Sexual reproduction of the snow alga Chloromonas fukushimae (Volvocales, Chlorophyceae) induced using cultured materials

Snow algae are microalgae, growing in melting snowpacks, and are thought to act as primary producers in the snow ecosystem. Chloromonas (Volvocales, Chlorophyceae) contains more than 15 snow-inhabiting species. Although vegetative cells and zygotes, or asexual cysts, of snow species of the genus are frequently collected in the field, sexual reproduction and zygote formation in culture have only been induced in C. tughillensis. Here we describe the sexual reproduction of another snow-inhabiting species, C. fukushimae, which was induced using both previously examined and newly established Japanese strains. Mating of isogamous gametes began after mixing two different strains, implying that C. fukushimae is an outcrossing species. Motile and nonmotile zygotes of the species were also described in this report. The nonmotile zygote of C. fukushimae was distinguishable from those of the other snow-inhabiting species of Chloromonas, based on the zygote shape and the presence of several large lipid bodies within the cell. In addition, C. fukushimae carried out sexual reproduction and produced zygotes even under the nitrogen-sufficient condition.


Introduction
Snow algae, microalgae growing in melting snowpacks, are thought to act as primary producers in the snow ecosystem [1,2]. Chloromonas (Volvocales, Chlorophyceae), a unicellular biflagellate genus, includes more than 15 snow-inhabiting species in addition to approximately 130 mesophilic species [2,3]. Before the snowpack melts completely, the snow-inhabiting species form zygotes or asexual cysts that later enter a dormant state [1]. Mature zygotes and cysts accumulate orange carotenoid pigments within the cell, due to a protective reaction against excessive visible and ultraviolet light. The blooms of snow-inhabiting Chloromonas species color snow green if vegetative cells, planozygotes (motile zygotes), or immature cysts or aplanozygotes (nonmotile zygotes) dominate. The snow is stained orange, brown or reddish if mature aplanozygotes and/or cysts are dominant within the blooms. In Antarctica, red snow caused by vegetative cells of C. rubroleosa, which contain amount of red-pigmented oil droplets within the cell, was also reported [4]. These phenomena are called "colored snow" (S1 Fig). To date, more than one hundred strains of Chloromonas originating from snow are available from public culture collections [e.g., Culture Collection of Algae at the University of Texas at Austin [5], Culture Collection of Cryophilic Algae [6], and Microbial Culture Collection at the National Institute for Environmental Studies (NIES) [7]]. Recently, several snowinhabiting species of Chloromonas, such as C. fukushimae, were described based on a combination of comparative morphological analysis and molecular data of cultured materials (i.e., polyphasic approaches) [8][9][10][11]. However, for most species, there is insufficient information about their aplanozygotes because sexual reproduction (including aplanozygote formation) in culture was only accomplished in C. tughillensis [12]. Chloromonas tughillensis is an outcrossing species. Its sexual reproduction was induced mixing different mating types (mt+ and mt−) that were maintained under nitrogen-depletion conditions for almost three weeks and placed under blue light during the final seven days of growth [12]. Nitrogen-depletion and blue light are thought to be classic gamete formation triggers in freshwater green algae [13,14]. The method also induced the sexual reproduction of a self-fertilizing snow species, C. chenangoensis [12]. Despite these successful inductions, aplanozygotes were not observed in the experiment. On the other hand, recent molecular data identified the aplanozygotes, or asexual cysts, of the four snow-inhabiting species (C. hindakii, C. miwae, C. muramotoi, and C. krienitzii) from field-collected materials [9,11,15].
Here, we induced the sexual reproduction of C. fukushimae, using both previously examined and newly established strains. The mating of gametes and the morphologies of planozygote and aplanozygote in this species were described, and aplanozygote morphologies were compared between C. fukushimae and the other snow-inhabiting species of Chloromonas. Additionally, we compared the induction rates of the gametes and the planozygotes in the sexual reproduction of C. fukushimae under nitrogen-sufficient and -deficient conditions.

Ethics statement
We collected colored snow from snowpacks in Mt. Hakkoda in Towada-Hachimantai National Park and in Oze National Park. No specific permission was required because collecting snowpacks containing microalgae or other protists from national parks is not legally restricted in Japan. In addition, we confirmed the samples did not contain protected organisms.

Induction of sexual reproduction
Sexual reproduction was induced according to the method for a mesophilic green alga Chlamydomonas reinhardtii (Volvocales, Chlorophyceae) [18], with some modifications. For comparison, the following experiments were performed using AF-6 medium (including 22 mg NH 4 NO 3 L −1 and 140 mg NaNO 3 L −1 ) [7,17], and AFM medium [19] containing very little nitrogen (2.2 mg NaNO 3 L −1 ), respectively. Each seven-day-old culture in 10 mL of AF-6 medium was centrifuged at 5˚C and concentrated to approximately 0.2 mL. The culture was washed with 4 mL of AF-6 or AFM medium on ice. After re-concentration, each culture's cell density was equalized at 1.5 × 10 6 cells mL −1 ; this emulated maximum population density in field-collected samples [1,12]. A pair of 0.5 mL of each culture was mixed in a well of Cellstar 24-well culture plate (Greiner Bio-One, Kremsmünster, Austria) in a cold experimental room at 5˚C and maintained as described above. For continuous observation, mating experiments were carried out just after the dark period.
To assign mating types (mt+ and mt−) of the examined strains, we conducted the tunicamycin sensitivity test [20]. Before induction of sexual reproduction, cells in each culture were treated by 2.5 μg mL −1 tunicamycin (Sigma-Aldrich, Missouri, USA) for overnight. Sexual reproduction of treated and untreated (positive control) cultures were induced as described above. Based on the previous study [20], a tunicamycin-sensitive mating type was assigned to mt+.

Morphological observation and cell counting
Phase contrast microscopy of living cells was carried out in the cold experimental room, using a Zeiss Primovert inverted light microscope (Zeiss, Oberkochen, Germany) and a ScopePad-550 digital camera (Ouyan International, Shenzhen, China). Cells were fixed for counting, using 0.4% of Lugol's iodine solution, at one, two, three, four, six, eight, 10, and 12 hours after mixing the strains in the experiments started just after the dark period. Standard glutaraldehyde solutions were not used in this study since they cause the loss of flagella from cells during fixation. Using an Olympus CKX41 microscope (Olympus, Tokyo, Japan) at room temperature, the fixed cells were categorized as vegetative cells or gametes, autosporangia with daughter cells, mating pairs, and quadriflagellate planozygotes. In respective times, 400-cell counts were conducted according to the previous study [21]. The proportion of cell types among 100 mating pairs was also calculated at each time. We compared the induction rates and the proportion of cell types under the nitrogen-sufficient condition (AF-6) with those under the nitrogen-deficient condition (AFM), using Welch's t test. P < 0.01 was considered significant. In addition, light and fluorescent microscopy of fixed gametes and planozygotes, and living aplanozygotes were also performed, at room temperature, using an Olympus BX51 light microscope equipped with Nomarski differential interference optics and an Olympus DP72 digital camera (Olympus). To detect lipid bodies within the cell, the aplanozygotes were stained by 1 μg mL −1 Nile Red for more than 48 hours at 5˚C.

Sexual reproduction of Chloromonas fukushimae
The strains of C. fukushimae used in this study (excluding NIES-3389) showed an outcrossing feature (i.e., heterothallism) and were classified into two complementary mating types. One contained HkCl-106, HkCl-108, HkCl-113, HkCl-117, HkCl-121 and HkCl-125, and the other included OzCl-11 and NIES-3390 (S1 Table). The remaining strain, NIES-3389, did not cross with the other strains examined. Since mating type-specific genes such as MID in Chlamydomonas reinhardtii [22] have not been reported in the genus Chloromonas, we conducted the tunicamycin sensitivity test [20] to assign mt+ and mt−. Unilateral effect of inhibiting gamete interactions by tunicamycin was observed in some outcrossing species of Chlamydomonas [20] and snow-inhabiting Chloromonas tughillensis [12]. The test using the representative strains of the complementary mating types exhibited that the strain HkCl-117 was more sensitive to the inhibition of aplanozygote formation by tunicamycin treatment than the strain NIES-3390 (Table 1). Based on the previous results [20], the mating type including the strain HkCl-117 was assigned to mt+ in this study.
In the sexual reproduction of C. fukushimae, pairing isogamous gametes with entwined flagella ( Fig 1A) began two to three hours after mixing mt+ and mt− strains. The gametes were elongate-bean-shaped and morphologically resembled the vegetative cells [8]. The gametes measured 4.0-6.5 μm wide and 12.8-16.1 μm long. Each gamete lacked the eyespot. Autosporangium formation was not observed from the mixing of mt+ and mt− strains to the beginning of gamete pairing. Polyvalent gamete interactions (i.e., agglutination or clumps) was occasionally found during the experiment (Fig 1B), as in the sexual reproduction of some volvocalean green algae [23] such as mesophilic Chlamydomonas reinhardtii [14] and snow-inhabiting Chloromonas tughillensis [12]. During the sexual reproduction process, one of the gametes released its cell wall and became spherical (Fig 1C and 1D; S2 Fig). Then, the other gamete also cast off the cell wall and became spherical (Fig 1E and 1F). Such morphological change of gametes was also reported in snow species of Chloromonas [12] as well as in several Chlamydomonas species [3,14,24]. Mating papillae, or fertilization tubules, were not recognized in the gametes. Finally, the pair of two spherical gametes fused and formed a quadriflagellate planozygote (Fig 1G and 1H). The planozygotes were broad ellipsoidal or almost spherical, lacked an eyespot, and measured 9.5-12.5 μm wide and 10.6-15.7 μm long. Within a few days, the planozygotes lost their flagella and became aplanozygotes (Fig 1I-1L). The aplanozygotes were ellipsoidal, broad ellipsoidal or almost spherical with the smooth cell wall, and measured 12.3-19.1 μm wide and 14.4-22.5 μm long (Fig 1I and 1K). Several large lipid bodies were observed within the cell (Fig 1J and 1L). Within two months, the aplanozygotes accumulated orange carotenoid pigments and appeared mature (Fig 1M and 1N). Neither aplanozygote meiosis nor vegetative cell germination from the aplanozygotes were observed in this study.

Time scale of mating under different nitrogen conditions
Mating pairs (Fig 1A-1F) and planozygotes (Fig 1G and 1H) were observed under both nitrogen-sufficient (AF-6) and -deficient (AFM) conditions, after two to three or eight hours, respectively, from mixing mt+ and mt− strains (Fig 2A). At each time, the induction rates of mating pairs and planozygotes under the nitrogen-sufficient condition were not significantly different from those under the nitrogen-deficient condition. In addition, in each of the times, the proportion of cell types among 100 mating pairs under the nitrogen-sufficient condition did not significantly differ from that under the nitrogen-deficient condition (Fig 2B). Under both conditions, the rate of elongate gamete pairs (Fig 1A) gradually decreased from six hours after mixing strains. In contrast, the rate of spherical and elongate gamete pairs (Fig 1C and  1D) and that of two spherical gamete pairs (Fig 1E and 1F) increased from six and 10 hours, respectively, after the mix of strains.

Discussion
The present method for inducing sexual reproduction in C. fukushimae is much simpler than the previous method applied to C. tughillensis strains [12]. Although sexual reproduction of outcrossing C. tughillensis was previously induced by nitrogen starvation, our method was also able to induce sexual reproduction with aplanozygote formation in this species even under the nitrogen-sufficient condition (S3 Fig). Characteristics of the sexual reproduction in C. tughillensis observed in our experiment were consistent with those reported in the previous study [12]. Thus, this simple method is effective and may be applicable to other snow-inhabiting species of Chloromonas, including a self-fertilizing C. chenangoensis. For induction of sexual reproduction in snow-inhabiting C. chenangoensis and C. tughillensis, optimal conditions of cell density, light intensity, photoperiod, and spectral composition were investigated [12,21,25,26]. The previous studies demonstrated that blue light and longer photoperiods significantly increased the induction rates of sexual reproduction of both species. Since the main objective of this study is to describe the sexual reproduction of C. fukushimae, we did not seek such optimal conditions for the species. However, the present induction rates of mating pairs and planozygotes in C. fukushimae (Fig 2) seems not to be much different from those in C. tughillensis [12].
In this study, pairing of gametes in C. fukushimae was observed after two to three hours from mixing of mt+ and mt− strains (Fig 2). If gametogenesis was already induced prior to the experiment or the cells were constitutively gametic, it would be expected that such the gamete paring has occurred quickly after the mix of the strains. In the C. tughillensis strains maintained under nitrogen-starvation conditions for three weeks, pairing of gametes was observed after 30 minutes from the mix of mt+ and mt− strains [12]. Therefore, the time lag found in C. fukushimae suggests that the gametogenesis was induced by mixing with complementary mating type and/or increasing cell density. In several snow-inhabiting species of Chloromonas, gametes are thought to be produced via gametangium formation [12,27]. Since new gametangium formation or multiple fission was not observed from the mix of the strains to the beginning of gamete pairing, we considered that gametes of C. fukushimae were differentiated directly from vegetative cells after the induction. Nevertheless, it is difficult to completely rule out the possibility that a part of cells in the C. fukushimae strains is constitutively gametic or produces gametes via gametangium formation prior to the experiments. Besides, the gametes of this species are morphologically indistinguishable from the vegetative cells (i.e., no gametespecific morphological characters such as mating papillae). Therefore, further studies, such as the expression analysis focusing on homologues of gamete-specific genes in Chlamydomonas, are required to reveal when the gametogenesis actually occurred in C. fukushimae.
The C. fukushimae strain NIES-3389 did not cross with the other strains examined in this study (S1 Table). NIES-3389 cannot be distinguished from NIES-3390 (mt−) by morphological and molecular data [8]. We hypothesize that NIES-3389 lost the ability to reproduce sexually during subculturing. However, we cannot reject the possibility that C. fukushimae actually contains more than one sexually isolated group (i.e., syngen).
Among the snow-inhabiting species of Chloromonas, the aplanozygote of C. fukushimae resembles that of C. tughillensis in that it lacks cell wall ornamentations, such as flanges or spines, and has almost spherical cell shape [12]. However, C. fukushimae possessed multiple well-defined lipid bodies within the cell and C. tughillensis does not (Fig 1I-1L) (S3F Fig; [12]). The aplanozygotes of C. fukushimae and C. tughillensis, described based on cultured materials, were never reported from field-collected materials in previous studies. Thus, future studies of snow-inhabiting Chloromonas should confirm that experimentally cultured aplanozygotes are morphologically identical to those formed in natural environments.
As in some freshwater green algae, nitrogen-depletion is thought to be related to shifting phases in the life cycle of snow-inhabiting Chloromonas [1]. On the contrary, we demonstrated that the sexual reproduction of C. fukushimae (and C. tughillensis) could be induced even under nitrogen-sufficient conditions (Fig 2; S3 Fig). The induction rates of mating pairs and planozygotes, as well as the proportion of cell types among the mating pairs in the sexual reproduction of C. fukushimae, were similar between nitrogen-sufficient and nitrogen-deficient conditions (Fig 2). Although further research is required, bark and leaf litters on and near colored snow might provide nutrients, including nitrogen, for snow-inhabiting species of Chloromonas, during the snowmelt season [1,28]. Such litter is frequently observed in snowpacks in Japanese mountainous areas. Therefore, sexual reproduction, regardless of nitrogen conditions, seems advantageous for outcrossing species of snow-inhabiting Chloromonas in temporal snowpacks since they must produce resting zygotes quickly before the snow completely melts away.