View metadata, citation and similar papers at core.ac.uk brought to you by CORE provided by Harvard University - DASH Coleochaete and the origin of sporophytes The Harvard community has made this article openly available. Please share how this access benefits you. Your story matters Citation Haig, D. 2015. “Coleochaete and the Origin of Sporophytes.” American Journal of Botany 102 (3) (March 1): 417–422. doi:10.3732/ ajb.1400526. http://dx.doi.org/10.3732/ajb.1400526. Published Version doi:10.3732/ajb.1400526 Citable link http://nrs.harvard.edu/urn-3:HUL.InstRepos:23017252 Terms of Use This article was downloaded from Harvard University’s DASH repository, and is made available under the terms and conditions applicable to Open Access Policy Articles, as set forth at http:// nrs.harvard.edu/urn-3:HUL.InstRepos:dash.current.terms-of- use#OAP 1 Haig: Coleochaete and the origin of sporophytes 2 1 3 Coleochaete and the origin of sporophytes 4 5 David Haig2,3 6 7 2. Department of Organismic and Evolutionary Biology, 8 Harvard University, 26 Oxford Street, 9 Cambridge MA 02138 USA. 10 11 1 1 1. Manuscript received _______; revision accepted ______. 2 3. Author for correspondence (e-mail: [email protected]) 2 1 Premise of study: Zygotes of Coleochaete are provisioned by the maternal thallus before 2 undergoing 3–5 rounds of division to produce 8–32 zoospores. An understanding of the 3 selective forces favoring post-zygotic divisions would be relevant not only to the life 4 history of Coleochaete but to the origin of a multicellular diploid phase in embryophytes. 5 Methods: Simple optimization models are developed of the number of zygotes per 6 maternal thallus and number of zoospores per zygotes. 7 Key results: Zygotic mitosis is favored once zygotes exceed a threshold size but natural 8 selection usually promotes investment in additional zygotes before zygotes reach this 9 size. Factors that favor production of fewer, larger zygotes include multiple paternity, 10 low fecundity and non-provisioning (accessory) costs of zygote production. Such factors 11 can result in zygotes exceeding the size at which zygotic mitosis becomes profitable. 12 Conclusions: Coleochaete may possess large zygotes that undergo multiple fission because 13 of accessory costs associated with matrotrophy (cellular cortex, unfertilized oogonia). 14 The unpredictability of fertilization on land is proposed to have increased accessory 15 costs from unfertilized ova and, as a consequence, to have favored the production of 16 larger zygotes that underwent postzygotic division to produce diploid sporophytes. 17 18 Key words: Coleochaete, sporophyte, alternation of generations, size-versus-number, 19 matrotrophy 3 1 Many nineteenth-century botanists considered the multicellular ‘fruits’ (zygospores) of 2 Coleochaete to be analogous, perhaps even homologous, to the sporophytes of land plants. 3 Supporters of both the homologous and antithetic theories of the origin of sporophytes 4 used the ‘fruit’ as a model but disagreed about how it should be interpreted, whether as 5 a modified asexual generation or as a novel interpolated structure (Haig 2008). 6 Coleochaete fell from favor in these debates after Allen (1905, 1906) concluded that the 7 first two divisions of its zygospore were the heterotypic and homotypic divisions (in 8 modern parlance, meiosis I and II). Since then, the ‘fruit’ has generally been interpreted 9 as a haploid rather than diploid structure. 10 Interest in Coleochaete has revived with recognition that it belongs among the 11 closest algal relatives of embryophytes (Ruhfel et al. 2014). The absence of a multicellular 12 diploid phase in streptophyte algae is now considered strong support for the antithetic 13 theory because it weakens the case for an ancestral isomorphic alternation of generations 14 as envisioned in modern versions of the homologous theory (Blackwell 2003; McManus 15 and Qiu 2008). Clearly, contemporary arguments about homologous versus antithetic 16 alternation of generations bear only a tenuous relation to the morphological questions at 17 the heart of the nineteenth-century debate (Haig 2008). Although the ‘fruit’ has lost favor 18 as an analogue of sporophytes, matrotrophy has gained prominence as a feature shared 19 by Coleochaete and embryophytes. Coleochaete zygotes increase in size and accumulate 20 reserves after syngamy, suggesting that the haploid maternal parent transfers resources 21 to the diploid product of fertilization (Graham and Wilcox 1983, 2010). 4 1 Although the occurrence of zygotic meiosis in Coleochaete is generally accepted, 2 evidence in support of this ‘common knowledge’ is thin. Allen (1905) was unable to 3 count chromosomes but concluded that the first two divisions of zygospores were 4 meiotic on the basis of differences in chromosome compaction. On the other hand, 5 Hopkins and McBride (1976) detected nuclei with eight times the unreplicated haploid 6 quantity of DNA (8C) within germinating zygospores. A division sequence that reduces 7 DNA levels from 8C to 1C corresponds to neither meiosis nor mitosis as conventionally 8 understood (Haig 2010). 9 This paper presents simple life-history models of the transition from a single- 10 celled zygote to a multicelled ‘fruit.’ These models are agnostic about the precise nature 11 of Coleochaete’s postzygotic divisions whether meiotic, mitotic, or something else. 12 Zygotes are assumed to develop attached to a multicellular maternal thallus. Therefore, 13 developmental mechanisms required for postzygotic multicellularity are assumed 14 already to be present and expressed in prezygotic parents (for a discussion of the origin 15 of these mechanisms see Niklas 2014). Although my focus is on understanding life- 16 history evolution and variation in Coleochaete, implications for early stages in the origin 17 of sporophytes in embryophytes will also be considered. 18 SIZE-VERSUS-NUMBER TRADEOFFS 19 Haploid parents will be called mums and dads to distinguish them from diploid 20 mothers and fathers (Haig 2013). Two size-versus-number tradeoffs will be considered. 21 The first is faced by mums: whether to produce a few large or many small zygotes. The 5 1 second is faced by zygotic offspring: how many zoospores to produce from a zygote’s 2 reserves. These interrelated questions can be conceptualized as asking how should a 3 mum allocate an amount Z among n zygotes each of which produces m zoospores. 4 Coleochaete filaments produce oogonia one at a time whereas the postzygotic 5 divisions involve successive bipartitions of the zygospore cytoplasm without an increase 6 in zygospore size (multiple fission or palintomy). Therefore, the number of zygotes will 7 be assumed to change by integral increments (n, n + 1, n + 2, …) but the number of 8 zoospores per zygote by successive doublings (m, 2m, 4m, …). My models address the 9 specific question under what conditions natural selection favors a change from 10 producing m to 2m zoospores per zygote. The fitness contribution of each zoospore will 11 be represented by a function, f(x), where x is a measure of the zoospore’s nutrient 12 reserves. Following Smith and Fretwell (1974), f(x) is assumed to increase with x subject 13 to diminishing marginal returns, i.e. f"(x) < 0 < f'(x), with some minimum positive value 14 of x below which f(x) = 0. Maternal fitness is mnf(x). Thus zoospores are assumed to 15 make independent contributions to maternal fitness determined by zoospore ‘size’ x. 16 Let maternal investment consist solely of zoospore reserves. A mum who invests 17 a total amount Z in zygote production invests X = xm in each of n = Z/X zygotes. Z is 18 optimally distributed when each zygote receives where is the investment per 19 zoospore at which marginal returns on investment equal average returns 20 21 Mums are predicted to respond to variation in Z by varying the number rather than the 6 1 size of zygotes (Smith and Fretwell 1974; Lloyd 1987). 2 Under the assumption that f"(x) < 0 < f'(x), there will be a critical investment x* 3 for which f(x*) = 2f(x*/2). For a zygote of size X, higher fitness would be obtained by 4 dividing X among m zoospores for X < mx*, but by dividing X among 2m zoospores for 5 X > mx*. However, the optimal size of zoospores is less than this critical size, < x* (Fig. 6 1). If mums always produced zygotes of size , then these zygotes would be 7 smaller than the ‘size’ at which an extra division becomes profitable. 8 Changes in Z and X are continuous but changes in m and n occur by integral 9 steps. At least one zoospore must receive more or less than if Z is not a precise 10 multiple of . Suppose that where . For ∆Z close to zero, 11 Z is better distributed evenly among n zygotes but, for ∆Z above some critical value, Z is 12 better distributed evenly among n + 1 zygotes. As ∆Z approaches this critical value, 13 optimal zoospore size approaches x' then abruptly decreases to x" as the mum switches 14 from investing in n to n + 1 zygotes where nf(x') = (n + 1)f(x"). As n becomes large, x' 15 and x" converge on . Conversely, low fecundity (small n) favors greater variation in 16 zygote size as Z fluctuates. The difference between x' and x" is maximal for n = 1 when 17 x' = x* and x" = x*/2. In the special case when Z = X* = mx*, three alternatives yield the 18 maximum return on investment (i) a single zygote that produces m zoospores of size x*; 19 (ii) two zygotes that each produce m zoospores of size x*/2; or (iii) a single zygote that 20 undergoes an extra division to produce 2m zoospores of size x*/2. 21 The above model predicts that adaptive adjustment of x will be achieved by 22 changing n (number of zygotes) rather than m (number of zoospores per zygote) except 7 1 when n is small.