G3: Genes|Genomes|Genetics Early Online, published on February 26, 2016 as doi:10.1534/g3.115.026229
1
Sequence of the Gonium pectorale mating locus reveals a complex and dynamic
2
history of changes in volvocine algal mating haplotypes
3
Takashi Hamaji1,*, Yuko Mogi1,†, Patrick J. Ferris‡, Toshiyuki Mori†, Shinya Miyagishima§,
4
Yukihiro Kabeya§, Yoshiki Nishimura**, Atsushi Toyoda§, Hideki Noguchi§, Asao Fujiyama§,
5
Bradley J.S.C. Olson††, Tara N. Marriage††, Ichiro Nishii‡‡, James G. Umen*, Hisayoshi Nozaki†.
6
*:
Donald Danforth Plant Science Center, St Louis, MO, USA
7
†:
Department of Biological Sciences, Graduate School of Science, the University of Tokyo,
8
Tokyo, Japan
9
‡:
Department of Ecology and Evolutionary Biology, University of Arizona, Tucson, AZ,
10
USA
11
§:
National Institute of Genetics, Mishima, Shizuoka, Japan
12
**:
Department of Botany, Graduate School of Science, Kyoto University, Kyoto, Japan
13
††:
Division of Biology, Kansas State University, Manhattan, USA
14
‡‡:
Faculty of Science, Nara Women's University, Nara, Japan
15
1: These authors equally contributed to this work.
16
DDBJ/ENA/GenBank accessions: LC062386 (G. pectorale DRG1 plus haplotype); LC062718
17
(MT- scaffold); LC062719 (MT+ scaffold).
18
1
© The Author(s) 2013. Published by the Genetics Society of America.
19
Running title: Green algae mating locus evolution
20
Keywords: Gonium, Chlamydomonas, Volvox, mating locus, evolution
21
Corresponding author:
22
Takashi Hamaji,
23
[email protected]
24
Donald Danforth Plant Science Center
25
975 N. Warson Road, St. Louis, MO 63132 USA
26
+1-314-587-1683
2
27
28
Abstract
Sex-determining regions or mating-type loci (MT) in two sequenced volvocine algal
29
species, Chlamydomonas reinhardtii and Volvox carteri exhibit major differences in size,
30
structure, gene content, and gametolog differentiation. Understanding the origin of these
31
differences requires investigation of MT loci from related species. Here we determined the
32
sequences of the minus and plus MT haplotypes of the isogamous 16-celled volvocine alga
33
Gonium pectorale that is more closely related to multicellular V. carteri than to C. reinhardtii.
34
Compared to C. reinhardtii MT, G. pectorale MT is moderately larger in size and has a less
35
complex structure with only two major syntenic blocs of collinear gametologs. However, the
36
gametolog content of G. pectorale MT has more overlap with that of V. carteri MT than with C.
37
reinhardtii MT, while the allelic divergence between gametologs in G. pectorale is even lower
38
than in C. reinhardtii. Three key sex-related genes are conserved in G. pectorale MT—GpMID
39
and GpMTD1 in MT-, and GpFUS1 in MT+. GpFUS1 protein exhibited specific localization at
40
the plus-gametic mating structure, indicating a conserved function in fertilization. Our results
41
suggest that the G. pectorale-V. carteri common ancestral MT experienced at least one major
42
reformation after the split from C. reinhardtii and that the V. carteri ancestral MT underwent a
43
subsequent expansion and loss of recombination after the divergence from G. pectorale. These
44
data begin to polarize important changes that occurred in volvocine MT loci and highlight the
45
potential for discontinuous and dynamic evolution in sex determining regions.
3
46
Introduction
47
Sex determining regions (SDRs) can exhibit complex patterns of molecular evolution
48
that are distinct from those of autosomes (Bachtrog et al. 2011). The origins of SDR architecture
49
and function have been studied in diverse species, but our understanding of the evolutionary
50
processes that shape SDRs is still limited. Recombination suppression, a common property of
51
both haploid and diploid SDRs, reinforces linkage between sex-related genes within the SDR and
52
promotes accumulation of sexually antagonistic alleles linked to the SDR (Bachtrog et al. 2011;
53
Immler and Otto 2015). SDRs often contain genomic rearrangements that may be either the
54
cause or consequence of recombination suppression between heteromorphic haplotypes or sex
55
chromosomes. The deleterious effects of suppressed recombination are also seen in SDRs where
56
Y or Z chromosomes in diploid mating systems may undergo degeneration (Charlesworth et al.
57
2005). Much less is known about the long-term evolutionary histories of SDRs in haploid
58
mating systems where SDRs also undergo degeneration, but where limited data are available
59
(Yamato et al. 2007; Bachtrog et al. 2011; Sun et al. 2012; McDaniel et al. 2013; Ahmed et al.
60
2014).
61
Volvocine algae offer unique advantages as a model for the evolution of SDRs. Like
62
many protists, volvocine algae are volvocine algae are haploid and capable of both asexual
63
(vegetative) and sexual reproduction. Environmental or hormonal cues trigger gametic
64
differentiation, which in heterothallic species is controlled by a heteromorphic mating locus
65
(MT) with two haplotypes, plus/minus or male/female (see below). Volvocine algae also
66
encompass a range of morphologies from unicellular genera such as Chlamydomonas, to Volvox
67
that have thousands of cells and exhibit germ-soma differentiation and other developmental
4
68
innovations that result in a functionally integrated multicellular colony (summarized in Figure 1).
69
Sexual reproduction coevolved with colony size in volvocine algae whose mating strategies
70
include isogamy in unicellular and small colonial genera (Chlamydomonas, Gonium and
71
Yamagishiella), anisogamy in intermediate-sized genera (Eudorina and Pleodorina) and oogamy
72
in the largest and most differentiated genus Volvox. Transitions between heterothallism (genetic
73
sex determination) and homothallism (self-mating) have also occurred within volvocine algae
74
(Coleman 2012) (Figure 1). The well established phylogenetic relationships between volvocine
75
species enable specific traits and innovations to be ordered and mapped onto a coherent
76
evolutionary framework (Nozaki et al. 2000; Herron and Michod 2008).
77
The full genome sequences including regions of both haplotypes of MT were
78
previously described for two heterothallic species—isogamous unicellular C. reinhardtii and
79
oogamous multicellular Volvox carteri (Figure 2) (Merchant et al. 2007; Ferris et al. 2010;
80
Prochnik et al. 2010). Importantly, the genetic and molecular bases of mating-type
81
differentiation in volvocine algae including functions of mating locus genes, have been
82
investigated (Ferris and Goodenough 1994, 1997; Ferris et al. 1996, 2002; De Hoff et al. 2013;
83
Geng et al. 2014). MT haplotypes (plus/minus or male/female) segregate as single Mendelian
84
traits but the loci themselves are multigenic regions in which recombination is suppressed. The
85
core of the mating locus is the rearranged (R) domain where gene order and arrangements
86
between the two haplotypes are non-colinear (Umen 2011). In C. reinhardtii, the plus and minus
87
haplotypes of MT reside on chromosome 6 (Ferris and Goodenough 1994; Ferris et al. 2002; De
88
Hoff et al. 2013) while V. carteri MT is located in a region of linkage group I that shares a
89
common origin with C. reinhardtii chromosome 6 (Ferris et al. 2010). Despite their shared
90
chromosomal location, V. carteri MT is much larger in size than C. reinhardtii MT and shows a
5
91
far higher degree of differentiation between gametologs (allelic gene pairs residing in MT)
92
(Ferris et al. 2010). This discrepancy raises many questions about why the two MT loci differ so
93
much in genetic divergence and recombination potential (Charlesworth and Charlesworth 2010;
94
Umen 2011). One important feature of C. reinhardtii MT that is not found in V. carteri MT is
95
low-frequency gene conversion between gametologs that acts to reduce allelic differentiation (De
96
Hoff et al. 2013).
97
The C. reinhardtii MID (minus-dominance) gene is present only in the MT-
98
haplotype and plays a critical role in determining mating type (Ferris and Goodenough 1997).
99
Recently, the V. carteri MID gene which is only found in the male MT haplotype (Ferris et al.
100
2010) was also shown to govern critical aspects of sex determination (Geng et al. 2014).
101
Moreover, MID genes, have been found in the minus or male mating haplotypes of other
102
heterothallic volvocine algae including several Gonium species (Hamaji et al. 2008, 2013a;
103
Setohigashi et al. 2011), and Pleodorina starrii (Nozaki et al. 2006) suggesting that the genetic
104
basis of sex or mating-type determination is conserved throughout the volvocine lineage (Figure
105
1). Other than the MID gene, no sex-related genes are conserved between the MT loci of V.
106
carteri and C. reinhardtii. The strikingly different sizes, gametolog contents, gametolog
107
differentiation and recombination behaviors of C. reinhardtii versus V. carteri MT loci raise
108
questions about how these two SDRs, which appear to have arisen from a common ancestral
109
SDR region, diverged so markedly from each other. To answer these questions requires
110
information on volvocine algal MT loci from additional species in order to begin reconstructing
111
ancestral states and polarizing changes within the lineage.
112
113
Gonium pectorale is a small isogamous colonial volvocine species that is more
closely related to V. carteri than is C. reinhardtii (Nozaki et al. 2000). As such it may represent
6
114
an informative taxon for reconstructing the evolution of MT loci in the volvocine lineage.
115
Previous work on G. pectorale identified MT- homologs of MID and of MTD1, genes that are
116
both also found in the MT- haplotype of C. reinhardtii (Hamaji et al. 2008, 2009). To date no
117
homologs of MT+ limited genes from C. reinhardtii have been found in other volvocine species,
118
including the gene FUS1 that encodes a membrane bound protein which localizes to the mating
119
apparatus or “fertilization tubule” and is required for fusion of plus and minus gametes (Ferris et
120
al. 1996, 1997; Misamore et al. 2003).
121
Here we report the sequence of G. pectorale MT+ and MT- haplotypes derived from
122
a combination of chromosomal walking and whole genome sequencing. The overall structure of
123
G. pectorale MT was found to differ from C. reinhardtii MT and V. carteri MT in terms of
124
structural rearrangements between haplotypes. It has fewer rearranged sequence blocs than C.
125
reinhardtii MT and no large autosomal insertions in either haplotype. G. pectorale shares in
126
common with C. reinhardtii the presence of sex-limited genes MID and MTD1 in its MT-
127
haplotype (Hamaji et al. 2008, 2009), and a FUS1 homolog in its MT+ haplotype whose gene
128
product, GpFUS1, localizes to the fertilization tubule indicating homologous function with
129
CrFUS1. Like the case in C. reinhardtii, gametolog differentiation in G. pectorale MT is
130
minimal. However, the gametolog gene complement of G. pectorale MT is more similar to that
131
of V. carteri MT than of C. reinhardtii MT. Taken together, these data enable us to reconstruct a
132
minimal set of changes in the MT region that led to their current states. Our findings suggest a
133
dynamic and most likely punctuated history of volvocine MT evolution that involves episodes of
134
structural reconfiguration in which gametolog content and syntenic blocs change, but where
135
mating-related genes remain linked to their ancestral MT haplotypes. The expansion, gametolog
136
differentiation, and loss of MTD1 and FUS1 genes that characterize V. carteri MT likely
7
137
occurred after the G. pectorale and V. carteri lineages split.
8
138
Materials and Methods
139
Strains used
140
The genomic DNA sequence data presented here come from previously described
141
strains of G. pectorale: Kaneko3 (minus) and Kaneko4 (plus) (Yamada et al. 2006; Hamaji et al.
142
2008, 2013b); K3-F3-4 (minus) and K4-F3-4 (plus) which are F3 hybrid progeny from Kaneko3
143
and Kaneko4 as described previously (Hamaji et al. 2013b).
144
BAC library construction
145
DNA plugs were prepared from both mating types (Kaneko4: plus; Kaneko3: minus)
146
in 1% SeaPlaque GTG agarose (Cambrex Bio Science Rockland, Inc.), treated with Pronase E
147
(Sigma) and washed thoroughly as described (Ferris et al. 2010). BAC library production was
148
performed at the Clemson University Genome Institute. The DNA plugs were partially digested
149
with EcoRI and then DNA fragments size-fractioned (approx. 120 kb) by pulse-field gel
150
electrophoresis were ligated into pIndigoBAC536. Single colonies of E. coli strain DH10b
151
transformed with G. pectorale genomic fragment-containing BACs were picked, spotted on
152
nylon membranes (or “BAC filters”) and stored frozen in glycerol at -80C (total BAC number,
153
plus: 27,648; minus: 18,432, estimated 24x and 16x coverage of G. pectorale plus and minus
154
genome, respectively).
155
BAC screening and chromosomal walking
156
BAC filters were hybridized with probes labeled and detected with CDP-Star (GE
157
healthcare) or DIG-labeling kit (Roche). Positive signals were confirmed by direct PCR with
9
158
probe-specific primers on the individual clones. BACs of true positives were column-purified
159
from 5 ml cultures and end-sequenced with M13 Forward and Reverse primers. Each end
160
sequence is designed for a pair of specific primers, which was used to determine the relative
161
locations of overlapping BACs. Primer pairs from both distal ends were selected to label probes
162
for the next round of chromosomal walking. Inverse PCR (Sambrook and Russell 2001) and
163
TAIL-PCR (Liu and Whittier 1995) were performed to identify flanking sequences when no
164
additional BAC clones could be identified. GpMTD1 probes were the same as used for the
165
GpMTD1 DNA gel-blot analysis (Hamaji et al. 2009). The PCR product of GPLEUFATG and
166
GPLEURTAA was labeled to screen for LEU1S alleles (Table S1). Two gametologs (WDR57
167
and DRG1) and their flanking regions, which were not obtained in the MT+ BAC assembly, were
168
obtained by genomic PCR using specific primers based on the MT- alleles. The linkage of MT+
169
DRG1 was determined by recombination scores described below. Abundant repetitive regions
170
flanking the MT assembly prevented further chromosomal walking to connect directly to
171
autosomal sequences.
172
Shotgun sequencing and assembly
173
BAC DNA was mechanically sheared either with nebulizers (for 30 secs) in the
174
TOPO shotgun sub-cloning kit (Invitrogen) or with sonication. Blunt-end fragments were
175
subcloned into pCR4, pCRII (Invitrogen) or pUC118 (Takara Bio Inc.). Shearing with
176
sonication and pUC118 subcloning were performed by the Kazusa DNA Research Institute.
177
Shotgun subclones were sequenced from both ends by the Research Resource Center (BSI,
178
RIKEN) or the Kazusa DNA Research Institute. Raw sequence data was base-called, vector-
179
trimmed for each, end-clipped to remove low-quality regions and assembled by CodonCode
180
Aligner (CodonCode Corp.: http://www.codoncode.com/aligner/). Assembled contigs were
10
181
queried against C. reinhardtii V4 protein models or V. carteri protein models (Ver. 2, JGI and
182
male and female MT, NCBI) using BLASTX to identify protein coding genes (Merchant et al.
183
2007; Ferris et al. 2010).
184
Annotation
185
Gene models were generated using Fgenesh (Salamov and Solovyev 2000)
186
(http://linux1.softberry.com/berry.phtml) with the “Chlamydomonas” option and then manually
187
edited based on similarity to C. reinhardtii (JGI V4 protein models: http://genome.jgi-
188
psf.org/Chlre4/Chlre4.home.html) or V. carteri (male and female MT genes available in
189
Genbank: http://www.ncbi.nlm.nih.gov/genbank/) homologs.
190
GpFUS1 identification and characterization
191
BLASTX search of the G. pectorale MT locus contigs using C. reinhardtii proteins
192
(ver. 4) identified a FUS1-like sequence in MT+. RT-PCR fragments of 5’-/3’-RACE were
193
cloned into TOPO TA vector to determine the full-length cDNA sequence. Subsequently,
194
genomic sequences were obtained from plus BAC 71M20. The G. pectorale FUS1 full-length
195
cDNA sequence was determined by the same method as GpMTD1 (specific primers used:
196
GpFUS1AF; GpFUS1AR; GpFUS1BF; GpFUS1BR; GpFUS1CF; GpFUS1CR). The initially
197
gapped genomic sequence of GpFUS1 was filled by genomic PCR. Hydrophobicity with
198
window size of 9 and theoretical protein isoelectric point (pI) / molecular weight (MW) were
199
calculated with ProtScale and Compute pI/Mw in ExPASY (Gasteiger et al. 2003)
200
(http://au.expasy.org/).
201
Anti-GpFUS1 antibody production, western blotting and immunofluorescence.
11
202
The coding region corresponding to amino acid residues 90 to 489 out of the
203
predicted 824-aa sequence of GpFUS1 protein was amplified by PCR with primers YM001F and
204
YM001R. The resultant DNA was cloned into pET100/D-TOPO vector (Invitrogen) and
205
expressed as a His-tag fusion protein. The recombinant protein was isolated with a HisTrap HP
206
column (GE Healthcare). Purified protein was injected into two rabbits for antibody production
207
(Riken, Saitama, Japan). The rabbit antisera were affinity-purified with a 6His-GpFUS1-coupled
208
HiTrap NHS-activated HP column (GE Healthcare), according to the manufacturer’s instructions.
209
The final concentration of antibody was ≫0.25 mg/ml.
210
To stain GpFUS1 protein expressed in gametes, indirect immunofluorescence
211
microscopy assay and Western blot analyses were performed as previously described after
212
dibutyryl-cyclic AMP (db-cAMP) treatment (Mogi et al. 2012).
213
Linkage mapping
214
DNA samples of 78 F1 progeny (haploid recombinants from F1 meiosis) derived from G.
215
pectorale Mongolia 1x4 strains (described previously) (Hamaji et al. 2008) were genotyped for
216
markers designed based on polymorphic genome sequences. Primers, restriction enzymes and
217
accession numbers in PCR-RFLP analyses are listed in Table S1. The PCR conditions were 2
218
min 94°C, 35 cycles of 30 sec at 94°C, 30 sec at 53°C (for DRG1, 50°C) and 30 sec at 72°C,
219
followed by 7 min at 72°C with the same reaction composition as reported previously (Ferris et
220
al. 2010). Restriction digests of the PCR products were performed (Sambrook and Russell 2001)
221
to identify the two genotypes.
222
Molecular evolutionary analyses
223
Previously reported genes linked to or residing in MT loci of C. reinhardtii or V. carteri were
12
224
queried by TBLASTN searching (Altschul et al. 1990) of the G. pectorale de novo draft genome
225
assembly (Hanschen et al. in press). Genome-wide synteny comparison between C. reinhardtii
226
(v4 assembly) and G. pectorale was done using SyMAP (Soderlund et al. 2006, 2011).
227
MT shared genes (i.e. gametologs) were analyzed as previously described (Ferris et
228
al. 2010). Coding or genomic sequences were aligned using ClustalX 2.0 (Larkin et al. 2007)
229
and alignments adjusted manually. Divergence scores were calculated using yn00 from the
230
PAML4 package (Yang 2007); non-synonymous and synonymous site divergence (dN and dS,
231
respectively) of aligned coding sequences of gametologs for was estimated according to Yang
232
and Nielsen (Yang and Nielsen 2000) with equal weighting between pathways, and the same
233
codon frequency for all pairs. Sliding window analysis of each gametolog was performed using
234
DnaSP v5.10.1 (Librado and Rozas 2009) for Pi (Nei 1987) over the aligned genomic sequences
235
from both haplotypes from start to stop codons in 100 base-windows with a 25 base-interval.
13
236
Results and Discussion
237
Size, structure and gene content of G. pectorale MT
238
BAC contigs containing G. pectorale MT sequences comprised approximately 500
239
kb for MT- and 360 kb for MT+ (Figure 2A; Figure S1; Table 1). Although these contigs could
240
not be directly connected to autosomal scaffolds, based on their genetic contents and other
241
criteria outlined below it is likely they contain all or most of the rearranged (R) domain of G.
242
pectorale MT. Contigs from both haplotypes were sequenced and used for BLASTX and
243
TBLASTX queries against the C. reinhardtii proteome and against the V. carteri MT locus DNA
244
sequence respectively. These searches detected 24 G. pectorale MT genes, 21 of which are
245
gametolog pairs meaning that an allele is present in both mating haplotypes. Three additional
246
mating-type-specific genes, MID, MTD1 and FUS1 were also present, with MID and MTD1 in
247
MT-, and FUS1 in MT+ (Figure 2A; Table S2). The gene contents of G. pectorale MT are
248
described in more detail below.
249
Three rearranged sequence blocs (designated I-III in Figure 2A; Figure S2) define its
250
R domain of G. pectorale MT with the largest, bloc II, being a >200 kb inversion that retains
251
collinearity of its ten resident gametologs, and whose intergenic regions contain a few
252
interspersed repeats (Figure 2A; Figure S2). Bloc III contains a single gene, PTC1 that is
253
flanked by extensive non-coding repeats that comprise the majority of this sub-region. While the
254
coding regions of PTC1 are nearly identical between gametologs, the 3’-UTR of the MT- allele is
255
extended compared to the MT+ allele leading to a larger predicted gene size (Figure 2A). Bloc I
256
contains the TOC34 gene, whose gametolog pairs have low similarity between gametologs
14
257
compared with bloc II gametologs (Figure 2A; discussed below), and has almost no similarity in
258
intergenic regions (Figure 2A; Figure S2). Bloc IV (Figure 2A; Figure S2) is almost completely
259
collinear between haplotypes, and as described below, likely represents the centromere-proximal
260
autosomal flank of MT. The only significant difference between the two mating haplotypes in
261
bloc IV is at its proximal end next to the R domain where a large intron is present in the MT-
262
allele of SelEF at its 3’ end but absent from the MT+ allele, accounting for their different sizes
263
(Figure 2A; Figure S2). The distal-most gene from the R domain in bloc IV, ANK17, was
264
incompletely assembled in each haplotype with differences in the length of assembled sequences
265
accounting for the size difference between its two alleles (Figure 2A). On the other side of MT
266
(top end in Figure 2A) no additional sequences distal to TOC34 in MT- and PTC1 in MT+ were
267
found and it is likely that beyond this region are repeats and/or telomeric sequences as is the case
268
in V. carteri (Ferris et al. 2010). Interspersed between the syntenic blocs I-IV are three regions
269
(v-vii) of G. pectorale MT that are unique to one of the two haplotypes. Region vi of MT-
270
contains homologs of two C. reinhardtii sex determining genes, GpMTD1 and GpMID that have
271
been previously described (Hamaji et al. 2008, 2009). Region vii of MT+ contains a homolog of
272
FUS1, a C. reinhardtii fertilization protein. Characterization of GpFUS1 is presented below.
273
Region v initially appeared to be unique to MT- and contains a single gene, DRG1, whose
274
orthologs in C. reinhardtii and V. carteri are present as gametolog pairs in their respective MT
275
loci (Ferris et al. 2002, 2010; De Hoff et al. 2013) but which have no known function in the
276
sexual cycle. A MT+ linked allele of DRG1 was subsequently identified as described below.
277
Overall, the haplotype bloc structure of G. pectorale MT is notable for being
278
completely different from C. reinhardtii or V. carteri MT with fewer overall rearrangements than
279
in C. reinhardtii (Figure 2) (Ferris et al. 2010; De Hoff et al. 2013). Also apparently absent from
15
280
G. pectorale MT are large autosomal insertions and tandem repeats that are found in C.
281
reinhardtii MT (De Hoff et al. 2013). Although it is larger than C. reinhardtii MT, the structure
282
of G. pectorale MT is simpler with fewer sequence rearrangements and insertions, consistent
283
with a younger age. It is also very different from V. carteri MT that is characterized by much
284
more extensive sequence rearrangements and few collinear blocs of gametologs. Taken together,
285
our findings suggest a non-linear trajectory of MT evolution in volvocine algae which is
286
discussed further below.
287
Genomic location and completeness of the G. pectorale MT assembly by linkage
288
mapping.
289
Although we could not connect the G. pectorale MT assembly physically to other
290
scaffolds in the G. pectorale whole genome assembly, we identified a linked 2.8 Mb autosomal
291
scaffold (scaffold00010) that is syntenic to a portion of C. reinhardtii MT-containing
292
chromosome 6 (Table 2; Figure S3). Recombination was measured between each terminus of G.
293
pectorale scaffold00010 and MT, and revealed that one of the termini did not recombine MT
294
(0/78) but that the other one did with a genetic distance of 17 cM (13/78 recombinants). Based
295
on physical estimates from two independent mapping intervals (Figure S3; Table 2) the proximal
296
end of scaffold00010 is less than 100 kb from MT. Ferris et al. (2010) reported a common
297
chromosomal origin for regions containing MT loci of C. reinhardtii (Chr. 6) and V. carteri (Chr.
298
1). Our findings here further support this idea and show that despite large changes in structure
299
and gene content, MT loci appear to have remained stably associated with a single chromosomal
300
location throughout the volvocine lineage.
301
302
The near completeness of the G. pectorale R domain assembly was supported by our
finding gametolog partners in each mating haplotype for every gene other than the three genes
16
303
described above (MID, MTD1 and FUS1) that are already known to be sex-limited homologs in
304
C. reinhardtii and/or V. carteri. With the exception of DRG1, both gametolog copies are in MT
305
scaffolds. In addition, when we searched for G. pectorale homologs of genes present in and
306
around MT from C. reinhardtii and V. carteri we were able to locate nearly all of them in G.
307
pectorale autosomal scaffolds or within its MT assembly. Although we could not find a copy of
308
DRG1 in either the MT+ genome scaffolds or in the G. pectorale whole genome assembly
309
(Hanschen et al. in press), we reasoned that a MT+ allele might be in a region of the MT+
310
haplotype that was not assembled. Using PCR with primers designed to amplify the MT- copy of
311
GpDRG1, we identified a MT+ DRG1 allele (Materials and Methods) and found that it is tightly
312
linked to MT (0/78 recombinants). Besides DRG1, there are no other “orphan” gametologs in
313
our MT assembly and no additional sex-limited genes other than GpMID, GpMTD1, and
314
GpFUS1. Taken together, our data suggest that the MT assembly shown in Figure 2 is nearly
315
complete as is the inventory of G. pectorale MT-linked genes (Table S3).
316
317
318
Evolution of gene contents and structure of volvocine MT loci
G. pectorale homologs of many V. carteri and C. reinhardtii mating type genes (i.e.
319
gametologs) are autosomal meaning that in G. pectorale these genes reside outside of the MT
320
locus (Table S3). Overall, the gametolog contents of G. pectorale MT show much more overlap
321
with V. carteri gametologs (19 out of 20 in common) than with C. reinhardtii gametologs (2 out
322
of 20 in common) 4 that G. pectorale MT and V. carteri MT diverged more recently (Figure 1).
323
Besides the sex-limited genes already mentioned above (MTD1, MID, and FUS1 in C.
324
reinhardtii; MID in V. carteri) C. reinhardtii and V. carteri both have additional sex-limited
325
genes that we searched for in G. pectorale (Table S2). Homologs of EZY2 or MTA1 from C.
17
326
reinhardtii (Ferris and Goodenough 1994; Ferris et al. 2002) or FSI1 and HMG1 from V. carteri
327
(Ferris et al. 2010) could not identified anywhere in the G. pectorale whole genome sequence or
328
MT region. However, a G. pectorale homolog of the C. reinhardtii MT+ gene OTU2 (De Hoff et
329
al. 2013) was found on an autosomal scaffold (scaffold00113). In summary, compared to either
330
C. reinhardtii or V. carteri, G. pectorale MT is the most minimal in terms of structural
331
rearrangements and in its sex-limited gene content.
332
Molecular evolution of G. pectorale MT genes
333
Previous work on C. reinhardtii and V. carteri MT loci revealed very different
334
divergences between gametolog pairs, with high divergences for most V. carteri pairs and low
335
divergences for C. reinhardtii pairs (Ferris et al. 2010). It was subsequently found that low
336
frequency gene conversion between gametolog sequences plays a role in maintaining sequence
337
homogeneity between the nominally non-recombining regions of C. reinhardtii MT (De Hoff et
338
al. 2013). We calculated frequencies of synonymous (dS) and non-synonymous (dN)
339
substitutions between gametologs for G. pectorale MT genes and found, with two exceptions that
340
they were even lower than in C. reinhardtii (Table S4, Figure 3A) (Hiraide et al. 2013). These
341
findings suggest either higher rates of gene conversion between G. pectorale gametologs than
342
those in C. reinhardtii and/or a younger age for its haplotype blocs that may have had less time
343
to diverge. A potentially informative clue about haplotype divergence and the history of G.
344
pectorale MT was revealed from one exceptional region at the distal end of MT- (bloc I in Figure
345
2A) that encompasses TOC34 and part of neighboring gene HSP70B that resides in bloc II
346
(Figure 2A). The genic regions of TOC34 (dN=0.0623 and dS=0.5549; Figure 3B; Table S4)
347
and part of HSP70B (dN=0.0006 and dS=0.0678; Figure 3C; Table S4) have up to ten-fold
348
higher divergence values than all other gametolog pairs (Figure S4) (average
18
349
dN=0.00333±0.00288 and dS=0.0331±0.0257) with a boundary in the middle of HSP70B where
350
divergence between gametologs drops and resembles the remainder of bloc II (Figure 3C). High
351
divergences are also evident in the intergenic regions around TOC34 on both its proximal and
352
distal sides relative to HSP70B, none of which can be aligned between the two haplotypes
353
(Figure S2). By contrast, many of the intergenic sequences in blocs II and IV align well between
354
haplotypes (Figure S2).
355
The discontinuity of divergences in the TOC34-HSP70B region has at least two
356
possible explanations. One possibility is that the TOC34-HSP70B region is a “cold-spot” for
357
recombination and contains a sequence that blocks progression of gene conversion tracts near the
358
middle of HSP70B. There is precedent for inter-haplotype gene conversion in C. reinhardtii (De
359
Hoff et al. 2013), though no isolated cold-spots were found. A second possibility is that the
360
highly similar regions of G. pectorale MT are relatively young, and that the more diverged
361
region containing TOC34 is a remnant from a previous version of MT with older and more
362
diverged haplotypes. In this case the low-divergence regions of MT could have arisen as a result
363
of a replacement event in which bloc II from one of the two MT haplotypes replaced the same
364
region of the opposite mating haplotype (in inverted orientation) with a break point between
365
TOC34 and HSP70B on one end, and between WDR57 and PTC1 on the other end (Figure S5).
366
This resetting event might also have allowed renewed and continuing gene-conversion between
367
gametologs in bloc II, but would leave TOC34 gametologs in bloc I to continue diverging while
368
also restricting the extent of gene conversion for HSP70B that abuts bloc I. The two
369
explanations we propose for the discontinuous divergence patterns of bloc I versus blocs II-IV of
370
G. pectorale MT are not mutually exclusive, and in both cases they underscore the dynamic and
371
likely punctuated nature of volvocine algal MT evolution.
19
372
Characterization of GpFUS1
373
The G. pectorale FUS1 gene (GpFUS1) was identified based on a BLASTX search
374
with the G. pectorale MT region queried against the C. reinhardtii proteome. Overall GpFUS1
375
has a similar intron-exon structure as CrFUS1 (Figure 4A) with the addition of four additional
376
introns not present in CrFUS1. The protein sequences of GpFUS1 and CrFUS1 have 31% amino
377
acid identity but retain the invasion/intimin repeats that may be involved in mediating mate
378
recognition or membrane fusion (Figure S6; Figure S7) (Misamore et al. 2003) and a signal
379
peptide specifying membrane localization (Figure S8; Figure S9).
380
GpFUS1 mRNA expression was detected in gametes but not vegetative cells, and
381
was up-regulated in gametes by treatment with the mating-related signaling agonist db-cAMP
382
(Figure 4B) (Mogi et al. 2012), similar to its ortholog CrFUS1 (Pasquale and Goodenough 1987;
383
Goodenough et al. 2007). Polyclonal antibodies against recombinant GpFUS1 recognized a
384
protein on Western blots from plus but not minus gametes (Figure 4C). The anti-GpFUS1 signal
385
(70 kD) is smaller than the predicted molecular weight of GpFUS1 (~90kD) in western blotting
386
analysis suggesting that GpFUS1 might undergo post-translational processing.
387
Immunofluorescence using the same antibodies localized GpFUS1 to the plus fertilization tubule
388
in G. pectorale (Figure 4D) indicating a probable function in fertilization that is likely to be
389
similar to that for CrFUS1; no signal was observed in the minus fertilization tubule.
390
Together our results indicate that the FUS1 gene has been conserved as a mating type
391
plus gene in the volvocine lineage with an ancestral function in fertilization that has persisted at
392
least since the G. pectorale-C. reinhardtii split, and has been subsequently lost from the lineage
393
leading to V. carteri. Currently the minus gamete interacting partner for FUS1 is unknown, but
394
our results predict that it will also be conserved between C. reinhardtii and G. pectorale.
20
395
396
A Dynamic History of MT locus evolution in the volvocine lineage
Previous comparative studies of volvocine algal MT loci have raised many questions
397
about their evolutionary histories that this current study begins to address (Charlesworth and
398
Charlesworth 2010; Umen 2011). The C. reinhardtii and V. carteri MT occupy a homologous
399
chromosomal location and share a common ancestral origin, but differ in many respects
400
including gene content, gametolog divergence, locus size, syntenic bloc structure, and repeat
401
content (Ferris et al. 2010). Although the C. reinhardtii MT locus is presumed to more closely
402
reflect the ancestral state of volvocine algal MT loci than the V. carteri MT locus, without
403
additional information about MT loci from other volvocine species the evolutionary trajectory of
404
the mating type region cannot be unambiguously inferred. The structure and gene content of G.
405
pectorale MT described here begins to clarify the evolutionary history of volvocine algal MT loci.
406
In overall size, structure and gametolog divergence, G. pectorale MT resembles C. reinhardtii
407
MT: It has just a few syntenic blocs containing multiple genes (Figure 2; Figure S2), low
408
divergence between most gametolog pairs (Figure 3A), and alignable intergenic regions
409
interrupted in just a few positions with repeats (Figure S2). It also resembles C. reinhardtii in its
410
complement of key mating regulators--MID and MTD1 in MT- and FUS1 in MT+. However G.
411
pectorale MT differs from C. reinhardtii MT in its apparent lack of autosomal insertions into the
412
MT region and in the composition of its gametologs that have more in common with those in V.
413
carteri MT (Figure 1; Table S3). Intriguingly, a small region of G. pectorale MT in the TOC34-
414
HSP70B region shows elevated divergence between gametologs that approach divergence levels
415
found for V. carteri gametologs, and may represent an early stage of MT differentiation in this
416
lineage (Figure 3; Figure S5).
417
Based on these similarities and differences we can infer a minimal set of changes that
21
418
led to the generation of three distinct mating locus regions of C. reinhardtii, G. pectorale and V.
419
carteri (Figure 5). While the actual history of this region is likely to be more complicated than
420
in our depiction and awaits validation, our model creates a framework for classifying the types of
421
changes that occurred in volvocine MT loci and possible ancestral states. The common ancestor
422
of all volvocine algae most likely had a mating locus near its telomere on a chromosome that is
423
syntenic with Chromosome 6 in C. reinhardtii. The ancestral MT locus would have contained
424
MID and MTD1 homologs in its MT- haplotype and a FUS1 homolog in its MT+ haplotype. Its
425
gametologs were genes found on the current C. reinhardtii Chromosome 6 or V. carteri Linkage
426
Group I. After the G. pectorale and C. reinhardtii lineages split one or both of their MT loci
427
underwent a major reconfiguration in which a different group of gametologs became fixed
428
around the rearranged region of MT. During this time the process of gametolog gene conversion
429
was ongoing and maintained relative homogeneity between gametolog pairs between plus and
430
minus haplotypes in each lineage. The autosomal insertions in C. reinhardtii MT+ (De Hoff et
431
al. 2013) were probably not lost in the G. pectorale lineage but gained in the C. reinhardtii
432
lineage, although the latter possibility cannot be excluded (Figure 5). In either case the
433
appearance of G. pectorale MT is “younger” and less complex than that of C. reinhardtii, which
434
suggests a recent reconfiguration. Supporting the idea of reconfiguration is our finding a small
435
and possibly more ancient region containing TOC34 and part of HSP70B that underwent
436
substantial differentiation (Figure 3; Figure S2). By the time of the split between the G.
437
pectorale and V. carteri lineages their common ancestor had acquired many of the gametologs
438
that are still common between these two species. We cannot determine when the FUS1 gene was
439
lost from the V. carteri lineage, but the MTD1 gene may have been lost relatively recently as a
440
possible MTD1 pseudogene was identified in the male V. carteri MT haplotype (Ferris et al.
22
441
2010). In addition, V. carteri MT gained over a dozen male-limited or female-limited MT genes
442
that lack detectable homologs in G. pectorale or elsewhere suggesting that these additional genes
443
arose after the two lineages split. At some point after the G. pectorale-V. carteri split, gametolog
444
divergence in the V. carteri accelerated and the R domain of its MT locus underwent a large
445
expansion. Although we do not know when accelerated evolution and expansion began in the V.
446
carteri lineage, there is additional information on gametolog divergence for the MT gene MAT3,
447
whose sequences have been identified in P. starrii, an oogamous colonial species that is more
448
closely related to V. carteri than to G. pectorale (Figure 1). While the male and female MAT3
449
gametologs of V. carteri are highly diverged (Ferris et al. 2010), those in P. starrii have a low
450
divergence that is comparable to that found in G. pectorale (Hiraide et al. 2013). If the P. starrii
451
MAT3 data are representative of its MT locus as a whole, then the massive divergence observed
452
for V. carteri MT gametologs occurred relatively recently in the lineage.
23
453
Conclusion
454
While many questions remain about the history, timing and mechanism of volvocine
455
algal MT evolution, our data begin to elucidate the highly dynamic nature of their diversification
456
and to establish the trajectory of this process. Overall, there appears to be relative stasis in terms
457
of MT chromosomal location, and in the presence of the sex-determining gene MID. FUS1 and
458
MTD1 appear to have persisted in a large part of the volvocine lineage but became dispensable in
459
V. carteri and perhaps its close relatives. While MTD1 is not totally essential for mating-type
460
differentiation in C. reinhardtii, FUS1 is essential for fertilization (Goodenough et al. 2007; Lin
461
and Goodenough 2007). How and why the FUS1 function was dispensed with or replaced in V.
462
carteri (and perhaps other Volvocine species) and how this change relates to its sexual cycle are
463
interesting areas for future investigation.
464
It is unclear why MT remains associated with the same chromosome in the three
465
widely-separated volvocine species where it has been characterized, but is suggestive that inter-
466
chromosomal sequence exchanges are much less frequent than intra-chromosomal exchanges and
467
rearrangements. Large changes in MT haplotype structure, gametolog content, and gametolog
468
divergence have all occurred in volvocine algae. In all these areas there may be episodic
469
upheavals where the MT locus structure is reconfigured and where the competing effects of
470
autosomal insertions, gametolog gene conversion, and recombination suppression shape
471
haplotype differentiation. Data on haploid mating-type or sex chromosome evolution from other
472
lineages suggests that processes similar to what we documented for volvocine algae are likely to
473
take place elsewhere. For example gene conversion between gametologs has been documented
474
in fungal mating loci as have rearrangements and haplotype bloc formation associated with
24
475
recombination suppression (Sun et al. 2012). Much less is known about haploid SDR regions
476
outside of fungi though some data are now available for haploid bryophyte sex chromosomes
477
(Yamato et al. 2007; McDaniel et al. 2013), and for the SDR region of the brown alga
478
Ectocarpus sp. (Ahmed et al. 2014). Future comparative studies in these other groups and within
479
the volvocine lineage will help elucidate the mechanisms that contribute to mating locus and sex
480
chromosome evolution and how these mechanisms impact the evolution of sex and sexual cycles.
25
481
482
Acknowledgments
This work was supported by Grants-in-Aid for Scientific Research on Innovative
483
Areas “Genome Science” (grant number 221S0002), Scientific Research (A) (grant number
484
24247042 to H.N.), National Institutes of Health grant GM 078376 to JGU, Japan Society for the
485
Promotion of Sciences Research Fellows (19-7661 and 23-5499 to T.H.) from Ministry of
486
Education, Culture, Sports, Science and Technology of Japan/ Japan Society for the Promotion
487
of Sciences KAKENHI. T.H. is a Postdoctoral Fellow for Research Abroad (26-495) of the
488
Japan Society for the Promotion of Science.
26
489
References
490
Ahmed S., Cock J. M., Pessia E., Luthringer R., Cormier A., Robuchon M., Sterck L., Peters A.
491
F., Dittami S. M., Corre E., Valero M., Aury J.-M., Roze D., Van de Peer Y., Bothwell J.,
492
Marais G. A. B., Coelho S. M., 2014 A haploid system of sex determination in the brown
493
alga Ectocarpus sp. Curr. Biol. 24: 1945–1957.
494
495
496
497
498
499
500
501
502
Altschul S. F., Gish W., Miller W., Myers E. W., Lipman D. J., 1990 Basic local alignment
search tool. J. Mol. Biol. 215: 403–410.
Bachtrog D., Kirkpatrick M., Mank J. E., McDaniel S. F., Pires J. C., Rice W., Valenzuela N.,
2011 Are all sex chromosomes created equal? Trends Genet. 27: 350–357.
Charlesworth D., Charlesworth B., 2010 Evolutionary biology: the origins of two sexes. Curr.
Biol. 20: R519–R521.
Coleman A. W., 2012 A comparative analysis of the Volvocaceae (Chlorophyta). J. Phycol. 48:
491–513.
De Hoff P. L., Ferris P., Olson B. J. S. C., Miyagi A., Geng S., Umen J. G., 2013 Species and
503
population level molecular profiling reveals cryptic recombination and emergent
504
asymmetry in the dimorphic mating locus of C. reinhardtii. PLoS Genet 9: e1003724.
505
506
507
Ferris P. J., Goodenough U. W., 1994 The mating-type locus of Chlamydomonas reinhardtii
contains highly rearranged DNA sequences. Cell 76: 1135–1145.
Ferris P., Woessner J., Goodenough U., 1996 A sex recognition glycoprotein is encoded by the
27
508
509
510
511
512
513
514
515
plus mating-type gene fus1 of Chlamydomonas reinhardtii. Mol. Biol. Cell 7: 1235–1248.
Ferris P. J., Pavlovic C., Fabry S., Goodenough U. W., 1997 Rapid evolution of sex-related
genes in Chlamydomonas. Proc. Natl. Acad. Sci. U. S. A. 94: 8634–8639.
Ferris P. J., Goodenough U. W., 1997 Mating Type in Chlamydomonas Is Specified by mid, the
Minus-Dominance Gene. Genetics 146: 859–869.
Ferris P. J., Armbrust E. V., Goodenough U. W., 2002 Genetic structure of the mating-type locus
of Chlamydomonas reinhardtii. Genetics 160: 181–200.
Ferris P., Olson B. J. S. C., De Hoff P. L., Douglass S., Casero D., Prochnik S., Geng S., Rai R.,
516
Grimwood J., Schmutz J., Nishii I., Hamaji T., Nozaki H., Pellegrini M., Umen J. G.,
517
2010 Evolution of an expanded sex-determining locus in Volvox. Science 328: 351–354.
518
Gasteiger E., Gattiker A., Hoogland C., Ivanyi I., Appel R. D., Bairoch A., 2003 ExPASy: the
519
proteomics server for in-depth protein knowledge and analysis. Nucleic Acids Res. 31:
520
3784 –3788.
521
522
523
524
Geng S., De Hoff P., Umen J. G., 2014 Evolution of sexes from an ancestral mating-type
specification pathway. PLoS Biol 12: e1001904.
Goodenough U., Lin H., Lee J.-H., 2007 Sex determination in Chlamydomonas. Semin. Cell Dev.
Biol. 18: 350–361.
525
Hamaji T., Ferris P. J., Coleman A. W., Waffenschmidt S., Takahashi F., Nishii I., Nozaki H.,
526
2008 Identification of the minus-dominance gene ortholog in the mating-type locus of
527
Gonium pectorale. Genetics 178: 283–294.
28
528
Hamaji T., Ferris P. J., Nishii I., Nozaki H., 2009 Identification of the minus mating-type
529
specific gene MTD1 from Gonium pectorale (Volvocales, Chlorophyta). J. Phycol. 45:
530
1310–1314.
531
Hamaji T., Ferris P. J., Nishii I., Nishimura Y., Nozaki H., 2013a Distribution of the Sex-
532
Determining Gene MID and Molecular Correspondence of Mating Types within the
533
Isogamous Genus Gonium (Volvocales, Chlorophyta). PLoS ONE 8: e64385.
534
Hamaji T., Smith D. R., Noguchi H., Toyoda A., Suzuki M., Kawai-Toyooka H., Fujiyama A.,
535
Nishii I., Marriage T., Olson B. J. S. C., Nozaki H., 2013b Mitochondrial and plastid
536
genomes of the colonial green alga Gonium pectorale give insights into the origins of
537
organelle DNA architecture within the Volvocales. PLoS ONE 8: e57177.
538
Hanschen E. R., Marriage T. N., Ferris P. J., Hamaji T., Toyoda A., Fujiyama A., Neme R.,
539
Noguchi H., Minakuchi Y., Suzuki M., Kawai-Toyooka H., Smith D. R., Sparks H.,
540
Anderson J., Bakarić R., Luria V., Karger A., Kirschner M., Durand P. M., Michod R. E.,
541
Nozaki H., Olson B. J. S. C., in press Cell cycle regulation, group level adaptation, and
542
the evolution of multicellularity. Nat. Commun.
543
544
545
Herron M. D., Michod R. E., 2008 Evolution of complexity in the volvocine algae: transitions in
individuality through Darwin’s eye. Evolution 62: 436–451.
Hiraide R., Kawai-Toyooka H., Hamaji T., Matsuzaki R., Kawafune K., Abe J., Sekimoto H.,
546
Umen J., Nozaki H., 2013 The evolution of male–female sexual dimorphism predates the
547
gender-based divergence of the mating locus gene MAT3/RB. Mol. Biol. Evol. 30: 1038–
548
1040.
29
549
550
551
Immler S., Otto S. P., 2015 The evolution of sex chromosomes in organisms with separate
haploid sexes. Evolution 69: 694–708.
Larkin M. A., Blackshields G., Brown N. P., Chenna R., McGettigan P. A., McWilliam H.,
552
Valentin F., Wallace I. M., Wilm A., Lopez R., Thompson J. D., Gibson T. J., Higgins D.
553
G., 2007 Clustal W and Clustal X version 2.0. Bioinformatics 23: 2947–2948.
554
Librado P., Rozas J., 2009 DnaSP v5: a software for comprehensive analysis of DNA
555
556
polymorphism data. Bioinformatics 25: 1451–1452.
Lin H., Goodenough U. W., 2007 Gametogenesis in the Chlamydomonas reinhardtii minus
557
mating type is controlled by two genes, MID and MTD1. Genetics 176: 913–925.
558
Liu Y.-G., Whittier R. F., 1995 Thermal asymmetric interlaced PCR: automatable amplification
559
and sequencing of insert end fragments from P1 and YAC clones for chromosome
560
walking. Genomics 25: 674–681.
561
McDaniel S. F., Neubig K. M., Payton A. C., Quatrano R. S., Cove D. J., 2013 Recent gene-
562
capture on the UV sex chromosomes of the moss Ceratodon purpureus. Evolution 67:
563
2811–2822.
564
Merchant S. S., Prochnik S. E., Vallon O., Harris E. H., Karpowicz S. J., Witman G. B., Terry A.,
565
Salamov A., Fritz-Laylin L. K., Marechal-Drouard L., Marshall W. F., Qu L.-H., Nelson
566
D. R., Sanderfoot A. A., Spalding M. H., Kapitonov V. V., Ren Q., Ferris P., Lindquist E.,
567
Shapiro H., Lucas S. M., Grimwood J., Schmutz J., Cardol P., Cerutti H., Chanfreau G.,
568
Chen C.-L., Cognat V., Croft M. T., Dent R., Dutcher S., Fernandez E., Fukuzawa H.,
569
Gonzalez-Ballester D., Gonzalez-Halphen D., Hallmann A., Hanikenne M., Hippler M.,
30
570
Inwood W., Jabbari K., Kalanon M., Kuras R., Lefebvre P. A., Lemaire S. D., Lobanov A.
571
V., Lohr M., Manuell A., Meier I., Mets L., Mittag M., Mittelmeier T., Moroney J. V.,
572
Moseley J., Napoli C., Nedelcu A. M., Niyogi K., Novoselov S. V., Paulsen I. T., Pazour
573
G., Purton S., Ral J.-P., Riano-Pachon D. M., Riekhof W., Rymarquis L., Schroda M.,
574
Stern D., Umen J., Willows R., Wilson N., Zimmer S. L., Allmer J., Balk J., Bisova K.,
575
Chen C.-J., Elias M., Gendler K., Hauser C., Lamb M. R., Ledford H., Long J. C.,
576
Minagawa J., Page M. D., Pan J., Pootakham W., Roje S., Rose A., Stahlberg E.,
577
Terauchi A. M., Yang P., Ball S., Bowler C., Dieckmann C. L., Gladyshev V. N., Green
578
P., Jorgensen R., Mayfield S., Mueller-Roeber B., Rajamani S., Sayre R. T., Brokstein P.,
579
Dubchak I., Goodstein D., Hornick L., Huang Y. W., Jhaveri J., Luo Y., Martinez D.,
580
Ngau W. C. A., Otillar B., Poliakov A., Porter A., Szajkowski L., Werner G., Zhou K.,
581
Grigoriev I. V., Rokhsar D. S., Grossman A. R., 2007 The Chlamydomonas genome
582
reveals the evolution of key animal and plant functions. Science 318: 245–250.
583
Misamore M. J., Gupta S., Snell W. J., 2003 The Chlamydomonas Fus1 protein is present on the
584
mating type plus fusion organelle and required for a critical membrane adhesion event
585
during fusion with minus gametes. Mol Biol Cell 14: 2530–2542.
586
Mogi Y., Hamaji T., Suzuki M., Ferris P., Mori T., Kabeya Y., Miyagishima S., Nozaki H., 2012
587
Evidence for tubular mating structures induced in each mating type of heterothallic
588
Gonium pectorale (Volvocales, Chlorophyta). J. Phycol. 48: 670–674.
589
Nei M., 1987 Molecular Evolutionary Genetics. Columbia University Press.
590
Nozaki H., Misawa K., Kajita T., Kato M., Nohara S., Watanabe M. M., 2000 Origin and
591
evolution of the colonial Volvocales (Chlorophyceae) as inferred from multiple,
31
592
593
594
595
596
chloroplast gene sequences. Mol. Phylogenet. Evol. 17: 256–268.
Nozaki H., Mori T., Misumi O., Matsunaga S., Kuroiwa T., 2006 Males evolved from the
dominant isogametic mating type. Curr. Biol. 16: R1018–R1020.
Pasquale S. M., Goodenough U. W., 1987 Cyclic AMP functions as a primary sexual signal in
gametes of Chlamydomonas reinhardtii. J. Cell Biol. 105: 2279–2292.
597
Prochnik S. E., Umen J., Nedelcu A. M., Hallmann A., Miller S. M., Nishii I., Ferris P., Kuo A.,
598
Mitros T., Fritz-Laylin L. K., Hellsten U., Chapman J., Simakov O., Rensing S. A., Terry
599
A., Pangilinan J., Kapitonov V., Jurka J., Salamov A., Shapiro H., Schmutz J., Grimwood
600
J., Lindquist E., Lucas S., Grigoriev I. V., Schmitt R., Kirk D., Rokhsar D. S., 2010
601
Genomic analysis of organismal complexity in the multicellular green alga Volvox carteri.
602
Science 329: 223–226.
603
604
605
606
607
Salamov A. A., Solovyev V. V., 2000 Ab initio Gene Finding in Drosophila Genomic DNA.
Genome Res. 10: 516 –522.
Sambrook J., Russell D. W., 2001 Molecular Cloning: A Laboratory Manual. NY Cold Spring
Harbor Laboratory Press.
Setohigashi Y., Hamaji T., Hayama M., Matsuzaki R., Nozaki H., 2011 Uniparental Inheritance
608
of Chloroplast DNA Is Strict in the Isogamous Volvocalean Gonium. PLoS ONE 6:
609
e19545.
610
611
Soderlund C., Nelson W., Shoemaker A., Paterson A., 2006 SyMAP: A system for discovering
and viewing syntenic regions of FPC maps. Genome Res. 16: 1159–1168.
32
612
613
Soderlund C., Bomhoff M., Nelson W. M., 2011 SyMAP v3.4: a turnkey synteny system with
application to plant genomes. Nucleic Acids Res. 39: e68–e68.
614
Sun S., Hsueh Y.-P., Heitman J., 2012 Gene conversion occurs within the mating-type locus of
615
Cryptococcus neoformans during sexual reproduction. PLoS Genet 8: e1002810.
616
617
618
Umen J. G., 2011 Evolution of sex and mating loci: An expanded view from Volvocine algae.
Curr. Opin. Microbiol. 14: 634–641.
Yamada T. K., Nakada T., Miyaji K., Nozaki H., 2006 Morphology and molecular phylogeny of
619
Gonium multicoccum (Volvocales, Chlorophyceae) newly found in Japan. J. Jpn. Bot.
620
81: 139–147.
621
Yamato K. T., Ishizaki K., Fujisawa M., Okada S., Nakayama S., Fujishita M., Bando H.,
622
Yodoya K., Hayashi K., Bando T., Hasumi A., Nishio T., Sakata R., Yamamoto M.,
623
Yamaki A., Kajikawa M., Yamano T., Nishide T., Choi S.-H., Shimizu-Ueda Y., Hanajiri
624
T., Sakaida M., Kono K., Takenaka M., Yamaoka S., Kuriyama C., Kohzu Y., Nishida H.,
625
Brennicke A., Shin-i T., Kohara Y., Kohchi T., Fukuzawa H., Ohyama K., 2007 Gene
626
organization of the liverwort Y chromosome reveals distinct sex chromosome evolution
627
in a haploid system. Proc. Natl. Acad. Sci. 104: 6472–6477.
628
629
630
631
Yang Z., Nielsen R., 2000 Estimating synonymous and nonsynonymous substitution rates under
realistic evolutionary models. Mol. Biol. Evol. 17: 32 –43.
Yang Z., 2007 PAML 4: phylogenetic analysis by maximum likelihood. Mol. Biol. Evol. 24:
1586.
33
632
Figure legends
Gametes
+
MID
–
+
MID
–
♀
Isogamy
♂
MID
Anisogamy
♀
♂
MID
Oogamy
Vegetative
Chlamydomonas Gonium Yamagishiella
Pleodorina
reinhardtii
pectorale
unicocca Eudorina sp.
starrii
Volvox
carteri
633
634
Figure 1.
A schematic diagram for phylogenetic relationships of selected volvocine species
635
based on Nozaki et al. (2000); Herron and Michod (2008). The top row illustrates
636
gamete type and structure. Tubular mating structures in isogamous gametes are
637
indicated with red bars at the flagellar base. The possession of a MID gene is
638
shown next to the minus mating type or male gametes. The lower row of cartoons
639
depicts vegetative morphology (not to scale) for the indicated species.
640
34
A
C
Gonium pectorale
MT+
PTC1
II
WDR57
III
v
WDR57
DRG1
PTC1
vi
vii
Female MT
TOC34
HSP70B
ATPvC
MME6
LEU1S
METM
SPS1
EIF5Bb
GMTp03
MAT3
I
25 kb
MAT3
GMTp03
EIF5Bb
SPS1
METM
LEU1S
MME6
ATPvC
HSP70B
FUS1
TOC34
SelEF
GMTp14
RPL37A
AMPKR1
ATPvL1
FTT2
GMTp19
PGM6
ANK17
Volvox carteri
MT–
IV
MID
MTD1
SelEF
GMTp14
RPL37A
AMPKR1
ATPvL1
FTT2
GMTp19
PGM6
ANK17
TOC34
MTF0821 MTF0822
AIP1
MTF0870
PTC1
HSP70b
METM1
EFG8
MTF0991
MTF0992
MTF1026
MME6
ALB3.1
MTF1109
MAT3
HRGP1
DHC1b
B
MT+
EZY2-OTU2INT1 repeat
HRGP1
UTP1
MT0796
SRL
MT0828
MT0829
FUM1
FBX9
641
642
ATPvC1
SPL2
HMG1
MTF1320
Chlamydomonas reinhardtii
HDH1
TOC34
MTP0428
PDK1
CGL70
NMDA1
DRG1
DLA3
522872
LEU1S
SPL2
FUS1
LPS1
522875
MTA5
MTA4
MTA1
MTA3
294708
PKY1
MADS2
UBCH1
GCSH
PR46a
PR46b
MT0618
155027
Figure 2.
MT–
NMT1
AMPKR1
MTF1436
25 kb
HDH1
TOC34
522875
LPS1
PDK1
CGL70
NMDA1
DRG1
Ѱ-EZY2
OTU2a
155027
MT0618
PR46b
PR46a
GCSH
UBCH1
MADS2
PKY1
MID
DLA3
522872
MTD1
LEU1S
SPL2
HRGP1
UTP1
MT0796
MT0828
MT0829
FUM1
FBX9
Male MT
RPS3A2
MT0044 PRP4
CGL55 MT0045
PIP5K FAP75
MT0083
PAP1
(VPS53 repeats)
RGP1
MTF0674
MTF0684
MAPKK1
MTF0709 L7Ae ARP4
MTF0716
SAD1
MTF1458
MTF1484
SPS1
PRX1
25 kb
RPS3A2
PRP4 MT0044
MT0045 CGL55
FAP75 PIP5K
MT0083
PAP1
(VPS53 repeats)
SAD1
MTM0001 MTM0015
ARP4 L7Ae
MAPKK1
MTM0041
MTM0046
WDR57
MTM0097 LEU1S
UNC50
PTC1
EFG8
MTM0210
PSF2
METM1
HSP70b
MID
DRG1
MTM0349
Ѱ-MTD1
MTM0397
MTM0417
SPS1
MTM0441
AIP1
ATPvC1
UNC50
EIF5Bb
PSF2
BFR1
MTF1733
MOT41
DRG1
CRB1
WDR57
LEU1S
FA1
FTT2 UCP2
ATPvL1
PKY1
MT2177
MADS2
UBCH1
GCSH1
PR46a
PR46b
RPL37a
BBS2
MT2215
SelEF
MTF2030
FSI1
UTP1
MAT3
RGP1
MTM0564
CRB1
EIF5Bb
SelEF
MTM0629 MTM0637
MTM0638
TOC34
MTM0665
DHC1b
ALB3.1
MTM0761
SPL2
MTM0775
NMT1
MTM0832
MME6
MOT41
UTP1
MTM0897
ATPvL1 FTT2
UCP2
FA1
MTM0946
PKY1
PRX1
MT2177
BFR1
MADS2
AMPKR1
UBCH1
MTM1037
MTM1058
GCSH1
HRGP1
PR46a
PR46b
RPL37a
BBS2
MT2215
Schematic of volvocine mating loci for A) Gonium pectorale, B) Chlamydomonas
35
643
reinhardtii (modified from De Hoff et al. 2013), and C) Volvox carteri (modified from
644
Ferris et al. 2010) with rearranged (R) domains in light blue (minus/male) or pink
645
(plus/female). For G. pectorale MT, syntenic blocs are indicated by grey shading
646
and labeled with upper case roman numbers (I-IV), while sequences that are unique
647
to one of the two mating haplotypes are indicated with lower case roman numbers
648
(v-vii). Red and blue shading on gene names indicates plus/female and minus/male
649
specific genes, respectively. Grey dots beside gene names indicate those found in
650
the R domains of all three species; dark blue triangles indicate presence of gene in
651
R domain of G. pectorale and V. carteri only; green squares indicate presence of
652
gene in R domain of C. reinhardtii and V. carteri only. No R domain genes are
653
shared exclusively by G. pectorale and C. reinhardtii.
654
36
dN
0.7
dS
3.0
A
B
0.500
2.5
0.6
0.375
0.250
0.5
0.125
0
1.0
0.3
1.5
0.4
2.0
0
0.2
TOC34
427
853
1280
1706
TOC34
(within syntenic breakage)
C
0.1500
0.1
0.5
0.1125
HSP70B
0.0
0.0375
0.0
Chlamydomonas reinhardtii Gonium pectorale
N = 28
N = 21
0.0750
0
0
1165
Volvox carteri
N = 60
HSP70B
655
656
Figure 3.
2329
3494
4658
syntenic breakage
Molecular evolutionary analyses of volvocine algal gametologs. A) Box-whisker
657
plots comparing the distributions of dS (blue) and dN (red) values for
658
Chlamydomonas reinhardtii, Gonium pectorale and Volvox carteri. Open dots in G.
659
pectorale are values for indicated genes. B, C) Sliding window plots of gametolog
660
similarity (Pi) for TOC34 (B) or HSP70B (C). Position within each gene is indicated
661
on x axis by bp number beginning with the start codon.
662
663
37
C
plu
s
m
inu
B
s
A
D
plus
minus
664
665
Figure 4.
Identification and characterization of GpFUS1. (A) Comparison of FUS1 intron-exon
666
structures between Chlamydomonas reinhardtii and Gonium pectorale.
667
Homologous CDS (black boxes) sequences are connected by grey shading. Introns
668
are represented by thin lines. UTRs are represented by thick lines. (B)
669
Semiquantitative RT-PCR of GpFUS1 from MT+ G. pectorale strain. mRNAs were
670
isolated from vegetative cells, gametes, and sexually activated gametes with db-
671
cAMP (D-cAMP) treatment. EF-like gene was used as an internal control for RT-
672
PCR. (C) Immunoblot analysis of GpFUS1 expression of activated plus and minus
38
673
gametes using anti-GpFUS1 antibody. (D) Double immunofluorescence staining of
674
db-cAMP activated mating type plus and minus gametes using anti-GpFUS1 and
675
anti-actin antibody. Arrows indicate fertilization tubule, a mating apparatus that
676
contains actin filament bundles.
39
Isogamy Oogamy
Suggested evolutionary events
gene loss
rearrangement/
gene conversion
gene acquisition
MT female
MT male
MID
MT-
MT- MT+
MID&
MTD1
FUS1
MID
MTD1
MT+
MT- MT+
FUS1
MID
MTD1
volvocine ancestral MT Chlamydomonas reinhardtii MT
(hypothetical)
FUS1
Volvox carteri MT
Gonium pectorale MT
ongoing gene conversion
autosomal duplications
into MT
}
further reformation:
massive gene translocation
from MT/flanking region
}
FUS1 gene loss/
MTD1 pseudogene
}
complete loss of
recombination/
further rearrangement
acquisition of
novel mating type
specific genes
}
}
reformation:
flanking genes into rearranged domain
}
677
678
Figure 5.
Possible evolutionary history for volvocine MT loci based on minimal changes
679
necessary to explain observed results in this study and in previous studies (Ferris et
680
al. 2002, 2010; De Hoff et al. 2013). Each proposed event is indicated by a thick
681
line crossing the node accompanying a circle: open, rearrangement or gene
682
conversion event; filled, gene loss; red, gene acquisition.
683
684
40
685
Tables
686
Table 1:
687
Genome and Mating Locus Properties of Gonium pectorale, Chlamydomonas
reinhardtii and Volvox carteri.
Gonium pectorale
MT+
MT-
Chlamydomonas
reinhardtii
whole
MT+
MT-
Volvox carteri
whole
female
male
whole
Size (Mb)
0.366
0.499
148.8
0.310
0.204
111.1
1.51
1.13
131.1
%GC
Gene
density
[genes/Mb]
Introns /
gene
Average
intron
length (bp)
59.7
57.38
61.0
46.09
64.5
120.9
60
74.25
61
117.6
64.1
159.6
52
39
53
54
56.1
114.1
9.62
9.09
6.50
7.1
6.37
7.46
9
8
6.29
175.8
278.50
349.83
354.17
358.70
420.01
618
584
399.50
688
689
41
690
Table 2.
691
Linkage Mapping of Progeny from Gonium pectorale Mongolia 1 and 4
Crossing Revised from a Former Experiment (Hamaji et al. 2008).
692
Scaffold* 5
3
1
2
97
7
7
MT (phenotype) MT 10
†
Character* RABB MTF110 PRX UNC5 IDA5 PR46 ALB3. LEU1 MID MT
PGM SpoV
*
1
9
1
0
a
1
S
6
S
Cre.
2
6
6
6
13 6
6
6
6
6
6
Chr.***
RABB1
34/78
42/7 38/77 42/7 37/77 38/78 40/78 40/7 35/69
40/78 40/78
8
8
8
MTF1109
34/7 43/77 36/7 33/77 34/78 44/78 44/7 40/69
44/78 44/78
8
8
8
PRX1
34/77 40/7 32/77 26/78 38/78 38/7 36/69
38/78 38/78
8
8
UNC50
42/7 33/76 37/77 38/77 38/7 33/68
38/77 38/77
7
7
IDA5
39/77 36/78 48/78 48/7 42/69
48/78 48/78
8
PR46a
12/77 44/77 44/7 39/68
44/77 44/77
7
ALB3.1
40/78 40/7 38/69
40/78 40/78
8
LEU1S
0/78 0/69
0/78 0/78
MID
0/69
MT
0/78 0/78
0/69 0/69
PGM6
0/78
PLC
693
694
695
696
697
698
MT
10
PLC
6
35/7
8
49/7
8
41/7
8
39/7
7
49/7
8
45/7
7
41/7
8
13/7
8
13/7
8
12/6
9
13/7
8
13/7
8
* Scaffold in Gonium pectorale genome assembly.
** Genotypic or phenotypic characters assayed.
*** Physical positions in Chlamydomonas reinhardtii chromosome for homologs of genetic markers.
† Mating phenotype.
Ratio of recombinant progeny to total progeny
42