INTRODUCTION

The northern Yucatan Peninsula is characterized by shallow waters at the seaside, extended sandy beaches, and mangroves. In some places, subaquatic vegetation is abundant and dominated by seagrasses (Thalassia testudinum Banks ex König, Syringodium filiforme Kützing, and Halodule wrightii Ascherson) and green algae mainly of the genera Caulerpa J. V. Lamouroux, Udotea J. V. Lamouroux, Halimeda J. V. Lamouroux, Penicillus Lamarck, Batophora J. Agardh, and Avrainvillea Decaisne (^{Aguilar-Trujillo et al., 2014}, ²⁰¹⁷; ^{Okolodkov et al., 2014}). Along the coast, there are several marinas (Sisal, Chuburná, Uaymitún, Telchac, Dzilam and El Cuyo) and natural lagoons such as Chuburná and Río Lagartos, the latter with two fishing villages, San Felipe and Río Lagartos, on the shore. The marinas are subjected to anthropogenic influence, primarily fishery activities, and to accumulation of decomposing mangle residues on the bottom. Some of these sites, such as Chuburná and Dzilam, are characterized by nutrient-rich underground water (^{Álvarez-Góngora & Herrera-Silveira, 2006}; ^{Morales-Ojeda et al., 2010}; ^{Murgulet et al., 2020}) that discharges into the lagoon and the sea near the coast. There are no rivers along the Yucatan coast. These features are the marine part of a ring of cenotes (sinkholes with groundwater), better known as small continental water bodies in the Yucatan Peninsula, formed along the outer edge of the Chicxulub crater, produced by a meteorite collision during the Cretaceous about 65 million years ago (^{Connors et al., 1996}; ^{Pope et al., 1996}). Groundwater discharges causes minor salinity variations. Three meteorological seasons can be distinguished: a dry season from March to early June; a rainy season from June to October; and the “nortes” (northerly winds) season with short periods of storms and strong winds coming from the north from November to February (^{Herrera-Silveira, 1993}).

In August 2010, an unknown small-sized Peridiniales species caused an intense bloom in the Sisal marina. In subsequent years, it was occasionally found at other sites along the Yucatan coast. It was assigned to the family Kryptoperidiniaceae, a monophyletic group of dinoflagellates known as “dinotoms” (due to the presence of plastids derived from diatoms), although a second (eukaryotic) nucleus was not observed, thus sharing its thecal features with both Kryptoperidinium Lindemann (until now a monotypic genus; ^{Tillmann et al., 2023}) and Durinskia Carty et Cox (^{Okolodkov et al., 2020}). By comparing both morphological and ecological characteristics, it was concluded that this species was different from all known Kryptoperidinium and Durinskia species, as well as from the other genera (Blixaea Gottschling, Dinothrix Pascher, Galeidinium M. Tamura et T. Horiguchi and Unruhdinium Gottschling) of the same family. However, no preference was given to either of these taxa in identifying the genus, and a name was not given to the new species then. The main objective of the present study was to reveal its ecological preferences. In addition, a name has now been assigned to this species for further reference.

MATERIAL AND METHODS

Phytoplankton blooms were monitored from 13 August 2011 to 8 September 2014 (Fig. 1). Samples were collected with a 2-liter Van Dorn bottle and fixed with acidic Lugol’s solution. Nutrient concentrations (ammonium [NH₄], nitrates [NO₃], nitrites [NO₂], orthophosphates [PO₄], silicates [SiO₄] and urea [CH₄N₂O]) were determined according to spectrophotometric techniques (^{Strickland & Parsons, 1972}). Dissolved inorganic nitrogen (DIN) was calculated as the sum of NO₃, NO₂, and NH₄ concentrations. Surface water temperature (SWT), salinity, and dissolved oxygen (DO) were measured using a YSI Professional Plus multiparameter meter (Yellow Springs, Ohio, USA). Chlorophyll-a (chl-a) was determined by the extraction of pigments using a 90% acetone solution followed by spectrophotometric determination and recalculation formula (^{Jeffrey & Humphrey, 1975}; ^{Parsons et al., 1984}).

Figure 1 Sampling sites along the northern Yucatan Peninsula coast in 2011-2014. Small circles indicate stations near exposed coasts, and large circles indicate stations inside semi-enclosed marinas. 

Cells were counted in a 1-mL Sedgwick-Rafter chamber on an Olympus CK2 inverted microscope (Olympus Optical Co., Ltd., Japan) equipped with an LCAch N 40x/0.55 PhP objective. Fixed specimens were also examined in a JEOL JSM-7600F (JEOL, Ltd., Tokyo, Japan) scanning electron microscope (SEM) at a working distance of 15 mm, a voltage of 5.0 kV and a low secondary electron detector (LEI) after a preliminary wash in distilled water followed by dehydration in a series of ethanol solutions of increasing concentration (30, 50, 70, 90, and 100%). Specimens were then air dried on 0.5” aluminum mounts and sputter-coated with gold-palladium using a Polaron SC7640 High-Resolution Sputter Coater (Quorum Technologies, Newhaven, SXE, UK).

Permutational multivariate analysis of variance (PERMANOVA) and Euclidian distance analyses were applied to determine any significant differences among physicochemical variables: SWT (°C), salinity, DO (mg L^-1), phosphates (µM), silicates (µM), urea (µM), chl-a (mg m^-3) and DIN (µM). Years (2011, 2012, 2013, and 2014) and months were used as fixed factors for each zone (exposed coast and marinas) (^{Anderson, 2014}). PERMANOVA was also applied to evaluate spatio-temporal differences in cell abundances (cell L^-1) of this new Durinskia species among sampling sites and months, using the same fixed factors, the Bray-Curtis similarity distance, and Adonis function of the vegan R package (^{Oksanen et al., 2013}). Similarly, spatio-temporal variation of cell abundances was visualized using a bubble chart and ggplot2 function (^{Lee et al., 2020}).

To reveal relationships between the response variable (cell abundance) and the explanatory variables (physicochemical factors) (^{Wood, 2006}; ^{Zuur et al., 2007}), generalized additive models (GAM) with a negative binomial distribution and a log link function were used to adjust the data on cell density for the effect of the physicochemical variables. The models were constructed for each year (2011 to 2014) and zone (exposed coast and marinas) using gam function in the mgcv2 package, suggesting smooth effects of each explanatory variable on the response variable (the Durinskia sp. cell abundance). Previously, to avoid multicollinearity of the explanatory variables, the Pearson correlation was estimated, excluding the highly (r≥60%) correlated variables for each constructed model. In addition, the Akaike information criterion was used to choose the best-adjusted model (^{Symonds & Moussalli, 2011}). All analyses were performed using R (^{R Core Team, 2022}) and RStudio (^{RStudio Team, 2015}).

RESULTS

Taxonomy

Morphological description of the species has been previously published (^{Okolodkov et al., 2020}). It was not assigned to any genus of the family Kryptoperidiniaceae; however, as stated, by comparing both morphological and ecological characteristics, our species is different from all known Kryptoperidinium and Durinskia species in having a larger 1a plate compared to plate 2a (Fig. 2E). Morphological differences between our species and the genus Kryptoperidinium (Table 1) allowed us to assign the species to the genus Durinskia. Herein, additional SEM micrographs are given to complement the previously published description, accompanied by SEM and light microscope images.

Figure 2 Durinskia yucatanensis under SEM: A and B - cells in ventral view; C and D - the sulcal area in ventral view; E - dorsal view; F - left-side view. Symbols for thecal plates: 1’-4’ - the apical plates; 1”-6” - the precingular plates; 1a and 2a - the anterior intercalary plates; 1’”-5’” - the postcingular plates; 1””and 2”” - the posterior plates; APC - the apical pore complex; c1-c5 - the cingular plates; fp - the flagellar pore; PC - the pore cover plate; Po - the pore plate; Sa - the anterior sulcal plate; Sd - the right sulcal plate; Sp - the posterior sulcal plate; x - the canal plate. Scale bar: 1 μm. 

Table 1 Comparison of Durinskia yucatanensis (present study) with Kryptoperidinium triquetrum (based on ^{Saburova et al., 2012} - Durinskia spp., and ^{Tillmann et al., 2023} - K. triquetrum). 

Morphological features	D. yucatanensis	K. triquetrum
Cell length	12.5-23.7 μm	15-50 μm
Thecal pores	sparsely, irregularly located	densely located, often arranged in rows on some dorsal epithecal plates
Epitheca	slightly larger than hypotheca (up to 1.5 times)	equal to hypotheca
Plate formula	Po X 4’ 2a 6” 5c 4s(?) 5”’ 2””	Po X 4’ 2a 7” 5c 7s 5”’ 2””
Plate 1’	slightly asymmetrical, extended longitudinally, not subdivided	strongly asymmetrical, extended obliquely, from the apex toward the right side, subdivided into 1’ a and 1’ p
Precingulars	6”	7” (the 7” plate is L- or boot- shaped, very narrow)
Precingulars 1”-5”	1” and 5” are the highest	of similar height
Anterior intercalaries	1a > 2a	1a < 2a
Cingulum path	submedian (closer to the antapex), descending, displaced ¾-1 girdle width	median, circular, disconnected ventrally by plate 1’ p
Plate PC*	present	absent

*Note: plate PC is probably a modified plate Sd.

Phylum Dinoflagellata (Bütschli) ^{Fensome, Taylor, Norris, Sarjeant, Wharton et Williams, 1993}

Subphylum Dinokaryota ^{Fensome, Taylor, Sarjeant, Wharton et Williams, 1993}

Class Dinophyceae Pascher, 1914

Subclass Peridiniphycidae ^{Fensome, Taylor, Sarjeant, Wharton et Williams, 1993}

Order Peridiniales Haeckel, 1894

Suborder Peridiniineae Poche, 1913

Family Kryptoperidiniaceae Lindemann, 1926

Genus Durinskia Carty et Cox, 1986

Durinskia yucatanensis Okolodkov, Steidinger et Gárate-Lizárraga sp. nov.

Figs. 3-7 in ^{Okolodkov et al., 2020} and Figs. 2-3 herein.

Figure 3 Durinskia yucatanensis under SEM: A and B - cells in ventral view; C - cell in apical-dorsal view; D - the apex with adjacent plates in apical-dorsal view; E - cells in antapical (above) and apical views (below); F - cell in left-side view. For symbols, see Fig. 2. Scale bar: 1 μm. 

Description: Published in: ^{Okolodkov et al., 2020} (A Kryptoperidiniaceae species (Dinophyceae: Peridiniales) blooming in coastal Yucatan waters, Gulf of Mexico. Protistology 14(2): 58-69, 7 figs, 1 table 1; as Kryptoperidiniaceae species).

Affinities: Out of the six to seven Durinskia species that have been described so far, this species is morphologically (in cell size and thecal morphology) closer to Durinskia kwazulunatalensis N. Yamada, Sym et Horiguchi (^{Yamada et al., 2017}), described from a marine costal environment in South Africa. The latter is 18.4-27.4 μm long, with the median, descending, and displaced approximately by its own width, the 1a and 2a plates of almost the same size, the suture between them approximately central on the dorsal side of the epitheca, a trapezoidal 3” plate and a pentagonal 4” plate.

Holotype:Figure 6a in ^{Okolodkov et al. (2020: 64)} obtained from a formalin-fixed net sample from the station (st.) 1 taken on 18 August 2010, leg. Yuri B. Okolodkov. The sample is deposited in the collection of liquid samples of phytoplankton and benthic microalgae in the Laboratory of Marine Botany and Planktology at the Institute for Marine Sciences and Fisheries, University of Veracruz (Instituto de Ciencias Marinas y Pesquerías, Universidad Veracruzana (ICIMAP-UV), Boca del Río, Veracruz, Mexico).

Type locality: Sisal marina (21°9’40.22” N, 90°2’42.71” W), the northern coast of the Yucatan Peninsula, State of Yucatan, Mexico.

Habitat: Coastal marine, planktonic.

Etymology: The epithet yucatanensis refers to the Yucatan Peninsula.

Ecology

Our species was usually found at marine salinities between 32 and 40, at water temperatures between 28 °C and 33 °C, characteristic of late summer (August-September), in a wide range of macronutrient concentrations (Table 2). However, during the period 2011-2014, at the distance of 1-7 km from the coastline, in shallow waters with site depths of 1.2-6.1 m, the species was found with an abundance of 2.7 x 10⁴ - 8.0 x 10⁴ cells L^-1. Based on the bloom observed in August 2010 (with up to 3.75 x 10⁷ cells L^-1), it appears to prefer eutrophic conditions that are typical for marinas along the northern coast of Yucatan.

Table 2 Permutational multivariate analysis of variance (PERMANOVA) for the cell abundances of Durinskia yucatanensis transformed using Log+1 and the Bray-Curtis distance. Models for the different fixed factors, years, months, sampling sites and zones. 

	Years		Months		Sites		Zone		Interaction
	Pseu.F	P	Pseu.F	P	Pseu.F	P	Pseu.F	P
Abundance of D. yucatanensis (cells L-1)	12.12	0.001***	3.96	0.001***	3.16	0.001***	21.74	0.001***	Years-zones
									Years-months***

Significance codes: <0.001***, 0.001**, 0.01*, 0.05*.

The PERMANOVA analysis revealed significant differences between years and, months and sampling sites (Table 2). All physicochemical variables, except phosphates, showed a seasonal pattern, particularly between months and years. Orthophosphates showed a multi-year pattern (PERMANOVA, P=0.03; Table 2). The mean annual temperatures differed significantly between the two zones: 29.1±1.84 °C along the exposed coast and 27.92±1.9 °C in the marinas (PERMANOVA; P=0.001). In the exposed coastal zone, the highest temperatures (33.3 °C at st. 15 and 16) and salinity values (37.56-39.37) were measured in July-August, and the minimum temperature (23 °C) was registered in March for all four years; the minimal temperature (~22.7 °C) along the exposed coast was also observed in March. The highest DO concentrations (>10 mg L^-1) and salinity (39.93) were observed in marinas in 2014, in particular, in August-October at st. 7, 8, 2, 3, and 5). The highest concentrations of orthophosphates (10.61 µM L^-1, June 2011), urea (19.48 µM L^-1, June 2011), and silicates (95 µM L^-1, March 2014) along the exposed coast were measured in spring-early summer.

The highest DIN concentrations were measured along the exposed coast in January 2012, especially at st. 2 to 14 (23.49 a 72.49 µM L^-1), with annual peaks up to >40 µM L^-1 in marinas (in June 2011, April 2012, May 2013 and March 2014). Chl-a was the only parameter that presented the most significant spatio-temporal variability between years, months and sampling sites (P=0.001). The maximal value (131.67 mg m^-3) in marinas was measured in February 2012 when the highest DIN concentration (57.85 µM L^-1) occurred. Along the exposed coast, the highest chl-a value (34.06 mg m^-3) occurred in 2012, while the mean annual value for the other three years was 21.83 mg m^-3 (September 2011, October 2013, and January 2014; Fig. 4).

Figure 4 Annual mean values of physicochemical variables (mean ± SD) along the northern Yucatan coast in 2011-2014. 

Cell abundance of D. yucatanensis changed during the study period. A temporal pattern between months was found (PERMANOVA, P=0.001, Table 3). In 2011, 2012, and 2014, along the exposed coast, the highest abundances were observed in September 2011 (up to 4.6 x 10⁶ cells L^-1, mean 7.0 x 105 cells L^-1), November 2012 (up to 2.8 x 10⁶, mean 7.0 x 10⁵ cells L^-1) and November 2014 (up to 4.1 x 10⁵, mean 2.1 x 105 cells L^-1), in particular at st. 1 to 8 and 14 to 16 (Fig. 5). The seasonal pattern of the cell abundance of D. yucatanensis was similar between 2011 and 2012 when the species was found between June and November 2011 and between May and November 2012 (Fig. 5). In comparison to those years, 2013 and 2014 were different. In 2013 the species was frequent between January and July, and in 2014 during a much shorter period, from August to October. In addition, the number of species observations was much fewer in 2014 than in previous years.

Table 3 Summary of the best adjusted generalized additive models and total explained deviation. 

Zones/Years	Minimal adjusted model	Explained deviation (%)
Exposed coast
2011	Cell abundance ~ Temperature*+ Salinity + Phosphates***	28.9
2012	Cell abundance ~ Salinity***+ Phosphates + Silicates*+ DIN*	23.0
2013	Cell abundance ~ Temperature + Salinity*+ Urea* + Chlorophyll-a***+ DIN***	60.7
2014	Cell abundance ~ Temperature + DO*+ Silicates* + Urea + Chlorophyll-a	49.8
Marinas
2011	Cell abundance ~ Temperature + Urea*** + Chlorophyll-a***	73.7
2012	Cell abundance ~ Temperature* + DO + Phosphates** + DIN	64.1
2013	Cell abundance ~ Phosphates*** + DIN*	74.1
2014	Cell abundance ~ Urea* + DIN***	90.8

Significance codes: <0.001***, 0.001**, 0.01*, 0.05*.

Figure 5 Spatio-temporal variation in the cell abundance of Durinskia yucatanensis along the northern Yucatan coast in 2011-2014. The size of the circles is proportional to the cell abundances expressed in x 10⁴ cells L^-1. The ordinate axis indicates the monthly scale, the abscissa axis indicates sampling stations (1 - Celestún, 2 - Chuburná, 3 - Chelem, 4 - Punta Bass, 5 - Progreso, 6 - Chicxulub 1, 7 - Chicxulub 2, 8 - Uaymitún, 9 - Telchac, 10 - San Crisanto 1, 11 - San Crisanto 2, 12 - Chabihau 1, 13 - Chabihau 2, 14 - Santa Clara 1, 15 - Santa Clara 2, 16 - Dzilam, 17 - San Felipe). Different segments of the graph indicate the years of sampling (2011-2014). 

As for the spatial distribution of the cell abundances of D. yucatanensis, the highest values were observed at st. 5, 6, and 7 (the Progreso-Chicxulub area) along the exposed coast in 2011, and only at st. 17 (Dzilam) during all studied years in marinas (Fig. 5). In marinas, the maximum cell abundances were observed in August 2012 and February 2014 (5.1 x 10⁶ and 3.0 x 10⁷ cells L^-1). In 2011, the maximum abundance occurred in July (2.2 x 10⁶, mean 6.4 x 10⁵ cells L^-1) and in May 2013 (4.9 x 10⁶, mean 8.2 x 10⁵ cells L^-1) (Fig. 6, Table 4). DIN showed a positive correlation with cell abundances for both the exposed coast and marinas, especially in July-August, characterized by high water temperature (31-32 °C) (Table 2, Fig. 6).

Figure 6 Annual mean cell abundances of Durinskia yucatanensis in relation to dissolved inorganic nitrogen (DIN) along the northern Yucatan coast in 2011-2014. The letters added to the years in the parentheses indicate exposed coasts (a) and marinas (b). 

DISCUSSION

The species was placed into the family Kryptoperidiniaceae (^{Okolodkov et al., 2020}) despite the absence of the second (eukaryotic) nucleus that originated from a tertiary endosymbiont derived from a diatom (^{Dodge, 1971}; ^{McEwan & Keeling, 2004}; for additional references, see ^{Satta et al., 2020} and ^{Tillmann et al., 2023}) and characteristic of the Kryptoperidiniaceae, and it was classified according to ^{Fensome et al. (1993)} and ^{Kretschmann et al. (2018)}. It is noteworthy that the isolate of Kryptoperidinium triquetrum (Ehrenberg) Tillmann, Gottschling, Elbrächter, Kusber et Hoppenrath (= Kryptoperidinium foliaceum Lindemann) that caused a bloom event in South Carolina, USA, was mononucleate (^{Kempton et al., 2002}); the authors suggested a more transient endosymbiotic association than previously considered. Similar observations of the species were made from the Baltic Sea (^{Gottschling et al., 2019}). This species has caused recurrent blooms in southern Chile: 18% of the 220 registered bloom events that occurred from 1956 to 2021 (^{Barría et al., 2022}). The Kryptoperidiniaceae blooms appear to be spreading along the Latin American coasts. ^{Yamada et al. (2019)} performed a comparative study of the nuclear dynamics in three Durinskia species. They found that D. capensis Pienaar, H. Sakai et T. Horiguchi in culture keeps its captured diatoms temporarily, only for two months. The authors also suggest the unique nuclear dynamics of converting kleptoplasts to permanent plastids in D. kwazulunatalensis. In all cases, the host dinoflagellates control the diatom karyokinesis.

The highest silicate concentrations (96.46-97.22 µmol L^-1) found in Sisal marina and at 50 m from the coast in 2010 and 2011 are most likely derived from terrestrial environments. In this case, chl-a reflects the sum of the total phytoplankton biomass (mainly diatoms), extracellular chlorophyll, and undigested microalgae in fecal pellets of planktonic invertebrates (mainly crustaceans), rather than the biomass of the blooming species.

The GAM model analysis showed that the temperature (P=0.01) and salinity (P=0.01) could help to explain the changes in cell abundance (up to 60.7% of the total explained deviation), in particular, along the exposed coast (in 2011, 2013 and 2014) and in marinas (in 2011 and 2012) compared to changes in DIN and urea (P<0.001; Table 4). Similarly, models of the field data obtained from two Mediterranean coastal shallow lagoons in Sardinia, Italy, revealed a significant role of rainfall and DIN in the occurrence and blooms of Kryptoperidinium sp. (^{Satta et al., 2020}). Cell abundances of D. yucatanensis were directly proportional to an increase in temperature, especially when above 30 °C, particularly during 2013 along the exposed coast and 2012 in marinas (Fig. 7: 2013a and 2012b). The GAM model applied to the exposed coast in 2012 and 2013 showed the capacity of adaptation of the species to a wide range of salinities (Fig. 7: 2012a and 2013a). In 2013, in marinas, the increase in cell abundance was significantly related to phosphates (P<0.001; Table 4). Similarly, the major urea and DIN concentrations resulted in predicted increases in cell abundance, particularly in 2014 (Fig. 5, 6 and 7).

Figure 7 Generalized additive models (GAM) describing the cell abundance response of Durinskia yucatanensis to physicochemical variables: salinity, temperature (°C), urea and DIN (μM), during different years (2011-2014) along the northern Yucatan coast (a - exposed coasts, b - marinas). Bold lines represent GAM smoothed mean ratios, and shaded areas are 95% confidence intervals. 

CONCLUSIONS

Durinskia yucatanensis is a marine planktonic dinoflagellate. It differs from other known Durinskia species by a larger 1a plate compared to plate 2a. As for the cell size and thecal morphology, it is closer to D. kwazulunatalensis. It prefers semi-enclosed coastal environments characterized by more stagnant water and increased concentrations of macronutrients. Its cell abundance spatio-temporal distribution pattern along the northern Yucatan coast may change over the years; however, it prefers higher summer temperatures during the rainy season and is also common in autumn before the northerly wind season.