J. Microbiol. Biotechnol. (2013), 23(3), 289–296 http://dx.doi.org/10.4014/jmb.1203.03057 First published online November 24, 2012 pISSN 1017-7825 eISSN 1738-8872

Improved Methodology for Identification of : Combining Light Microscopy and PCR Amplification

Xia, Shuang1,2, Yingyin Cheng3, Huan Zhu1,2, Guoxiang Liu1*, and Zhengyu Hu1

1Key Laboratory of Algal Biology, Institute of Hydrobiology, Chinese Academy of Sciences, Wuhan 430072, China 2Graduate School of Chinese Academy of Sciences, Beijing 100039, China 3Center for Water Environment and Human Health, Institute of Hydrobiology, Chinese Academy of Sciences, Wuhan 430072, China Received: March 26, 2012 / Revised: August 6, 2012 / Accepted: October 16, 2012

Cryptomonads are unicellular, biflagellate algae. Generally, Key words: , fixative, glutaraldehyde, Lugol’s cryptomonad cells cannot be preserved well because of solution, morphology, PCR amplification their fragile nature, and an improved methodology should be developed to identify cryptomonads from natural habitats. In this study, we tried using several cytological Cryptomonads are unicellular, biflagellate algae, which are fixatives, including glutaraldehyde, formaldehyde, and cosmopolitan in distribution, living as important primary their combinations to preserve field samples collected producers in both freshwater and marine habitats [4, 15, from various waters, and the currently used fixative, 25, 26, 32, 37]. They often assume dominant phytoplankton Lugol’s solution was tested for comparison. Results status in various waters [27], and sometimes even form showed that among the fixatives tested, glutaraldehyde nuisance blooms [1, 9, 34]. However, the cell density of preserved the samples best, and the optimal concentration cryptomonads was low in preserved samples and their of glutaraldehyde was 2%. The cell morphology was well importance was often neglected. Cryptomonads cells cannot preserved by glutaraldehyde. Cells kept their original color, be preserved well since their cells either rupture or distort volume, and shape, and important taxonomic features drastically when fixatives such as formaldehyde are added such as furrow/gullet complex, ejectosomes, as well as [25]. This is probably because their cell shape is maintained flagella could be observed clearly, whereas these organelles by a delicate surface termed periplast, which is probably frequently disappeared in Lugol’s solution preserved proteinaceous [5]. samples. The osmotic adjustments and buffers tested could Ever since the 1980s, the of Cryptophyta has not preserve cell density significantly higher. Statistical been drastically changing, with new genera set up and old calculation showed the cell density in the samples preserved ones eliminated [7, 18, 39]. Specifically, when phylogenetic by 2% glutaraldehyde remained stable after 43 days of analyses based on DNA sequences were incorporated into the fixation procedure. In addition, DNA was extracted taxonomic studies, new hypotheses on the taxonomy of from glutaraldehyde preserved samples by grinding with Cryptophyta were raised [6, 11, 19-21]. However, those liquid nitrogen and the 18S rDNA sequence was amplified works were focused on a relatively small group collected by PCR. The sequence was virtually identical to the and cultured in a few western counties. There are numerous reference sequence, and phylogenetic analyses showed unrecognized genera and to be described in the very close relationship between it and sequences from the natural habitats [2, 35]. same organism. To sum up, the present study demonstrated At the present time, Lugol’s solution is routinely used to that 2% unbuffered glutaraldehyde, without osmotic preserve cryptomonads samples from natural habitats adjustments, can preserve cryptomonads cells for identification, because it provides the least distortion of cell shape [27]. in terms of both light microscopy and phylogenetic analyses However, Lugol’s solution has two obvious disadvantages based on DNA sequences. in the morphological study. For one thing, it colors cells purple because of the iodine in it, but cell color is a main *Corresponding author Phone: +86-027-68780576; Fax: +86-027-68780123; generic feature of cryptomonads; for another, some important E-mail: [email protected] morphological features of cryptomonads, such as ejectosomes 290 Xia et al. and flagella, frequently disappear in samples fixed with The osmotic adjustments of sucrose and sorbitol as well as Lugol’s solution. More recently, PCR were successfully the buffer of PBS and Hepes were tested. Different performed on protists and microalgae from plankton samples methods of DNA extraction and PCR amplification were preserved in Lugol’s solution, but a complicated washing performed on preserved specimens. Phylogenetic analyses step with thiosulfate solution was needed to overcome were performed. Results of the present study clearly showed PCR inhibition caused by iodine, and there was no that glutaraldehyde without any buffer or osmotic adjustment information about DNA damage [3]. could preserve the morphology of cryptomonads well for In addition, experimenting on live cells collected from light microscopy at the final concentration of 2%, and cell natural habitats is often impractical, because the change of lysis due to fixation was negligible after 43 days of the temperature during transportation can distort or even fixation procedure. For the first time, DNA extraction disrupt the cells [27]. Thus, an improved fixation procedure and PCR amplification were successfully performed on proper for the identification of cryptomonads should be cryptomonads samples preserved with glutaraldehyde. developed. Formaldehyde is a conventional fixative for phytoplankton [44], whereas glutaraldehyde has been used widely as a MATERIALS AND METHODS fixative in electronic microscopy. More recently, glutaraldehyde was also utilized to preserve plankton Samples of seven cryptomonad species were used in the present samples for light microscopy because it was considered study. Samples were collected by phytoplankton nets from various to be able to preserve delicate cell structures [31, 48]. kinds of water bodies in Hubei Province of China from January 2009 to July 2012. Samples of Campylomonas reflexa were collected However, there was no record about fixation of cryptomonads o o from Lake Nanhu (30 29'46''N, 114 20'12''E) in the city of Wuhan. with glutaraldehyde for light microscopy and DNA tetrapyrenoidifera was collected from a small pond extraction from algae preserved with glutaraldehyde up till (30o29'09''N, 114o23'14''E) in a campus, where it had become a now. Besides, the combination of glutaraldehyde and dominant species. Cryptomonas pyrenoidifera was collected from a formaldehyde has been tried to preserve several algal eutrophic fishpond (30o31'47''N, 114o21'16''E), where it bloomed. samples, including Raphidophytes, and positive results Cryptomonas sp., , and sp. were were recorded [12, 24]. Thus, it may be practical to use collected from Lake Donghu (30o32'55''N, 114o21'16''E). glutaraldehyde or the combination of glutaraldehyde and sp. was collected from the reservoir of the Three Gorges Dam o o formaldehyde to preserve cryptomonad samples. (31 07'21''N, 110 47'01''E), where it bloomed and the water presented Moreover, buffer and osmotic adjustment are two a red-brown color. important factors that affect the preservation of cell Glutaraldehyde (25~28% solution, BR grade) and formaldehyde (37~40% solution, AR grade) was purchased (Sinopharm Chemical morphology during the fixation procedure [24]. PBS o Reageal Company, China) and stored at 4 C in a refrigerator. (phosphate-buffered saline) is a buffer solution commonly Lugol’s solution was prepared as follows: 5 g iodine (I2) and 10 g used in biological research. Hepes [N-(2-hydroxyethyl)- potassium iodide (KI) were added to 85 ml of distilled water. Then piperazine-N-(2-ethanesulfonic acid)] has been successfully the mixture was stirred with a magnetic bar for dissolution. PBS was used as a buffer for the fixation of algae [28]. Sorbitol and prepared as follows: 8 g sodium chloride (NaCl), 0.2 g potassium sucrose have been used to avoid breakage induced by chloride (KCl), 1.44 g disodium hydrogen phosphate (Na2HPO4), fixation, and sucrose has been applied to the fixation of and 0.24 g potassium dihydrogen phosphate (KH2PO4) were added algae [24]. to 800 ml of distilled water, and the pH was adjusted to 7.2 with In the present study, in order to develop a simple and hydrochloric acid (HCl). convenient procedure for the identification of cryptomonads, At first, samples of Campylomonas reflexa and Cryptomonas pyrenoidifera various fixatives, including glutaraldehyde, formaldehyde, were preserved using the combinations of glutaraldehyde and the combinations of them with different concentration and formaldehyde with different concentration ratios at the final concentration of 1% and 2%, respectively (Experiment A). The final ratios were tested to preserve field samples. The optimal concentrations of the combined fixatives tested are shown in Table 1. concentration of the fixative was found by gradient tests. The fixatives were prepared just before use. After 24 h of fixation, The currently used fixative, Lugol’s solution, was tested cells were observed under light microscopy. To evaluate the effects for comparison both qualitatively and quantitatively. Fixed of the fixation on cell morphology, undistorted cell bodies as well as samples were observed for 53 days to evaluate cell lysis. cells retaining flagella were counted. Live cells were counted for

Table 1. Final concentrations of fixatives (%) tested in Experiment A (using the combination of glutaraldehyde and formaldehyde with different concentration ratios to preserve cryptomonads samples). Group A1 B1 C1 D1 E1 F1 G1 A2 B2 C2 D2 E2 F2 G2 Formaldehyde 1.0 0.9 0.7 0.5 0.3 0.1 0 2.0 1.8 1.4 1.0 0.6 0.2 0 Glutaraldehyde 0 0.1 0.3 0.5 0.7 0.9 1.0 0 0.2 0.6 1.0 1.4 1.8 2.0 IMPROVED METHODOLOGY FOR IDENTIFICATION OF CRYPTOMONADS 291 comparison before the fixation procedure. When counting live cells, E.Z.N.A. Gel Extraction Kit (Omega, USA), and then cloned into low melting-point agarose was added to prevent cells from swimming. the pMD18-T vector (Takara, China). Clones were sequenced by One-way analysis of variance (ANOVA) was carried out to test the universal sequencing primer M13. Sequence was deposited in difference in cell densities among the treatments with a discrimination GenBank under the Acession No. JX453451. level of p < 0.05. Nuclear 18S rDNA sequences of putative relatives were downloaded Then, samples of Cryptomonas tetrapyrenoidifera were preserved from GenBank. Sequences were aligned with ClustalX (ver. 1.83) using glutaraldehyde at different final concentrations and Lugol’s [43]. The alignments were refined manually by MEGA (ver. 4.0) solution at final concentration of 1% (Experiment B). The final [42], and phylogenetic trees were constructed using the neighbor- concentrations of glutaraldehyde in the samples were, respectively, joining (NJ) method in the same software. Bootstrap analyses with 0.1%, 0.5%, 1.0%, 2.0%, and 4.0%. Before and after fixation, the 1,000 replicates were calculated to estimate statistical reliability. length and width of 50 cells were measured. ANOVA was carried In order to confirm that this fixation procedure can be applied to out to test the difference before and after fixation with a other natural cryptomonad populations from various kinds of habitats, discrimination level of p < 0.05. After 24 h of fixation, undistorted the samples of Cryptomonas sp., Komma caudata, Chroomonas sp., cell bodies and cells retaining flagella were counted. To evaluate and Plagioselmis sp. collected from various habitats were preserved cell lysis caused by fixation, the cells were observed and counted with glutaraldehyde at the final concentration of 2%. Cell morphology for 53 days. ANOVA was carried out to test the difference in cell was observed under light microscopy. densities among the days. The effects of buffers and osmotic adjustments were evaluated (Experiment C). Two buffers were tested: PBS (final concentration RESULTS of 0.1 M, pH 7.2) and HEPES (Calbiochem; final concentration of 30 mM). Two osmotic adjustments were tested: sorbitol and sucrose The results of Experiment A showed that glutaraldehyde (Sinopharm Chemical Reageal Company; saturated). The fixatives preserved cell morphology the best among the fixatives were prepared just before use. The samples of Campylomonas reflexa were used. After 24 h of fixation, cells were observed and tested (Figs. 1, 2). In groups A1 and A2, there were no counted under light microscopy. ANOVA was carried out to test the cryptomonads cells observed, which indicated that difference in cell densities among the treatments with a formaldehyde caused severe cell loss. When fixed with the discrimination level of p < 0.05. combination of formaldehyde and glutaraldehyde, the For extraction of the total DNA and PCR amplification higher the concentration of glutaraldehyde used, the more (Experiment D), specimens of Cryptomonas pyrenoidifera were intact cells were counted. There was no significant o preserved with 2% glutaraldehyde and stored at 4 C. Three days difference between the numbers of undistorted cell bodies later, pellets of cells were harvested by centrifugation, washed in in group F1 and groups G1, F2, and G2 (p > 0.05), but in ddH2O 3 times, and ground in liquid nitrogen. Then the total DNA groups G1 and G2, the number of cells retaining flagella was extracted using the E.Z.N.A. Plant DNA Kit (Omega, USA) was significantly higher than that in groups F1 and F2 following the manufacturer’s instructions. (p < 0.05). In groups G1 and G2, 86.5 ± 4.2% and 92.0 ± To amplify the nuclear 18S rDNA sequence, we used the combination of the primers EK82f and Proto5r [3]. The PCR 5.1% (mean ± standard deviation) of cells retained flagella, program started with 5 min at 94oC, followed by 35 cycles of 1 min whereas in groups F1 and F2, 72.4 ± 8.8% and 84.3 ± 5.0% at 94oC, 1min at 55oC, and 1.5 min at 72oC, ending with a final of the cells did. Cells that retained flagella in other groups hold of 10 min at 72oC. All PCR amplicons were cleaned using the were even fewer. The cell density in samples fixed with

Fig. 1. Cell density of Campylomonas reflexa before and after Fig. 2. Cell density of Cryptomonas pyrenoidifera before and fixation with combinations of glutaraldehyde and formaldehyde. after fixation with combinations of glutaraldehyde and formaldehyde. Columns and error bars indicate mean and standard deviation, respectively. Columns and error bars indicate mean and standard deviation, respectively. 292 Xia et al.

Fig. 3. Cell density of Cryptomonas tetrapyrenoidifera fixed with different concentrations of glutaraldehyde and 1% Lugol’s solution. GA = glutaraldehyde. Columns and error bars indicate mean and standard deviation, respectively.

1% glutaraldehyde was 52.2 ± 2.5% of live cells, whereas the cell density in samples fixed with 2% glutaraldehyde Fig. 4. Micrographs of Campylomonas reflexa and Cryptomonas tetrapyrenoidifera. was 96.5 ± 5.1%. E = ejectosomes; F = flagella. Scale Bars = 10 µm. A, B, C show The results of Experiment B showed that, after 24 h of Campylomonas reflexa; D, E, F show Cryptomonas tetrapyrenoidifera; A, fixation, the cell density in the Cryptomonas tetrapyrenoidifera D show live cells; B, E show cells fixed with glutaraldehyde; C, F show samples fixed by 2% glutaraldehyde was significantly cells fixed with Lugol’s solution. higher than that in samples fixed by glutaraldehyde with other concentrations (p<0.05), and there was no significant difference between the cell density in samples fixed with colored cells purple, and few flagella or ejectosomes could 2% glutaraldehyde and in samples fixed with Lugol’s be observed (Figs. 4C, 4F). solution (p>0.05). However, the number of cells retaining The cell density in the samples fixed with 2% glutaraldehyde flagella in samples fixed with 2% glutaraldehyde was and Lugol’s solution was observed for 53 days. In the significantly higher than that in samples fixed with samples fixed with 2% glutaraldehyde, there was no Lugol’s solution (p<0.05). In the samples fixed with 2% significant difference in cell density (p > 0.05) until 43 glutaraldehyde, 88.4 ± 3.6% of cells retained flagella, whereas days after the fixation procedure. After 53 days, the cell in the Lugol’s solution fixed samples, only 30.3 ± 2.0% of density was significantly lower than that at the beginning cells did (Fig. 3). Experiment B also revealed that glutaraldehyde preserved cryptomonads morphology better than the currently used fixative, Lugol’s solution. The cell length of Cryptomonas tetrapyrenoidifera was 24.8 ± 1.87 µm and 24.3 ± 1.96 µm before and after fixation, and cell width was 13.5 ± 1.34 µm and 13.0 ± 1.06 µm respectively. There was no significant difference between the cell length and width measured before and after fixation in glutaraldehyde-preserved samples (p < 0.05), which indicated that glutaraldehyde caused neither swelling nor shrinkage of cell volume. In glutaraldehyde-fixed samples, Campylomonas reflexa and Cryptomonas tetrapyrenoidifera kept their original olive green color, and two flagella could be observed (Figs. 4B, 4E). Ca. reflexa kept its characteristic sigmoid cell shape with a recurved posterior. Several rows of ejectosomes lay Fig. 5. Changes in cell density of Cryptomonas tetrapyrenoidifera in the ventral anterior (Fig. 4B). Specific features of Cr. after fixation. The white circles show the cell density in samples fixed by 1% Lugol’s tetrapyrenoidifera, four prominent shelled pyrenoids, solution, and black circles show that by 2% glutaraldehyde. The error bars could be seen clearly. Two rows of ejectosomes lined show the standard deviation. Lines were obtained by smoothing spline beside the furrow (Fig. 4E). In contrast, Lugol’s solution regression analysis. IMPROVED METHODOLOGY FOR IDENTIFICATION OF CRYPTOMONADS 293

The results of Experiment C demonstrated that when the fixative was buffered by PBS, cell precipitation occurred, which made counting impracticable. There was no significant difference between the cell density in samples fixed with the buffer of Hepes and in samples fixed without buffer (p > 0.05) (Fig. 6). Moreover, the osmotic adjustments of sorbitol and sucrose caused more cell loss (Fig. 6). The cell densities of samples added with sorbitol and sucrose were 62.5 ± 8.3% and 76.3 ± 7.4% of that in the samples without osmotic adjustments, respectively. In other field samples preserved with 2% glutaraldehyde, the cell morphology of cryptomonads could be observed Fig. 6. Cell density of Campylomonas reflexa samples added clearly under light microscopy (Fig. 7). Komma caudata with different buffers and osmotic adjustments. was blue-green, comma shaped, with a central pyrenoid Columns and error bars indicate mean and standard deviation respectively. and two subequal flagella; the longer was about 2/3 of the cell length. Plagioselmis sp. has a red-brown (p < 0.05), and 86.0 ± 1.9% cells were left. As to the samples chloroplast, and an acute posterior end; the flagella were a fixed with Lugol’s solution, cell loss was negligible. The little longer than the cell body. Chroomonas sp. was blue- cell density after 53 days of fixation was not significantly green, with a round posterior and a prominent shelled lower than that at the beginning (p > 0.05), and it was pyrenoid; the longer flagellum was nearly equal to the cell significantly higher than that in samples fixed with 2% length. Cryptomonas sp. was brown in color, with a round glutaraldehyde after 53 days (p < 0.05) (Fig. 5). posterior and a rostrate anterior; the flagella were about half of the cell length. The 18S rDNA sequences of Cryptomonas pyrenoidifera amplified from samples preserved with glutaraldehyde and from live cells were 1,548 bp long, and only one site was different between them. A T-C mutation happened. Most phylogenies reconstructed from the 18S rDNA dataset had strongly defined backbone topologies and

Fig. 8. Phylogenetic tree of 18S rDNA sequences of cryptomonads Fig. 7. Micrographs of cryptomonads fixed with 2% glutaraldehyde. and relatives. Scale bars: A, 10 µm; B, C, D, 5 µm. A: Cryptomonas sp.; B: Komma Bootstrap support from neighbor joining (NJ) are presented on the nodes. caudata; C: Plagioselmis sp.; D: Chroomonas sp. Values above 50 are shown. The sequence obtained in our study is shaded gray. 294 Xia et al. well-supported internal clades (Fig. 8). In the 18S rDNA rates of 0.0% to less than 0.1%, the least among cytological phylogenetic trees, species of the Cryptomonas pyrenoidifera fixatives [13]. In the present study, we tried several formed a well-supported group (99 for NJ), with the methods to extract the total DNA from samples preserved sequence obtained from the present study included. They with glutaraldehyde (data not shown), such as the CTAB represented a sister lineage to C. marssonii. method [14], digestion with proteinase K, grinding with liquid nitrogen, and several commercial DNA extraction kits. Only the method with grinding in liquid nitrogen DISCUSSION succeeded. This may also be because the rigid denatured periplast, which was composed of numerous cross-linked More than 10 kinds of fixatives with different concentrations proteins, was hard to disrupt by chemicals like CTAB or were tested to preserve cryptomonads samples in this proteinase K, and only physical grinding achieved it. The study, such as paraformaldehyde, chloroform, acetone, DNA sequence amplified from glutaraldehyde-preserved ethanol, glacial acetic acid, methanol, ether, picronitric specimens was virtually identical to the sequence amplified acid, and different combinations of them (data not shown), from live cells (one mutation in 1,548 nucleotides sequenced), but none gave better results than glutaraldehyde with final which corresponded with previous works [13]. The concentration of 2%. phylogenetic analyses showed it was very close to other Currently, fixation of cryptomonads is carried out 18S rDNA sequences of the same organism. Thus, the mostly with Lugol’s solution [27], and its main disadvantages DNA sequence amplified from glutaraldehyde-preserved have been mentioned above. Meanwhile, fixation of specimens can be used reliably in phylogenetic analyses. phytoplankton has been performed mainly with formaldehyde Cell density in the samples fixed by 2% glutaraldehyde for qualitative study [44], which is destructive to was stable until 43 days after the fixation procedure. cryptomonads cells, as shown in this study and in previous Accordingly, for short-term taxonomical study, glutaraldehyde works [26]. This study clearly demonstrated that fixation is an ideal fixative for cryptomonads. However, glutaraldehyde with 2% glutaraldehyde was practical for microscopy. gradually volatilized, and its concentration in the samples First, the cell shapes were not altered, and neither swelling changed, which probably caused cells lyses in the samples nor shrinking of cell volume happened; second, the cell preserved with glutaraldehyde some days after the fixation kept its original color; and finally important taxonomic procedure. In contrast, samples fixed by Lugol’s solution features such as ejectosomes, furrow/gullet, and flagella could hold on longer (at least 53 days in our experiment). were retained. Besides, glutaraldehyde was cheaper than It is believed that cells preserved in Lugol’s solution may Lugol’s solution when used to preserve samples of the remain intact for more than two years [22]. Hence, for long- same volume according to our calculations. term study of identified cryptomonads, Lugol’s solution is When used as a fixative, glutaraldehyde caused certain still preferable. kinds of algal cells and ciliates to shrink or swell in previous Besides cryptomonads, this fixation procedure was also works [28, 31, 41, 48], but the change of cell volume was tested on other Raphidophytes in the present study, some not observed in the present study. This is probably because of which do not have a cellulose cell wall and cannot the cell shape of cryptomonads is maintained by a be preserved well by formaldehyde. Samples of Euglena proteinaceous periplast [5]. The -CHO groups of glutaraldehyde sp. (Euglenophyta), Trachelomonas sp. (Euglenophyta), combined with protein nitrogens and caused cross-linking Peridinopsis sp. (Pyrrophyta), and Gymnodinium eucyaneum [33], making the periplast rigid. (Pyrrophyta) were preserved by glutaraldehyde at final An interesting phenomenon in our experiment was that concentration of 2%, and their taxonomic features could be the osmotic adjustments of sorbitol and sucrose caused observed clearly in preserved samples. Accordingly, it was more cell loss. The possible reason was the chemical probably practical that glutaraldehyde be used for the reaction of glutaraldehyde was relatively slow (minutes to fixation of delicate Raphidophyte cells, which may be hours) [36]. When sorbitol and sucrose penetrated into the distorted or disrupted by formaldehyde. However, glutaraldehyde cells, which were still osmotically responsive, dehydration is volatile and toxic. Effects from exposure include skin occurred, which results in the rupture of the periplast. sensitivity resulting in dermatitis, and irritation of the eyes Previous studies have proved that DNA can be extracted and nose with accompanying rhinitis [8, 17, 23, 46]. It is and amplified from samples of microalgae and aquatic recommended that operation be done in a fume cupboard. animals preserved with formaldehyde [16, 29, 38, 40, 45], but a relatively high rate of DNA damage was reported [10, 13, 47], whereas very few works have focused on Acknowledgments DNA extraction and amplification from glutaraldehyde- preserved samples. Glutaraldehyde treatments were reported This work was financially supported by the National Natural to produce DNA damage in tissues of plants and fungi at Science Foundation of China (Grant No. 30970501). We IMPROVED METHODOLOGY FOR IDENTIFICATION OF CRYPTOMONADS 295 appreciate the staff of the Center for Algal Biology and species identification: Studies on fixatives and algal growth Applied Research, Institute of Hydrobiology, Chinese stages. Harmful Algae 1: 375-382. Academy of Sciences, for their cordial and helpful suggestions. 17. Hansen, K. S. 1983. Glutaraldehyde occupational dermatitis. Contact Derm. 9: 81-82. 18. Hill, D. R. A. 1991. A revised circumscription of Cryptomonas () based on examination of Australian strains. REFERENCES Phycologia 30: 170-188. 19. Hoef-Emden, K. 2007. Revision of the Cryptomonas 1. Andreoli, C., C. Tolomio, N. Rascio, and R. Talarico. 1986. (Cryptophyceae) II: Incongruences between the classical Some observations on a Cryptophyceae responsible for a winter morphospecies concept and molecular phylogeny in smaller red bloom. G. Bot. Ital. 120: 70-71. pyrenoid-less cells. Phycologia 46: 402-428. 2. Andersen, R. A. 1992. Diversity of eukaryotic algae. Biodivers. 20. Hoef-Emden, K., B. Marin, and M. Melkonian. 2002. Nuclear Conserv. 1: 267-292. and SSU rDNA phylogeny in the Cryptophyta and 3. Auinger, B. M., K. Pfandl, and J. Boenigk. 2008. Improved the evolution of cryptophyte diversity. J. Mol. Evol. 55: 161-179. methodology for identification of protists and microalgae from 21. Hoef-Emden, K. and M. Melkonian. 2003. Revision of the plankton samples preserved in Lugol’s iodine solution: genus Cryptomonas (Cryptophyceae) I: A combination of Combining microscopic analysis with single-cell PCR. Appl. molecular phylogeny and morphology provides insights into a Environ. Microbiol. 74: 2505-2510. long hidden dimorphism. Protist 154: 371-409. 4. Barone, R. and L. Naselli-Flores. 2003. Distribution and 22. Hötzel, G. and R. Croome. 1999. A Phytoplankton Methods seasonal dynamics of cryptomonads in Sicilian water bodies. Manual for Australian Freshwaters. Land and Water Resources Hydrobiologia 502: 325-329. Research and Development Corporation, Green Words & Images, 5. Brett, S. J., L. Perasso, and R. Wetherbee. 1994. Structure and Canberra. development of the cryptomonad periplast: A review. Protoplasma 23. Jordan, W. P., M. V. Dahl, and H. L. Albert. 1972. Contact 181: 106-122. dermatitis from glutaraldehyde. Arch. Dermatol. 105: 94-95. 6. Cavalier-Smith, T., J. A. Couch, K. E. Thorsteinsen, P. Gilson, 24. Katano, T., M. Yoshida, J. Lee, M. S. Han, and Y. Hayami. J. A. Deane, D. R. A. Hill, and G. I. McFadden. 1996. 2009. Fixation of Chattonella autiqua and C. marina Cryptomonad nuclear and nucleomorph 18S rRNA phylogeny. (Raphidophyceae) using Hepes-buffered paraformaldehyde and Eur. J. Phycol. 31: 315-328. glutaraldehyde for flow cytometry and light microscopy. 7. Clay, B. L., P. Kugrens, and R. E. Lee. 1999. A revised Phycologia 48: 473-479. classification of Cryptophyta. Bot. J. Lin. Soc. 131: 131-151. 25. Klaveness, D. 1988. Ecology of the Cryptomonadida: A first 8. Corrado, O. J., J. Osman, and R. J. Davies. 1986. Asthma and review. pp. 105-133. In C. D. Sandgren (ed.). Growth and rhinitis after exposure to glutaraldehyde. Hum. Toxicol. 5: 325- Reproductive Strategies of Freshwater Phytoplankton. Cambridge 328. University Press, New York. 9. Dame, R., M. Alber, D. Allen, M. Mallin, C. Montague, A. 26. Klaveness, D. 1988. Biology and ecology of the Cryptophyceae: Lewitus, et al. 2000. Estuaries of the South Atlantic coast of Status and challenges. Biol. Oceanogr. 6: 257-270. North America: Their geographical signatures. Estuar. Coast. 27. Kugrens, P. and B. L. Clay. 2003. Cryptomonads, pp. 715-755. 23: 793-819. In J. D. Wehr and R. G. Sheath (eds.). Freshwater Algae of 10. De Giorgi, C., M. F. Sialer, and F. Lamberti. 1994. Formalin- North America. Academic Press, San Diego. induced infidelity in PCR-amplified DNA fragments. Mol. Cell. 28. Leakey, R. J. G., P. K. Burkhill, and M. A. Sleigh. 1994. A Probes 8: 459-462. comparion of fixatives for the estimation of abundance and 11. Deane, J. A., I. M. Strachan, G. M. Saunders, D. R. A. Hill, and biovolume of marine planktonic ciliate populations. J. Plankton G. I. McFadden. 2002. Cryptomonad evolution: Nuclear 18S Res. 16: 375-389. rDNA phylogeny versus cell morphology and pigmentation. J. 29. Marín, I., A. Aguilera, B. Reguera, and J. P. Abad. 2001. Phycol. 38: 1236-1244. Preparation of DNA suitable for PCR amplification from fresh 12. Doughty, M. J., J. P. Bergmanson, and Y. Blocker. 1995. Impact or fixed single dinoflagellate cells. Biotechniques 30: 88-90. of glutaraldehyde versus glutaraldehyde-formaldehyde fixative 30. McFadden, G. I. and M. Melkonian. 1986. Use of Hepes buffer on cell organization in fish corneal epithelium. Tissue Cell 27: for microalgal culture media and fixation for electron microscopy. 701-712. Phycologia 25: 551-557. 13. Douglas, M. P. and S. O. Rogers. 1998. DNA damage caused 31. Menden-Deuer, S., E. J. Lessard, and J. Satterberg. 2001. Effect by common cytological fixatives. Mutat. Res. 401: 77-88. of preservation on dinoflagellate and diatom cell volume and 14. Doyle, J. J. and E. E. Dickson. 1987. Preservation of plant consequences for carbon biomall predictions. Mar. Ecol. Prog. samples for DNA restriction endonuclease analysis. Taxon 36: Ser. 222: 41-50. 715-722. 32. Menezes, M. and G. Novarino. 2003. How diverse are 15. Gillot, M. 1990. Phylum Cryptophyta (Cryptomonads), pp. planktonic cryptomonads in Brazil? Advantages and difficulties 139-151. In L. Margulis, J. O. Corliss, M. Melkonian, and D. of a taxonomic-biogeographical approach. Hydrobiologia 502: J. Chapman (eds.). Handbook of Protoctista. Jones Bartlett 297-306. Publishers, Boston. 33. Monsan, P., G. Puzo, and H. Marzarguil. 1975. Etude du 16. Godhe, A., D. M. Anderson, and A. Rehnstam-Holm. 2002. mecanisme d’etablissement des liaisons glutaraldehyde-proteines. PCR amplification of microalgal DNA for sequencing and Biochimie 57: 1281-1292. 296 Xia et al.

34. Nishijima, T. 1990. Growth characteristics of Plagioselmis sp. 42. Tamura, K., J. Dudley, M. Nei, and S. Kumar. (2007). MEGA4: (strain 87) causing freshwater red tide in the lower part of the Molecular evolutionary genetics analysis (MEGA) software Nakasuji River, Japan. Nippon Suisan Gakkaishi. 56: 353-359. version 4.0. Mol. Biol. Evol. 24: 1596-1599. 35. Novarino, G. A. 2003. Companion to the identification of 43. Thompson, J. D., T. J. Gibson, F. Plewniak, and D. G. Higgins. cryptomonad (Cryptophyceae, cryptomonadea). 1997. The CLUSTAL_X Windows interface: Flexible strategies Hydrobiologia 502: 225-270. for multiple sequence alignment aided by quality analysis tools. 36. Paljarvi, L., J. H. Garcia, and H. Kalimo. 1979. The efficiency Nucleic Acids Res. 25: 4876-4882. of aldehyde fixation for electron microscopy: Stabilization of rat 44. Throndsen J. 1978. Preservation and storage, pp. 69-74. In A. brain tissue to withstand osmotic stress. Histochem. J. 11: 267- Sournia (ed.). Phytoplankton Manual. United Nations Educational, 276. Scientific and Cultural Organization, Paris. 37. Reynolds, C. S. 1978. Notes on the phytoplankton periodicity of 45. Voo, K. S. and M. Britta. 1998. Extraction and amplifi-cation of Rostherne Mere, Cheshire, 1967-1977. Br. Phycol. J. 13: 329- mitochondrial DNA from formalin-fixed deep-sea mollusks. 335. Biotechniques 24: 243-247. 38. Richlen, M. and P. Barber 2005. A technique for the rapid 46. Wiggins, P., S. A. McCurdy, and W. Zeidenberg. 1989. Epistaxis extraction of microalgal DNA from single live and preserved due to glutaraldehyde exposure. J. Occup. Med. 31: 854-856. cells. Mol. Ecol. Notes 5: 688-691. 47. Williams, C., F. Pontén, C. Moberg, P. Söderkvist, M. Uhlén, J. 39. Santore, U. J. 1984. Some aspects of taxonomy in the Pontén, et al. 1999. A high frequency of sequence alterations is Cryptophyceae. New Phycol. 98: 627-646. due to formalin fixation of archival specimens. Am. J. Pathol. 40. Shiozawa, D., J. Kudo, R. P. Evans, S. R. Woodward, and R. N. 155: 1467-1471. Williams. 1992. DNA extraction from preserved trout tissues. 48. Woelfl, S. and B. A. Whitton. 2000. Sampling, preservation and West. N. Am. Naturalist 52: 29-34. quantification of biological samples from highly acidic 41. Stoecker, D. K., D. J. Gifford, and M. Putt. 1994. Preservation environments (pH≤3). Hydrobiologia 433: 173-180. of marine planktonic ciliates: Losses and cell shrinkage during fixation. Mar. Ecol. Prog. Ser. 110: 293-299.