Development of an indirect immunofluorescence assay for detecting ciguatoxin-producing microalgae

The seafood illness, ciguatera poisoning (CP) is a significant public health concern in tropical and subtropical regions worldwide. In Japan, CP is particularly prevalent in Okinawa Prefecture, where is the subtropical region of Japan, due to the consumption of fish contaminated with Pacific ciguatoxins (P-CTXs). The benthic dinoflagellate Gambierdiscus polynesiensis is known to produce P-CTXs in the South Pacific, but P-CTX-producing microalgae have not been identified in Japanese waters. A practical method applicable to field samples containing diverse benthic microalgae is therefore required. To identify a P-CTX producer in Japan, an indirect immunofluorescence (IIF) assay using P-CTX-specific monoclonal antibodies that can be applied to various benthic microalgae such as diatoms and dinoflagellates was developed. The optimal conditions for cell membrane permeabilization were determined using the IIF assay with an anti-actin antibody on benthic diatoms and dinoflagellates treated with various concentrations of Triton X-100. Treatment of the cells with 5.0% (v/v) Triton X-100 resulted in the highest fluorescence and the lowest cell loss. When an IIF assay with an anti-P-CTX antibody (8H4) was performed on cells of Gambierdiscus australes, G. jejuensis, G. polynesiensis, and G. silvae treated with 5.0% Triton X-100, only G. polynesiensis showed fluorescence. The simultaneous application of two anti-P-CTX antibodies (8H4 and 10C9) doubled the fluorescence compared to 8H4 alone. Extending the antibody reaction time from one hour to 16 hours increased the fluorescence by about sevenfold. These optimized conditions enabled visualization of P-CTXs-derived fluorescence in G. polynesiensis cells under both not only a confocal laser scanning microscope but also a conventional inverted fluorescence microscope having deeper depths of focus. Combining the IIF assay with observing the cells under an inverted microscope provides a practical tool for detecting P-CTX producers in samples of various sizes of benthic microalgae collected from coastal waters. This will contribute to identifying unknown P-CTX-producing microalgae in Japanese and other coastal waters in the future.

Ciguatera poisoning (CP) is foodborne illness with a low mortality rate that primarily occurs in tropical and subtropical regions due to the consumption of toxic fish. CP is the world’s most prevalent foodborne illness caused by a natural marine toxin, affecting the digestive, circulatory, and nervous systems. The symptoms include thermoanesthesia (dry-ice sensation), diarrhea, nausea, vomiting, abdominal pain, hypotension, bradycardia, fatigue, joint pain, and itching (Lehane and Lewis, 2000; Lewis, 2001; Noguchi, 2003; Chinain et al., 2010; Oshiro, 2010; Oshiro et al., 2011; Yogi et al., 2011; Shiomi and Nagashima, 2013; Tamanaha, 2013). CP frequently occurs in regions such as the islands of French Polynesia, the coast of Queensland, Australia, the U.S. Virgin Islands, Miami, Florida, USA, and Hawaii (Oshiro, 2010). In Japan, CP cases were primarily reported in Okinawa Prefecture, where is the subtropical region of Japan, since the 1980s, and have been associated with the consumption of toxic reef fish such as Variola louti and Lutjanus bohar, and Oplegnathus punctatus (Oshiro et al., 2010). However, in recent years, cases of CP linked to poisonous fish caught from temperate region of the Pacific coast of Japan, have been sporadically observed (Oshiro, 2010). There is growing concern that CP incidents may increase in Japan’s temperate zones due to rising sea temperatures associated with global warming (Oshiro, 2010; Oshiro et al., 2011).

The toxin that causes CP has been reported to be ciguatoxins (CTXs), potent neurotoxins (Yasumoto et al., 1977). CTXs are classified by geographical origin and structural differences. Those found in the Pacific Ocean are categorized as Pacific-ciguatoxins (P-CTXs) (Sasaki et al., 2023). P-CTXs can be further classified into two types based on their chemical structure: ciguatoxin-1B (P-CTX1B) and ciguatoxin-3C (P-CTX3C) (Hirama et al., 2001). The P-CTX1B is synthesized in fish through the further oxidation of 54-deoxy P-CTX1B and 52-epi-54-deoxy P-CTX1B. These compounds are produced through the oxidation of P-CTX4A and its structural isomer, P-CTX4B, within the fish’s body (Lewis, 2001; Caillaud et al., 2010; Hamilton et al., 2010). Additionally, 51-hydroxy P-CTX3C and M-seco P-CTX3C are synthesized through the oxidation of P-CTX3C (Satake et al., 1993).

In the South Pacific, P-CTXs are believed to originate in the food from species of the dinoflagellate genus Gambierdiscus, which grow associated with the surface of macroalgae. Gambierdiscus polynesiensis produces P-CTX4A, P-CTX4B, P-CTX3C, P-CTX3B, M-seco-3B/C, and M-seco-4A/B (Yasumoto, 2005; Chinain et al., 2010; Yogi et al., 2011; Murray et al., 2024). Recently a second species in the Pacific region, Gambierdiscus holmesii, was found to produce chemically detectable CTXs (M-seco-CTX4A/B) (Murray et al., 2025). Although the occurrence of various Gambierdiscus species including Gambierdiscus australes, Gambierdiscus scabrosus, Gambierdiscus jejuensis, Gambierdiscus sp. type 3, and Gambierdiscus silvae has been reported in Japanese coastal waters, none of them have been found to produce P-CTXs (Murray et al., 2024; Uchida et al., personal communication). Additionally, a comprehensive metabarcoding analysis of numerous macroalgal-associated microbial community samples collected from coastal areas of Okinawa Prefecture and Kochi Prefecture (temperate region) in Japan did not detect G. polynesiensis or G. holmesii (Funaki et al., 2022). Through metabarcoding, Funaki et al. (2022) discovered non-CTX-producing Gambierdiscus phylotypes and species, as well as G. caribaeus and four new Gambierdiscus phylotypes along the Japanese coast. The CTX-producing capabilities of the latter species and phylotypes remain unknown because culture strains have not been established. Furthermore, it is possible that undiscovered microalgae within this genus or other genera possess CTX-producing capabilities. To find uncharacterized CTX-producing organisms, it is important to use a method that can specifically detect them and provide morphological information enabling their isolation and cultivation.

Honsell et al. (2022) developed an indirect immunofluorescence (IIF) assay using the anti-P-CTX antibody 8H4, developed by Tsumuraya et al. (2006), to detect ‘CTX-like compounds’ in thin cell sections of G. australes using a confocal laser scanning microscope. However, they failed to detect ‘CTX-like compounds’ in whole cells, indicating that the anti-P-CTX antibody could not penetrate the cell membrane under the applied cell processing conditions. These results suggest that the IIF assay is not suitable for detecting CTX-producing microalgae from mixed community environmental samples because preparing thin cell sections of various microorganisms present in field samples is impractical. Furthermore, a confocal laser scanning microscope has a shallow depth of focus and is therefore unsuitable for observing cells of various sizes collected from coastal waters simultaneously. Therefore, to apply an IIF assay to field samples containing various microalgal cells to detect P-CTX-producing microorganisms, it is necessary to reexamine the membrane penetration conditions of the surfactant used in the IIF assay. Additionally, an IIF assay using conventional inverted fluorescence microscopy with a deeper depth of focus that can simultaneously observe cells of various sizes must be developed.

The goal of this study was to develop a new IIF assay that can detect the cells of P-CTX-producing species from environmental samples collected in CP endemic areas. First, we investigated the conditions under which antibodies could permeate cells by treating various benthic microalgae with surfactants. As a control, we performed an IIF assay using an antibody that binds to actin, a protein present in all eukaryotic microalgae, on cells of various benthic microalgal strains. Next, we performed an IIF assay with an anti-P-CTX antibody under optimized membrane permeation conditions using three non-CTX-producing Gambierdiscus species strains occurring in Japanese coastal waters, as well as a P-CTX-producing strain G. polynesiensis. This was done to determine whether fluorescence derived from P-CTXs in G. polynesiensis could be specifically observed. Finally, we attempted to develop an IIF assay that can observe P-CTXs-derived fluorescence using conventional inverted fluorescence microscopy, which allows for the rapid and easy observation of fluorescence in cells of different sizes. To this end, we investigated conditions to enhance fluorescence intensity to facilitate the observation of P-CTXs-derived fluorescence under conventional inverted fluorescence microscopy, which has generally lower fluorescence detection sensitivity and a deeper depth of focus than confocal laser scanning microscopy.

Strains and culture conditions

The culture strains used in this study are listed in Supplementary Table 1. They included the benthic diatoms: Navicula sp. OKS200804-D11, Helicotheca sp. ONCA20, and Odontella sp. ONCA28, and the dinoflagellates: Ostreopsis sp. ISK200207-O1(#494), Prorocentrum lima NIES-4558, Vulcanodinium rugosum CAWD188, Gambierdiscus australes SO8911_1, Gambierdiscus jejuensis MO808_2, Gambierdiscus silvae ISK-M4, and Gambierdiscus polynesiensis CAWD267. These strains were cultured in polypropylene-capped test tubes for 30 days under a light intensity of 90-100 μmol photons m -2 s -1 with 14 h light/10 h dark cycle, with 30 µL of the media at the temperatures shown in Supplementary Table 1. Cell at the late log/early stationary phases were used for the assay described below.

Effect of Triton-X100 on membrane permeability in benthic algae

Five mL of culture medium from Navicula sp., Helicotheca sp., Odontella sp., Ostreopsis sp., P. lima, V. rugosum, and G. polynesiensis were collected and centrifuged at 1,000× g for 5 minutes to harvest the cells. The resulting cell pellets were fixed with phosphate-buffered saline (PBS) containing 10% (w/v) paraformaldehyde (final concentration of 4.0%, w/v, pH 7.8). The cells were then incubated for 15 minutes at room temperature. The fixed cells were washed three times with PBS by centrifugation at 1,000× g for 5 minutes each time. The cells were then resuspended in 1 mL of PBS. A 10 µL aliquot of this suspension was placed on boundary slides (S6114, TGK, Tokyo, Japan), and the cells were counted under an inverted microscope (IX-70, Olympus, Tokyo, Japan).

Membrane permeabilization was evaluated by adding Triton X-100 to reach final concentrations of 0.1%, 5.0%, and 10% (v/v). The cells were then incubated for 15 minutes at room temperature. The cells were washed three times with PBS, counted as previously described. The number was then compared with that before permeabilization to estimate the “processed-to-pre-cell ratio.” Blocking was performed with PBS containing 5.0% (w/v) bovine serum albumin (BSA) for 30 minutes at room temperature. The primary antibody (mouse monoclonal anti-actin, clone C4, MP Biomedicals, 0869100-CF, USA) was added at a dilution of 1:100 and incubated for 60 minutes at room temperature. The cells were then washed three times with PBS. The secondary antibody, Multi-rAb™ CoraLite ® Plus 488 Goat Anti-Mouse Recombinant Secondary Antibody (H+L) (Proteintech, USA), was then added at a dilution of 1:300 and incubated for 60 minutes at room temperature. The samples were then washed three times with PBS, resuspended in 30 µL of PBS, and examined under a confocal scanning laser microscope, as described later (section 2.4). The fluorescence intensities of cells treated with 0.1%, 5.0%, and 10% were obtained by subtracting the fluorescence intensities of the negative control cells (cells treated with only secondary antibody) from those of the tested cells (treated with anti-actin antibody and secondary antibody), which were shown in Supplementary Figures.

IIF assay using anti-P-CTX antibodies

For immunofluorescence using anti-P-CTX antibodies, 5 mL culture of Gambierdiscus australes, G. jejuensis, G. silvae, and G. polynesiensis were collected. The cells were harvested by centrifugation, fixed with 5.0% (v/v) Triton X-100 for 15 minutes at room temperature, and washed three times with PBS. Blocking was carried out according to the aforementioned procedure. Anti-P-CTX antibodies (8H4 and/or 10C9; Tsumuraya et al., 2006) were applied individually or simultaneously at a final dilution of 1:2,500. The cells were incubated for 60 minutes at room temperature or 16 hours at 4 °C. After incubation, the cells were washed three times with PBS. The secondary antibody described previously was then used under identical incubation conditions. This protocol is referred to as Ts-IIF assay, the IIF assay developed in this study. For comparison, the procedure reported by Honsell et al. (2022) was used and designated as a Honsell’-IIF assay (H-IIF). In the H-IIF assay, the cells were fixed with 4.0% paraformaldehyde at room temperature for 30 minutes. Then, the cells were washed three times with gentle stirring in PBS for 15 minutes, followed by centrifugation at 180× g for 5 minutes. The cells were then treated with 0.1% (w/v) Triton X-100 at 4 °C for 30 minutes. The cells were then incubated in PBS containing 1% (w/v) BSA at 4 °C for 30 minutes. The 8H4 antibody was diluted 1:500 and incubated at 37 °C for 90 min. The secondary antibody (Alexa Fluor 488-conjugated anti-mouse IgG, Jackson ImmunoResearch, 115-545-006, Italy) was added at a dilution of 1:500 and incubated at 37 °C for 60 min. After each antibody treatment, the samples were exposed to 0.1% Triton X-100 in PBS for 15 minutes and washed three times by centrifugation at 180× g for 5 minutes. The samples were resuspended in 30 µL of PBS, and examined under a confocal laser scanning microscope, as described later (section 2.4).

Fluorescence observation by confocal laser scanning microscopy

Ten microliters of the 30 µL cell suspension prepared for the IIF assay were placed on a microscope slide (Matsunami Glass, Osaka, Japan) and covered with a micro cover glass. Fluorescence imaging was performed using a confocal laser scanning microscope (LSM700, Carl Zeiss, Jena, Germany). The observation parameters were set as follows: an excitation wavelength of 488 nm, an emission wavelength of 490–519 nm, and a laser intensity of 10.0, which corresponds to the detection of green fluorescence. Quantitative fluorescence analysis was carried out on 10 randomly selected cells for each experimental condition. Fluorescence within each cell was measured using image analysis software (ZEN 3.3; Carl Zeiss, Jena, Germany). Cell boundaries were manually delineated, and an adjacent background area of equivalent size was defined in a cell-free region. Background fluorescence was subtracted from the measured cellular fluorescence. The calculated fluorescence values obtained from the 10 cells were averaged to determine the mean fluorescence per cell. Autofluorescence derived from chlorophyll was detected under excitation at 639 nm, with an emission range of 640–700 nm, and a laser intensity of 2.0. The fluorescence intensities of cells tested were obtained by subtracting the fluorescence intensities of the negative control cells treated with the secondary antibody from those of the tested cells treated with the anti-CTX antibodies and the secondary antibody shown in Supplementary Figures. Differences in fluorescence between samples subjected to the Ts-IIF assay and the H-IIF assay were evaluated using a t -test (p < 0.05). Multiple comparisons among other experimental conditions were adjusted using a t -test with Bonferroni correction (p < 0.05).

Fluorescence observation by conventional inverted fluorescence microscopy

To quench the autofluorescence of G. polynesiensis cells, the cells were exposed to an autofluorescence quenching device (Tiyo, NEPAGENE, Japan), for three or six hours. Then, fluorescence of the cells were observed using an inverted fluorescence microscope (IX-70, Olympus, Japan) and compared to that of cells not exposed to the device.

To observe the fluorescence of G. polynesiensis cells treated with the IIF assay under an inverted fluorescence microscope, the cells were first fixed and then treated with 5.0% Triton X-100 as described above. Then, they were exposed to an autofluorescence quenching device (Tiyo) for six hours. This was followed by blocking and antibody treatment, as described above. The resulting cell suspension was then transferred to a 48-well plate (Thermo Scientific, 150687, USA). Fluorescence images were obtained using an inverted fluorescence microscope. The observation conditions included excitation at the NIBA filter for green fluorescence detection. Autofluorescence originating from chlorophyll was also recorded under excitation using WIG filter to detect red fluorescence. The fluorescence images were captured using a digital microscope camera (DP23M, Evident, Japan) and processed using cellSens Standard software (Evident, Japan).

Effect of Triton-X concentration on membrane permeability for the IIF assay with anti-actin antibody to various benthic microalgae

The optimal concentration of Triton X-100 required for the anti-actin antibody to permeate the membranes of various benthic microalgal cells was investigated by evaluating the reactivity of cells treated with different concentrations of Triton X-100 using the IIF assay with the anti-actin antibody. The cells of three diatom strains (Navicula sp., Helicotheca sp., and Odontella sp.) and three dinoflagellate strains (Ostreopsis sp., Prorocentrum lima, and Vulcanodinium rugosum) exhibited significantly higher fluorescence of the cells treated at 5.0% and 10% of Triton X-100 compared to those treated at 0.1% treatment (Figure 1; Supplementary Figure 1). However, only the cells of G. polynesiensis treated with 5.0% Triton X-100 showed significantly higher fluorescence than the 0.1% treatment.

To evaluate the reduction in cell numbers after Triton X-100 treatment compared to before treatment, the processed-to-pre cell ratio was determined by comparing the cell counts before and after treatment. Treatment with 10% Triton X-100 significantly decreased the number of morphologically intact cells of Navicula sp., Helicotheca sp., Odontella sp., Ostreopsis sp., and G. polynesiensis (Supplementary Figure 2). Although treatment with 5.0% as well as 10% Triton X-100 caused a significant decrease in the number of Helicotheca sp. cells (Supplementary Figure 2), treatment with 5.0% Triton X-100 was the most suitable for permeabilizing cell membranes, yielding strong fluorescence and moderate cell lysis. Figure 2 shows representative fluorescence images of seven strains of cells that were treated with 5.0% Triton X-100 and then subjected to the IIF assay with the anti-actin antibody.

Results of the Ts-IIF and H-IIF assays on four species of the genus Gambierdiscus

Four species, G. polynesiensis, G. australes, G. jejuensis, and G. silvae, were examined using the Ts-IIF or H-IIF assay with the anti-P-CTX antibody 8H4. Significant differences were observed when comparing the fluorescence between T1 (G. polynesiensis cells treated by Ts-IIF) and P1 (G. polynesiensis cells treated by H-IIF), or between T2 and P2 (Figure 3). No significant difference was found between T3 and P3. However, a significant difference was observed between T4 and P4 (Figure 3). Among T1, T2, T3, and T4 or P1, P2, P3, and P4, T1 and P1 exhibited significantly higher fluorescence compared with the other tested cells (Table 1). Green fluorescence, red autofluorescence, and bright images of G. polynesiensis cells subjected to the Ts-IIF assay or the H-IIF assay with the 8H4 antibody were shown in Figure 4. Faint green fluorescence was observed in the cells subjected to the H-IIF assay (Figure 4A). Strong fluorescence outlining the cell periphery and weak intracellular fluorescence were observed in cells subjected to the Ts-IIF assay (Figure 4D).

Four species, G. polynesiensis, G. australes, G. jejuensis, and G. silvae, were examined using the Ts-IIF assay with the anti-P-CTX antibody 10C9. The fluorescence observed in G. polynesiensis (T5) was significantly higher than in G. australes (T6), G. jejuensis (T7), and G. silvae (T8) (Figure 5). Additionally, the fluorescence of G. silvae in T8 was higher than that of G. australes and G. jejuensis in T6 and T7 (Figure 5).

Enhancement of immunofluorescence obtained from P-CTXs in G. polynesiensis cells

To evaluate whether fluorescence derived from P-CTXs could be enhanced by simultaneous use of two anti-P-CTX antibodies (8H4 and 10C9) on extended incubation times of the antibodies. The fluorescence in T9 (8H4 and 10C9 for 16 hours) was significantly greater than that in T10 (8H4 and 10C9 for one hour), T11 (10C9 for one hour), and T12 (8H4 for one hour) (Figure 6). The fluorescence in T10 was approximately twice that in T12 and significantly higher than in T11 and T12. T9 exhibited about seven-fold higher fluorescence than T10 (Figure 6).

Visualization of the fluorescence derived from the binding of anti-P-CTX antibodies

When G. polynesiensis cells were treated with Tiyo, an autofluorescence quenching device, for six hours, autofluorescence decreased slightly compared to cells treated for 1 hour and decreased significantly compared to untreated cells (Supplementary Figure 6). When the green fluorescence derived from the anti-P-CTX antibodies and red fluorescence derived from autofluorescence were assessed under conventional inverted fluorescence microscopy, no visible green fluorescence was observed in the cells tested in T12 (8H4 for one hour) (Figure 7A). The cells that were tested in T10 (8H4 and 10C9 for one hour) exhibited weak green fluorescence that was localized to inside the cell (Figure 7D). In contrast, cells tested in T9 (8H4 and 10C9 for 16 hours) displayed clear visible green fluorescence derived from P-CTXs in granular organelle-like structures (Figure 7G). After autofluorescence quenching, little to no chlorophyll fluorescence was observed in the cells (Figures 7B, C, E, F, H, I, K, L, N, O, Q, R). Furthermore, no visible green fluorescence was observed in the negative control cells (Figures 7J, M, P).

To detect P-CTX-producing microalgae in natural samples based on the presence or absence of fluorescence obtained through an IIF assay, high antibody-derived fluorescence is desirable to ensure reliable visualization. In the present study, G. polynesiensis cells subjected to the Ts-IIF and H-IIF assay using an anti-P-CTX antibody exhibited stronger antibody-derived fluorescence with the Ts-IIF than with H-IIF assay, making them easier to visualize under a microscope.

One possible reason for the higher fluorescence intensity observed with the Ts-IIF assay than with the H-IIF assay is that the increased concentration of Triton X-100, from 0.1% to 5.0%, improved antibody penetration into cells. This would increase the amount of anti-P-CTX antibodies that bind to P-CTXs. In relation to this result, Honsell et al. (2022) reported that G. australes whole cells treated with 0.1% Triton X-100 did not allow sufficient antibody penetration when subjected to the H-IIF assay. This study investigated cell membrane permeabilization conditions for an IIF assay using anti-actin antibodies against three benthic diatom strains and four dinoflagellate strains. The results showed that antibody-derived fluorescence was obtained with as little as 0.1% Triton X-100 treatment. However, 5.0% Triton X-100 treatment yielded significantly higher antibody-derived fluorescence intensity than cells treated with 0.1%. These results suggest that increasing the concentration of Triton X-100 enhances antibody penetration into cells. Previous studies have generally used 0.1% Triton X-100 for the immunostaining of diatoms and dinoflagellates (Mashini et al., 2022; Vona et al., 2023). Our results suggest that using 5.0% Triton X-100 may improve the efficiency of future immunostaining of newly examined diatom and dinoflagellate species in the future. The second reason for the higher fluorescence intensity obtained with the Ts-IIF assay than with the H-IIF assay is thought to be the simultaneous addition of two anti-P-CTX antibodies (8H4 and 10C9) in the Ts-IIF assay. The 8H4 antibody binds to the right-wing structures of P-CTX1B, 54-deoxyP-CTX1B, P-CTX3C, and 51-hydroxyP-CTX3C, and the 10C9 antibody binds to the left-wing structures of P-CTX3C and 51-hydroxyP-CTX3C (Tsumuraya et al., 2006). The simultaneous binding of these two antibodies to different sites on the same CTXs molecule was thought to increase the amount of secondary antibodies that bind to them, resulting in higher antibody-derived fluorescence intensity. The third reason for the higher fluorescence intensity obtained with the Ts-IIF assay than with the H-IIF assay is the longer antibody reaction time, which increased from one hour used in the H-IIF assay to 16 hours. Consistent with this finding, it has been reported that extending the antibody incubation time from 30 minutes to 20 hours improves the staining of target antigens in immunostaining using antibodies that bind to proteins recognizing and eliminating foreign substances such as invading bacteria or viruses (Brandtzaeg, 1981). In addition, in immunostaining using an anti-RBM3 antibody that binds to the mouse RNA-binding protein RBM3, it has been reported that the binding affinity in antigen–antibody reactions continues to increase up to three hours after the start of the reaction (Dubois et al., 2013). Based on the above, it was suggested that extending the antibody reaction time would effectively enhance the fluorescence intensity obtained during an IIF assay.

Honsell et al. (2022) reported that no intracellular fluorescence was observed when G. australes whole cells were treated with the H-IIF assay. In this study, slight autofluorescence was detected when whole cells of the Japanese G. australes strain were treated with the H-IIF assay, but no fluorescence derived from 8H4 antibody binding was observed. Honsell et al. (2022) also noted that fluorescence was detected within G. australes cells after thin cell sections were treated with the H-IIF assay. This suggests that the G. australes strain used may produce ‘CTX-like compounds’. In contrast, when Japanese G. australes cells were treated with 5.0% Triton X-100 to increase cell membrane permeability before the Ts-IIF assay treatment, slight autofluorescence was observed again, but no fluorescence from 8H4 antibody binding was detected. These differing results may be due to the Japanese G. australes strain lacking the ability to produce ‘CTX-like compounds’.

As mentioned above, autofluorescence was sometimes detected when cells of Gambierdiscus strains were treated with the Ts-IIF assay, Therefore, when cells that possess autofluorescence are subjected to an IIF assay, it is important to evaluate the results by measuring the fluorescence intensity of cells treated with P-CTX antibodies and secondary antibody, as well as cells treated only with secondary antibody for the negative control and to compare the reactivity. Additionally, when applying an IIF assay to field samples containing various benthic microalgae, it is important to observe the presence of autofluorescence or not under an inverted fluorescence microscope before applying the sample to an IIF assay. It is desirable that samples containing cells with autofluorescence are treated using an autofluorescence quenching device, such as Tiyo (NEPAGENE, Japan).

Honsell et al. (2022) reported that, in G. australes cells subjected to the H-IIF assay, antibody-derived fluorescence was localized in intracellular regions consisting of spherical aggregations of tubular structures. In contrast, in the present study, when G. polynesiensis cells were subjected to the Ts-IIF assay and observed by conventional inverted fluorescence microscopy, strong green fluorescence was observed in regions presumed to be granular intracellular organelles. These observations suggest that P-CTXs may be produced in these granular organelle-like structures.

No fluorescence was observed using the Ts-IIF assay with the 8H4 antibody in G. silvae, whereas fluorescence was detected when the IIF assay with the 10C9 antibody was used. In relation to these findings, LC–MS/MS analysis of this cultured strain revealed no known CTXs (Uchida et al., personal communication). However, a Neuro-2A cytotoxicity assay demonstrated CTXs-like activity (Uchida et al., personal communication). As previously mentioned, the 8H4 antibody binds to the right-wing structures of P-CTX1B, 54-deoxyP-CTX1B, P-CTX3C, and 51-hydroxyP-CTX3C. In contrast, the 10C9 antibody recognizes the left-wing structures of P-CTX3C and 51-hydroxyP-CTX3C (Tsumuraya et al., 2006). Based on these structural binding specificities, the present results suggest that G. silvae may produce a compound with the left-wing structural moiety of P-CTX3C or 51-hydroxyP-CTX3C, or P-CTX3C- or 51-hydroxy P-CTX3C-like compounds with a modification group at the right-wing structural moiety that the 8H4 antibody cannot access and can be metabolized in fish, such as Variola louti. However, it is unclear whether such compounds contribute to the occurrence of CP in Japan. Further studies are necessary to elucidate the structure of the substances produced by this strain, as well as their accumulation in fish and potential toxicity to humans.

In previous reports (Honsell et al., 2022), cells subjected to the IIF assay have been observed using confocal laser scanning microscopy, which features a shallow depth of focus and high fluorescence detection sensitivity. However, confocal laser scanning microscopes are unsuitable for observing fluorescence within cells of varying sizes, or for determining the presence or absence of green fluorescence localized within cells. Therefore, observing cells with various sizes in field samples using confocal laser scanning microscopes is difficult. On the other hand, a general-purpose inverted fluorescence microscope is less expensive, and its deep depth of focus allows simultaneous observation of fluorescence within cells of different sizes without changing the focus. However, when cells subjected to the conventional method were observed using an inverted fluorescence microscope, their green fluorescence intensity was insufficient for detection. In contrast, cells processed using the Ts-IIF assay exhibited strong green fluorescence within the cells, enabling detection using an inverted fluorescence microscope.

Anti-P-CTX antibodies used in the Ts-IIF assay specifically binds to P-CTX3C and 51-hydroxyP-CTX3C (Tsumuraya et al., 2006). G. polynesiensis is known to produce these toxins. Given its distribution in the South Pacific, including French Polynesia and the Great Barrier Reef in Australia (Chinain et al., 2010; Murray et al., 2025) (Supplementary Table 2), this method could be used to detect the cells and assess their occurrence in samples collected from these regions. Furthermore, in Japan, P-CTX3C has been detected in fish (Oplegnathus punctatus) from temperate area (Yogi et al., 2011), although the causative organism responsible for the toxification remains unknown. This method developed in this study could be useful for identifying it. Additionally, similar to Japan, CP occurs in Pacific areas, including South Asia, where the causative organism is also unknown. This method is considered useful for identifying the causative organisms in these regions.

This study has clarified the optimal condition for antibody penetration through cell membranes, applicable to diverse benthic microalgae collected from coastal areas. By applying the indirect immunofluorescence (IIF) assay developed in this study to cells treated under the optimal condition, it became possible to detect green fluorescence derived from P-CTXs in the whole cells that were not detected with conventional methods. This simultaneously revealed the localized presence of green fluorescence within the cells. Furthermore, by simultaneously using two types of anti-P-CTX antibodies and extending the antibody reaction time during the IIF assay, we succeeded in increasing the fluorescence intensity derived from CTXs within cells by more than tenfold. Therefore, the use of the indirect immunofluorescence (IIF) assay developed in this study and observation of cells treated under a general-purpose inverted fluorescence microscope represents a new method enabling the detection and identification not only of cells of CTX-producing Gambierdiscus species, but also of cells of novel CTX-producing species exhibiting morphology distinct from Gambierdiscus in the samples collected from Japanese coasts as well as other regions such as South Asia where CP causative organisms have not been fully elucidated.