Tetrapeptide from microbial coculture exhibiting a unique algicidal mechanism against Alexandrium fundyense

The outbreak of harmful algal blooms (HABs) is exacerbated by ocean eutrophication and climate change, driven by human activities. These outbreaks have shown a global poleward shift¹, causing severe damage to aquaculture and posing significant risks to marine ecosystems and human health^1,2. HABs produce toxins and deplete oxygen, leading to high mortality rates among marine organisms. Aerosolized HABs toxins can also harm human health, such as by reducing lung function². Paralytic shellfish toxins produced by Alexandrium species accumulate in shellfish tissues, with detoxification periods ranging from days to over a year, emphasizing the urgent need for efficient HABs management strategies³.

While chemical and physical treatments, including copper sulfate, hydrogen peroxide, and clay spraying, are effective for HABs control, they are often non-selective, environmentally harmful, and costly^4,5. These limitations have spurred interest in natural algicidal compounds as selective and eco-friendly alternatives. Bioactive peptides, in particular, have emerged as promising candidates due to their environmental safety, biodegradability, low cost, and rapid algicidal effects^6,7,8. These peptides can be sourced from marine organisms through immune system production, chemical hydrolysis of by-products, or co-culture with microorganisms^9,10. However, challenges in isolation and limited research have resulted in only a small number of bioactive peptides being identified for effective HABs control¹¹.

Existing natural algicidal peptides primarily target algal cell membranes, exploiting hydrophobic interactions for membrane disruption^{6,12,13,14,15,16}. However, this approach is less effective against thecate dinoflagellates, such as Alexandrium species, which possess rigid cell walls that obscure exposed cell membranes^17,18. This gap underscores the need for studies exploring alternative mechanisms of action for controlling thecate harmful algae.

This study focuses on understanding the mode of action and toxicity of the tetrapeptide F32-1, previously isolated from a mackerel waste-microorganism co-culture, which has demonstrated algicidal activity against Alexandrium species. Unlike most known peptides that rely on membrane disruption, this research investigates whether F32-1 employs a distinct mechanism. Biochemical and ultrastructural analyses were conducted to reveal the cellular changes induced by F32-1, while modifications to specific amino acids were made to evaluate the influence of physicochemical properties on its activity. Additionally, the toxicity of F32-1 was assessed to determine its environmental safety and potential applicability for sustainable HABs management.

Algal cultivation

Samples from the harmful alga Alexandrium fundyense (LIMS-PS-2657), obtained from the Library of Marine Samples in the Korea Institute of Ocean Science and Technology (Geoje, Korea), were used. A. fundyense was cultivated using a modified f/2-si medium (Sigma-Aldrich, St. Louis, MO, USA) with a salinity of 30 practical salinity units. The harmful algae were then cultivated at 20 ± 2 °C under 150 μmol/m²/s of light intensity with a 14:10 h light:dark cycle. The cultivated algal cell density was maintained at an exponential growth phase using a hemocytometer (Marienfeld, Lauda-Königshofen, Germany) and a Sedgewick chamber (Marienfeld) under the light microscope.

Algal cell viability assay of F32-1

A linear tetrapeptide F32-1 (NH-KMNF-COOH) possessing an algicidal effect was used in this experiment. The F32-1 was obtained from biodegraded mackerel waste using identical separation methods described in previous studies^19,20. The tetrapeptide was designated as F32-1 because it was isolated from the reverse-phase chromatography fraction 32-1 that showed algicidal activity¹⁹. The mackerel waste was biodegraded with a mixed microbial culture, followed by separation using size-exclusion and reversed-phase chromatography^19,20. The sequence of the isolated F32-1 was confirmed using liquid chromatography-electrospray ionization-tandem mass spectrometry, following the methods described in previous studies¹⁹.

To determine how algal cell viability was affected by the F32-1, a fluorescein diacetate (FDA; Alfa Aesar, Ward Hill, MA, USA) was used as a viability assay reagent, as it is a nonpolar compound that can penetrate cells. Once the FDA entered the cells, it was cleaved into acetate and fluorescein by esterases retained within the cells. The F32-1 (1.4 mg/mL) was added to 24-well plates (SPL, Seoul, Korea) that contained the exponential-growth-phase algal cells, and the plate was then incubated for 0, 0.5, 2, and 3 h under equal growth conditions. The fresh culture medium was used as a control. The algal cells were collected after centrifugation at 377×g for 10 min and rinsed twice with filtered seawater. The supernatant was then discarded. Afterward, the algal cells were stained with 2 μM FDA, incubated for 20 min in a dark room, and washed twice with filtered seawater.

The image of the stained algal cells was observed with a confocal laser scanning microscope (CLSM; LSM 700, Zeiss, Oberkochen, Germany). The stained algal cells were mounted with a ProLong Diamond Antifade Mountant (Invitrogen, Carlsbad, CA, USA) on a precleaned slide with a cover glass (Marienfeld, 18 × 18 mm). The slides were observed under the CLSM. Fluorescein isothiocyanate (FITC) was used as a laser filter, while the excitation and emission wavelengths were 488 nm and 530 nm, respectively. As auto-fluorescence occurs at wavelengths above 560 nm, signals beyond this range were excluded to minimize background interference. Images were processed with the Zen 2009 software (Zeiss, https://www.zeiss.com/microscopy/en/products/software/zeiss-zen.html, accessed 19 Oct 2025) and ImageJ v1.54i (National Institute of Mental Health, Bethesda, USA, https://imagej.net/ij/index.html, accessed 19 Oct 2025).

Generation of intracellular reactive oxygen species in damaged algal cells

To examine the algicidal mechanism, intracellular reactive oxygen species (ROS) generation was observed in algal cells treated with F32-1. To analyze the intracellular generation of ROS, dihydrorhodamine 123 (Sigma-Aldrich) was used²¹. The treated algal cells were washed twice with a phosphate buffer (pH 8.0) and stained using dihydrorhodamine 123 (final concentration, 15 μM) for 1 h in a dark room. The stained algal cell pellet was washed once with the phosphate buffer (pH 8.0). The intracellular generation of ROS was observed from the mounted algal cells on a precleaned slide under the CLSM using a FITC filter at 488/530 nm (excitation/emission wavelength).

Ultrastructural changes in damaged algal cells

A transmission electron microscope (TEM) was used to observe the change in the intracellular structure of the damaged algal cells when the algal cells were exposed to F32-1. The algal cells exposed to F32-1 for 3 h were centrifuged at 377×g for 10 min. After the supernatant was discarded, the algal cells were fixed with 2.5% glutaraldehyde dissolved in a phosphate-buffered saline (PBS) solution at 4 °C overnight. The pellets were post-fixed with 1% osmium tetroxide in a 0.01 M phosphate buffer containing 16.730 g/L of potassium phosphate dibasic and 0.523 g/L of potassium phosphate monobasic (pH 8.0) at 4 °C for 2 h and then washed at 4 °C using the 0.01 M PBS buffer. The washed cells were dehydrated for 10 min using graded ethanol solutions (50, 60, 70, 90, and 100%); dehydration was terminated after treatment with 100% acetone for 10 min. The collected algal cells were embedded using Epon 812 (Sigma-Aldrich), whereas the embedded block was baked at 60 °C for 2 days. The embedded block was trimmed and thinly sectioned in the range of 60-80 nm using an ultramicrotome (EM UC6, Leica, Wetzlar, Germany). The thinly sectioned parts were placed on copper grids and stained with a TI blue kit (Nisshin EM, Shinjukuku, Tokyo, Japan) and Reynolds lead citrate for 5 min. Lastly, the ultrastructure of the algal cell was observed using H-7500 (HITACHI, Chiyodaku, Tokyo, Japan).

Membrane penetration and intracellular localization of the tetrapeptide

To observe its membrane penetration and intracellular localization, F32-1 was conjugated to 5-carboxyfluorescein (FAM). The FAM-conjugated peptide (FAM-F32-1) was synthesized at Peptron (Daejeon, Korea) via a solid-phase peptide synthesis using ASP48S (Peptron). Trt-Cl resin (GL Biochem, Shanghai, China) was used to attach the C-terminal residue. Fmoc-L-amino acids (> 99.9%, GL Biochem) were coupled with O-benzotriazole-N,N,N′,N′-tetramethyl-uronium-hexafluorophosphate (99%, GL Biochem) and N,N′-diisopropylcarbodiimide (98%, Sigma-Aldrich), and were then dissolved in N,N-Dimethylformamide (99.5%, Daejung, Siheung, Korea) and 5(6)-FAM (90%, Ana-Spec, Fremont, CA, USA). A mixture of 87.5% (v/v) trifluoroacetic acid (99.5%, Alfa Aesar, Ward Hill, MA, USA), 5% (v/v) distilled water, 5% (v/v) thioanisole (99%, Alfa Aesar), and 2.5% (v/v) 1,2-ethandithiol (98%, Alfa Aesar) was used to attach, deprotect, and isolate the peptides from the resin at 25 °C for 3 h. Afterward, the crude peptide was washed with diethyl ether (99%, Daejung), dried under vacuum, and purified via reverse-phase high-performance liquid chromatography (Shimadzu Prominence HPLC, Nakagyoku, Kyoto, Japan) using a Vydac Everest C18 column (250 mm × 22 mm, 10 μm, USA). Elution was performed with a linear gradient of water and acetonitrile from 10 to 75% (v/v) of acetonitrile (99.9%, Tedia, Fairfield, OH, USA) containing 0.1% (v/v) trifluoroacetic acid. The conjugation of FAM with F32-1 was confirmed by measuring the molecular weight difference using liquid chromatography–mass spectrometry (Shimadzu LCMS-2020). Lastly, a powder-type FAM-F32-1 was obtained from the lyophilization using FDT-12012 (Operon, Kimpo, Korea).

FAM-F32-1 was dissolved in dimethyl sulfoxide (DMSO; Junsei, Chuoku, Tokyo, Japan) and diluted up to 1.41 mg/mL with filtered seawater. Subsequently, the prepared FAM-F32-1 was exposed to harmful algal cells for different time windows (0.25, 0.5, and 1 h).

The FAM-F32-1-treated algal cells were washed twice with filtered seawater, and a droplet of the washed algae was placed on a precleaned slide and mounted with ProLong Diamond Antifade Mountant (Invitrogen). The slides were observed under CLSM using a FAM laser filter (with an emission at 488 nm). The fluorescence emission of FAM-F32-1 was detected at a wavelength of 510 nm, with the removal of the emission wavelength over 560 nm. The images were processed with the Zen 2009 (Zeiss) software and ImageJ program.

Modification of physicochemical properties of tetrapeptide

To compare the mode of action of the F32-1 with other algicidal peptides reported in the literature, the physicochemical properties (hydrophobicity and structure) of the F32-1 were modified using sequence-based computational tools designed for antimicrobial peptides. Among various sequence-based computational tools, iAMPpred (http://cabgrid.res.in:8080/amppred/index.html)²¹ was selected as one of the most advanced tools for this study. Using this tool, specific amino acids in F32-1 were substituted with alternative residues. Sequence-based computational tools play a pivotal role in designing potential antimicrobial peptides²². To maintain the antimicrobial activity of F32-1, amino acid substitutions were performed based on supervised machine learning algorithms (antimicrobial activity probability > 0.5). The modified F32-1, with altered physicochemical properties, was synthesized by Peptron, and its algicidal activity was subsequently evaluated.

Algicidal activity assay

To analyze the algicidal activity of the modified F32-1s, they were exposed to the harmful algae A. fundyense (LIMS-PS-2657). One milligram per liter of the modified F32-1s was dissolved in f/2-si medium and placed on 24-well culture plates that contained the exponential-growth harmful algal cells. After 0.5, 2, 3, and 5 h treatment under the algal cultivation conditions mentioned above, the algal cells were observed and counted using a hemocytometer and Sedgewick chamber. Each experiment was conducted using 24-well tissue culture test plates with 1 mL of algal cells per well. After the viable algal cells were counted for each sample, the algicidal rate was calculated based on the following equation:

where Nc and Nt represent the number of algal cells in the control and treatment groups, respectively.

Toxicity test of algicidal peptide

The toxicity assessment of F32-1 was conducted with a few modifications to ISO guideline 10253 (ISO, 2016). Skeletonema costatum (LIMS-PS-0848) and Phaeodactylum tricornutum (LIMS-PS-0962) were obtained from the Library of Marine Samples for use in the study. The algae were cultured in ISO 10253 (2016) medium prepared with filtered natural seawater obtained from Goseong Deep Sea Water Industry Foundation (Goseong, Korea) under the conditions of 20 ± 2 °C and 3000 lx with a 12 h:12 h light:dark cycle. Algal cultures at a concentration of 10⁴ cells/mL were inoculated into vials and exposed to peptides at five concentrations (0.23, 0.35, 0.7, 1.4, 2.1, and 2.8 mg/mL) in medium. The control group was treated with an equal amount of medium, and algal cells were observed and counted using a hemocytometer and Sedgewick chamber after 24, 48, 72, and 96 h. The experiment was conducted in triplicate to obtain confidence intervals.

Statistical analysis

For multiple treatment group comparisons (Fig. 5), Tukey’s HSD test was employed, and distinct significance groups (a, b) were assigned based on the corrected p-values. To analyze the differences between the treatment groups and the control, Dunnett’s test was conducted after verifying the assumptions of normality (Shapiro-Wilk test) and homoscedasticity (Levene’s test). The significance level was set at α = 0.05. Bonferroni correction was applied to control for multiple comparisons. No Observed Effect Concentration (NOEC) was the highest concentration with no statistically significant difference from the control, and Lowest Observed Effect Concentration (LOEC) was the lowest concentration with a statistically significant difference. Statistical analyses were performed using Stata 15 (Stata Corp, College Station, TX, USA), and the data are presented as mean ± standard deviation.

Viability in the treated algal cells

The viability of algal cells treated with F32-1 on A. fundyense at different time points was observed under CLSM, as shown in Fig. 1 and Supplementary Fig. 1. Intracellular esterase activity was used as an indicator to assess the algicidal effects of F32-1²³. Esterase is a critical enzyme involved in cellular metabolism, and its activity reflects the metabolic integrity of the cell²⁴. A decline in esterase activity indicates compromised cellular function, leading to eventual cell death²⁴. This approach is a sensitive and reliable method to measure of the efficacy of F32-1 as an algicidal agent²⁴.

Previous studies showed that F32-1 effectively targets thecate harmful algae, particularly A. fundyense (Lethal Concentration, LC₅₀ = 0.69 mg/mL, LC₉₀ = 1.41 mg/mL)¹⁹. In this study, esterase activity began to decline significantly after 2 h of treatment, with complete loss of activity by 3 h (Fig. 1C,D). In contrast, control cells retained high esterase activity throughout (Fig. 1A). The loss of motility in treated cells immediately after exposure further underscores the peptide’s rapid effect, with full cell death confirmed by 5 h post-treatment (Supplementary Fig. 1). These findings indicate that F32-1 exerts its algicidal effects efficiently within 3 h, highlighting its potential for rapid mitigation in real-world scenarios.

The rapid action of F32-1 is particularly advantageous when compared to other biologically active agents, which typically require 24 h to exhibit algicidal effects. For instance, Pseudoruegeria sp. M32A2M disrupts Alexandrium catenella cell membranes²⁵, Mangrovimonas yunxiaonensis LY01 induces algal cell disruption in Alexandrium tamarense²⁶, and Shewanella Y1 demonstrates activity against Alexandrium pacificum²⁷. The faster action of F32-1 offers a distinct advantage for immediate response strategies in HAB management. The rapid action of F32-1 aligns with the common characteristic of algicidal peptides, which are known to exhibit swift effects on harmful algae. This feature makes algicidal peptides including F32-1 particularly suitable for scenarios where prompt algal bloom suppression is critical to reducing environmental and economic damage.

Generation of intracellular ROS in treated algal cells

Algal cells treated with F32-1 exhibited a reduction in esterase activity, but no membrane disruption or collapse was observed during the extended period (up to 8 h). The reduction in esterase activity by F32-1 without observable membrane disruption suggested an intracellular mode of action. To investigate this, intracellular ROS levels were monitored, as ROS are critical mediators of stress-induced cell death and play pivotal roles in apoptosis and necrosis pathways^28,29.

Excessive ROS production was observed in treated cells within 3 h of treatment, with fluorescence indicating ROS generation in the cytoplasm and multiple organelles, including mitochondria and chloroplasts (Fig. 2B). Control cells exhibited negligible ROS production (Fig. 2A). ROS accumulation disrupts intracellular homeostasis and triggers oxidative stress, leading to cell death. After 24 h of F32-1 treatment, cell destruction was observed under optical microscopy, characterized by significant morphological changes, including cell shrinkage and fragmentation. These findings confirm the progressive damage caused by ROS accumulation, which initiates oxidative stress and culminates in cell lysis.

This ROS-mediated mechanism aligns with that of malformin C, a cyclic peptide derived from Aspergillus niger, which is known to induce intracellular ROS accumulation and algal cell death¹⁸. However, it differs significantly from the mechanisms of algicidal compounds such as HPA3NT3³⁰, hybrid peptide Hn-Mc¹⁷, cyclic lipopeptide surfactins from Bacillus tequilensis strain D8¹⁴, and Mastoparan B³¹, which primarily target athecate algae like Heterosigma akashiwo by relying on hydrophobic interactions to disrupt the exposed plasma membrane.

Despite the limited number of known algicidal peptides, which exhibit mechanisms such as membrane disruption, or intracellular ROS generation, F32-1 provides a distinct example of a peptide specifically utilizing ROS-mediated pathways to overcome the barriers posed by thecate algae. The rarity of such findings in peptides likely stems from challenges in isolating and characterizing bioactive compounds with precise intracellular targets.

Ultrastructural changes in treated algal cells

Based on the observed ROS accumulation, ultrastructural changes in algal cells treated with F32-1 for 3 h were examined using TEM (Fig. 3). Ultrastructural changes provide sensitive and specific phenotypic insights into cell death, often complementing molecular and biochemical analyses^32,33. Unlike molecular indicators such as DNA fragmentation or biochemical changes like loss of membrane asymmetry^{29,34,35,36,37,38,39,40}, ultrastructural observations directly reveal the physical degradation of cellular components.

In F32-1-treated cells, degradation of intracellular organelles, particularly chloroplasts, was evident after 3 h of treatment (Fig. 3D-F). TEM images clearly showed swollen and damaged chloroplasts, consistent with the oxidative stress induced by ROS. Mitochondria and other organelles also exhibited structural degradation. Notably, despite the extensive damage to intracellular organelles, the cell membrane remained intact throughout the observation period, indicating that F32-1 triggers intracellular degradation without disrupting the plasma membrane. In contrast, control cells maintained intact organelles and cytoplasmic structures, with no observable ultrastructural changes (Fig. 3A-C). These findings indicate that F32-1-induced ROS accumulation triggers intracellular degradation pathways leading to cell death.

ROS are well-documented for playing a critical role in various cell death pathways, including necrosis, by disrupting cellular homeostasis. However, in this study, TEM images indicated a distinct pathway resembling autophagy rather than necrosis. Unlike necrosis, which often involves cell envelope rupture and uncontrolled cytoplasmic leakage, F32-1-treated cells showed intact cell membranes, with selective degradation of intracellular organelles. Similarly, apoptosis, typically characterized by highly organized processes such as cell shrinkage, budding, chromatin condensation and apoptotic body formation, was not consistent with the observed ultrastructural features. These findings suggest that F32-1-induced cell death involves a unique intracellular mechanism distinct from classical necrosis or apoptosis. Unlike previously studied peptides, which appear to induce cell death primarily through membrane disruption, F32-1’s action resembles autophagy-like processes. The presence of rigid thecal plates in algae such as A. fundyense, along with limited studies on intracellular death pathways and oxidative stress-related enzymes such as SODs and GSTs^41,42, complicate efforts to understand algal cell death. Notably, enzymatic activity could not be reliably detected, presumably due to limited probe permeability through the theca and methodological constraints associated with current assay systems⁴¹. Additional investigation will be necessary to clarify the molecular mechanism of F32-1’s action and determine whether autophagy-like processes are predominantly involved.

Penetration of the algicidal peptide

F32-1’s algicidal effects, characterized by intracellular ROS generation and organelle degradation without membrane disruption, suggested a unique mode of action. To confirm whether these effects result from peptide penetration into algal cells, a fluorescein-conjugated form of F32-1 (FAM-F32-1) was synthesized and analyzed using CLSM at a concentration equivalent to LC₉₀ (Fig. 4).

CLSM revealed no fluorescence in untreated control cells (Fig. 4A). In treated cells, FAM-F32-1 fluorescence was detected near the cell envelope within 15 min, indicating initial peptide-cell interaction (Fig. 4B). By 30 min, fluorescence was more broadly distributed around the periphery (Fig. 4C), and at 1 h, the peptide was localized throughout the cytoplasm and intracellular organelles (Fig. 4D). These results confirm that F32-1 penetrates A. fundyense cells, enabling its intracellular action and initiating algicidal effects through ROS production and organelle breakdown.

Cell-penetrating peptides (CPPs), characterized by their cationic, hydrophobic, or amphiphilic properties, facilitate intracellular delivery via non-covalent interactions with the cell membrane⁴³. While many CPPs act by disrupting membranes, F32-1 represents a rare short peptide with a non-membrane-disruptive penetration mechanism⁴⁴. The phenylalanine residue at F32-1’s C-terminus enhances hydrophobic interactions with the lipid bilayer, while lysine residues enable binding to negatively charged cell surfaces, facilitating penetration^44,45. The short peptide identified from mackerel hydrolysate represents the first discovery of cell-penetrating peptides capable of penetrating algal cells.

Algicidal activity of modified F32-1s

To assess whether F32-1 could exert its algicidal effect through a mechanism similar to that of known membrane-disrupting algicidal peptides, its physicochemical properties were further optimized to resemble those of peptides, which act through electrostatic and hydrophobic interactions. Antimicrobial peptides, including those with algicidal activity, exert their effects via non-covalent interactions with negatively charged microbial and algal cell membranes, driven by factors such as amino acid sequence, net charge, hydrophobicity, and structural features (e.g., helical or stapled structures). These attributes are critical in determining peptide activity, influencing both the interactions with and disruption of target membranes. To predict antimicrobial potential and guide peptide design, several machine learning programs have been developed based on these characteristics²². Following this approach, F32-1 was modified using the sequence-based computational tool iAMPpred^21,22 to systematically adjust its hydrophobicity and structural properties, enhancing its similarity to known algicidal peptides while preserving its efficacy. The results sorted by score are provided in Table 1. Among high-scoring candidates, peptides with physicochemical properties clearly distinct from their parent sequences were designated as modified peptides. For this purpose, one modification involved substitution with small residues (e.g., Gly, Ser) to reduce molecular weight and increase structural flexibility by lowering steric hindrance, and the other aimed to enhance hydrophobicity to promote membrane disruption. These modified peptides were designated as F32-1-M1 and F32-1-M2, respectively, and synthesized for further evaluation (Tables 1 and 2).

The synthesized F32-1-M1 and F32-1-M2 were tested for their algicidal effects on the harmful algae A. fundyense. As shown in Fig. 5, both peptides immediately stopped algal movement after treatment, with no cell lysis observed, demonstrating algicidal effects distinct from those of known algicidal peptides. The algicidal rates of F32-1-M1 and F32-1-M2 were 46.67% and 51.72%, respectively, compared to 63.46% for the original F32-1. Statistical analysis indicated that the algicidal rate of F32-1 was significantly higher than that of F32-1-M1 (p = 0.016) and F32-1-M2 (p = 0.010), whereas no significant difference was observed between F32-1-M1 and F32-1-M2 (p = 0.894). F32-1-M1 was modified to reduce molecular weight and increase flexibility for enhanced membrane disruption, aiming for better interaction with the cell membrane based on structure–activity relationship analysis⁴⁶. However, its algicidal activity was not improved. F32-1-M2 was designed to substantially enhance hydrophobicity to test whether membrane disruption could improve its algicidal efficacy. Despite these modifications, algal cell lysis was not observed, and its algicidal efficacy was reduced.

F32-1 was confirmed to exhibit algicidal effects through a penetration-based mechanism, distinguishing it from membrane-disrupting algicidal peptides, which primarily target athecate harmful algae and demonstrate limited efficacy against thecate species. Table 3 summarizes previous findings, while variations in endpoints (e.g., immotility, lytic, ecdysis) and methodologies mean that potency values are not shown. Park et al. (2011) reported that peptides such as HPA3 and HPA3NT3 exert strong algicidal activity by disrupting the plasma membrane, with increased hydrophobicity enhancing their potency³⁰. Further modification through hybridization with melittin (Hn-Mc and its analogs) accelerated membrane disruption, leading to enhanced algicidal effects against harmful algal blooms (HABs)¹⁷. These membrane-disrupting peptides have been shown to be particularly effective against athecate dinoflagellates, including H. akashiwo, Chattonella sp., Chattonella marina, and Cochlodinium polykrikoides. However, peptides such as Mastoparan B and surfactins-C13, while inducing membrane lysis in C. marina and C. polykrikoides, exhibited minimal algicidal effects against thecate species such as A. tamarense, even at high concentrations, due to the protective cellulose theca surrounding their plasma membranes^14,18.

These findings indicate that increased hydrophobicity enhances interactions between peptides and the plasma membranes of athecate dinoflagellates. However, this approach does not confer algicidal efficacy against thecate species, where the plasma membrane is shielded by a rigid thecal layer. Given that F32-1 exhibited algicidal activity without inducing membrane lysis, its physicochemical properties and mechanism of action differ from those of previously reported algicidal peptides. This suggests that F32-1 functions through a unique penetration mechanism rather than membrane disruption, enabling it to exert algicidal effects against thecate dinoflagellates despite the presence of a cellulose theca. The distinct mode of action of F32-1 highlights its potential as a novel agent for controlling harmful algal blooms, particularly those caused by thecate species that are resistant to conventional membrane-disrupting peptides.

Toxicity of algicidal peptide

For an algicidal peptide to be applicable in environmental management, it must effectively control HABs while minimizing adverse effects on non-target algal species. To assess the environmental safety of F32-1, its toxicity was evaluated using Skeletonema sp. and P. tricornutum, two marine diatoms commonly employed as model organisms in toxicity studies. These species are primary producers in marine ecosystems and play essential roles in estuarine and coastal food webs^47,48. Their high sensitivity and reproducibility in chemical toxicity assessments have led to their designation as standard test species under the ISO 10253:2016 protocol^48,49,50.

To evaluate the potential impact of F32-1 on non-target algal species, five different concentrations were selected according to ISO 10253. Although high concentrations were applied, determining the endpoint for each algal species remained challenging, suggesting that only extremely high concentrations could induce toxicity. The LOEC of Skeletonema sp. and P. tricornutum was observed at concentrations equal to or above the EC₉₀ for harmful thecate algae⁵¹. Skeletonema exhibited greater sensitivity, with NOEC and LOEC calculated as 0.7 mg/mL and 1.4 mg/mL (p = 0.034), and higher concentrations were also significant (p < 0.01). P. tricornutum showed NOEC and LOEC values of 1.4 mg/mL and 2.1 mg/mL (p = 0.015), with significance reinforced at 2.8 mg/mL (p < 0.01) (Fig. 6).

While some algicidal compounds exhibit broad-spectrum toxicity, others are selective toward harmful algal groups. For instance, deinoxanthin, a secondary metabolite produced by Deinococcus sp. Y35, exhibited algicidal activity against A. tamarense, while a higher concentration was required to inhibit the growth of S. costatum and P. tricornutum, as observed for F32-1⁵². Additionally, culture extracts from Bacillus sp. AB-4 exhibited algicidal activity against harmful algae while having minimal toxicity toward non-harmful species, including S. costatum and P. tricornutum, demonstrating the feasibility of developing environmentally selective algicidal agents⁵³.

However, research on the environmental release of peptides and the toxicity of algicidal peptides on non-target algal species remains highly limited. Given that bioactive peptides are generally biodegradable and have low environmental persistence, their long-term ecological impact may be lower than that of more structurally stable compounds, such as synthetic chemicals or toxins^6,7,8. By evaluating the effects of F32-1 on Skeletonema sp. and P. tricornutum, this study provides critical insights into the potential ecological risks associated with algicidal peptide applications. However, further investigation is required to elucidate its toxicity mechanisms and ecological impact, particularly regarding its degradation and bioaccumulation potential in marine ecosystems. Such assessments are essential for ensuring the safe and sustainable application of F32-1 in harmful algal bloom management.

This study reveals the unique algicidal mechanism of F32-1, a tetrapeptide derived from microbial coculture of mackerel waste. F32-1 penetrates thecate dinoflagellate cells and induces intracellular oxidative stress, leading to selective degradation of organelles while preserving plasma membrane integrity. This membrane-independent mechanism is particularly effective against Alexandrium fundyense, which resists conventional membrane-targeting agents due to its rigid thecal structure. Although bioactive peptides are considered ideal for environmentally safe algal control due to their biodegradability and low persistence, few have been studied for thecate species. The present findings demonstrate that F32-1 acts through a non-membrane-disruptive, intracellular mechanism and induces algal death within 3 h of treatment, which is faster than most biological agents, while showing negligible toxicity to the non-target species Skeletonema sp. and Phaeodactylum tricornutum, with growth inhibition observed only at concentrations 1.7- and 3.4-fold higher than the EC₉₀ for A. fundyense, respectively.

These results position F32-1 as a fast-acting, selective, and low-residue candidate for sustainable algal bloom management. By combining waste reutilization with a novel mechanistic insight, this work contributes to the development of peptide-based tools for ecologically responsible marine biocontrol.

HABs:

Harmful algal blooms

FDA:

Fluorescein diacetate

CLSM:

Confocal laser scanning microscope

FITC:

Fluorescein isothiocyanate

ROS:

Reactive oxygen species

TEM:

Transmission electron microscope

PBS:

Phosphate-buffered saline

FAM:

5-Carboxyfluorescein

HPLC:

High-performance liquid chromatography

DMSO:

Dimethyl sulfoxide

LC₅₀ :

Lethal Concentration 50%

CPPs:

Cell-penetrating peptides

NOEC:

No observed effect concentration

LOEC:

Lowest observed effect concentration

1. Gobler, C. J. Climate change and harmful algal blooms: insights and perspective. Harmful Algae 91, 101731. https://doi.org/10.1016/j.hal.2019.101731 (2020).

   Article  PubMed  Google Scholar 

2. Heil, C. A. & Muni-Morgan, A. L. Florida’s harmful algal bloom (HAB) problem: escalating risks to human, environmental and economic health with climate change. Front. Ecol. Evol. 9, 646080. https://doi.org/10.3389/fevo.2021.646080 (2021).

   Article  Google Scholar 

3. Brown, A. R. et al. Assessing risks and mitigating impacts of harmful algal blooms on mariculture and marine fisheries. Rev. Aquac. 12(3), 1663–1688. https://doi.org/10.1111/raq.12403 (2020).

   Article  Google Scholar 

4. Humbert, J. F. & Quiblier, C. The suitability of chemical products and other short-term remedial methods for the control of cyanobacterial blooms in freshwater ecosystems. Front. Environ. Sci. 176, 1–4. https://doi.org/10.3389/fenvs.2019.00176 (2019).

   Article  Google Scholar 

5. Shumway, S. E., Frank, D. M., Ewart, L. M. & Evan Ward, J. Effect of yellow loess on clearance rate in seven species of benthic, filter-feeding invertebrates. Aquac. Res. 34(15), 1391–1402. https://doi.org/10.1111/j.1365-2109.2003.00958.x (2003).

   Article  Google Scholar 

6. Quan, H., Zhang, Y., Yin, P. & Zhao, L. Effects of two algicidal substances, ortho-tyrosine and urocanic acid, on the growth and physiology of Heterosoigma akashiwo. Environ. Pollut. 284, 117004. https://doi.org/10.1016/j.envpol.2021.117004 (2021).

   Article  CAS  PubMed  Google Scholar 

7. Su, J. Q. et al. Isolation and characterization of a marine algicidal bacterium against the toxic dinoflagellate Alexandrium tamarense. Harmful Algae 6(6), 799–810. https://doi.org/10.1016/j.hal.2007.04.004 (2007).

   Article  CAS  Google Scholar 

8. Li, H. et al. Natural algicidal compounds: strategies for controlling harmful algae and application. Plant Physiol. Biochem. 215, 108981. https://doi.org/10.1016/j.plaphy.2024.108981 (2024).

   Article  CAS  PubMed  Google Scholar 

9. Ahmed, I. et al. Exploring marine as a rich source of bioactive peptides: Challenges and opportunities from marine pharmacology. Mar. Drugs 20(3), 208. https://doi.org/10.3390/md20030208 (2022).

   Article  CAS  PubMed  PubMed Central  Google Scholar 

10. Nguyen, T. T., Heimann, K. & Zhang, W. Protein recovery from underutilised marine bioresources for product development with nutraceutical and pharmaceutical bioactivities. Mar. Drugs 18(8), 391. https://doi.org/10.3390/md18080391 (2020).

    Article  CAS  PubMed  PubMed Central  Google Scholar 

11. Balaji-Prasath, B. et al. Methods to control harmful algal blooms: a review. Environ. Chem. Lett. 20(5), 3133–3152. https://doi.org/10.1007/s10311-022-01457-2 (2022).

    Article  CAS  Google Scholar 

12. Nakamura, N., Nakano, K., Sugiura, N. & Matsumura, M. A novel cyanobacteriolytic bacterium, Bacillus cereus, isolated from a eutrophic lake. J. Biosci. Bioeng. 95(2), 179–184. https://doi.org/10.1016/S1389-1723(03)80125-1 (2003).

    Article  CAS  PubMed  Google Scholar 

13. Sakata, T., Yoshikawa, T. & Nishitarumizu, S. Algicidal activity and identification of an algicidal substance produced by marine Pseudomonas sp. C55a–2. Fish. Sci. 77(3), 397–402. https://doi.org/10.1007/s12562-011-0345-8 (2011).

    Article  CAS  Google Scholar 

14. Shao, X. et al. Algicidal characteristics of novel algicidal compounds cyclic lipopeptide surfactins from Bacillus tequilensis strain D8 in eliminating Heterosigma akashiwo blooms. Front. Microbiol. 13, 1066747. https://doi.org/10.3389/fmicb.2022.1066747 (2022).

    Article  PubMed  PubMed Central  Google Scholar 

15. Volk, R. B. Screening of microalgal culture media for the presence of algicidal compounds and isolation and identification of two bioactive metabolites, excreted by the cyanobacteria Nostoc insulare and Nodularia harveyana. J. Appl. Phycol. 17(4), 339–347. https://doi.org/10.1007/s10811-005-7292-7 (2005).

    Article  CAS  Google Scholar 

16. Wang, X. et al. Algicidal activity of rhamnolipid biosurfactants produced by Pseudomonas aeruginosa. Harmful Algae 4(2), 433–443. https://doi.org/10.1016/j.hal.2004.06.001 (2005).

    Article  CAS  Google Scholar 

17. Park, S. C. et al. Algicidal effect of hybrid peptides as potential inhibitors of harmful algal blooms. Biotechnol. Lett. 38(5), 847–854. https://doi.org/10.1007/s10529-016-2052-0 (2016).

    Article  CAS  PubMed  Google Scholar 

18. Li, X. et al. Malformin C, an algicidal peptide from marine fungus Aspergillus species. Ecotoxicology 30(5), 996–1003. https://doi.org/10.1007/s10646-021-02389-3 (2021).

    Article  CAS  PubMed  Google Scholar 

19. Cho, J. Y. & Kim, J. K. Isolation and identification of a novel algicidal peptide from mackerel muscle hydrolysate. J. Chromatogr. B 1093, 39–46. https://doi.org/10.1016/j.jchromb.2018.06.056 (2018).

    Article  CAS  Google Scholar 

20. Cho, J. Y., Jung, H. Y. & Kim, J. K. Biodegraded mackerel wastewater selectively inhibits harmful algal blooms. J. Hazard. Mater. 364, 349–355. https://doi.org/10.1016/j.jhazmat.2018.10.053 (2019).

    Article  CAS  PubMed  Google Scholar 

21. Meher, P. K., Sahu, T. K., Saini, V. & Rao, A. R. Predicting antimicrobial peptides with improved accuracy by incorporating the compositional, physico-chemical and structural features into Chou’s general PseAAC. Sci. Rep. 7(1), 1–12. https://doi.org/10.1038/srep42362 (2017).

    Article  CAS  Google Scholar 

22. Agüero-Chapin, G. et al. Emerging computational approaches for antimicrobial peptide discovery. Antibiotics 11(7), 936. https://doi.org/10.3390/antibiotics11070936 (2022).

    Article  CAS  PubMed  PubMed Central  Google Scholar 

23. Lin, Z., Chen, B. & Zhao, L. Fluorescence-based bioassays with dose–response curve and relative potency in measuring algicidal virulence of Bacillus sp. B1 exudates against Heterosigma akashiwo. Sci. Total Environ. 724, 137691. https://doi.org/10.1016/j.scitotenv.2020.137691 (2020).

    Article  CAS  PubMed  Google Scholar 

24. Park, S. Y. et al. Ecotoxicological effects of suspended sediments on marine microalgae using flow cytometry and pulse-amplitude modulation (PAM) fluorometry. Mar. Pollut. Bull. 208, 116968. https://doi.org/10.1016/j.marpolbul.2024.116968 (2024).

    Article  CAS  PubMed  Google Scholar 

25. Cho, S. et al. Elucidation of the algicidal mechanism of the marine bacterium Pseudoruegeria sp M32A2M against the harmful alga Alexandrium catenella based on time-course transcriptome analysis. Front. Mar. Sci. 8, 728890. https://doi.org/10.3389/fmars.2021.728890 (2021).

    Article  Google Scholar 

26. Li, Y. et al. Toxicity of algicidal extracts from Mangrovimonas yunxiaonensis strain LY01 on a HAB causing Alexandrium tamarense. J. Hazard. Mater. 278, 372–381. https://doi.org/10.1016/j.jhazmat.2014.06.032 (2014).

    Article  ADS  CAS  PubMed  Google Scholar 

27. Chen, X. et al. Algicidal effects of a high-efficiency algicidal bacterium Shewanella Y1 on the toxic bloom-causing dinoflagellate Alexandrium pacificum. Mar. Drugs 20(4), 239. https://doi.org/10.3390/md20040239 (2022).

    Article  CAS  PubMed  PubMed Central  Google Scholar 

28. Rezayian, M., Niknam, V. & Ebrahimzadeh, H. Oxidative damage and antioxidative system in algae. Toxicol. Rep. 6, 1309–1313. https://doi.org/10.1016/j.toxrep.2019.10.001 (2019).

    Article  CAS  PubMed  PubMed Central  Google Scholar 

29. Johnson, J. G., Janech, M. G. & Van Dolah, F. M. Caspase-like activity during aging and cell death in the toxic dinoflagellate Karenia brevis. Harmful Algae 31, 41–53. https://doi.org/10.1016/j.hal.2013.08.005 (2014).

    Article  CAS  PubMed  Google Scholar 

30. Park, S. C., Lee, J. K., Kim, S. W. & Park, Y. Selective algicidal action of peptides against harmful algal bloom species. PLoS ONE 6(10), e26733. https://doi.org/10.1371/journal.pone.0026733 (2011).

    Article  ADS  CAS  PubMed  PubMed Central  Google Scholar 

31. Seo, J. G. et al. The algicidal effect of antimicrobial peptide, Mastoparan B. J. Fish Pathol. 16(3), 193–201 (2003).

    Google Scholar 

32. Barreto Filho, M. M., Bagatini, I. L. & Durand, P. M. How shall we measure programmed cell death in eukaryotic microalgae?. Eur. J. Phycol. 58, 13–34. https://doi.org/10.1080/09670262.2022.2041731 (2022).

    Article  CAS  Google Scholar 

33. Durand, P. M. The Evolutionary Origins of Life and Death (University of Chicago Press, 2020).

34. Pokrzywinski, K. L., Tilney, C. L., Warner, M. E. & Coyne, K. J. Cell cycle arrest and biochemical changes accompanying cell death in harmful dinoflagellates following exposure to bacterial algicide IRI-160AA. Sci. Rep. 7(1), 1–13. https://doi.org/10.1038/srep45102 (2017).

    Article  CAS  Google Scholar 

35. Ray, J. L. et al. Virus infection of Haptolina ericina and Phaeocystis pouchetii implicates evolutionary conservation of programmed cell death induction in marine haptophyte–virus interactions. J. Plankton Res. 36(4), 943–955. https://doi.org/10.1093/plankt/fbu029 (2014).

    Article  ADS  CAS  PubMed  PubMed Central  Google Scholar 

36. Vavilala, S. L., Gawde, K. K., Sinha, M. & D’Souza, J. S. Programmed cell death is induced by hydrogen peroxide but not by excessive ionic stress of sodium chloride in the unicellular green alga Chlamydomonas reinhardtii. Eur. J. Phycol. 50(4), 422–438. https://doi.org/10.1080/09670262.2015.1070437 (2015).

    Article  CAS  Google Scholar 

37. Bidle, K. D. The molecular ecophysiology of programmed cell death in marine phytoplankton. Ann. Rev. Mar. Sci. 7, 341–375. https://doi.org/10.1146/annurev-marine-010213-135014 (2015).

    Article  PubMed  Google Scholar 

38. Du, W. et al. Characterization of cell death in harmful Karenia mikimotoi under algicidal activity of Marinobacter sp O-7. J. Sea Res. 191, 102326. https://doi.org/10.1016/j.seares.2022.102326 (2023).

    Article  Google Scholar 

39. Zuppini, A., Gerotto, C. & Baldan, B. Programmed cell death and adaptation: Two different types of abiotic stress response in a unicellular chlorophyte. Plant Cell Physiol. 51(6), 884–895. https://doi.org/10.1093/pcp/pcq069 (2010).

    Article  ADS  CAS  PubMed  Google Scholar 

40. Jiménez, C. et al. Different ways to die: cell death modes of the unicellular chlorophyte Dunaliella viridis exposed to various environmental stresses are mediated by the caspase-like activity DEVDase. J. Exp. Bot. 60(3), 815–828. https://doi.org/10.1093/jxb/ern330 (2009).

    Article  CAS  PubMed  PubMed Central  Google Scholar 

41. Bui, Q. T. N. & Ki, J. S. Two novel superoxide dismutase genes (CuZnSOD and MnSOD) in the toxic marine dinoflagellate Alexandrium pacificum and their differential responses to metal stressors. Chemosphere 313, 137532. https://doi.org/10.1016/j.chemosphere.2022.137532 (2023).

    Article  CAS  PubMed  Google Scholar 

42. Park, H. et al. Two novel glutathione S-transferase (GST) genes in the toxic marine dinoflagellate Alexandrium pacificum and their transcriptional responses to environmental contaminants. Sci. Total. Environ. 915, 169983. https://doi.org/10.1016/j.scitotenv.2024.169983 (2024).

    Article  CAS  PubMed  Google Scholar 

43. Ouyang, J., Sheng, Y. & Wang, W. Recent advances of studies on cell-penetrating peptides based on molecular dynamics simulations. Cells 11(24), 4016. https://doi.org/10.3390/cells11244016 (2022).

    Article  CAS  PubMed  PubMed Central  Google Scholar 

44. Bhunia, D. et al. Spatial position regulates power of tryptophan: Discovery of a major-groove-specific nuclear-localizing, cell-penetrating tetrapeptide. J. Am. Chem. Soc. 140(5), 1697–1714. https://doi.org/10.1021/jacs.7b10254 (2018).

    Article  ADS  CAS  PubMed  Google Scholar 

45. Patel, S. G. et al. Cell-penetrating peptide sequence and modification dependent uptake and subcellular distribution of green florescent protein in different cell lines. Sci. Rep. 9(1), 6298. https://doi.org/10.1038/s41598-019-42456-8 (2019).

    Article  ADS  CAS  PubMed  PubMed Central  Google Scholar 

46. Rončević, T. et al. Antibacterial activity affected by the conformational flexibility in glycine–lysine based α-helical antimicrobial peptides. J. Med. Chem. 61(7), 2924–2936. https://doi.org/10.1021/acs.jmedchem.7b01831 (2018).

    Article  CAS  PubMed  Google Scholar 

47. Wang, X. et al. Acute toxic effect of typical chemicals and ecological risk assessment based on two marine microalgae, Phaeodactylum tricornutum and Platymonas subcordiformis. Environ. Toxicol. Pharmacol. 85, 103649. https://doi.org/10.1016/j.etap.2021.103649 (2021).

    Article  CAS  PubMed  Google Scholar 

48. Feng, P. et al. Effects of environmental concentrations of sulfamethoxazole on Skeletonema costatum and Phaeodactylum tricornutum: Insights into growth, oxidative stress, biochemical components, ultrastructure, and transcriptome. Ecotoxicol. Environ. Saf. 283, 116851. https://doi.org/10.1016/j.ecoenv.2024.116851 (2024).

    Article  CAS  PubMed  Google Scholar 

49. Ríos, F., Lechuga, M., Lobato-Guarnido, I. & Fernández-Serrano, M. Antagonistic toxic effects of surfactants mixtures to bacteria Pseudomonas putida and marine microalgae Phaeodactylum tricornutum. Toxics 11(4), 344. https://doi.org/10.3390/toxics11040344 (2023).

    Article  CAS  PubMed  PubMed Central  Google Scholar 

50. ISO 10253:2016; Water Quality. Marine Algal Growth Inhibition Test with Skeletonema sp. and Phaeodactylum tricornutum (International Organization for Standardization, 2016).

51. Suzuki, G. et al. Emission of dioxin-like compounds and flame retardants from commercial facilities handling deca-BDE and their downstream sewage treatment plants. Environ. Sci. Technol. 55(4), 2324–2335. https://doi.org/10.1021/acs.est.0c06359 (2021).

    Article  ADS  CAS  PubMed  Google Scholar 

52. Li, Y. et al. The first evidence of deinoxanthin from Deinococcus sp. Y35 with strong algicidal effect on the toxic dinoflagellate Alexandrium tamarense. J. Hazard Mater. 290, 87–95. https://doi.org/10.1016/j.jhazmat.2015.02.070 (2015).

    Article  ADS  CAS  PubMed  Google Scholar 

53. Kim, Y. S., Lee, D. S., Jeong, S. Y., Lee, W. J. & Lee, M. S. Isolation and characterization of a marine algicidal bacterium against the harmful raphidophyceae Chattonella marina. J. Microbiol. 47, 9–18. https://doi.org/10.1007/s12275-008-0141-z (2009).

    Article  CAS  PubMed  Google Scholar 

Download references

Authors and Affiliations

1. Department of Biotechnology, School of Marine, Fisheries and Life Science, Pukyong National University, Busan, 48513, Republic of Korea

   Ja Young Cho & Joong Kyun Kim

2. Prokaryote Research Division, Nakdonggang National Institute of Biological Resources, Sangju-Si, 37242, Republic of Korea

   Ja Young Cho

Contributions

Conceptualization, Writing-Review & Editing, Supervision: J.K.K.; Methodology, Validation, Formal analysis, Writing-Original Draft, Project administration, Funding acquisition: J.Y.C.; Investigation, Visualization: J.K.K., J.Y.C. All authors reviewed the manuscript.

Corresponding author

Correspondence to Joong Kyun Kim.

Competing interests

The authors declare no competing interests.

Publisher’s note

Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

Open Access This article is licensed under a Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 International License, which permits any non-commercial use, sharing, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if you modified the licensed material. You do not have permission under this licence to share adapted material derived from this article or parts of it. The images or other third party material in this article are included in the article’s Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article’s Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://creativecommons.org/licenses/by-nc-nd/4.0/.

Reprints and permissions