, Volume 63, Issue 2, pp 47–57

Dynamics of the growth, life history transformation and photosynthetic capacity of Oophila amblystomatis (Chlorophyceae), a green algal symbiont associated with embryos of the northeastern yellow spotted salamander Ambystoma maculatum (Amphibia)


DOI: 10.1007/s13199-014-0287-x

Cite this article as:
Bishop, C.D. & Miller, A.G. Symbiosis (2014) 63: 47. doi:10.1007/s13199-014-0287-x


The recent discovery that the unicellular green alga Oophila amblystomatis, invades embryonic tissues and cells of the salamander Ambystoma maculatum prompted us to investigate the growth and life history transformations of the algal symbionts in egg capsules. During embryonic development, symbionts were first detected microscopically as a cohesive population of swimming cells in the vicinity of the blastopore around embryonic stage 17. This population of cells grew and at embryonic stage 25, a fraction of the population began to affix to the inside of the egg capsule. Cells then underwent syngamy, lost flagella, and transformed into non-motile cells. We observed a linear increase in the accumulation of such capsule-associated cells from embryonic stage 25 to 40. The population of zoospores did not grow over this period and showed a declining trend between stage 39 and 40. We verified the population growth by measuring relative chlorophyll a content and also measured quantum yield (QY) of photosystem II (PS II) using pulse amplitude modulated (PAM) fluorometry. The population, but not the cell size, of non-motile capsule membrane-bound cells increased modestly during a one-month period after hatching, and continued to contain high levels of chlorophyll a and photosynthetic capacity. We conclude that O. amblystomatis undergoes a life history transition in egg capsules and speculate that many of these symbionts become zygotes, rather than invading the embryo.


Symbiosis Oophila amblystomatis Ambystoma maculatum 



Photosystem II


Quantitative yield


Pulse amplitude modulated fluorometry




Post-flash depression

1 Introduction

The unicellular green alga Oophila amblystomatis, closely related to Chlamydomonas moeuwsii (Kerney et al. 2011), has been known for 125 years to associate with egg capsules of some amphibian taxa, including those of the northeastern yellow spotted salamander Ambystoma maculatum (Orr, 1888; Kerney 2012 for review). Over the last 70 years, progress has been made in describing the benefits of this association to the algae and host (e.g., Gilbert 1942, 1944; Hutchison and Hammen 1958; Goff and Stein 1978; Bachmann et al. 1986; Pinder and Friet 1994; Tattersall and Spiegelaar 2008; Bianchini et al. 2012; Graham et al. 2013). The postulated benefits for the embryos are that: (i) at least during daytime, algal photosynthesis increases the partial pressure of oxygen in egg capsules (Pinder and Friet 1994), (ii) removal of excreted ammonia by the algae (Goff and Stein 1978), (iii) removal of CO2 by algal photosynthesis (preventing acidification by H2CO3 production), (iv) transfer of photosynthates from algae to embryos and (v) production of anti-microbial compounds (Graham et al. 2013). Conversely, postulated benefits for the algae are the provision of ammonia and CO2 by hosts (Goff and Stein 1978), and protection from predation from heterotrophs in pond water.

Oxygen derived from algae benefits embryos that would otherwise develop more slowly and with higher rates of abnormalities under the oxygen gradient that forms from the edge to the center of egg masses due to oxygen diffusion limitations under natural conditions (Gilbert, 1944; Pinder and Friet 1994; Valls and Mills 2007; Tattersall and Spiegelaar 2008). Survival rates of embryos decreased and stage at hatching increased when egg masses were cultured in the dark (Goff and Stein 1978, 2008), supporting this contention.

Kerney et al. (2011) recently showed that algal cells invade host embryonic tissues and cells. This discovery has inspired investigations in the taxonomy of O. amblystomatis (Lewis et al. 2013; Bishop et al. unpublished) and photosynthate exchange between symbionts and host (Graham et al. 2013). Oviducal washes by Gilbert (1944) produced no evidence of algae in the oviduct. However, Kerney et al. (2011) demonstrated the presence of O. amblystomatis 18S rDNA in two out of three oviducts sampled from adult female salamanders, suggesting that intergenerational transmission of algal symbionts is possible. Three aspects of the growth of algae within the egg capsule provided the impetus for this study.

Firstly, algal growth within capsules is likely to be regulated. Algal growth is minimal in egg capsules after embryos are removed (Gilbert, 1944), suggesting that the embryo, not the egg capsule per se, promotes algal growth. Moreover, algae growing unchecked would at a certain point reach a population size at which the rate of oxygen production per algal cell would become light limited. This would reduce the amount of O2 available to the embryo and the effect could be exacerbated under chronic low light conditions. Egg capsules at the centre of intact egg masses under a 14:10 light:dark cycle at 10 °C in tap water, underwent fluctuations from hyperoxic to anoxic conditions during light and dark cycles Pinder and Friet (1994). Secondly, as observed by Goff and Stein (1978), many of the flagellated swimming symbionts (presumably zoospores) accumulate on the inside surface of the egg capsule whereupon they lose their flagella and affix themselves to the inner membrane of the egg capsule. Thus part of the life history of the symbiont occurs in the egg capsule but the dynamics of this process have not been characterized. Thirdly, only a small fraction of algae that populate the capsule enter into the embryo (Kerney et al. 2011) but the fate of the remainder of the algae after the larva hatches and vacates the egg capsule has not been established.

In order to understand the life history of the algal symbionts better, we documented aspects of the growth, life history and photosynthetic capacity of algae that occur inside the egg capsule during later stages of embryonic development. We also tracked the fate of algae associated with the inner membrane of the capsule for 1 month after hatching.

2 Materials and methods

2.1 Organisms

Egg capsules for laboratory-based cell counting, immunocytochemistry and fluorometry were from an egg mass collected on May 12th, 2012 from Mira River, Cape Breton Island, Nova Scotia (46.017°N, 60.015°W). Isolated capsules were cultured at 8 °C within the range of temperatures that embryos would encounter naturally, at least during early development (Hutchison and Hammen 1958; CDB and AGM unpublished results). Egg capsules were maintained under constant illumination of 50 μmol photons.m2.s−1 (PAR) provided by 25 W, cool-white fluorescent bulbs.

Egg masses for determination of Fm from field-collected specimens were collected May 9-19th, 2013 from a pond in the vicinity of Antigonish, Nova Scotia (45.684 N, 61.922 W). We chose this site over the Mira River site because of its proximity to the laboratory. Based upon 18S rDNA sequence, algae from Ambystoma maculatum egg capsules sampled across the host range host are O. amblystomatis (Kim et al. 2014). Using a clean razor blade, we removed a portion of each egg mass (approximately 1/4 of the total) and placed it into a sealable plastic container filled to the brim with pond water. We placed each portion of an egg mass into a separate container, and transported them to the laboratory for analysis. On the same day of collection we subjected several egg capsules isolated from each egg mass portion to PAM fluorometry and then fixed the embryos from those egg capsules in 4 % formaldehyde, the developmental stages of which were determined at later date. Every 2–3 days over a period of 10 days, we sampled additional portions of the same egg masses for a total of four portions. All collections were conducted by permit from Department of Natural Resources of Nova Scotia.

2.2 Suspended and capsule membrane-bound cell counts and area measurements

To estimate algal growth kinetics among capsules within a single egg mass, we sampled five capsules for each of six different time points during development (Stages 25,35,36,37,39,40). At each time point embryos were staged under an Olympus SZX16 stereomicroscope according to Harrison (1969). Because of the tendency of flagellated cells to lose their flagella and attach to the inner wall of the egg capsule, we developed a method for counting samples of both suspended and capsule membrane-bound algal cells. Thirty intact capsules were manually liberated from the egg mass (Fig. 1a) and arrayed in a 24-well polystyrene tissue culture plate. Each well contained five egg capsules. At each time point, suspended cells from five capsules were counted by shearing capsule membranes with a plastic pipette, having a narrower bore than the capsule, in 1.0 mL of distilled water in a separate well of a 24 well polystyrene tissue culture plate. Capsule membranes, held by fine forceps, were washed gently before removing them from the water and then placing them in a separate well containing 1.0 mL water (Fig. 1b). For each sample, suspended algae were mixed well and then a 50 μL aliquot was added to a hemacytometer. Algae present in 25 squares (corresponding to 0.1 μL) were counted. This number was multiplied by 104 to generate the number of suspended algae . mL−1. We accounted for the intracapsular fluid (estimated to be 60 uL) released from egg capsules, based on the application of a measured capsule radius of 2.5 mm to the formula for the volume of a sphere (4/3πr3). Thus, the total volume used for calculation of algal number per capsule was 1,060 μL.
Fig. 1

Preparations of egg capsules and capsule membranes for cell counts and cell area measurements. a An intact egg capsule isolated from an egg mass, containing an embryo at stage 39 and a population of algal cells. b A collection of egg capsules from which embryos had been liberated with a narrow bore plastic pipette. Arrowhead points to a tear in the capsular membrane through which the embryo was drawn. c An example of an egg capsule membrane preparation used to count capsular membrane-bound cell number and measure cell area, volume and Fm values. d A higher magnification view of capsule membrane-bound cells bound to the inner capsule membrane. Scale bars = 1 mm in a and c; 2.5 mm in b; 30 μm in d

Capsule membrane-bound cells were counted by first manually halving a shorn capsule using fine forceps and then mounting one half of the capsule on a standard microscope slide. This mount was then covered with a coverslip, thereby flattening the capsular membrane into a single layer and ensuring that all cells were on a similar plane (Fig. 1 c,d). Under a 10X objective lens, a region of the capsule was chosen randomly and photographed digitally (Olympus BX51, D72 digital camera, CellSens Dimension image capture software, v1.6). For preparations of capsule membranes 0–4 weeks after hatching (by June 19th all embryos had hatched), the same procedure was used, except we photographed and then counted algae contained along a transect of five fields of view. Cells in each image were counted using the Analyze/Cell Counter plug-in for Image J64 (v.1.46r) using the following procedure.

Each image was converted to an 8-bit grayscale image. The “threshold” function was then used to identify cells based upon a threshold contrast difference between the foreground and the background. The ‘Analyze Particles’ function was used, which reported cell counts and the area of each counted object. Because some images varied in brightness and contrast, as well as the number of non-algal particles, some initial experimentation in brightness and contrast and “size exclusion” and “circularity” settings for identified particles was necessary. Automated counts were then verified by comparing a subsample of these counts to manual counts. For most images, counts that included only particles with a size of >20 pixels2 and a circularity of 0.5–1.0 accurately counted the number of cells. These parameters were conservative in that they more often failed to detect cells (false negatives) than detecting non-cellular particles (false positives). Thus, because this effect would be amplified as the total number of capsular cells increased, cell counts for later time points during the embryonic development series would tend to slightly underestimate the total number of membrane-bound cells. The “Overlay Masks” function was also invoked, which showed particles that had been automatically counted. In the few cases where cells had not been counted automatically, the remaining cells were manually counted. Generally, cells that were not counted were those that were sufficiently close enough to a neighbouring cell to violate the circularity criterion, or were at the edge of the field (in which case they were specifically ignored), or had not been identified by the initial “color threshold” function. For each image, total number of capsular cells was estimated from raw count data by multiplying the size of the field of view (1.4 mm2) by the surface area of a sphere (4πr2).

For area measurements, we set ‘Analyze Particles’ function to record the area of every particle that was counted, as well as to report the summed area of all particles. It is important to note that we estimated cell size by measuring the area of cells at the particular focal plane at which they were positioned when the image was collected. Therefore, since all cells do not occupy the same axial position (relative to the light path), only a fraction of the equators of cells (the point at which the diameter is maximum) in a particular preparation will be in focus. For this reason, our automated methods for estimating area (within and among preparations) systematically underestimate actual area. Area was reported by the software in pixels2, which was then converted to μm2 by multiplying the former value by 0.10239 μm/pixel, the digital resolution of the camera used to capture the image.

We also calculated the total cell volume for both suspended and capsule membrane-bound cells. For suspended cell volume calculations, the volume of a biflagellated cell (771.28 μm3) was calculated and then multiplied by the cell count data generated above. Because the automated cell counting procedure did not generate diameter measures required to calculated cell volumes, we returned to images of capsule membrane preparations and subsampled them. The diameter of 51 cells for each capsule membrane preparation (five preparations/day) was manually measured, using a measuring tool (CellSens Dimension image capture software, v1.6). These measurements were used to calculate the mean radius and then the mean volume. The mean volume was then multiplied by the capsule membrane-bound cell count data generated by automated cell counts.

2.3 Anti-acetylated tubulin immunocytochemistry

On May 30th 2012 capsule membranes were prepared as above and then fixed for 1 h in 4 % formaldehyde, prepared from paraformaldehyde formulated in 0.45 μm filter sterilized pond water. Fixed capsules were rinsed several times in phosphate buffered saline (PBS) pH 7.4 and then blocked in PBS containing 2.5 % (w:v) bovine serum albumin and 0.1 % (v:v) Triton-X-100 before incubation in a 1:400 dilution of anti-acetylated tubulin (IgG clone 6-11B-1; Sigma-Aldrich). Primary antibody was rinsed out with blocking buffer and then replaced with goat-anti mouse (IgG) at 1:1,000 dilution for 2 h at room temperature. Specimens were mounted in 70 % (v:v) glycerol in PBS and imaged under epi-fluorescence illumination (Olympus BX51, D72 digital camera, CellSens Dimension image capture software, v1.6).

2.4 Measurement of algal growth as chlorophyll a fluorescence emission

Notwithstanding concerns about variation in chlorophyll a to biomass ratios under different growth conditions (Felip and Catalan 2000), the total amount of chlorophyll a is commonly used as proxy for algal biomass (e.g. Cullen, 1982; Mayer et al., 1997; Boyer et al., 2009). Chlorophyll a fluorescence emission, a relative measure of chlorophyll a concentration in algal cells, was measured from photosystem II (PSII) with a pulse amplitude modulated (PAM) fluorometer (Heinz Walz, Effeltrich, Germany) using a 660 nm light emitting diode and an RG9 optical filter (transmission >700 nm) in front of the photodiode (Schreiber et al. 1986). The relative chlorophyll a content of the algae bound to the capsule preparation was measured at the Fo stage (Fig. 2b), using the modulated monitoring beam (16,000 Hz) of the PAM fluorometer at a very weak integrated light intensity of about 0.2 μmol photons.m−2.s−1. This low light intensity ensures that the primary electron acceptor of PSII, the quinone QA, is kept in the non-reduced Fo state (van Kooten and Snel 1990).
Fig. 2

Mean algal cell counts and volume measurements per egg capsule, as a function of time. Embryonic stages corresponding to time points at which algal cells were counted or measured are shown along the top (modified from Harrison 1969). For panels (a)-(d) squares = suspended algae; diamonds = capsule membrane-bound algae. a counts of suspended and capsule membrane-bound algae, b counts of total algae and c ratio of suspended to capsule membrane-bound algae. d Calculated volume of suspended cells and measured volume of capsule membrane-bound cells, e total cell volume and f ratio of total capsule membrane-bound to suspended cell volume. (n = 5)

For each sample the maximum fluorescence (Fm) was also determined in order to calculate the potential quantum yield (QY) of charge separation in PSII. The QY, (calculated as \( \frac{F\mathrm{m}\hbox{-} \mathrm{Fo}}{F\mathrm{m}} \) is a measure of the photosynthetic potential of PSII. For three of the six sampling dates in 2012 six capsular preparations were measured and four on the other three. Due to low chlorophyll a content in the intact capsules at the beginning of algal growth in the field in 2013, the fluorescence emission at the Fm stage (see Fig. 6b) was measured instead of at Fo, increasing the measurement sensitivity at least 6 fold (Fig. 6b). The Fm stage was maintained by incubating the capsules in solution of 50 μM of 3-(3,4-dichlorophenyl)-1,1-dimethylurea (DCMU) for 10 min. DCMU blocks the light-driven flow of electrons past QA, keeping QA in its fully reduced state and resulting in maximum chlorophyll a fluorescence emission (i.e. Fm; Schreiber et al. 1986; van Kooten and Snel 1990). Five intact capsules were used at each sampling time, chosen haphazardly by gently swirling the container and using the capsule nearest to each of five marks on the bottom.

2.5 Statistical analyses

Time series data for algal counts, area, volume, Fm and QY values were subjected to a one-way analysis of variance with Tukey HSD post-hoc tests. The null hypothesis was that there were no differences among the means of these values as a function of time. Time points were treated as independent cases. For area values, statistical analyses were conducted on pooled mean pixel values within replicates and thus the ANOVA was conducted on the means of mean areas with day as the factor. Graphed data are presented as means ± S.E.M. In cases where data were graphed as ratios, the fractional S.E.M. was calculated by squaring the fractional standard error for each mean, adding them and then taking the square root of the sum (i.e. \( \sqrt{\left(\frac{EA}{A}\right)2}+\sqrt{\left(\frac{EB}{B}\right)2} \) where EA and EB = the standard error for each term and A and B = the values used to generate ratios).

3 Results

3.1 Suspended and capsule membrane-bound cell counts, volume and area measurements

Despite the increase in green color of egg capsules over time, after their initial increase in number (i.e. prior to stage 25), the population of suspended algae did not increase significantly over the period observed and in fact began to decline after embryonic stage 37 (Fig. 2a). By one-way ANOVA this decline was not statistically significant, but strongly trending (F = 2.12; p = 0.098). In contrast, during the same period, the number of capsule membrane-bound algae increased linearly from 0 to 2.5 × 104 cells . capsule−1 (F = 7.45; p = 0.0002; Fig. 2a and Online Resource 1). Due to the higher number of suspended algae, the change in the total number of cells . capsule−1 is dominated by the decrease in the population of suspended algae. The ratio of suspended vs. capsule membrane-bound algae (Fig. 2c) demonstrates the change in the number of suspended vs. capsule membrane-bound algae. The summed cell volume of suspended and capsule membrane-bound cells (Fig. 2d) indicates that for most of development, the suspended cell volume exceeds that of capsule membrane-bound cells. Between embryonic stage 39 and 40 this relationship is reversed (Fig. 2f). Thus, later in embryonic development, the summed volume of capsule membrane-bound cells exceeded that of suspended cells by a factor of approximately 3:1.

Here it is important to note that, while the life cycle of O. amblystomatis has not been formally described (either within the egg capsule or in culture), we assume it to be comparable to that of one of its closest relatives Chlamydomonas moeuwsii (Kerney et al. 2011, Kim et al. 2014), whose life cycle conforms with that of other Chlamydomonads (Lee, 2008). Thus, algae associated with suspended and capsule membrane preparations were assumed to be zoospores (or gametes) and zygotes, respectively. However, results from anti-acetylated tubulin immunostaining (see below) indicated that some capsular cells were bi-flagellated, some quadri-flagellated, and some non-flagellated. Because we automated count and area measurements from capsule membrane preparations and therefore did not microscopically evaluate the morphology of every cell, capsule membrane-bound cells per se cannot be taken to be synonymous with zygotes. Likewise, motile cells could be either zoospores or gametes. Thus, we retain the terminology of ‘suspended’ and ‘capsule membrane-bound’ cells when referring to cells associated with these two fractions. As the embryos approached the hatching stage, virtually all capsule-membrane associated cells were large and non-flagellated.

Using laboratory-cultured egg masses, we explored cell growth dynamics among capsule membrane-bound algal cells by measuring changes in the mean cell area. The mean area of capsular cells increased three fold during embryonic stage 25 to 40, but the only significant change in area among individual time points occurred between stage 39 and 40 (F = 19.12; p < 0.0001; Fig. 3a). The distribution of cell areas, over the time corresponding to the interval in which significant change occurred, is depicted in Fig. 3b and Online Resource 2.
Fig. 3

Mean and distribution of the area of capsule membrane-bound algae over the same period as in Fig. 1. a a and b denote significant differences in mean cell area among time points. Day 0 was not included in the statistical analysis because some replicates contained zero capsular algae, creating unequal sample sizes. b Distribution of cell areas for day 0 and day 22, binned in 15 μm increments. (n = 5)

When larvae hatched at stage 46, twenty capsules were retained and arrayed in 24-well polystyrene dishes in 1.0 mL of sterile filtered pond water. The number and area of capsular cells was measured once per week for 4 weeks. While the total area of capsular cells did not change after hatching, the total number of cells increased by 23 % (F = 3.32; p = 0.041; Fig. 4a,b).
Fig. 4

Area and counts of capsular algae during a one-month post-hatching period. a cell area. b cell counts. (n = 5). *denotes a significant difference (see text)

To verify the progression of the life history transition occurring in the capsule, we immunostained algae associated with capsules using anti-acetylated tubulin. Most cells were spherical and were not immunoreactive, but there were numerous examples of bi-flagellated zoospores either isolated or in aggregations, and larger quadri-flagellated, cells (Fig. 5a-d).
Fig. 5

Anti-acetylated tubulin immunostaining of algae associated with egg capsule membranes prepared from Stage 39 embryos, the stage corresponding to day 15 in Figs. 2, 3b-f, 4 and 6a. Fluorescent images were captured with a low level of DIC illumination to show cell boundaries and the depression in the capsular membrane in which large non-flagellated cells reside. a a quadri-flagellated algal cell in the vicinity of several zygotes and jettisoned flagella. b a putative mating aggregation of several flagellated algae. c a flagellated alga in the vicinity of a putative zygote indicating the difference in cell size. d a quadri-flagellated alga, presumably post-syngamy, adjacent to a zygote. Scale bar = 30 μm

3.2 Estimation of algal growth by measurement of relative chlorophyll a content

In order to verify the algal count data, we measured Fm, at the same developmental time points from which cell count, area and volume data had been collected (Fig. 6a). Fm was used as a relative measure of chlorophyll a in the egg capsule. The quantity of chlorophyll a increased over time, peaking at 85.7 ± 18.9 at stage 39 before dropping to 54.1 ± 22.5 by stage 40 (F = 4.11; p = 0,003; Fig. 6b). Using PAM fluorescence, we also measured growth rates of algae growing in egg masses in ponds. In Spring 2013 we sampled five egg masses four times over a period of 10 days (May 9th to May 19th; see Methods). We observed a steady increase in Fm values that was consistent among egg masses that were living under non-identical light and temperature regimes (Fig. 6c; environmental data not shown). Likewise, the mean (±S.D.) stages of embryos in egg masses from which Fm values were collected were 30.0 ± 0.7 (May 9th), 37.7 ± 0.2 (May 12th), 38.8 ± 0.1 (May 15th) and 39.1 ± 0.2 (May 19th) indicating relatively synchronous development.
Fig. 6

Increase in relative chlorophyll a content of algae in intact capsules as a function of time and as a function of cell volume. a The left trace is a record of the chlorophyll a fluorescence of algae in intact capsules at 8 °C, showing the two fluorescence parameters Fo and Fm from which quantum yield (QY; a measure of photon use efficiency of PSII) is calculated. The placement of the capsules in the weak intensity monitoring beam (MB) gives rise to Fo (relative units; see Methods) and a subsequent saturating flash (SF) gives rise to the transient Fm (relative units). The right trace demonstrates a typical record of the quenching of Fo post-flash depression(PFD) that occurs among dark-adapted preparations at 20 °C (see text). b Change in Fm from algae from intact capsules cultured in the laboratory over the same time period as in Fig. 2. c Fm from intact capsules from egg masses newly collected from ponds (n = 4–6). Each time point represents capsules taken from successively from the same egg mass over a period of 10 days. d Fm values for suspended algae as a function of cell volume over time. e Fm values for capsule membrane-bound algae as a function of cell volume over time. f Fm as a function of total algal cell volume per capsule

We expressed Fm values as a function of cell volume, for both suspended and capsule-bound cell fractions. Over developmental time, on a mean per unit volume basis, chlorophyll a content rose in suspended cells while declining in capsule membrane-bound cells (Fig. 6d,e). Expressed as a function of total cell volume, chlorophyll a content did not increase significantly over the period observed (Fig. 6f).

3.3 Photosynthetic capacity of the algal symbiont during and after host embryonic development

An important aspect of photosynthetic capacity is the ability with which photons can be used to reduce the first stable acceptor Qa of PSII. The reduction of Qa at the expense of oxidation of phaeophytin a results in a voltage gradient, the so-called charge separation of PSII across the thylakoid membrane. QY is therefore also known as the quantum yield of charge separation of PSII, and the average value for vascular plants is 0.83. This means that one absorbed photon causes 83 % of the charge separation. Here, the QY of PSII in algal symbionts changed during embryonic development. QY was relatively static at 0.77 ± 0.01 until embryonic stage 39, at which point it dropped significantly to 0.65 ± 0.06 (F = 14.57; p < 0.0001) before returning to 0.72 ± 0.01 at stage 40 (Fig. 7a).
Fig. 7

The effect of temperature on quantum yield (QY) and post-flash depression (PFD) for algal symbionts. a QY of isolated intact capsules over developmental time (n = 4). b Fo, Fm, QY and PFD values as a function of temperature (expressed as a percentage of the peak PFD value). Fo and Fm are values from which QY is calculated (see text for details). Closed squares = Fm; open squares = QY; open diamonds = Fo; closed diamonds = PFD

Experiments measuring QY values (Fig. 7a) were initially done at 8 °C to match the culture temperature. In these experiments the fluorescence, after presentation of the saturating flash, required to measure QY, decayed rapidly to the pre-flash level. Experiments later performed at higher temperatures revealed a post-flash depression (PFD) \( {\mathrm{F}}_{\mathrm{o}} \), which recovered within 20s (Fig. 6a). Calculated as (Fo − Fo)/(Fo − 0) PFD at 20 °C averaged (±S.D.) 0.62 ± 0.08 (n = 64; where n = a separate capsule preparation). Such a drastic PFD was unexpected. The onset of PFD slightly overlaps with peak fluorescence (i.e. Fm). However, the overlap did not significantly affect the determination of Fm (data not shown). We are currently investigating the mechanism underlying this novel PFD. The QY value was somewhat dependent on temperature, dropping from 0.73 at 2 °C to 0.57 at 34 °C. The observed drop in QY with increased temperature was due both to a drop in Fm and a rise in Fo (Fig. 7b).

4 Discussion

A feature of algal growth in the egg capsules of A. maculatum and A. gracile (Goff and Stein 1978) is the occurrence of a life history transition from motile to non-motile cells that become fixed to the capsule membrane. Based upon the chlamydomonad life cycle (Lee, 2008) and our observations, we conclude that motile bi-flagellated cells are either zoospores or gametes, that quadri-flagellated capsule membrane-bound cells are cell in the process of mating and that large non-flagellated capsule membrane-bound cells are zygotes. It is also possible that the latter of these life history stages are haploid “mother” cells capable of undergoing mitosis to produce zoospores. However, having examined many capsule membranes and suspended cells at different host developmental stages, we have never observed sporangia (Mihara and Hase 1971). In contrast, sporangia in O. amblystomatis culture isolates were readily observed (Kim et al. 2014). Among capsule membranes cultured for up to 1 month post-hatching we observed no increase in mean algal cell area; we did observe a modest (23 %) increase in the number of cells, but only between 2 weeks and 1 month post-hatching. Moreover, among capsule membranes kept in laboratory culture conditions for extended periods after hatching, capsule membrane-bound zygospores, identifiable by elaborate secondary cell wall material, were observed. Taken together, and with the caveat that the complete life history of O. amblystomatis has not been documented yet, these results strongly suggest that the majority of the cells on the capsule membrane are zygotes.

We find it intriguing that some but not all suspended cells undergo the transformation from motile to non-motile stages and when the larvae hatch out of their egg capsules, a puff of suspended algae is released. Recalling the observation that no sporangia were observed in any of our preparations, this suggests that only a portion of the life cycle is occurring within A. maculatum egg capsules and, per force, only once. This is consistent with a life history transition that is co-ordinated with hatching of the host. However, this raises the question as to the means by which the population of zoospores increases over time. Kerney et al. (2011) showed that numerous algae encapsulated by a thick wall, and containing chloroplasts, are associated with the innermost egg membrane surface prior to the detection of suspended algae and that the ova contain algal DNA. It is not clear what stage of the life cycle these cells represent but an intriguing possibility is that they are zygotes that then produce zoospores by meiosis, thus contributing to the expanding population of suspended cells during embryonic development. Thus, an invasion of egg masses from environmentally acquired zoospores (Gilbert 1944, Gatz, 1973) may not be the sole source of the expanding population of zoospores, supporting the contention (Kerney et al. 2011) that algae are transferred intergenerationally.

Over a two-week period, from stage 25–39, no significant change in the population of suspended algal cells was observed, whereas between stage 39 and 40, corresponding to a one-week period, a substantial decline in suspended cells was observed, albeit not significantly in a one-way ANOVA. This suggests that a portion of the ‘greening’ of the capsule in later stages is due to the accumulation of capsule membrane-bound cells. Indeed, we observed a linear increase in the population of capsule membrane-bound cells, yet only observed a significant increase in algal cell area between stage 39 and 40. Therefore, while the number of motile cells that settled on the inner surface of the capsule appeared to increase at a constant rate, the majority of the increase in cell size of subsequent capsule membrane-bound cells did not. This suggests synchronous growth of cell size at this phase of the life history. One source of synchrony may be passage of the host through a particular embryonic stage, possibly releasing a chemical on which algae cue. Alternatively, the embryo may pass through peak catabolism of yolk proteins, with a corresponding decline in ammonium production, well before hatching. Nitrogen starvation triggers mating and zygosporulation in Chlamydomonas reinhardtii (Sager and Granick 1954). Future work will need to verify whether this time-structured growth in the size of capsule-membrane bound cells occurs under natural conditions.

It is not clear why O. amblystomatis cells are retained on the capsule membrane after the host hatches. If capsule membrane-bound cells are zygotes, retention of their photosynthetic capacity by delaying the onset of zygosporulation may allow cells to increase storage of sugars in preparation for a dormant period. Kerney et al. (2011) observed that, relative to the total number of algae in the capsule, a small proportion of algae invade the embryo, and fewer still invade embryonic cells. What is the fate of the algae that do not enter into the embryo? Free-living O. amblystomatis have been identified by 18S rDNA sequence from salamander breeding habitat (Lin and Bishop, unpub. obs.), lending support to the hypothesis that O. amblystomatis may be acquired environmentally (Gilbert 1942). However, it is unknown whether this population of cells invades A. maculatum egg capsules. In summary, there are two possible fates for the initial zoospore population: 1) invasion of the salamander embryos, and 2) conversion into zygospores that repopulate the ponds where egg masses are deposited.

4.1 Growth by fluorescence measurements

During the same embryonic period over which we observed growth and transformation of zoospores to capsule membrane-bound cells, chlorophyll a content increased in both laboratory-cultured and field-collected preparations. Chlorophyll a, as a function of suspended cell volume, appeared to increase over developmental time. In synchronous in vitro cultures of C. reinhardtii and C. eugametos, zoospores increase in volume 5–6 fold and about 10 fold, respectively, after release from zoosporangia (Su et al. 2001, Molendijk et al. 1992). We observed an increase in chlorophyll a per unit cell volume, but this is likely due to having applied a single measure of suspended cell volume to calculate total cell volume. This is consistent with our observation that the suspended cell number does not increase but that the volume of individual suspended cells (zoospores or gametes) does.

In contrast, as a function of mean cell volume, the amount of chlorophyll a among capsule-membrane bound cells decreased over developmental time. We observed a six-fold decrease in chlorophyll a per unit mean cell volume, but only a three-fold increase in mean cell volume. Therefore, the increase in mean cell volume only partially accounts for the decrease in chlorophyll a, suggesting the possibility of its degradation as capsule membrane-bound cells grow.

Under our conditions, capsule membrane-bound cells continue to have substantial photosynthetic capacity, for at least 4 weeks after hatching of the embryo. If these cells are zygotes, they are quite unlike Chlamydomonas reinhardtii in which photosynthetic capacity drops as the newly formed zygote transforms into a dormant zygospore within 12–24 h (Baldan et al. 1991). Zygote formation in C. reinhardtii cultures is induced by N-limitation, and under these conditions zygospore formation follows rapidly after syngamy of the gametes (Sager and Granick 1954). By 48 h after syngamy, the QY had dropped to 0.24, a value only about 34 % the average value presented by zoospores (Baldan et al. 1991). Within days following syngamy the zygospore of C. reinhardtii undergoes dedifferentiation of the chloroplast and the chlorophyll is degraded (Cavalier-Smith 1976). The QY  of zygotes, which had completed their development (i.e. zygosporulation), was zero (Baldan et al. 1991). The respiration rate of dormant zygospores is also very low compared to that of germinated zygsopores (Hommersand and Thimann 1965).

While determining Fm with a saturating flash, we noticed that after Fm was reached, a return to original Fo value was intervened by a very significant quenching of fluorescence below Fo. We refer to this as a post-flash depression (PFD). This deep, but temporary quenching of fluorescence below the Fo level, has to our knowledge not previously been reported. We surveyed 82 articles reporting the measurement of QY in chlamydomonads using a saturating flash, an in no case was quenching of this order of magnitude below Fo observed. We have since confirmed this pattern of quenching with O. amblystomatis in field measurements, but the biological significance remains unknown.

Finally, we wish to emphasize the experimental advantages provided by the salamander embryo symbiosis. A single species of a unicellular alga can be studied over time, with respect to life cycle shifts and associated physiological transformations. The salamander egg capsules represent a microcosm in which a synchronous and relatively pure algal population can be examined, both in nature (while still in the egg mass), or, upon isolation from the egg mass, under controlled laboratory conditions.


This work was supported by the Natural Sciences and Engineering Research Council (NSERC) of Canada through a Discovery Grant to CDB and a NSERC Research Tools and Instruments Grant to AGM. Ainslie Cogswell is thanked for finding and delivering the Mira River egg mass used in this study. David Garbary, two anonymous reviewers, and the editor are thanked for providing critical feedback.

Supplementary material

13199_2014_287_MOESM1_ESM.docx (23.7 mb)
Online Resource 1Representative images from capsular preparations, depicting the accumulation of capsule membrane-bound algae as a function of time. With the exception of cases in which the number of algae was sufficiently low to count manually, these images were subjected to automated counts (see Methods). Preparations are from (a) May 18 (b) May 22 (c) May 27 (d) May 30 (e) June 6. Scale bar = 0.5 mm (DOCX 24,246 kb)
13199_2014_287_MOESM2_ESM.docx (384 kb)
Online Resource 2The distributions of cell area, expressed as pixels2, for all preparations of capsular algae as a function of time. Each histogram represents area counts from a single image and a column of histograms represents five different capsules for a given date. The black vertical line represents the median bin (DOCX 383 kb)

Copyright information

© Springer Science+Business Media Dordrecht 2014

Authors and Affiliations

  1. 1.Department of BiologySt. Francis-Xavier UniversityAntigonishCanada

Personalised recommendations