Changes in valve morphogenesis of Aulacoseira islandica by γ-tubulin inhibitor gatastatin

Texto integral

1. Introduction

Diatoms are unicell eukaryotes belonging to the kingdom Chromista (Cavalier-Smith, 2018); the morphology of their silica frustules is widely various (Round et al., 1990). Diatom systematics are based on the frustule structure and phylogeny of marker genes such as 18S rRNA and rbcL (Medlin and Kaczmarska, 2004; Theriot et al., 2010). Silica frustules consist of two valves and a ring of several overlapping girdle bands (Pickett-Heaps et al., 1990). Frustule details are synthesized sequentially during the cell cycle; after mitosis, valve morphogenesis occurs and girdle bands are formed throughout interphase.

The efforts of many researchers are aimed at searching for genetic and cell mechanisms providing species specific differences in diatom frustule symmetry and structure. It is known that valve morphogenesis is under cytoskeleton control, and the microtubule role is the most studied (Tesson and Hildebrand, 2010; Bedoshvili and Likhoshway, 2021). The correct functioning of microtubules is based on the dynamic polymerization/depolymerization of α- and β-tubulin (Mitchison and Kirschner, 1984; Caudron et al., 2005). It was shown that treatment of diatom cells with microtubule inhibitors led to various valve anomalies (Oey and Schnepf, 1970; Cohn et al., 1989; Van de Meene and Pickett-Heaps, 2002; Van de Meene and Pickett-Heaps, 2004; Kharitonenko et al., 2015; Bedoshvili et al., 2018). It was shown that colchicine can stop nucleus division; however, for some species, valve morphogenesis still occurred; thus, under the influence of colchicine, the cells of centric Cyclotella cryptica and Aulacoseira islandica (Oey and Schnepf, 1970; Bedoshvili et al., 2018) formed “lateral” valves whose morphology was similar to that of normal valves, but their only face surface was adjacent to mature girdle bands. It is supposed that such morphology is possible when microtubules disrupt under the colchicine influence at the mitosis beginning (Bedoshvili et al., 2018). Due to the disruption of karyokinesis and cytokinesis after colchicine treatment, cells with lateral valve formation have a single irregularly shaped nucleus.

The microtubule formation (nucleation) does not occur randomly in the cells; there are specialized nucleation sites, mostly in the Microtubule Organizing Center (MTOC). Diatoms have a specific acentriolar MTOC with atypical features (De Martino et al., 2009). There are variations in the MTOC morphology; however, γ-tubulin underlies all these structures forming large complexes with other proteins (Zheng et al., 1995; Oegema et al., 1999). These complexes do not include α- or β-tubulins, and γ-tubulin is an essential condition for microtubule nucleation. Earlier, it was shown that diatom genomes contain α-, β-, and γ-tubulins (De Martino et al., 2009; Aumeier, 2012; Findeisen et al., 2014). More detailed analysis allows to identify structural features of the predicted amino acid sequences of diatom γ-tubulin (Khabudaev et al., 2022), but its significance for silica valve morphogenesis was not shown.

Gatastatin is a recently synthesized anti-cancer drug, which is able to bind with γ-tubulin and block microtubule nucleation without influencing α- and β-tubulin (Hayakawa et al., 2012). The use of gatastatin makes it possible to demonstrate the role of γ-tubulin in the diatom valve morphogenesis. The freshwater diatom Aulacoseira islandica turned out to be a convenient model for studying abnormal valve morphology due to its valve with a high mantle and large girdle bands. The main purpose of this study was to study γ-tubulin role in the valve morphogenesis of Aulacoseira islandica.

2. Materials and methods

2.1. Cultivation

The Aulacoseira islandica Mr553 strain was isolated from a natural population in Lake Baikal and cultivated in the DM medium (Thompson et al., 1988) at 4°С with natural light and a day–night cycle. The procedure of isolation and cultivation were made according to the protocol described earlier (Zakharova et al., 2020).

2.2. Scanning electron microscopy (SEM)

The frustules were cleaned according to the previously described protocol (Kharitonenko et al., 2015). Suspensions of cleaned frustules were pipetted onto cover glasses, dried, and mounted on SEM stubs with carbon double-sided adhesive tape (SPI Supplies, West Chester, USA). Morphological analysis among the 200 frustules encountered was carried out using a QUANTA 200 scanning electron microscope (FEI Company, Hillsboro, USA).

2.3. Gatastatin treatment

The Aulacoseira islandica culture was incubated for 48 hours in the presence of gatastatin at concentrations selected according to literature data and manufacturer’s recommendations (10, 3, 0.3, and 0.03 μM) and in the presence of PDMPO (Thermo Fisher Scientific, USA) to visualize the forming silica frustules. All experiments were carried out in triplicate. At the end of the experiment, the cells were fixed with paraformaldehyde (4% in the culture medium) for 2 hours. After fixation, the cells were washed with phosphate buffer (0.1 M, pH 7.0) and mounted on glass slides in Mowiol® 40-88 (Sigma-Aldrich, Germany) for fluorescent and laser scanning microscopy.

Cells forming parts of the frustules (valves immediately after division or girdle bands during interphase) were counted among 100 randomly encountered cells. Cells without internal contents and without chloroplast fluorescence or PDMPO staining were considered dead. Counting was made with an optical microscope Axiovert 200 equipped with a blue filter for light with a wavelength of 546 nm.

2.4. Immunostaining and laser scanning microscopy (LSM)

To localize α-tubulin, a previously proposed modified protocol was used (Pasternak et al., 2015). Cell cultures were fixed with a 2% paraformaldehyde solution with the addition of 0.1% Triton X-100 in a microtubule stabilizing buffer (MTSB - PIPES 0.1M, EGTA 0.01M, MgSO₄*7H₂O 0.01M, KOH 0.1M, pH 7.0) for 30 min, washed with the same buffer, and permeabilized with a solution of 1% Triton X-100 and 10% DMSO for 20 min at 37 °C. Blocked with 2% BSA solution in MTSB. To localize alpha-tubulin, monoclonal antibodies to the fragment 426-450 a.a. of the chiken α-tubulin were used. Antibodies were conjugated with Alexa488 (Alexa Fluor® 488 Anti-alpha Tubulin antibody [DM1A] (ab195887), Abcam). Cells were incubated with antibodies at a dilution of 1:200 for 2 h at 37°C, after which the cells were washed with buffer, additionally stained with DAPI (10 μg/ml for 10 min), and mounted on glass using Mowiol® 4-88 (Sigma-Aldrich, Germany).

The samples were examined using a laser scanning microscope LSM 710 (Zeiss) equipped with a Plan-Apochromat 63×/1.40 Oil DIC M27 immersion lens (Zeiss). Chloroplast autofluorescence was excited with a 561 nm laser; emission was registered in the range of 650–723 nm. Alexa488 fluorescence was excited with a 488 nm laser; emission was registered in the range 496-647 nm. DAPI fluorescence was excited with a 405 nm laser; emission was registered in the ranges of 410-492 nm. Three-dimensional reconstructions were obtained from 100 optical sections (z-thickness 15–30 μm). PDMPO fluorescence was excited by a laser with a wavelength of 405 nm, and emission was recorded in the range of 441–587 nm. The 3D reconstruction was obtained from 100 optical sections (z-thickness 15-70 µm).

3. Results

Figure 1 shows electron microscopy of the frustules of Aulacoseira islandica strain 3Mr553. The strain studied was characterized by the formation of valves of different lengths in sister cells (Fig. 1A). All control valves have spatulate connecting spines and ordered rows of areoles (Fig. 1B). The girdle bands covering the valves from the beginning of formation until the next division are sometimes so densely packed on the valves that they remain there even after harsh multi-stage chemical treatment (Fig. 1C).

 

Fig.1. Fine structure of valves and girdle bands of Aulacoseira islandica strain 3Mr553 (SEM). Scale: A – 5 µm; B, C – 1 µm.

 

In the cells of the studied strain, α-tubulin was localized in interphase and at the early stage of valve formation (Fig. 2). Interphase cells are characterized by a star-shaped distribution of microtubules; large bundles of microtubules are irradiated from a center that is in close connection with the nucleus and indicate indirectly the location of the microtubule center (Fig. 2A). Microtubule strands are localized in A. islandica in a thin layer of cytoplasm, and often a single optical layer is sufficient to visualize them. During the early stages of valve morphogenesis, the microtubule packing was not dense enough to be visible in transmitted light, so large strands of microtubules were not observed (Fig. 2B).

 

Fig.2. Visualization of α-tubulin at the early stage of valve formation and in interphase in one optical section (A, B) and 3D-reconstructions (A-1, B-1) in the cells of A. islandica (LSM). Green – fluorescence of Alexa-488 after staining of α-tubulin, blue – fluorescence of DAPI, red – autofluorescence of chlorophyll. Scale bar: A, B – 10 µm; A-1, B-1 – 2 µm.

 

3D-reconstructions of the valves and girdle bands of Aulacoseira islandica formed during the experiment and stained with PDMPO are presented in Figure 3. During the experiment, the cells that formed the girdle bands were able to synthesize from one to three of them. Abnormal morphology was observed among valves formed in the presence of gatastatin. Figure 3 shows the main anomalies encountered among them: anomalies of connecting spines (or their absence) and, in rare cases, disruption of striation (disordered areolae).

 

Fig.3. Valves (left) and girdle bands (right) of A. islandica formed under the influence of different concentrations of gatastatin (indicated on the left) (LSM, 3D-reconstruction). Scale bar – 2 µm.

 

The studied strain was characterized by the presence of at least 30% dead cells in the culture (Fig. 4). Most cells formed actively both valves and girdle bands. At all gatastatin concentrations, the number of dead cells was higher than in the control. The number of colored valves was lowest at gatastatin concentrations of 3 and 10 µM, while at gatastatin concentrations from 0.03 to 3 µM, the number of girdle bands remained approximately the same, but less than in the control.

 

Fig.4. Morphogenesis of valves and girdle bands in the cells of A. islandica under the influence of gatastatin.

 

For A. islandica, it was shown that the number of forming valves and girdle bands decreased with an increase in gatastatin concentration in the medium. At gatastatin concentrations of 0.3 and 0.03 µM, the number of formed valves and girdle bands was comparable to control samples (Fig. 4). At concentrations of 3 and 10 µM, all found staining valves had changes in morphology.

4. Discussion

Earlier, it was shown that diatom γ-tubulin is present in their microtubule center (Craticula cuspidata – Aumeier, 2012). However, even in the large cells of some diatom species, visualization of γ-tubulin is hampered by its small size. Of all the diatom tubulins, γ-tubulin was the least conserved (Khabudaev et al., 2022), therefore poorly visualized by antibodies to γ-tubulin in other organisms. Despite the fact that the main part of the microtubule center is γ-tubulin and proteins of the γ-tubulin complex, the localization of α-tubulin allows indirect detecting of the microtubule center localization in the cells of Aulacoseira islandica and to show the microtubule polymerization degree. It was shown that microtubule bundles that were described earlier for large cells of Coscinodiscus granii (Tesson and Hildebrand, 2010) localized not so much during valve morphogenesis as in the interphase cell, when the valve was already formed. It is likely that such a difference is due to the small volume of cytoplasm in A. islandica cells compared to the previously studied C. granii.

Tubulins are highly conserved proteins, and the diatom α-tubulin sequence fragment (426-450 а.a.) is identical to the homologous sequence of chicken tubulin, from which the antibodies used were obtained. Thus, although the α-tubulin sequence of A. islandica is not available, there is no doubt about the identity of the localized protein (Fig. 2).

The structure of mature valves can be described using electron microscopy; however, different stages of valve morphogenesis remain mostly inaccessible because forming valves are closed by girdle bands during the cell cycle for research unless the cells are treated with harsh reagents to remove organic matter. Due to the short exposure time of gatastatin (48 hours), most of the valves formed under its influence remain hidden for examination using SEM. The use of intravital fluorescent dyes and laser scanning microscopy enables to observe the developing valves as 3D-reconstructions. This method allows to determine accurately the valves formed specifically in the current experiment.

Previously it was shown that under the influence of colchicine cells of A. islandica formed valves with the only center of symmetry, which was the cause of the failed cytokinesis. In this case, the valve morphogenesis occurs and new forming valve is positioned like a giant girdle band (Bedoshvili et al., 2018). Colchicine did not cause high cell mortality at the high concentration of 20 and 40 µg/mL, unlike gatastatin. The results of the study show that the effect of gatastatin caused not so much the formation of abnormal valves but rather a stop in cell division, accompanied by an arrest of the valve morphogenesis. Furthermore, under the influence of gatastatin, lateral valve morphogenesis did not occur indicating that valve formation in A. islandica was impossible without the participation of the microtubule center.

It is known that gatastatin is able to bind not only to γ-tubulin, but also to α- and β-tubulin, although, the dissociation constant with the latter is more than ten times higher (Chinen et al., 2015). It is obvious that anomalies in the structure of the valves are precisely associated with the ability of gatastatin to bind to all types of tubulins, and the formation of spines in A. islandica occurs throughout the entire morphogenesis of the valve; therefore, they are the most sensitive structure to various aberrations (Bedoshvili et al., 2018).

5. Conclusion

Specific staining enabled to localize α-tubulin in Aulacoseira islandica cells during valve morphogenesis and in interphase. The microtubule nucleation center was visualized in close association with the nucleus, which was a characteristic of a microtubule center. Inhibition of γ-tubulin using the specific inhibitor gatastatin showed a decrease in the number of valve-forming cells suggesting that a properly functioning microtubule center was required to initiate valve morphogenesis. Thus, the results obtained suggest that the microtubule center is an important structure not only for cell division, but is also necessary for the onset of valve morphogenesis.

Acknowledgements

This research was funded by the Russian Science Foundation, grant number 22-24-00080. The microscopy studies were performed at the Electron Microscopy Center of the Shared Research Facilities “Ultramicroanalysis” of Limnological Institute, https://www.lin.irk.ru/copp/.

Sobre autores

Ye. Bedoshvili

Limnological Institute of the Siberian Branch of the Russian Academy of Sciences

Autor responsável pela correspondência
Email: bedoshvilied@list.ru
Rússia, Ulan-Batorskaya Str., 3, Irkutsk, 664033

E. Bayramova

Limnological Institute of the Siberian Branch of the Russian Academy of Sciences

Email: bedoshvilied@list.ru
Rússia, Ulan-Batorskaya Str., 3, Irkutsk, 664033

Yu. Zakharova

Limnological Institute of the Siberian Branch of the Russian Academy of Sciences

Email: bedoshvilied@list.ru
Rússia, Ulan-Batorskaya Str., 3, Irkutsk, 664033

Bibliografia

Arquivos suplementares