An ulvophycean marine green alga produces large parthenogenetic isogametes as predicted by the gamete dynamics model for the evolution of anisogamy

Abstract

In eukaryotes, gamete size difference between the two sexes (anisogamy) evolved from gametes of equal size in both mating types (isogamy). The gamete dynamics (GD) model for anisogamy evolution combines gamete limitation and competition and predicts that if gametes of both mating types can develop parthenogenetically (i.e. without fusing with the opposite mating type), large isogamy can evolve under gamete-limited conditions. Ulvophycean marine green algae that have been claimed to exhibit various gametic systems from isogamy to anisogamy are important models for testing such theories. However, in most previous papers, whether a species is isogamous or anisogamous has not been examined statistically. Caution is necessary regarding claims of slight anisogamy because of gamete size variation. We reveal (i) that the gametic system of Struvea okamurae is large isogamy using a generalized linear mixed model, which accounted for the variation of gamete size among individual gametophytes, and (ii) that gametes of this alga can actually develop parthenogenetically, contrary to a previous report. Its habitat environments and protracted duration of gamete release suggest that this alga might experience gamete-limited conditions. Struvea okamurae seems to produce large parthenogenetic isogametes following GD model predictions, as an adaptation to deep waters.

1. Introduction

Anisogamy, the gamete size difference defining the two sexes, underlies sexual selection in eukaryotes. Its evolution is one of the most fundamental phenomena in biology: it is plausibly claimed to have played an important role in generating morphological and behavioural differences between males and females through male–male competition [1–7]. Anisogamy is generally considered to have evolved from isogamy with gametes of equal size in both mating types (e.g. [8–11]) though this can infrequently reverse secondarily [12].

Evolutionary explanations for anisogamy consist of two unrelated types: (i) those relating to intracellular conflicts between nuclear and cytoplasmic genes (reviewed by Lessells et al. [13]), and (ii) those relating to the dynamics of fertilization, coupled with a gamete size or number trade-off and the need to provision the zygote, referred to hereafter as ‘gamete dynamics’ (GD) models. The first GD models relied on group or population selection [14,15] and showed that under gamete limitation (i.e. where fusion probability is low), a higher rate of gametic fusions could be achieved in an anisogamous population than an isogamous one. The first individual-based selection GD model was that of Parker et al. [8], which relied on competitive fusion between gametes, without gamete limitation. Smaller gametes are more numerous, and hence competitively claim more fusions with larger, high-resource gametes. Larger gametes provide zygotes with additional resources, increasing their viability or reproductive success, especially under less favourable conditions. Consequently, gamete size is disruptively selected, generating two sexes. Later work showed that gamete limitation can also generate anisogamy under individual selection (e.g. [16]), and Lehtonen & Kokko [17] made the first attempt to merge the two selective forces of gamete competition and gamete limitation into one general GD model; for anisogamy to evolve mainly by gamete limitation, gamete competition must be relatively weak or absent (see also [18]).

Simplest versions of the GD model typically predict isogamy with small gametes (small isogamy) if the requirement for zygote provisioning is low or anisogamy with microgametes if the provisioning requirement is high, as in complex multicellular organisms (see [19]). However, isogamy with large gametes (large isogamy) has been more difficult to explain. To our knowledge, there have been four different GD model contexts that predict large isogamy. Large isogamy can evolve when (i) reproductive reserves are divided by discrete cell divisions (rather than by continuous gamete volumes) in optimizing trait values [11], (ii) there is low gamete competition, high zygote requirement and efficient fertilization [17,20], (iii) in complex volvocine algae, under gamete-limited conditions a large chloroplast (constituting most of the gamete’s volume) is necessary to provision the gamete, constraining drive toward anisogamy (for which evidence exists [21]), and (iv) there is high gamete limitation, high zygote requirement and unfused gametes can develop parthenogenetically [22]. Here, we provide evidence that appears to support context (iv) in an ulvophycean alga.

Ulvophycean marine green algae are important models for testing theories of anisogamy evolution, mainly because they exhibit a variety of gametic systems [23]. In particular, they conform exactly to important GD model assumptions (i.e. broadcast fertilization and no parental care), thus differing from many other organisms (e.g. Volvocales freshwater green algae, in which sperm are released in packets, and zygotes are protected by the mother, not released into the medium) [23]. Some ulvophyceans have parthenogenetic gametes that can develop without fusing with a gamete of the different mating types [24]. The earliest unicellular and isogamous green alga might have produced parthenogenetic gametes [22]. Parthenogenesis might significantly increase the reproductive success of gametes particularly when sexual fusion is difficult. Two recent theoretical GD analyses add parthenogenetic gametes into GD theory [22,25], one predicting that while high gamete encounter rates result in small isogamy or anisogamy, large isogamy can evolve under gamete-limited conditions [22]. Evolutionary transitions to and from isogamy and parthenogenesis are examined in an important recent paper on brown algae [12] but without considering isogamete size, a feature of considerable theoretical significance [11,22].

Several previous papers claim that some ulvophyceans actually produce large parthenogenetic isogametes (see [22]) but only provide micrographs of gametes of similar size in two mating types as evidence for isogamy. However, we advocate carefully analysing gametic systems using appropriate statistical methods. Ulvophyceans include many slightly anisogamous species [23]. Due to gamete size variability within mating types, fusions may involve gametes of unequal size even in isogamous species, giving the appearance of slight anisogamy. Therefore, we should take gamete size variation among individual gametophytes into account when analysing gametic systems. Are large parthenogenetic isogametes really produced by ulvophyceans as the GD model predicts? Our investigation seeks to find an example that answers this question.

We focus on the ulvophycean marine green alga, Struvea okamurae (Cladophorales, Ulvophyceae). It is reported (i) that this alga has an isomorphic haplo-diplontic life cycle, that is, the haploid dioecious gametophyte and the diploid sporophyte are both multicellular and identical in form (Struvea plants), (ii) that gametophytes produce biflagellate isogametes that do not develop parthenogenetically in either mating type, and (iii) that sporophytes produce quadriflagellate zoospores through meiosis [26] (figure 1). However, no statistical analysis of this gametic system exists. We analyse the gametic system of this alga using statistical modelling and also carefully re-examine whether gametes develop parthenogenetically using culture experiments. Here, we confirm that it shows large isogamy but show that its gametes can develop parthenogenetically and suggest that it is likely to be gamete limited, hence conforming to recent GD predictions for large isogamy.

Figure 1.

Isomorphic haplo-diplontic life cycle of S. okamurae with a haploid dioecious gametophyte generation and a diploid sporophyte generation. In this study, we examine the gametic system of this alga (*1) and re-examine whether their gametes can or cannot develop parthenogenetically (*2). 1n: haploid stage; 2n: diploid stage.

2. Material and methods

(a). Collection of field materials and culture experiments

We collected haploid gametophytes and diploid sporophytes of S. okamurae in early spring (March 2020) at a depth of approximately 3 m off the coast of Ginoza, Okinawa, Japan (26°47′ N, 127°78′ E; electronic supplementary material, figure S1 for field photos). In our culture experiments, we aimed to obtain gametes produced by gametophytes, which are likely to be genetically different in each mating type, and examined the fate of unfused gametes. We established the uni-algal culture strains of these gametophytes (So-4, So-6, So-14, So-21 and So-27) and sporophyte (So-1) using a protoplast formation method [27]. Protoplasts were cultured in 1 : 4 concentration Provasoli’s enriched seawater medium [28] at 22°C under a long day condition (14 h : 10 h = light : dark) and roughly 15 μmol photons m⁻² s⁻¹ [29] in a culture chamber (LH-220S; NK system, Osaka, Japan). We isolated gametophyte or sporophyte Struvea plants developed from a protoplast individually into 12 well plastic plates (IWAKI, Tokyo, Japan) and induced gametogenesis or zoosporogenesis under an eutrophic condition by changing the medium [30]. The mating type of gametes was determined by a crossing test. We maintained gamete cultures (without mixing them with gametes of the opposite mating type) and zoospores that were produced through meiosis under the same condition as protoplasts. We obtained four more gametophytes (So-1.1.1c1, So-1.1.2b1, So-1.1.4a3 and So-1.1.4b4) that were developed from zoospores produced by the sporophyte (So-1) and also induced gametogenesis for them. These gametophytes are also likely to be genetically different.

(b). Biovolume estimation of gametes

We estimated the volume of S. okamurae gametes more accurately than previous studies on other ulvophycean species (e.g. [31]) by eliminating the error due to discrepancies between the calculation of biovolume based on geometric approximation and that based on real shapes [32]. Some gametes were fixed with 1% glutaraldehyde immediately after they were released. We observed them on a cover glass coated with poly-l-lysine (0.01 w/v%) and took a micrograph of each gamete using a differential interference microscope (Axio Imager A1; Zeiss, Oberkochen, Germany). We obtained successive vertical cross-sections through each pear-shaped gamete using the micrograph, calculated the volumes of each section with one pixel height and summed the volumes of all sections (see [32] for more detail).

(c). Statistical analysis of gamete size

The relationship between the sex and the volume of gametes was analysed by a generalized linear mixed model (GLMM). In this study, gamete size (volume) produced by individual gametophytes may vary due to individual differences, which can be explained by GLMM (see electronic supplementary material, text S1). We utilized two models with a gamma distribution and log link function that included different fixed effects: (1) sex (mating type A or B) and (2) no fixed effects (null model). The linear predictor of each model was as follows:

(i) ${\displaystyle \log \left(u_i\right)={\beta }_1+{\beta }_2{x}_i+r_i}$ ;

or (ii) ${\displaystyle \log \left(u_i\right)={\beta }_1+r_i}$ ,

where $u_i$ is the mean gamete volume of gametophyte i, $x_i$ is sex (mating type A or B), ${\beta }_1$ and ${\beta }_2$ are fixed-effect parameters and $r_i$ is a random-effect parameter among gametophytes with a normal distribution. ${\beta }_1$ , ${\beta }_2$ and the variance of $r_i$ (=s) were estimated based on a maximum-likelihood criterion. We assessed the fit of the models using Δ Akaike’s information criterion (AIC). The result of the model selection based on AIC was confirmed by the likelihood ratio test. All statistical analyses were performed with R 3.4.2 and the ‘lme4’ package [33].

3. Results

(a). Gamete release and gamete crossing tests

Gametes were released from matured gametophytes up to approximately 7 h after the light in the culture chamber was turned on (3.8 ± 1.3 h, mean ± s.d., n = 19). Released gametes had an eyespot and showed positive phototaxis (electronic supplementary material, figure S2). From our crossing tests, we determined that So-4, So-27, So-1.1.1c1 and So-1.1.2b1 strains have one mating type (labelled type A) and that So-6, So-14, So-21, So-1.1.4a3 and So-1.1.4b4 strains have the other mating type (labelled type B).

(b). Comparison of gamete volumes between mating types

The volume of gametes in S. okamurae is shown in figure 2. The mean volume per gamete released from mating type A was 387.3 ± 128.3 µm³ (mean ± s.d.; n = 445, number of gametophytes = 4) and that from mating type B was 407.8 ± 125.6 µm³ (n = 464, number of gametophytes = 5). Notably, gamete size varied widely within each mating type. In our statistical analysis, the model (2) that included no fixed effects was selected as the best model by the GLMM, based on AIC (AIC_best = 1161.7, s = 0.007, ΔAIC = 1.8). This implies that gamete volume is not different between the two mating types (p = 0.63, likelihood ratio test).

Figure 2.

Volume of gametes from the two mating types (A and B). Each box plot corresponds to an individual gametophyte. Boxes indicate the interquartile range (IQR) between the first and third quartiles, and the bold horizontal line inside each box indicates the median. Vertical lines represent the lowest and highest values within 1.5× IQR from the first and third quartiles, respectively. Circles indicate raw data. n.s., no significant differences between the mating types indicated by the likelihood ratio test and the GLMM analysis (see Results for more details).

(c). Parthenogenetic development of gametes

In both mating types (A and B), some of the gametes developed into Struvea plants even without fusing with a partner (mating type A: figure 3a–f ; mating type B: figure 3g–l ). These plants produced quadriflagellate zoospores (figure 3d,j ) within a few months after gametes were released, suggesting that they are parthenosporophytes; these zoospores later developed into Struvea plants (figure 3e,f,k,l ).

Figure 3.

Parthenogenetic development of gametes of S. okamurae. (a–f) Mating type A. (g–l) Mating type B. (a,g) Gametes. (b,h) One week after the gametes were released. (c,i) Roughly six weeks after the gametes were released. (d,j) Quadriflagellate zoospores released from a parthenogenetically developed Struvea plant. Numbers (1–4) indicate individual flagella. (e,k) Roughly one week after the quadriflagellate zoospores were released. (f) Roughly six weeks after the quadriflagellate zoospores were released. (l) Roughly two months after the quadriflagellate zoospores were released. Arrowheads indicate eyespots. Scale bars: (a,g) 5 µm, (b,h) 10 µm, (c,i) 100 µm, (d,j) 5 µm, (e,k) 10 µm, (f) 100 µm and (l) 1 mm.

4. Discussion

We have revealed three major points on the gametic system of S. okamurae. First, this alga is isogamous (gametes from gametophytes of different mating types are not different in size), as claimed in several other ulvophycean marine green algae [11]. Why its gamete varies so much in size remains under consideration and may relate to variation in the degree of gamete limitation and zygote resource requirement across its microhabitats. Second, the isogametes are larger than many other ulvophyceans (see electronic supplementary material, table S1, figures S3, S4 and text S2). Finally, contrary to [26], they are parthenogenetic. They can survive to develop into parthenosporophytes without fusing with a partner, as parthenogenetic gametes observed in other ulvophyceans [34]. Thus, the gametic system of S. okamurae appears to be large parthenogenetic isogamy.

Are the two conditions (gamete limitation and the need for high zygote provisioning) for large parthenogenetic isogametes as required by one of four existing GD model contexts fulfilled in S. okamurae? Field data on gamete limitation are unavailable, but informed speculation is possible from gamete release traits in natural habitats. Ulvophyceans including species that inhabit shallow waters at the upper intertidal zone often have mechanisms to release gametes synchronously at daytime low tide, and their released gametes show positive phototaxis to facilitate gametic encounter prior to fusion just under the two-dimensional sea surface [23]. They therefore probably avoid high gamete limitation. Their isogametes are much smaller than the isogametes produced by Cladophorales, including S. okamurae (see electronic supplementary material, table S1, figures S3 and S4 and text S2). Although S. okamurae gametes showed some positive phototaxis (electronic supplementary material, figure S2), it would be difficult for them to reach the sea surface by this method since S. okamurae inhabits deep waters in the subtidal zone. In many ulvophyceans that inhabit such deep waters, gamete release is also synchronized by light irradiation [23]. In contrast, it is unlikely that S. okamurae releases gametes synchronously to use phototaxis effectively, since the time interval for gametes to be released from gametophytes after illumination was long and highly variable. Additionally, the population density of its gametophytes is low in the field [35]. Struvea okamurae is therefore likely to be gamete limited in nature, and a recent paper pointed out that parthenogenesis might be a backup plan under such conditions [12]: failure to achieve fertilization can be somewhat compensated by parthenogenesis. Further, large resources would be required by zygotes and parthenogenetic gametes due both to the harsh low light conditions in their deep water habitat and their growth requirement to a complex multicellular form. Thus, our observations are consistent with the conditions for the evolution of large parthenogenetic isogamy under GD theory [22].

Could parthenogenesis also be adaptive for shallow water ulvophycean species with small isogametes? Many have gametes that switch phototaxis from positive to negative immediately after fusion, activating a return to their parental habitat (e.g. [36]). Any unfused gametes showing positive phototaxis would be drifted out to deep waters where they have insufficient resources to survive so that small isogamy with facultative parthenogenesis might not work well as a backup plan.

GD models without parthenogenesis can also explain the evolution of large isogamy [11,17,20,21]. Which model (with or without parthenogenesis) is more appropriate for shallow water ulvophyceans remains under consideration, since the frequency of parthenogenesis in nature is still largely unknown. However, S. okamurae seems likely to have adapted to the gamete-limited conditions in its deep water habitat by producing large parthenogenetic isogametes, though we cannot rule out other GD routes to large isogamy.

Ethics

This work did not require ethical approval from a human subject or animal welfare committee.

Data accessibility

All data are available from the electronic supplementary material and the Dryad Digital Repository [37].

Supplementary material is available online [38].

Declaration of AI use

We have not used AI-assisted technologies in creating this article.

Authors’ contributions

T.T.: conceptualization, data curation, formal analysis, funding acquisition, investigation, methodology, project administration, resources, software, supervision, validation, visualization, writing—original draft, writing—review and editing; K.N.: data curation, investigation, writing—review and editing; K.M.: data curation, investigation, writing—review and editing; G.A.P.: supervision, validation, writing—original draft, writing—review and editing; Y.H.: formal analysis, funding acquisition, investigation, visualization, writing—review and editing.

All authors gave final approval for publication and agreed to be held accountable for the work performed therein.

Conflict of interest declaration

We declare we have no competing interests.

Funding

This work was supported by grants-in-aid from the Japan Society for the Promotion of Science (grant numbers 22K20644 and 23K14260) and Fujiwara Natural History Public Interest Incorporated Foundation.