Intercellular and extracellular amino acids of different bloom species in the Mediterranean Sea

Extracellular algal production is considered a significant part of the algal primary production (Myklestad 2000). Despite the fact that the pool of algal dissolved free amino acids (DFAAs) in seawater represents a small fraction of dissolved organic matter (DOM) (Thornton 2014), its exploration is critical for understanding primary producers’ nitrogen demand during periods of insufficient dissolved inorganic nitrogen concentrations (Tyler et al. 2003), as well as the growth and other biological processes of living organisms (Vidoudez & Pohnert 2012) affecting the global carbon cycle (Kujawinski 2011). The release of NH₄⁺ through oxidation of AAs provides a competitive advantage to algae species capable of uptake (Tyler et al. 2005; Thornton 2014). While a wide range of algal taxa can release and consume amounts of DFAAs (Hounshell et al. 2017), the extent to which algae can use DFAAs as a source of nitrogen is still unclear (Flynn et al. 2008; Li et al. 2009). Furthermore, the quantification of AAs production in nature is difficult due to the interference of several specific factors that modulate their concentration such as a more ‘leaky’ cell membrane (Veldhuis et al. 2001), direct cell contact that can cause interference (Vardi et al. 2002), viral infection (Bettarel et al. 2005), feeding by predators (Møller 2007), physiological status (Flynn et al. 2008), cell death (Orellana et al. 2013), short residence time and rapid turnover, growth curve (Sarmento et al. 2013), and environmental conditions (Grosse 2017).

The present study attempts to characterize the autochthonous algal AAs exudation to define its role as an alternative nutrient source that can contribute to the development of harmful algal blooms (HABs) in the environment. The experiment was designed to simulate algal bloom scenarios to characterize the extracellular AA production of different algal species from different classes during the active growth phase, i.e. the exponential phase. A set of laboratory experiments was established for: i) monocultures of both S. costatum Cleve (centric diatom) and S. trochoidea (Stein) Loeblich (thecate dinoflagellate), ii) healthy thalli of U. fasciata Delile (Chlorophyta) and C. officinalis Linnaeus (Rhodophyta), and iii) co-cultures of the tested phytoplankton species with different weights of macroalgae. This is the first research work to date in Alexandria waters aimed at expanding our understanding of bloom dynamics as a necessary step toward predicting and mitigating HABs. The selected phytoplankton species repeatedly formed blooms in Alexandria waters, for example S. costatum blooms were recorded there three times in 2000 (Mikhail & Labib 2014) and elsewhere (Horner et al. 2005), while S. trochoidea blooms were recorded in 1998 and 2014 (Labib 2002; El Shafay et al. 2019). Skeletonema costatum and Scrippsiella trochoidea can form harmful blooms. The harmful effects of S. costatum blooms include their large biomass, which causes various ecological disturbances such as lack of oxygen, i.e. hypoxia (Zhang et al. 2016), and water discoloration (Ibrahim et al. 2021). Moreover, the adverse effects include production of reactive aldehydes that negatively affect hatching copepods (Chen et al. 2021), as well as arsenite and dimethyl arsenic, which can block important biochemical pathways in algae, as arsenic is chemically similar to nitrogen and phosphorous (Howard et al. 1995). The first blooms of Scrippsiella trochoidea, documented in 1890 and 1913 under the species name Glenodinium trochoideum, were associated with fish kills in Sydney Harbour (Whitelegge 1891). Under unique conditions in sheltered bays, blooms become very dense and cause death of fish and invertebrates from the depletion of oxygen (Hallegraeff 2002). A red tide in China, Japan, Korea, and Russia, observed by Sakamoto et al. (2021), killed fish and caused serious economic losses. The studied macroalgae occasionally form green tide mats (Labib et al. 2015), and co-occurrence of micro-macroalgal blooms is a common phenomenon (Hosny 2016).

Water samples and macroalgae species were collected during summer 2018 from the Qaitbey area located in the western vicinity of the Port of Alexandria, Egypt. The area has been identified as eutrophic with recurrent annual algal blooms (Labib 2002).

After isolation, cells of S. costatum and S. trochoidea were maintained in stock cultures in the f/2 medium prepared with aged and filtered offshore autoclaved seawater (Guillard & Ryther 1962) at 25°C, salinity 32–33 PSU, pH 8–8.2, under continuous light illumination of ~100 μE m^-2 s^-1, and bubbled with sterile filtered air. Derrien et al. (1998) recommended the use of continuous light illumination to ensure faster growth of S. trochoidea rather than the light/dark cycle. The growth was followed starting from the second day of incubation, and the daily cell density at approximately the same time was measured using a haemocytometer. Subsamples of each previously gently shaken cell suspension reaching ~1000 cells ml^-1 for S. costatum and 300 cells ml^-1 for S. trochoidea, which almost resemble their initial concentrations during early bloom stages recorded in the same area (Mikhail and Labib 2014), were distributed into polystyrene bottles with 150 ml of filtered seawater enriched with f/2 solution and cultured under the previously mentioned conditions. Exponential and stationary growth phases were identified daily by direct observation under a microscope (Kester et al. 1967). Microalgae samples of 3 ml were preserved with acidic Lugol’s iodine; then samples were settled overnight, and counts were performed under a microscope (160 ×) in random fields using a haemocytometer to determine the phytoplanktonic algal growth (Utermöhl 1958). The early stage of the exponential growth phase was detected on day 2 in S. costatum cultures, and analyses were performed during four consecutive days. On the other hand, it took a relatively long time in gently shaken S. trochoidea cultures, which appear to require more time due to their slower growth rate (Dixon & Syrett 1988). The cell-free culture medium of both species was collected during the exponential phase after filtration through 25 mm Whatman GF/C glass fiber filters under gentle vacuum (50 mm Hg) and stored in acid-washed vials at –20°C for a maximum of 7 days (Granum et al. 2002) until analysis of extracellular amino acids. Cell division per day of both species was calculated Guillard (1973).

Fresh, adult U. fasciata and C. officinalis thalli were carefully washed in the field with seawater to dislodge attached biota and other extraneous matter and transported moist to the laboratory within 30 min. Morphological identification of species followed the checklist of Aleem (1993). In the laboratory, samples were gently brushed, rinsed with distilled water, and dried with absorbent paper at room temperature. Prior to experiments, each macroalgal species (10–15 g fresh weight) was grown for 10 h in 1 l of fresh seawater under adjusted temperature (14–27°C) and salinity (26–32 PSU), with optima at 23°C and 32 PSU (Xiao et al. 2016), and light regime, and cultures were bubbled with air at a rate of 2 l min^-1, followed by control cultures for each species in F/2 medium. The progressive growth during the first week was determined by biomass estimation (Li et al. 2016). Algae during the growth phase were removed from the medium by filtration, and the filtrate was kept frozen until amino acids were analyzed.

Microalgae and macroalgae co-culture experiments consisted in adding macroalgae of different weights (0.5 g, 1 g, and 2 g) to stock phytoplankton cultures obtained in the exponential phase, which were grown under the same previously mentioned conditions. After filtration of samples between days 3 and 5, the supernatant of the filtrated media with free amino acids was analyzed. The role of bacteria cannot be excluded as their existence reflects the natural environmental conditions. Dissolved free amino acids (DFAA) were analyzed in the Central Laboratory at the National Institute of Oceanography and Fisheries using High-Performance Liquid Chromatography (HPLC) according to a method published in Agilent Application Note (Henderson et al. 2000). Briefly, DFAA samples and amino acid standards solutions were derivatized with o-phthalaldehyde (OPA) and analyzed by RP-HPLC (Agilent 1260 Infinity II chromatographic system) with ultraviolet-visible detection (diode array detector, DAD). After derivatization, an amount equivalent to 2.5 μl of each sample was injected onto a Zorbax SB-C18 (5 μm, 150 × 4.6 mm, Agilent) at 40°C, and detection was performed at λ = 338 nm. Mobile phase A was 40 mM NaH₂PO₄, while mobile phase B was acetonitrile/methanol/water (45/45/10 v/v/v). The separation was achieved at a flow rate of 2 ml/min using a gradient program that allowed 1.9 min at 0% B, followed by an increase in eluent B to 57% at 18.1 min, and then reaching 100% B at 18.6 to 22.3 min. Finally, the eluent B level dropped to 0% between 23.2 to 26 min. The concentrations of each DFAA in the samples were calculated from peak areas on chromatograms using response factors obtained from standards of amino acids (analytical grade, Sigma Aldrich, Louis Mo, USA), 17 amino acids in 0.1M hydrochloric acid, 2.5 mM each: alanine, arginine, aspartic acid, glutamic acid, glycine, histidine, isoleucine, leucine, lysine, methionine, phenylalanine, proline, serine, threonine, tyrosine, valine, and 1.25 mM cystine.

During the active (exponential) growth phase, the AA analysis is an attempt to resemble an algal bloom situation, where such information is still scarce. Determining the specific growth phase where the largest production of AAs occurs is still a matter of debate and the mechanisms of AAs excretion by algae are still largely unknown (Franklin et al. 2006). The exponential phase is the fastest growing phase among others (Bruckner et al. 2011; Orellana et al. 2013; Sarmento et al. 2013). S. costatum grew faster, exhibited a shorter exponential phase, and showed higher production of the dominant AAs compared to S. trochoidea, estimated at 4.25-fold for alanine and 3.06-fold for each lysine and threonine. The composition of AAs was uniform, more or less stable, but tyrosine acid was found exclusively in S. costatum, accounting for about one-fourth of total AAs (Martin-Jezequel et al. 1988). As previously reported (Vidoudez and Pohnert 2012; Sarmento et al. 2013), the variation in the amount and proportion of the released AAs is mainly determined by the species age and culture conditions. However, the AA relative frequency sequence differs from that reported for S. costatum by Martin-Jezequel et al. (1988), where glycine, aspartic, and glutamine acids dominate, respectively. However, these results are consistent with Mannino and Harvey (2000), indicating the significant contribution of alanine and threonine acids and lesser importance of aspartic and glutamic acids. Our data indicate a significant contribution of threonine acid in Skeletonema, which contradicts the data reported by Sarmento et al. (2013), similar to that for arginine acid in S. trochoidea (Meksumpun et al. 1994). Since arginine acid is a precursor to saxitoxin and gonyautoxin acids (Anderson & Harrison 1988), considering its rapid, lethal effects on shellfish and larvae (Tang & Gobler 2012), the current changes in the composition may offer an advantage in preventing dinoflagellate red tide outbreaks.

The study shows more diverse AA composition in macroalgae, homogeneity, predominance of aspartic, glutamine, alanine, and leucine acids, and limited changes in concentrations, slightly higher in C. officinalis (7.64%). The dominance of aspartic and glutamic acids was proven in Alexandria waters for the same examined macroalgae (Mustafa and Eladel 2015; Ismail 2017), previously documented elsewhere (Akgül et al. 2015).

The main difference that distinguishes this work from others was the determination of AA production in multi-taxa cultures of phytoplankton and macroalgae species, where, to our knowledge, no experimental or field comparative studies are available for this combination of taxonomic groups. This study revealed considerable variation in the content, composition, and relative contribution of the released AAs compared with cultures of individual taxa. The substantial changes were similar to those reported for Ulva intestinalis by Lourenco et al. (2004). The sharp reduction in the number of AAs compared with monocultures of each species during their exponential phase (control) was estimated at 43.75% for U. fasciata, 25% for S. costatum, and 47.06% for S. trochiodea. A positive trend was observed between the weight increase in Ulva and C. officinalis in co-cultures with S. costatum, while it was less pronounced in co-cultures of Ulva with S. trochoidea, which may reflect some resistance. Kolmakovaa & Kolmakov (2019) reported significantly higher AA content in macroalgae compared to phytoplankton species. The co-culture experiments lasted only five days in an attempt to reduce and/or avoid potential causes responsible for the observed variations, which often remain elusive. Another cause may be the allelopathic effect of U. fasciata on the growth of S. costatum and S. trochiodea (El-Sheekh et al. 2010; EL Shafay et al. 2019), as U. fasciata is characterized by strong and quick allelopathic activity against these two potentially harmful species, and the growth inhibition occurred after six days of co-incubation. Since the algal cultures used were not axenic, the variation in AA production was probably modified by co-occurring prokaryotes, which is consistent with Bruckner et al. (2011) for diatom and bacteria cultures. The differential uptake and selectivity of amino acids proved similar among macroalgae and phytoplankton (Tyler et al. 2003), which may also be a significant cause (Kolmakovaa & Kolmakov 2019). Furthermore, the expected decrease in inorganic nitrogen content in the medium with no additional supply may also be responsible for the observed variation (Granum et al. 2002). Thus, such variation is mainly a direct result of the balance between uptake and release of AAs, however, the mechanisms affecting the release of dissolved AAs in such multispecific cultures need further studies.

Since the coastal waters of Alexandria and elsewhere are affected by recurrent massive monospecific and/or multispecific blooms of phytoplankton and macroalgae, it seems likely that the release of AAs in the water column and the production of ammonia may contribute to the existence and long duration of these recurrent blooms. Alanine, lysine, and threonine acids released in monocultures of Skeletonema costatum and/or Scrippsiella trochoidea may contribute to the formation of mucilaginous phytoplankton blooms, with negative ecological consequences. Therefore, research on bloom dynamics is very important for coastal zone management. During the exponential phase in the co-culture of Skeletonema costatum and Scrippsiella trochoidea with Ulva fasciata and Corallina officinalis, a sharp reduction in the number of amino acids and inhibition growth of microalgae occurred, while weights of macroalgae increased, which implies that macroalgae are characterized by a strong and rapid allelopathic activity against these two potentially harmful species. Thus, these data can shed light on bloom dynamics, consequently leading to better implementation of optimal bloom prevention and control measures. Our findings may potentially have far-reaching ecological implications.