BHAB species G. toxicus Adachi et Fukuyo from the Greater Caribbean are most likely misidentifications; true G. toxicus has been found only in the Pacific (^{Litaker et al., 2010}). Coolia monotis Meunier, earlier reported as a cosmopolitan species (^{Steidinger & Tangen, 1996}), has also turned out in later studies to comprise a species complex. Various strains assigned to C. monotis were later reassigned to C. malayensis Leaw, Lim et Usup, C. santacroce Karafas, Tomas et York, and C. palmyrensis Karafas, Tomas et York, based on molecular data and subtle morphological differences (^{Karafas et al., 2015}). ^{Momigliano et al. (2013)} suggested that C. monotis is geographically restricted to the Mediterranean Sea and the eastern Atlantic, representing the original European clade.

About 60 benthic dinoflagellate species from 18 genera have been identified in Mexican waters (Table 2; number of species indicated in parentheses): Amphidiniella Horiguchi (1), Amphidiniopsis Wołoszyńska (1), Amphidinium (6), Ankistrodinium M. Hoppenrath, Shauna Murray, Sparmann et Leander (1), Blixaea Gottschling (1), Bysmatrum M.A. Faust et Steidinger (3), Cabra Shauna Murray et Patterson emend. Chomérat, Couté et Nézan (2), Coolia (5), Durinskia S. Carty et E.R. Cox (1), Fukuyoa (1), Gambierdiscus (6), Gymnodinium F. Stein (1), Ostreopsis (6), (?)Pileidinium Tamura et Horiguchi (1), Plagiodinium M.A. Faust et Balech (1), Prorocentrum (14), Sinophysis Nie et Wang (5), Thecadinium Kof. et Skogsb. (1), (?)Togula Flø Jørgensen, Shauna Murray et Daugbjerg (1), and Vulcanodinium Nézan et Chomérat (1). Critical taxonomy and identification of species in field samples and establishment of new benthic dinoflagellate species are challenging without molecular confirmation, and therefore identifications based only on morphology should be considered tentative (^{Leaw et al., 2016}).

Molecular diversity. In Mexico, studies on the molecular taxonomy of benthic dinoflagellates began less than 15 years ago when sequences of LSU and SSU rDNA subunits were obtained from Gambierdiscus isolate CM515 from Cancún by Donald R. Tindall, from Southern Illinois University, Illinois, USA (cited in ^{Richlen et al., 2008}). The Gambierdiscus species, however, was not identified. Sequencing of the D8-D10 domain of the LSU rDNA of the same strain later identified it as G. carolinianus Litaker, Vandersea, M. A. Faust, Kibler, W. C. Holland et P. A. Tester (^{Litaker et al., 2010}), but a GenBank accession number for the sequences was not provided. Fourteen LSU and SSU rDNA sequences of Prorocentrum species, including three planktonic and the questionably benthic species, P. rhathymum from the Pacific coast of Mexico were also published around this time (^{Cohen-Fernández et al., 2010}). ^{Herrera-Sepúlveda et al. (2013)} applied a capillary electrophoresis DNA fingerprint technique (CE-SSCP) to rDNA fragments as an alternative rapid identification of harmful dinoflagellates. After application to 10 species, including P. rhathymum and P. minimum (Pavill.) J. Schill., they concluded that this method could serve as an element of an early warning system for HABs in coastal waters of Baja California Sur. Unfortunately, this method has not been systematically applied and validated for benthic dinoflagellates in Mexico.

Within the last five years, several molecular studies and publications have focused on benthic dinoflagellates from Mexico. Based on morphology and LSU rDNA sequences, ^{Sepúlveda-Villarraga (2017)} analyzed strains of Ostreopsis, Coolia, Amphidinium and Prorocentrum, identifying C. malayensis, O. lenticularis and P. rhathymum; once again, however, these sequences are not available for comparison in the GenBank database. ^{Méndez-Torres (2019)} identified three species by morphological characters (O. lenticularis, O. heptagona and O. cf. siamensis) from Isla Contoy and Puerto Morelos, Quintana Roo; only the morphological identification of O. lenticularis was corroborated by partial sequences of internal transcribed spacers (ITS) of the rRNA gene operon. ^{Ramos-Santiago et al. (2022)} carried out both morphological and molecular identification of two isolates of Coolia from Bahía de La Paz, Baja California Sur. Morphological descriptions and partial sequences of LSU rDNA agreed; both strains (OM177218 and OM177219) were identified as C. malayensis, previously considered a synonym to C. monotis (^{Hoppenrath et al., 2014}). Based on the phylogenetic analysis of D1-D2 LSU rDNA sequences and morphological differences in thecal structure, C. malayensis has since been shown to be an independent species, forming a separate lineage within the genus Coolia (^{Ho & Nguyen, 2014}; ^{Leaw et al., 2016}). These results agree with a morphological and molecular study performed by ^{Morquecho et al. (2022)}, who isolated 16 strains of Coolia from Bahía de La Paz and identified 14 of these with partial sequences of D1-D2 rDNA. Four strains belonged within the C. canariensis M. A. Faust complex; eight strains in the C. malayensis clade; one strain aligned with C. palmyrensis, and one strain belonged to C. tropicalis M. A. Faust. Comparative analysis of sequences of the ITS1/5.8S/ITS2 region of nine strains identified three as C. malayensis, one as C. palmyrensis, four belonging to the C. canariensis complex, and one as C. tropicalis. The authors of this study proposed a review and reclassification of Coolia strain CMPV-1, previously identified as C. monotis; the new sequences indicate that this strain may belong to C. canariensis.

The first sequence data for genus Amphidinium from Mexico was provided from the study to determine the toxigenicity and cytotoxic effects of A. operculatum in culture (^{Mejía-Camacho et al., 2021}). The identification of Amphidinium strain AA60 as A. operculatum was confirmed by SEM and sequencing of the LSU and ITS regions of rDNA. Comparative sequences of A. massartii, A. carterae and A. operculatum from Mexico are also available in the GenBank database (https://www.ncbi.nlm.nih.gov/nucleotide/; Table 3).

A study on phylogeography and diversity of the Prorocentrum lima and P. hoffmannianum species complexes, including P. mexicanum/P. rhathymum, was published by ^{Cembella et al. (2021)}. In this work, 69 strains were identified to species, based on morphology and selected sequences that included the LSU, SSU and ITS regions of rDNA, with complementary base-pair analysis of secondary structures. These sequences are available in GenBank, and details of the taxonomic and phylogenetic analysis are provided in the Supplementary materials of the paper (^{Cembella et al., 2021}).

As of November 2022, 702 DNA/RNA gene sequences of dinoflagellates from Mexico are archived in GenBank. Only about 10% correspond to epibenthic dinoflagellates, including 68 sequences of LSU, SSU and ITS regions of rDNA, and one sequence of cytochrome b. As shown in Table 3, with GenBank accession numbers when available, most benthic dinoflagellate gene sequences (42) belong to Prorocentrum, and of these the majority (29) correspond to P. lima. Other benthic genera are less well represented - 4 sequences belong to Amphidinium, whereas 24 correspond to Coolia.

Benthic phycotoxin vectors and toxigenicity. Most studies on benthic dinoflagellates in Mexico have been directed towards species known or suspected to be responsible for BHABs on a global scale and which are common in Latin America (^{Durán-Riveroll et al., 2019a}). Phycotoxins have drawn attention worldwide because of their adverse effects on public health, marine faunal mortalities and economic losses. Toxic events occur through the accumulation of phycotoxins in seafood species via direct or indirect consumption of toxigenic microalgae (mainly dinoflagellates). Within vector species, phycotoxins are often bio-converted into more potent compounds and thereby transferred to higher trophic levels (^{Núñez-Vázquez et al., 2019}). Phycotoxin transfer through the food web to seafood consumed by humans is responsible for many cases of seafood poisoning, such as CFP linked to benthic dinoflagellates around the globe in tropical and subtropical regions (^{FAO, 2004}).

For Mexico in particular, the paucity of reports of BHABs does not necessarily signify the absence of confirmed poisoning incidents or respiratory distress caused by such blooms. Significant economic losses in aquaculture facilities have occurred due to HABs of dinoflagellates, particularly on the northern Pacific coasts (^{Alonso-Rodríguez & Páez-Osuna, 2003}; ^{Orellana-Cepeda et al., 2004}; ^{Gárate-Lizárraga et al., 2007}).

Many studies have been conducted worldwide to determine which species act as the origin and vectors of phycotoxins in seafood for human food consumption (for references, see ^{Deeds et al., 2008}). Mexican researchers have been mainly concerned with the effects of toxic dinoflagellates on diverse marine species in an effort to understand the physiological responses and toxin transfer mechanisms, especially in species of commercial interest. Confirmed links between particular BHAB species and toxic events associated with the etiology of toxic syndromes such as CFP are rare in Mexico.

There remains a substantial gap in knowledge on the toxigenicity of benthic dinoflagellates from Mexico, including which species produce toxins that impact human health and whether or not toxigenicity is limited to particular genetic strains or geographical populations of the species or is conditional upon prevailing environmental or substrate habitat types. Benthic dinoflagellates reported from Mexico include numerous species known to synthesize a wide array of phycotoxins and structurally related natural bioactive compounds. In most cases, such as the confirmation of CTX3C-bioactivity in Gambierdiscus caribaeus Vandersea, Litaker, M. A. Faust, Kibler, Holland et Tester from Cancún, Mexico (^{Litaker et al., 2017}), the potent ciguatoxic response in the mouse neuroreceptor-binding assay CBA-N2a has not been accompanied by chemical structural analysis. Structural confirmation of toxin composition among clonal isolates of most toxigenic benthic dinoflagellate species is usually unavailable or remains unpublished to date (Table 4). There is even less information on the chemodiversity of natural benthic populations in Mexico.

Toxigenic BHAB species commonly co-occur and occasionally in high cell abundance in natural epibenthic dinoflagellate assemblages. In Mexican coastal waters, these assemblages most commonly include Amphidinium, Coolia, Fukuyoa, Gambierdiscus, and Prorocentrum species. It has been previously assumed that these species may collectively and synergistically contribute to the CFP syndrome (^{Wu et al., 2020}). In most CFP cases in Mexico, there are circumstantial inferences towards Gambierdiscus/Fukuyoa based on what is known about the global toxigenicity of BHAB species and the etiology of toxin syndromes from other regions, but the culprit species are rarely determined.

Knowledge of the toxigenicity of benthic dinoflagellate species in Mexico is scarce, and research has begun only within the last few years (^{Durán-Riveroll et al., 2019a}). Lately, some studies related to toxigenicity and toxicity of benthic dinoflagellates from the Mexican coasts have been presented at local or national conferences (e.g., SOMEFAN, October 2022). Other studies have been published but remain as gray literature, primarily as conference abstracts (e.g., ^{ICHA, 2021}), brief proceedings (e.g., ^{Band-Schmidt & Rodríguez-Gómez, 2022}), and summaries, or are archived only in graduate theses within academic libraries and are not easily accessible.

Members of the genera Gambierdiscus and Fukuyoa are known producers of large polyether ciguatoxins (CTX) and maitotoxins (MTX) associated with CFP, particularly in tropical and subtropical ecosystems of the continental shelf, islands and archipelagos of the Pacific Ocean, eastern Atlantic margins and the Greater Caribbean (^{Fleming et al., 1998}; ^{FAO, 2004}; ^{Loeffler et al., 2021}). Species of these genera are commonly reported in Mexico (Table 2), mainly from the Caribbean Sea, but knowledge of their specific toxigenicity is generally lacking (note general absence from Table 4).

According to ^{Almazán-Becerril et al. (2021)}, from 1997 to 2017, at least 28 CFP events were registered, with 262 cases mainly in Cozumel and Isla Mujeres, Quintana Roo. To avoid these poisoning incidents, the consumption of potentially ciguateric fish, such as barracuda (Sphyraena barracuda Edwards), has been banned in the Mexican Caribbean area. Recently, ciguatoxicity has been also detected in waters of the states of Campeche, Yucatán and Quintana Roo along the coast of the Yucatán Peninsula (^{Barón-Campis et al., 2014}; ^{Ley-Martínez et al., 2014}; ^{Ley-Martínez, 2016}, ²⁰¹⁸; ^{Poot-Delgado et al., 2022}). Carnivorous fish from the families Carangidae, Carcharhinidae, Centropomidae, Haemulidae, Labridae, Lutjanidae, Sciaenidae, Scorpaenidae, Serranidae, Sphyraenidae and Sphyrnidae were studied; ciguatoxicity was reported for Campeche for the first time (^{Ley-Martínez, 2018}).

^{Núñez-Vázquez et al. (2019)} reported a study on commercial carnivorous fish from Campeche in the southern Gulf of Mexico, including seven fish species: barracuda (S. barracuda), crevalle jack (Caranx hippos (L.), hogfish (Lachnolaimus maximus (Walbaum), bonnethead (Sphyrna tiburo (L.), common snook (Centropomus undecimalis (Bloch), lane snapper (Lutjanus synagris (L.) and the scrawled cowfish (Acanthostracion quadricornis (L.). The mouse bioassay (MBA) for lipophilic toxins, the official method in Mexico for ciguatoxicity, the brine shrimp Artemia (Crustacea: Anostraca: Artemiidae) bioassay, and the neuro-receptor binding assay (RBA) have all been applied in Mexico to determine ciguatoxicity, which was detected in 13 of 23 fish samples tested by MBA and RBA. The Mexican regulatory limit for CTX-related toxicity is 2.5 MU (mouse units) 100 g^-1 of tissue (Norma Oficial Mexicana NOM-242-SSA1-2009; ^{DOF, 2009}). This first report of ciguatoxicity in Campeche (^{Núñez-Vázquez et al., 2019}) yielded more than 10 MU 100 g^-1, primarily from samples of fish viscera.

The lionfish Pterois volitans (L.) is an invasive species posing an ecological problem in the Caribbean and the Gulf of Mexico. Official management programs have included sponsored eradication, fishing tournaments, and gastronomic exhibitions to encourage fishing and consumption. However, lionfish have been proven to be a CTX-retaining species. ^{Ley-Martínez (2016)} reported CTX in lionfish for the first time in Mexican waters, specifically in the Mexican Caribbean. ^{Almazán-Becerril et al. (2021)} analyzed lionfish and barracuda from Isla Contoy, Puerto Morelos and Isla Cozumel, in the north of Quintana Roo (Fig. 1). They found 68% of the lionfish positive for ciguatoxicity in muscle tissue, with >50% of the lionfish and 57% of the barracuda containing sublethal ciguatoxicity levels. There was an apparent latitudinal gradient for the ciguatoxicity levels, with more positive samples from the north (Isla Contoy).

Data on CFP events are considered underestimated in Mexico, partly because of the reluctance or inability of victims to report them (^{Dickey & Plakas, 2010}). In many cases, the CFP victims do not report to medical services. True CFP cases are often mistaken as common “food poisoning” from bacterial causes and mistreated with antibiotics (^{McKee et al., 2001}). Furthermore, a large proportion of the putative CFP cases reported in ^{Almazán-Becerril et al. (2021)} were from among national and international tourists, for whom follow-up diagnoses, treatment, and medical reports cannot be confirmed.

Several Coolia species produce bioactive polyether metabolites (^{Holmes et al., 1995}; ^{Rhodes et al., 2010}; ^{Karafas et al., 2015}), but putative cooliatoxins have not been fully structurally characterized (Junqueira de ^{Azevedo Tibiriçá et al., 2020}). Some of the proposed “cooliatoxins” have proven to be toxic to mice, human cell lines in vitro, and some crustaceans, such as Artemia Leach (^{Rhodes et al., 2010}; ^{Karafas et al., 2015}; ^{Leung et al., 2017}), but toxicity has been registered only for a few strains of Coolia. Several authors have reported that C. malayensis, C. palmyrensis, C. santacroceandC. tropicalis M. A. Faust strains may also produce yessotoxin (YTX) analogs, whereas other strains of the same species were found to be non-toxic (^{Wakeman et al., 2015}; ^{Phua et al., 2021}). Coolia species occur occasionally in high cell abundance in epibenthic dinoflagellate assemblages in Mexico, commonly together with potentially toxigenic members of Gambierdiscus, Fukuyoa, Amphidinium and Prorocentrum. Still, there are no confirmed reports of Coolia involved in BHAB events in Mexico. No information is available on their specific toxicity or toxin composition. However, there is one report on C. malayensis and C. monotis from Mexican coastal waters with proven toxicity against Artemia salina L. nauplii and against several cancer cell lines (^{García Santos Reyes, 2022}).

Species of the genus Ostreopsis produce palytoxins (PLTXs), which are amongst the most potent non-peptide toxins (^{Ramos & Vasconcelos, 2010}; ^{Ajani et al., 2017}). Accumulation of these toxins and their derivatives by fish and mollusks has been reported from Ostreopsis blooms in Europe (^{Munday, 2008}). The species O. ovata Fukuyo is most widely known as responsible for causing respiratory distress from aerosol inhalation during blooms, particularly in the Mediterranean Sea and southern European waters (^{Ciminiello et al., 2006}). In Mexico, a few reports on this species have been published (^{Cortés-Altamirano et al., 2011}; ^{Gallegos-Mendiola et al., 2017}), but with little information on toxigenicity or human health risks. Ostreopsis siamensis has been found in the southern part of the Gulf of California (^{Cortés-Lara et al., 2005}), and O. heptagona in the Veracruz Reef System in the southern Gulf of Mexico (^{Okolodkov et al., 2007}), but neither toxicity nor human or marine fauna poisonings have been associated with the genus.

Members of the genus Amphidinium have been recognized as producers of bioactive compounds for more than three decades (^{Yasumoto et al., 1987}). No toxic events have been definitively linked to Amphidinium blooms and their confirmed toxins to date in Mexico. This is in spite of the fact that A. carterae and A. operculatum occur frequently in Mexico, and both species are known to produce ichthyotoxic compounds (^{Kobayashi & Tsuda, 2004}; ^{Kobayashi & Kubota, 2007}; ^{Murray et al., 2015}). Amphidinium blooms have been linked to rare fish kills globally, and at least four species (A. carterae, A. gibossum, A. massartii, and A. operculatum) have been reported as ichthyotoxic (^{Murray et al., 2012}, ²⁰¹⁵; ^{Pagliara & Caroppo, 2012}). Ichthyotoxicity of Amphidinium species is likely due to the production of amphidinols (AM) (^{Wellkamp et al., 2020}), hairpin-shaped polyketides forming difeferent subgroups with the same core unit. Preliminary analysis of a few Amphidinium isolates from Mexico - three isolates from Baja California Sur (AA39, AA40) and the Veracruz Reef System (AA60), did not detect any AM derivatives by advanced mass spectrometry techniques (^{Wellkamp et al., 2020}). A more comprehensive recent survey of multiple isolates from the Gulf of California (La Paz, Baja California Sur), Gulf of Mexico (Veracruz Reef System, Veracruz and Laguna de Términos, Campeche), southern Mexican Pacific (Isla de La Roqueta, Acapulco, Guerrero), and the northern Mexican Pacific (San Quintín, Baja California) from A. eilatiensis J. J. Lee, A. massartii, A. operculatum, and A. theodorei revealed a high diversity of AM analogs. These AM derivatives include the known AM02, 04, 05, 06, 07, 09, 11, 14, 15, and 17, as well as four recently described by ^{Wellkamp et al. (2020)}, and three new variants. This study showed that the A. eilatiensis strains from the northern Pacific coast of Mexico produced a higher diversity of AMs and total quantities per cell than any previously analyzed strains (Durán-Riveroll, pers. comm.). The non-AM-producing Amphidinium isolates (AA39, AA40, and AA60) have shown significant cytotoxic activities against cancer cell lines (^{Mejía-Camacho, 2020}; ^{Mejía-Camacho et al., 2021}), meaning that this bioactivity could be expressed by unknown metabolites other than AMs.

On a global scale, species of Prorocentrum are widely known as the proximal source of polyketide toxins associated with DSP, primarily okadaic acid (OA) and various related dinophysistoxin (DTX) derivatives (^{Durán-Riveroll et al., 2019a}). Collectively, these are referred to as diarrhetic shellfish toxins (DSTs), although the potency and relative toxicity of many of these analogs have not been confirmed. Prorocentrum species that produce DSTs are almost exclusively (epi)benthic forms, but they can occasionally be found as members of the tychoplankton via spontaneous detachment or through disturbance of their benthic habitat, e.g., by winds, currents, and other mechanical agitation. In fact, there are no confirmed cases of any truly planktonic Prorocentrum known to produce DSTs.

Epibenthic Prorocentrum are common throughout the tropical and subtropical coastal waters of Latin America (^{Durán-Riveroll et al., 2019a}), but toxigenicity has only been confirmed for a few populations from cultured isolates from Mexico (^{Cembella et al., 2021}). Prorocentrum lima from Isla El Pardito, Baja California Sur, was suspected as a possible cause of poisoning of five local fishermen after consumption of the liver of the Colorado snapper (Lutjanus colorado Jordan et Gilbert (Eupercaria: Lutjanidae) and the sawtail grouper (Mycteroperca prionura Rosenblatt et Zahuranec (Perciformes: Serranidae). Indeed, a cultured P. lima isolate from this event was found to produce DSTs and was toxigenic in bioassays (^{Heredia-Tapia et al., 2002}). The symptoms of the poisoned fisherman, however, were more similar to those of CFP, lasted a few days, and involved diarrhea, numbness, vomiting, weakness, pruritus, desquamation, hyperesthesia, lip and tongue paralyses, and in one case, convulsions - not classic symptoms of DSP. Benthic species other than P. lima were likely involved in this event because Prorocentrum species are not known to produce CTXs. No further effort has been made to isolate epibenthic dinoflagellates from this site to confirm a relationship to fish poisoning events.

DST toxigenicity among epibenthic Prorocentrum species has been intensively investigated among multiple cultured isolates from various locations in Mexico, including the Gulf of California, the Caribbean Sea and the Gulf of Mexico (^{Tarazona-Janampa et al., 2020}; ^{Cembella et al., 2021}). The application of advanced liquid chromatography coupled with tandem mass spectrometry (LC-MS/MS) (^{Krock et al., 2008}; ^{Nielsen et al., 2013}) permitted the resolution of the DST compositional profile and cell quota of 67 isolates from Mexico - the largest chemodiversity analysis of polyketide-derived toxins from a benthic dinoflagellate genus (^{Cembella et al., 2021}). All P. lima and P. hoffmannianum (except one) isolates were toxigenic, but the total cell DST content was not apparently well linked at the species level. The species complexes defined by morphological and molecular criteria to separate P. lima from P. hoffmannianum sensu lato could be distinguished by the relative composition of some analogs (OA, OA-D8, DTX1, DTX1a and DTX1a-D8), including two new undescribed DTX isomers. In any case, toxin composition exhibited no clear associations with substrate type or geographical origin, in spite of some clear affiliations within geographical populations.

Toxicity and species interactions. The functional role and allelochemical effects of BHAB species in chemical defense responses and competitive species interactions are still poorly understood.

There are a few experimental studies on strains of epibenthic dinoflagellates from Mexico. Most have focused on Prorocentrum that may be associated with the production of DSTs, and hence pose a potential (but unconfirmed) risk of DSP from shellfish consumption (Table 5). In fact, the first known culture of a benthic dinoflagellate from Mexican coastal waters was a strain of P. lima established from Isla El Pardito, Baja California Sur, after the poisoning of five local fishermen by consumption of fish liver from the area.

After the isolation of the aforementioned first strain of P. lima, many studies have been conducted to determine toxic responses of exposure in several vertebrate and invertebrate species. ^{Heredia-Tapia et al. (2002)} confirmed the toxicity and toxigenicity of P. lima by bioassays with adults of the brine shrimp Artemia franciscana Kellogg and by MBA of lipophilic toxins of lipophilic toxins. When exposed to a P. lima cell density of 2 x 10³ cells mL^-1, the crustaceans died within 2 h. The MBA showed typical symptoms for DSP toxins in mice (diarrhea, hind limb paralysis, dyspnea, respiratory failure), followed by death after 1 h after i.p. administration of a more concentrated cell extract (3.35 x 10⁴ cells mL^-1). The possible occurrence of DSTs in the cell extract was indicated by thin-layer chromatography (TLC), and the presence of OA and DTX1 were later confirmed by LC-MS.

Antioxidant enzymes have been used to evaluate oxidative stress induced by toxic microalgae in crustaceans and bivalve mollusks (^{Oyaneder-Terrazas et al., 2022}; ^{Tchivelekete et al., 2022}). ^{Campa-Córdova et al. (2009)} compared the changes in superoxide dismutase (SOD) activity in the hepatopancreas and muscle of juveniles of the white leg shrimp, Litopenaeus vannamei (Boone) (Crustacea: Decapoda: Penaeidae) and in the giant lion-paw scallop, Nodipecten subnodosus G. B. Sowerby I (Bivalvia: Ostreoida: Pectinidae), after exposure to different cell densities of P. lima. The results indicated the activation of antioxidant defenses, especially in the hepatopancreas, with higher levels of SOD in shrimp than in scallops. A cell concentrate of 500 cells mL^-1 of P. lima was enough to trigger SOD activity in both species, indicating that oxidative stress is induced by application of P. lima extracts, but the specific toxin content is unknown.

The Pacific oyster, Magallana gigas (Thunberg) (= Crassostrea gigas (Thunberg) (Bivalvia: Ostreoida: Ostreidae), a species of high commercial interest in Mexico, was introduced from Japan in 1973 (^{Arizpe, 1996}). Since then, it has become distributed in many coastal areas of Mexico (^{Paniagua-Chávez & Acosta-Ruiz, 1995}; ^{Arizpe, 1996}; ^{Cáceres-Martínez et al., 1998}, ^{García-Rico et al., 2001}). Among other suspension-feeding bivalves, this species is widely known to accumulate DSTs (^{Mafra et al., 2015}; ^{Karlson et al., 2021}), although primarily from pelagic blooms of Dinophysis spp. Contamination of bivalve mollusks with DSTs occurs only rarely from epibenthic Prorocentrum (^{Lawrence et al., 2000}; ^{Gayoso et al., 2002}). ^{Romero-Geraldo & Hernández-Saavedra (2014)} determined the effect of P. lima strain PRL-1 isolated from Isla El Pardito on stress response genes in M. gigas. Gene expression of glutamine synthetase (GS), glutathione S-transferase (GST), heat shock protein 70 (HSP70) and 90 (HSP90), Cu/Zn SOD and melanogenic peroxidase (POX) were assayed by semi-quantitative reverse-transcription polymerase chain reaction (RT-PCR). In general, an increased expression was observed at all P. lima cell densities tested, as a rapid response to exposure (from 0 to 3 h), whereas lower expression levels than the control were observed after 6 h exposure. The most noticeable differences in the expression levels of the analyzed genes were observed in juvenile oysters at 14 days of exposure, suggesting that the P. lima cell densities affect oyster metabolism irreversibly under sub-chronic exposure. Oysters are thus considered highly susceptible to the presence and persistence of BHABs of P. lima. The expression of the key genes involved in the first line of defense is significantly affected, making juvenile oysters vulnerable to the cumulative impact of multiple stressors (environmental, chemical, or biological). The authors (^{Romero-Geraldo & Hernández-Saavedra, 2014}) conclude that this explains the mass mortalities observed in oyster farms, particularly in newly sown oyster seed of 3-5 mm length in the field.

To further understand the acute and sub-chronic exposure effects of P. lima cells on Magallana gigas juveniles (3-5 mm),^{Romero-Geraldo et al. (2014)} followed the effects of P. lima on oysters by analyzing the expression levels of three genes involved in cell cycle regulation and one gene involved in the immune response. The analyses by PCR and real-time quantitative PCR (qPCR), detected changes in expression patterns with exposure time and P. lima cell concentration. The highest expression levels (fold above baseline) were found in oysters fed 3 x 10³ cells mL^-1 at 168 h for the cycle regulator p21 protein (9-fold), chromatin assembly factor 1 p55 subunit (8-fold), elongation factor 2 (2-fold) and lipopolysaccharide/b-1,3 glucan-binding protein (13-fold). Additionally, the transcript level of all the genes decreased in oysters fed with 30 x 10³ cells mL^-1 of the planktonic P. micans Ehrenb. after 72 h, and was lowest for the chromatin assembly factor 1 p55 subunit (0.9-fold below baseline). In M. gigas, whole-cell ingestion of P. lima caused a clear mRNA modulation expression of the genes involved in cell cycle regulation and the immune system. Acute exposure caused an alteration of the transcript levels of all the studied genes, indicating an immediate or early stress response. Sub-chronic exposure generated a higher expression level in all the genes, causing a significant impact that could be related to DNA damage and control loss of the cell cycle. This genomic instability might lead to diseases, severe pathologies, or death in oysters. The increased Cg-LGBP expression level shows an activation of the innate immune system as the first line of defense in M. gigas against P. lima cells (or their DSTs and other bioactives), suggesting it was recognized by the oyster as a pathogenic agent. The authors conclude that overexpression of the genes could be related to DNA damage, disturbances in the cell cycle continuity, and was probably a genotoxic effect, as well as an activator of the innate immune system as the first line of defense.

Continuing this line of research, ^{Romero-Geraldo et al. (2016)} reported the results of M. gigas exposed to P. lima cells to determine behavioral changes, histopathological alterations, and changes in expression patterns of the genes involved in the cell cycle (p21, cafp55, p53), cytoskeleton (tub, act) and the inflammatory process (casp1). The results demonstrated that P. lima under a realistic BHAB cell density of 3 x 10³ cells mL^-1 affected the feeding behavior of M. gigas, causing hemocyte accumulation and infiltration, structural loss of the digestive gland epithelium tubules, and inducing tissue-specific gene expression pattern modifications. These changes were found at the transcript level in a set of time-dependent tested genes (p53, cafp55, ef2, act, tub and casp1), suggesting the presence of P. lima in the digestive gland of M. gigas might trigger an inflammatory tissue process and cytoskeleton disruption. The authors also pointed out that a disruption of the hyperphosphorylation process generated by the presence of the OA by P. lima cells probably resulted in eliminating some cell cycle control points, which compromised the tissue process, disturbed the cell cycle and cytoskeleton and represented a risk to oyster integrity. The presence of DSTs in P. lima cells decreased the clearance rate, induced structural loss, significantly decreased the tubule area of the digestive gland and up-regulated the expression of all genes.

Following the same research theme, ^{García-Lagunas et al. (2019)} compared the effect of the paralytic shellfish toxin-producing (PST) planktonic dinoflagellate, Gymnodinium catenatum H. W. Graham, and the DST-producer, P. lima, on the differential expression of the immune response genes in M. gigas spat. Again, the consumption of toxic dinoflagellates by the oyster spat caused changes at the expression level of the genes involved in the immune response. Combined diets of both dinoflagellates generated changes in the feeding behavior of oysters, with a greater preference for feeding on G. catenatum. Expression levels of lipopolysaccharide (LPS)-binding protein 1,3-glucan were higher in oysters exposed to both dinoflagellates, which was probably related to an activation of the oyster immediate immune response during the first 24 h. Protein 44 interferon-induced gene expression level was repressed in treatments with the highest dinoflagellate concentration and overexpressed in the diet with equal dinoflagellate concentration. Interaction protein-Toll and immunoglobulin gene transcript levels reached the highest values on day 7 in oysters exposed to combined diets. The results indicated that the immune defense appeared activated in oyster spat as a response to DSTs versus PSTs (or other bioactives) and/or extracellular compounds produced by the dinoflagellates.

These experiments demonstrate the high susceptibility of the immune defense of three common invertebrate species from the Gulf of California - the shrimp Litopenaeus vannamei, the giant lion-paw scallop Nodipecten subnodosus and the oyster Magallana gigas - when exposed to low cell abundances of P. lima. Exposure likely makes them susceptible to other biological stressors and contributes to mortalities observed in aquaculture facilities. Unfortunately, these studies have been performed only with one epibenthic dinoflagellate species. As the number of available benthic dinoflagellate strains with different toxin profiles increases, such experimental studies can be better targeted to address the effects of benthic dinoflagellates on the physiology of diverse marine fauna, including key seafood species.

Several studies have demonstrated strong interactions between marine bacteria and epibenthic dinoflagellates (^{Sakami et al., 1999}; ^{Wang et al., 2018}). The extension of these close bacterial interactions with the phycosphere are still being elucidated. ^{Tarazona-Janampa et al. (2020)} studied the effect of culturable associated bacteria on the growth and toxigenicity of nine strains of the P. lima species complex from Isla Verde (Veracruz Reef System), Veracruz, and Puerto Morelos, Quintana Roo. Twenty-one bacterial genera belonging to Proteobacteria, Actinobacteria and Bacteroidetes were identified by amplifying the 16S rRNA gene marker from clonal Prorocentrum cultures, with gamma-Proteobacteria comprising the dominant class. A positive correlation was found among the bacterial genera associated with two Prorocentrum clones and the esterified toxin analog DTX1a-D8, but there was no apparent correlation between the other dinoflagellate clones and their associated bacteria with the other five detected DSTs. Additionally, there was no significant correlation between Prorocentrum cell volume, growth rate, bacterial cell counts, or cellular toxin concentration over the entire time-series culture cycle.

These results provided little confirmatory evidence that extracellular bacteria played a critical role in the regulation of DSP toxin production in Prorocentrum or modulation of their growth under non-nutrient limited conditions in culture. Nevertheless, the percentage of culturable bacteria is low compared to the total number of species found in the environment (^{Joint et al., 2001}); an approach considering non-culturable bacteria found in the natural environment and associated with BHAB species will give us a further understanding of this interaction.

^{Ramos-Santiago et al. (2022)} determined the effects of the extracts from the brown macroalga Dictyota dichotoma (Hudson) J. V. Lamour. (Phaeophyceae: Dictyotales: Dictyotaceae) on the growth of two strains of C. malayensis isolated from Bahía de La Paz, BCS, southern Gulf of California. Both isolates were cultured in GSe medium with and without macroalgal extracts: 1) GSe medium, 2) GSe plus a crude extract of D. dichotoma, 3) GSe plus the carbohydrate fraction of D. dichotoma, and 4) GSe with the addition of the non-precipitable fraction of D. dichotoma. The results indicated that the different fractions of D. dichotoma neither favor nor limit the growth of C. malayensis.

A biotechnological approach based on physiological changes in Amphidinium carterae is currently underway to explore the potential for biofuel production (^{Mendoza-Flores et al., 2022}). A strain of A. carterae isolated from a bloom in Bahía Todos Santos, Ensenada, Baja California, was exposed to five irradiance levels (50, 100, 150, 200 and 250 µmol photons m⁻² s⁻¹), and the effect on the growth rate, photosynthesis and the content in protein, carbohydrate, lipid and pigment was determined. The highest cell concentration was found in the cultures grown at 150 µmol photons m⁻² s⁻¹ (130 × 10³ cells mL⁻¹), and the lowest value at 50 µmol photons m⁻² s⁻¹ (49 × 10³ cells mL⁻¹). Cultures maintained under the lowest irradiance had the highest cell yield by organic and inorganic dry weight. The protein and carbohydrate content changed significantly with the irradiance level, with the highest values (1600 and 560 pg cell⁻¹, respectively) at 200 µmol photon m⁻² s⁻¹. The lipid content was modified by the effect of irradiance, with the highest values (6920 pg cell⁻¹) at the lowest irradiance. As a general trend, the highest irradiances increased the photosynthetic rates. These findings demonstrate that although the strain of A. carterae can grow well at higher irradiances (100 to 250 µmol photons m⁻² s⁻¹), a significant increase in the lipid content occurred at low irradiances. The authors suggest that this increase makes this species a candidate for use in biodiesel or ethanol production.

Concluding remarks and future perspectives. Due to negative impacts of benthic dinoflagellates causing BHABs in marine ecosystems with major consequences on human health and socioeconomic activities, studies on taxonomy, species diversity, allelochemical and toxic interactions and ecodynamics have increased during recent years in Mexico. The primary focus on a handful of key BHAB species has been necessary because of resource and personnel limitations, but has led to a relative neglect of cryptic, rare and non-BHAB taxa - hence a likely underestimate of diversity of benthic dinoflagellates in Mexican waters. In most cases, taxonomic studies have simply corroborated original descriptions based on morphological traits and strengthened local knowledge for accurate species identification (^{Cortés-Altamirano & Sierra-Beltrán, 2003}; ^{Hernández-Becerril & Almazán-Becerril, 2004}; ^{Almazán-Becerril et al., 2015}, ^2016a, ^b). A few more profound taxonomic investigations have indeed contributed to reclassification of genera or species within newly created genera or to resolving species complexes (^{Tarazona-Janampa et al., 2020}; ^{Cembella et al., 2021}). In Mexico, the need for and importance of reinforcing the identification of toxigenic species with molecular genetic data to minimize health risks has been previously emphasized by ^{Núñez-Vázquez et al. (2019)}. The point is worth repeating herein because studies focused on molecular taxonomy of these BHAB dinoflagellates remain scarce, in particular for biogeographical investigations integrating morphological and molecular descriptors with toxin phenotype and composition (e.g., ^{Cembella et al., 2021}). This review presents limited molecular genetic information available for toxigenic species of the genera Amphidinium, Coolia, Gambierdiscus, Ostreopsis and Prorocentrum. However, there are no molecular data available for populations of Fukuyoa, Vulcanodinium and other BHAB genera found in Mexico, highlighting a window of opportunity to confirm or reject previous identifications based only on morphological criteria, as well as to corroborate the presence of cryptic species and to discard misidentifications of species with high morphological similarities within a group. The latter problem has occurred frequently with species such as Ostreopsis ovata, O. siamensis, O. fattorussoi Accoroni, Romagnoli et Totti and O. rhodesiae Verma, Hoppenrath et Shauna Murray that are difficult to distinguish only by light microscopy (^{Carnicer et al., 2020}). In a broader context, application of ecogenomics technology - e.g., a DNA approach - will allow for genotyping in situ natural populations of benthic dinoflagellates for exploring their diversity, cryptic speciation and their functional role in marine ecosystems in Mexican waters.

Ecosystem level studies on benthic dinoflagellates with molecular tools have barely scratched the surface to date. As for habitats for epibenthic dinoflagellates, the few studies done in Mexico have focused on macroalgae and seagrasses. Other natural solid substrates, such as rocks, corals (both alive and dead) and other invertebrates, vertebrate species, mangrove roots, sediments (e.g., sandy or silty) and anthropogenic artificial substrates have not been paid much attention hitherto.

In a previous review of data published from 1940 to 2011 on the status of HAB studies in Mexico (^{Band-Schmidt et al., 2011}), it was pointed out that most research was conducted during short-term research programs based on sampling from limited local field sites and focused on few selected HAB-forming species. The authors also commented that many research topics had not been adequately addressed: allelopathy, cyst and life-history studies, advective transport of species and blooms, effect of climate change on HABs, socioeconomic analyses and mitigation strategies, among others. Even a decade later these comments remain equally valid for Mexico. In the recent past most attention has focused on the classic phycotoxins affecting human health and marine ecosystem functioning, while relatively neglecting the fact that many benthic dinoflagellates produce bioactive natural products with allelochemical effects on species interactions. In particular, the polyketide-derived polyether metabolites with uncharacterized functional roles (^{Durán-Riveroll et al., 2019a}) warrant closer scrutiny. Several of these compounds are under investigation in Mexico for biotechnological potential as anticancer or cell division inhibitors and as future therapeutants.

The claim of inadequate studies on HABs for the above topics remains valid on a global basis - not just for Mexico, where in fact the skills, knowledge, and infrastructure to address these issues have improved dramatically in recent years. Global research initiatives on benthic dinoflagellate blooms (BHABs), initiated more than a decade ago via international coordinating programs, such as GEOHAB (Global Ecology and Oceanography of Harmful Algal Blooms; since 2001) and its successor GlobalHAB (Global Harmful Algal Blooms; 2016-2025) of IOC-SCOR UNESCO, have yielded a more clearly defined perspective on the natural phenomena and the causative mechanisms and impacts. Nevertheless, science and monitoring in Mexico has not kept pace with other major regions faced with acute BHAB events (e.g., South Pacific, Caribbean Sea, Mediterranean Sea, including islands and archipelagos). In the near future, Mexico may experience even more cases of human intoxication, ecosystem disruption and crisis events (e.g., of ciguatera), caused by climate change or direct anthropogenic effects on marine ecosystems. The BHAB problem is not in crisis state in Mexico, cases are few, sporadic and widely dispersed, but the current status of scientific knowledge is clearly inadequate to describe and explain the extent of the phenomena, much less to predict and design appropriate monitoring and mitigation strategies in collaboration with the seafood industry and public health agencies. A comprehensive BHAB strategy will require more than fine-tuning morphotaxonomy for accurate identification of benthic dinoflagellates and describing new species and toxins.

From the present review it is evident that studies on epibenthic dinoflagellates in Mexico have begun to emerge more frequently in recent years. Hitherto, most studies on BHAB species and their effects have been crude investigations based on acute mortalities and toxicities of seafood species, e.g., on the effect of P. lima on shrimps and bivalves. There is ongoing work with international partners and collaborators that will lead to a greater understanding of these important epibenthic assemblages in the country. There are now several species banks including live strains of benthic dinoflagellates isolated from diverse coastal regions of Mexico available for laboratory research. Collaborations with international colleagues and global research initiatives with access to advanced technology platforms support the required studies on toxicity, allelopathy, ecological interactions, effects on mariculture species, biosynthesis of metabolites and biotechnological potential that remain to be addressed for benthic dinoflagellates in Mexico.