Available via license: CC BY-NC 4.0
Content may be subject to copyright.
1
Functional and evolutionary integration of a fungal gene with a bacterial 1
operon 2
3
Liang Sun1,2, Kyle T. David3, John F. Wolters1,2, Steven D. Karlen1, Carla Gonçalves2,3,4, Dana A. 4
Opulente1,2,5, Abigail Leavitt LaBella3,6, Marizeth Groenewald7, Xiaofan Zhou3,8, Xing-Xing 5
Shen3,9, Antonis Rokas3, Chris Todd Hittinger1,2,* 6
7
1DOE Great Lakes Bioenergy Research Center, Wisconsin Energy Institute, University of 8
Wisconsin-Madison, Madison, WI 53726, USA 9
2Laboratory of Genetics, Center for Genomic Science Innovation, Wisconsin Energy Institute, J. 10
F. Crow Institute for the Study of Evolution, University of Wisconsin-Madison, Madison, WI 11
53726, USA 12
3Evolutionary Studies Initiative and Department of Biological Sciences, Vanderbilt University, 13
Nashville, TN 37235, USA 14
4UCIBIO, Department of Life Sciences, NOVA School of Science and Technology, 15
Universidade NOVA de Lisboa, Caparica, Portugal 16
5Biology Department, Villanova University, Villanova, PA 19085, USA 17
6Department of Bioinformatics and Genomics, University of North Carolina at Charlotte, 18
Charlotte, NC 28223 19
7Westerdijk Fungal Biodiversity Institute, 3584 CT Utrecht, The Netherlands 20
8College of Agriculture and Biotechnology and Centre for Evolutionary & Organismal Biology, 21
Zhejiang University, Hangzhou 310058, China 22
9Guangdong Province Key Laboratory of Microbial Signals and Disease Control, Integrative 23
Microbiology Research Center, South China Agricultural University, Guangzhou 510642, China 24
25
*Corresponding author: cthittinger@wisc.edu 26
27
.CC-BY-NC 4.0 International licenseavailable under a
(which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made
The copyright holder for this preprintthis version posted November 21, 2023. ; https://doi.org/10.1101/2023.11.21.568075doi: bioRxiv preprint
2
Abstract 28
Siderophores are crucial for iron-scavenging in microorganisms. While many yeasts can uptake 29
siderophores produced by other organisms, they are typically unable to synthesize siderophores 30
themselves. In contrast, Wickerhamiella/Starmerella (W/S) clade yeasts gained the capacity to 31
make the siderophore enterobactin following the remarkable horizontal acquisition of a bacterial 32
operon enabling enterobactin synthesis. Yet, how these yeasts absorb the iron bound by 33
enterobactin remains unresolved. Here, we demonstrate that Enb1 is the key enterobactin 34
importer in the W/S-clade species Starmerella bombicola. Through phylogenomic analyses, we 35
show that ENB1 is present in all W/S clade yeast species that retained the enterobactin 36
biosynthetic genes. Conversely, it is absent in species that lost the ent genes, except for 37
Starmerella stellata, making this species the only cheater in the W/S clade that can utilize 38
enterobactin without producing it. Through phylogenetic analyses, we infer that ENB1 is a fungal 39
gene that likely existed in the W/S clade prior to the acquisition of the ent genes and 40
subsequently experienced multiple gene losses and duplications. Through phylogenetic topology 41
tests, we show that ENB1 likely underwent horizontal gene transfer from an ancient W/S clade 42
yeast to the order Saccharomycetales, which includes the model yeast Saccharomyces cerevisiae, 43
followed by extensive secondary losses. Taken together, these results suggest that the fungal 44
ENB1 and bacterial ent genes were cooperatively integrated into a functional unit within the W/S 45
clade that enabled adaptation to iron-limited environments. This integrated fungal-bacterial 46
circuit and its dynamic evolution determines the extant distribution of yeast enterobactin 47
producers and cheaters. 48
49
.CC-BY-NC 4.0 International licenseavailable under a
(which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made
The copyright holder for this preprintthis version posted November 21, 2023. ; https://doi.org/10.1101/2023.11.21.568075doi: bioRxiv preprint
3
Introduction 50
Most organisms on earth require iron as a redox mediator that is key to numerous 51
biological processes, including respiration; protection from reactive oxygen species; 52
photosynthesis; and the biosynthesis of amino acids, lipids, deoxyribonucleotides, and sterols. 53
Although iron is abundant in the crust of Earth, its availability to living organisms is generally 54
low. This poor bioavailability occurs because iron exists mainly in its oxidized ferric (Fe3+) state, 55
which is largely insoluble in water in neutral and basic pH environments. To circumvent iron 56
shortage, microorganisms have evolved the capacity to scavenge iron from environmental stocks 57
by secreting and taking up siderophores, a chemically diverse group of secondary metabolites 58
that bind to Fe3+ with very high affinity and specificity (Haas et al. 2008; Kramer et al. 2020). 59
These siderophore-mediated iron acquisition systems require active synthesis and secretion of 60
siderophores in iron-free form, as well as specific transporter mechanisms to recognize and 61
selectively take up the Fe3+-siderophore complex. 62
In contrast to most bacteria and filamentous fungi, yeasts of the subphylum 63
Saccharomycotina (hereafter, yeasts) are generally unable to synthesize siderophores with a few 64
exceptions, such as Kluyveromyces lactis, which produces pulcherrimin (Krause et al. 2018). The 65
model yeast Saccharomyces cerevisiae, however, can utilize iron bound to various structurally 66
distinct siderophores produced by other microbial species through two distinct systems (Yun et 67
al. 2001). One system depends on the FRE gene family, which encodes plasma membrane 68
reductases (Fre1, Fre2, or Fre3) that reduce and dissociate iron from siderophores at the cell 69
surface; iron is then translocated through the plasma membrane by the high-affinity ferrous iron 70
(Fe2+) transporter complex (Ftr1 and Fet3) (Stearman et al. 1996). Another system depends on 71
.CC-BY-NC 4.0 International licenseavailable under a
(which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made
The copyright holder for this preprintthis version posted November 21, 2023. ; https://doi.org/10.1101/2023.11.21.568075doi: bioRxiv preprint
4
four siderophore transporters of the major facilitator superfamily (MFS) that are distinct in 72
substrate specificity and named Arn1, Arn2, Sit1 (also known as Arn3), and Enb1 (also known as 73
Arn4) (Lesuisse et al. 1998; Heymann et al. 1999; Heymann et al. 2000a; Heymann et al. 2000b). 74
This system allows a Fe3+-siderophore complex to enter the cells prior to any reduction step with 75
high affinity and high specificity. Among these siderophore transporters, Enb1 is highly specific 76
to the siderophore enterobactin and has previously only been reported in yeasts belonging to the 77
genus Saccharomyces, whereas Sit1 transporters, which recognize a wide variety of 78
siderophores, are abundant across the yeast phylogeny (Dias and Sá-Correia 2013). From an 79
ecological and evolutionary perspective, siderophores can be shared between cells as public 80
goods, and their production typically represents a form of cooperation (Julou et al. 2013; Weigert 81
and Kümmerli 2017). However, yeasts with siderophore transporters can act as cheaters and 82
exploit the benefits of siderophore production without paying the cost of synthesis and, thus, may 83
outcompete the producers to a certain extent (Griffin et al. 2004; Dumas and Kümmerli 2012). 84
The potential fitness benefits of being a cheater may have driven the evolution of siderophore 85
transporter diversity in yeasts. While the biochemical properties of siderophore transporters have 86
been examined by a wealth of studies in S. cerevisiae and other model yeasts (Yun et al. 2000; 87
Ardon et al. 2001; Heymann et al. 2002), their evolutionary origins are largely unknown. 88
The exponentially rising amount of genomic sequence data have made yeast comparative 89
genomics a powerful tool to explore the genetic bases and evolutionary histories of important 90
traits (Butler et al. 2004; Hittinger and Carroll 2007; Marcet-Houben and Gabaldón 2015; 91
Vakirlis et al. 2016; Gonçalves and Gonçalves 2019; Krause and Hittinger 2022). One of the 92
most striking evolutionary events brought to light by comparative genomics is the horizontal 93
operon transfer (HOT) of an enterobactin biosynthesis pathway from an ancient bacterium in the 94
.CC-BY-NC 4.0 International licenseavailable under a
(which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made
The copyright holder for this preprintthis version posted November 21, 2023. ; https://doi.org/10.1101/2023.11.21.568075doi: bioRxiv preprint
5
order Enterobacteriales into the Wickerhamiella/Starmerella (W/S) clade of yeasts (Kominek et 95
al. 2019). All the ent genes required for enterobactin biosynthesis were maintained and 96
functionally expressed in at least some of the yeast species examined, which constitute a newly 97
discovered and surprising group of enterobactin producers. However, genes encoding bacterial 98
ATP binding cassette (ABC) transporters for enterobactin uptake were not found in these yeasts. 99
The capacity to lock iron away from competitors would be counterproductive if the producer 100
species were unable to utilize the iron bound to enterobactin, which suggests the presence of one 101
or more alternative transporter mechanisms in the W/S clade yeasts. 102
Siderophore-mediated iron uptake is crucial for fungal iron homeostasis and virulence, but 103
it has rarely been studied in yeasts other than S. cerevisiae and the opportunistic pathogen 104
Candida albicans (Haas et al. 2008). Here, we investigate iron transport mechanisms using the 105
enterobactin-producing yeast Starmerella bombicola as a model for targeted gene disruption and 106
show that this species uses the Enb1 transporter for the uptake of enterobactin-bound iron. 107
Through phylogenomic analyses, we show that ENB1 is a fungal gene that was likely present in 108
the most recent common ancestor (MRCA) of the Dikarya, the fungal clade that contains the 109
phyla Basidiomycota and Ascomycota. This fungal gene was then functionally integrated with 110
the horizontally acquired bacterial ent operon and evolved to operate coordinately in the W/S-111
clade yeasts as an adaptation to varying iron availability. Finally, we infer that a horizontal 112
transfer of the ENB1 gene from an ancestor of the W/S clade (order Dipodascles) to an ancestor 113
of the order Saccharomycetales (which includes S. cerevisiae), along with multiple independent 114
gene duplications and losses, shaped the patchy phylogenetic distribution of enterobactin 115
cheaters in yeasts. This study highlights the intriguing interplay between fungal and bacterial 116
.CC-BY-NC 4.0 International licenseavailable under a
(which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made
The copyright holder for this preprintthis version posted November 21, 2023. ; https://doi.org/10.1101/2023.11.21.568075doi: bioRxiv preprint
6
genes, which formed a functional unit in some yeasts to produce a valuable resource, and later 117
separated through dynamic evolution as cheaters arose to exploit that resource. 118
119
Results and Discussion 120
Enb1 imports enterobactin in Starmerella bombicola 121
While the HOT event and the subsequent genetic adaptations enabling bacterial gene 122
expression and enterobactin synthesis in the W/S-clade yeasts are well-documented (Kominek et 123
al. 2019), the mechanism by which these yeasts uptake the iron bound by enterobactin remains 124
unknown. In the model bacterium Escherichia coli, the Fe3+-enterobactin complex is first 125
translocated into the periplasm by the outer-membrane siderophore receptor FepA; it is then 126
transported into the cytoplasm via the ATP binding cassette (ABC) transporter encoded by three 127
genes: fepC, fepD, and fepG, which are parts of the E. coli ent operon (Usher et al. 2001; 128
Raymond et al. 2003). However, neither fepA, fepC, fepD, nor fepG was found in the genomes of 129
any W/S-clade yeasts, suggesting the presence of one or more alternative transport mechanisms. 130
We examined the possible mechanisms by which W/S-clade yeasts utilize enterobactin-131
bound iron using St. bombicola as a genetically tractable model species. Functionally related 132
genes are often located in genomic proximity, even in eukaryotes (Hurst et al. 2004). The 133
intercalation of genes encoding a eukaryotic ferric reductase (Fre) and an uncharacterized 134
transmembrane protein (Tm) between two ent genes in a subset of the W/S-clade yeasts 135
suggested a possible role in reductive uptake for this FRE-TM gene pair (Figure 1A). We also 136
identified a homolog of the S. cerevisiae ENB1 (ScENB1) in St. bombicola with 52% amino acid 137
sequence identity and 96% coverage, which suggested the possibility that its function might be 138
.CC-BY-NC 4.0 International licenseavailable under a
(which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made
The copyright holder for this preprintthis version posted November 21, 2023. ; https://doi.org/10.1101/2023.11.21.568075doi: bioRxiv preprint
7
conserved as well. To determine which, if any, of these candidate genes were involved in the 139
utilization of enterobactin-bound iron, we engineered two St. bombicola mutants by deleting 140
either the ScENB1 homolog (enb1
D
) or the FRE-TM (fre-tm
D
) gene cluster and compared their 141
growth to the wildtype (WT) strain in both regular and low-iron media. The fre-tmΔ mutant 142
exhibited comparable growth patterns to the WT in both media (Figure 1B), suggesting the FRE-143
TM gene pair is not essential for iron acquisition or could be functionally redundant with other 144
ferric reductase genes, including homologs of FRE1-FRE8 (Yun et al. 2001). Although it grew 145
normally with regular iron concentrations, the enb1Δ strain showed minimal growth in low-iron 146
conditions (Figure 1B). This result implies that the Enb1 homolog plays a crucial role in iron 147
uptake in iron-limited conditions. 148
To delve into the function of the Enb1 homolog, we measured the levels of extracellular 149
enterobactin. We found that all three strains produced substantially higher amounts of 150
enterobactin in low-iron medium compared to regular medium (6.5-9.6-fold increase) (Figure 151
1D). This result indicates that enterobactin biosynthesis in St. bombicola increases under 152
conditions of iron scarcity and might have evolved as an integral part of the regulatory network 153
governing iron hemostasis. Moreover, the fre-tm
D
strain amassed similar levels of enterobactin 154
as the WT in both culturing conditions, whereas cells of the enb1
D
strain accumulated 1.7-fold 155
and 2.5-fold more enterobactin than the WT in regular medium and low-iron medium, 156
respectively. An orthogonal colorimetric assay also showed a notably stronger enterobactin 157
signal from the enb1
D
strain as compared to the WT strain (Figure 1C), which suggests that the 158
mutant experiences severe iron starvation in low-iron conditions because it cannot effectively 159
import enterobactin. Thus, we conclude that the ENB1 homolog encodes a functional Enb1 160
transporter that is required for normal uptake of enterobactin-bound iron in St. bombicola. 161
.CC-BY-NC 4.0 International licenseavailable under a
(which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made
The copyright holder for this preprintthis version posted November 21, 2023. ; https://doi.org/10.1101/2023.11.21.568075doi: bioRxiv preprint
8
Interestingly, disruption of neither ENB1 nor FRE-TM noticeably affected the export of 162
enterobactin. In E. coli, entS, as part of the ent operon, encodes the primary enterobactin 163
exporter, which, together with the outer membrane channel TolC, mediates the secretion of 164
enterobactin (Furrer et al. 2002). Since no homolog of entS could be identified in St. bombicola 165
and other W/S-clade yeasts (Kominek et al. 2019), how the enterobactin synthesized in the cells 166
is secreted to the medium remains an unanswered question. 167
168
The origin of ENB1 169
To determine whether ENB1 was horizontally acquired by the W/S-clade yeasts from 170
bacteria like the ent genes, we performed BLASTp searches against the NCBI non-redundant 171
sequences (nr) database using the amino acid sequence of St. bombicola Enb1 (StbEnb1) as the 172
query. Proteins from the Saccharomycetales, including ScEnb1, were retrieved with E values 173
equal to or lower than 5e-132, above which point proteins from filamentous fungi started to be 174
recovered (Dataset S1-1). The E values obtained from some filamentous fungi hits were as low 175
as 2e-117, suggesting strong sequence similarity with the StbEnb1. Among the top 2,626 176
BLASTp hits recovered using an E value cutoff of e-60, only a few were from archaea or 177
bacteria, while all the others were fungal proteins (Dataset S1-2). According to the maximum 178
likelihood (ML) phylogeny constructed with these top hits (Figure S1), StbEnb1 and its close 179
homologs from the Saccharomycetales clustered monophyletically on a long branch with 100% 180
bootstrap support, whereas the proteins from archaea and bacteria scattered across the larger 181
gene tree on distant branches. Hence, we conclude that closely related Enb1 homologs are only 182
.CC-BY-NC 4.0 International licenseavailable under a
(which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made
The copyright holder for this preprintthis version posted November 21, 2023. ; https://doi.org/10.1101/2023.11.21.568075doi: bioRxiv preprint
9
found in fungi, and the StbENB1 gene is unlikely to have been acquired horizontally from 183
bacteria. 184
The high sequence similarity of StbEnb1 to proteins from filamentous fungi inspired us 185
to examine the evolutionary history of Enb1 in the kingdom Fungi. For this purpose, we 186
conducted a HMMER scan for Enb1 homologs against a dataset containing 1,644 published 187
fungal genomes (Li et al. 2021) using the aligned sequences of ScEnb1 and StbEnb1 as queries; 188
we obtained 1,061 protein sequences by employing a bit-score cutoff of 300 and a E-value cutoff 189
of 0.05 (Dataset S2). The resulting proteins were exclusively from the Dikarya; the vast majority 190
of them were recovered from the subphyla Saccharomycotina (491) and Pezizomycotina (553), 191
while only 17 of them were from the subphyla Agaricomycotina, Wallemiomycotina, and 192
Taphrinomycotina. After accounting for putative gene duplication events that gave rise to 193
distantly related paralogs (e.g. genes encoding Arn1, Arn2, and Sit1), the ML phylogenetic tree 194
constructed using the 1,061 protein sequences (Figure S2) was generally consistent with the 195
reported genome-scale species phylogeny (Li et al. 2021). Notably, ScEnb1 and StbEnb1, as well 196
as Enb1 homologs from many yeasts belonging to the W/S clade and the order 197
Saccharomycetales, formed a monophyletic group with 12 proteins from diverse fungi at 98% 198
bootstrap confidence. The 12 closest relatives of yeast Enb1 were recovered from species that 199
were scattered sparsely, but widely, in the Dikarya across the subphyla Pezizomycotina, 200
Agaricomycotina, Wallemiomycotina, and Taphrinomycotina. Since this sparse clade roughly 201
matched the species phylogeny, we propose that it comprises the extant Enb1 orthologs. Under 202
this model, ENB1 was likely present in the MRCA of the Dikarya, while its sparse and dispersed 203
distribution is likely due to extensive losses in the various lineages. 204
.CC-BY-NC 4.0 International licenseavailable under a
(which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made
The copyright holder for this preprintthis version posted November 21, 2023. ; https://doi.org/10.1101/2023.11.21.568075doi: bioRxiv preprint
10
Interestingly, the non-Saccharomycotina fungal species are either known plant pathogens, 205
including Eutypa lata, Rhizoctonia solani, Passalora fulva, and Ceratobasidium theobromae 206
(Thomma et al. 2005; Ajayi-Oyetunde and Bradley 2018; Ali et al. 2019; Lolas et al. 2020); or 207
they can exhibit yeast-like growth, including three fission yeasts (Schizosaccharomyces 208
japonicus, Schizosaccharomyces octosporus, and Schizosaccharomyces cryophilus), three 209
dimorphic fungi (Zymoseptoria brevis, Zymoseptoria ardabiliae, and Prillingera fragicola) 210
(Quaedvlieg et al. 2011; Takashima et al. 2019), and a basidiomycetous yeast (Apiotrichum 211
gamsii) (James et al. 2016). This unusual phylogenetic distribution suggests a potential role of 212
Enb1 in the evolution of fungal virulence and unicellular yeast growth, which is consistent with 213
the previous finding that siderophore-mediated iron uptake is crucial for fungal iron homeostasis 214
and virulence (Haas et al. 2008). 215
216
Horizontal gene transfer of ENB1 within yeasts 217
The presence of ENB1 in St. bombicola and S. cerevisiae, which belong to two distantly 218
related orders that diverged over 300 million years ago (Shen et al. 2018; Groenewald et al. 219
2023), prompted us to more thoroughly investigate its species distribution within the subphylum 220
Saccharomycotina. We performed a HMMER search against 345 publicly available yeast 221
genomes (Shen et al. 2018; Kominek et al. 2019) for Enb1 homologs using aligned sequences of 222
Enb1 from S. cerevisiae (ScEnb1) and St. bombicola (StbEnb1) as the query. In addition to 223
ScEnb1 and StbEnb1, 32 proteins, exclusively from the order Saccharomycetales and W/S clade 224
of yeasts, were retrieved up to E value of 3.9e-170 and bit score of 565.5, after which point the E 225
value ramped up and hit score dropped drastically (Dataset S3). All 34 closely related Enb1 226
homologs clustered monophyletically on a very long branch with 100% bootstrap support in the 227
.CC-BY-NC 4.0 International licenseavailable under a
(which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made
The copyright holder for this preprintthis version posted November 21, 2023. ; https://doi.org/10.1101/2023.11.21.568075doi: bioRxiv preprint
11
maximum likelihood (ML) phylogeny (Figure 2 & S3). Besides Enb1, we recovered the other 228
three siderophore transporters, Arn1, Arn2, and Sit1, as well as the glutathione exchanger Gex1 229
and its paralog Gex2 (Dhaoui et al. 2011) from S. cerevisiae. Consistent with previous 230
observations (Diffels et al. 2006; Dias and Sá-Correia 2013), S. cerevisiae Arn1, Arn2, Gex1, 231
and Gex2, as well as all their homologs from other Saccharomycetales species, formed a single 232
clade in the tree with 100% bootstrap support. 233
From these searches, we discovered clear ENB1 orthologs in species from the genera 234
Saccharomyces, Torulaspora, and Lachancea. Surprisingly, these ENB1 orthologs from the order 235
Saccharomycetales were nested within those of the W/S clade (Figure 2), which suggests the 236
possibility of a horizontal gene transfer (HGT) event from the W/S clade in the order 237
Dipodascales to the order Saccharomycetales. This model is consistent with the fact that 238
previously known ENB1 homologs were identified exclusively in the sub-telomeric regions of 239
yeasts belonging to the genus Saccharomyces (Dias and Sá-Correia 2013), which constitute a 240
favorable chromosomal location for the acquisition of foreign genes (Kellis et al. 2003; Novo et 241
al. 2009). To better illuminate the evolution of Enb1 within yeasts, we identified three more 242
Enb1 homologs from Wickerhamiella shivajii and Starmerella stellata, whose genomes were 243
recently published (Opulente et al. 2023). Using maximum likelihood, we carefully examined the 244
phylogenetic relationships between the 37 yeast Enb1-like protein sequences, as well as the 7 245
closest relatives of Enb1 from fungal outgroups. As expected, the yeast Enb1-like transporters 246
formed a monophyletic clade with 100% bootstrap support (Figure 3A, Figure S2). However, 247
there were at least 3 instances where the Enb1 tree did not recover the expected phylogenetic 248
relationships between yeast taxa (Figure 3A &4). Most notably, Enb1 homologs from the 249
Saccharomycetales were nested within the W/S clade and indeed were sister to the distantly 250
.CC-BY-NC 4.0 International licenseavailable under a
(which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made
The copyright holder for this preprintthis version posted November 21, 2023. ; https://doi.org/10.1101/2023.11.21.568075doi: bioRxiv preprint
12
related species St. stellata with 96% bootstrap support. This observation suggests the acquisition 251
via HGT of an ENB1 gene by an ancient Saccharomycetales from an ancestor of W/S clade 252
yeasts. To test this hypothesis, we performed topology tests where we compared the likelihood of 253
the ML-supported (unconstrained) topology against the likelihood of an alternative constrained 254
topology that enforced the reciprocal monophyly of the Saccharomycetales- and W/S-clade 255
homologs (Fig. 3B). Both weighted Shimodaira–Hasegawa (wSH) and approximately unbiased 256
(AU) tests significantly rejected the null model of vertical inheritance and instead supported 257
HGT (Fig. 3C). 258
Second, the St. stellata Enb1 homolog did not associate as expected with other 259
Starmerella branches of the tree in the Enb1 phylogeny (Figure S4A). Instead, St. stellata Enb1 260
formed a monophyletic clade with the Saccharomycetales, Wickerhamiella pararugosa, and W. 261
shivajii homologs. This topology suggests another possible HGT event from this Wickerhamiella 262
lineage into an ancestor of St. stellata. To test for additional HGT events between 263
Wickerhamiella and Starmerella, we performed a series of topology tests. First, we enforced the 264
placement of St. stellata Enb1 either as an outgroup of the Starmerella Enb1 clade (Constrained 265
Topology 2, Figure S4B) or to its exact location in the species phylogeny (Constrained Topology 266
3, Figure S4B). The wSH and AU tests of Constrained Topology 3 robustly rejected the null 267
model of vertical inheritance, while the same tests of Constrained Topology 2 did not reject the 268
null model (Figure S4C). Since Constrained Topology 3 incorporates additional incongruences 269
between the gene and species phylogenies, we conclude that HGT from an ancient 270
Wickerhamiella into an ancestor of St. stellata is one of many evolutionary models that are 271
consistent with the data, including more complex scenarios involving multiple HGTs, cryptic 272
gene duplication, or incomplete lineage sorting within the W/S clade. Additionally, the topology 273
.CC-BY-NC 4.0 International licenseavailable under a
(which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made
The copyright holder for this preprintthis version posted November 21, 2023. ; https://doi.org/10.1101/2023.11.21.568075doi: bioRxiv preprint
13
tests did not significantly support a possible HGT associated with the Wickerhamiella versatillis 274
Enb1 (Constrained Topology 1, Figure S4B, Figure S4C). Given the evolutionary distances 275
involved, we conclude that ENB1 has experienced a complex history in some W/S-clade 276
lineages, including the lineage leading to St. stellata. We further conclude that one of these 277
ancient W/S lineages likely acted as the donor of the Saccharomycetales ENB1. 278
279
Duplications and losses of ENB1 and its functional integration with a bacterial operon 280
To get more insight into the evolutionary mechanisms leading to the distribution of ENB1 281
orthologs in yeasts, we mapped its presence, absence, and copy numbers onto a genome-scale 282
species phylogeny inferred in a recent publication (Opulente et al. 2023) (Figure 4). Given its 283
broad distribution in the clade, ENB1 was almost certainly present in the MRCA of the W/S 284
clade. It could have been acquired from other fungi by HGT, or ENB1 could have been present in 285
the MRCA of Saccharomycotina and then lost in all other known yeast lineages (at least 9 286
independent losses by maximum parsimony). The St. bombicola enb1
D
mutant phenotype 287
(Figure 1) indicates that expression of an Enb1 transporter is essential to for normal uptake of 288
iron bound to enterobactin. Thus, ENB1 would convey a selective advantage in iron-limited 289
environments where enterobactin is present, including enterobactin made by other microbes, but 290
especially for organisms making their own enterobactin. Therefore, we hypothesize that the 291
ENB1 gene was actively expressed in the MRCA of the W/S-clade yeasts prior to the HOT event 292
that introduced the ent operon into yeasts. Under this scenario, the fepC/D/G genes from the 293
operon, which would have encoded the bacterial ABC-type enterobactin transporters, could have 294
been lost easily due to functional redundancy with fungal ENB1. 295
.CC-BY-NC 4.0 International licenseavailable under a
(which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made
The copyright holder for this preprintthis version posted November 21, 2023. ; https://doi.org/10.1101/2023.11.21.568075doi: bioRxiv preprint
14
The phylogenetic distribution of ENB1 copy numbers and inference by maximum 296
parsimony suggest at least two duplication events in the W/S clade: one that occurred before the 297
divergence of Wickerhamiella versatilis, Wickerhamiella domercqiae, and the genus 298
Starmerella; and a second that occurred in the last common ancestor of Starmerella geochares, 299
Starmerella vaccinii, and Starmerella sorbosivorans. This model is generally consistent with the 300
Enb1 protein tree (Figure 3A). Additionally, maximum parsimony suggests multiple loss events 301
occurred both prior to and following the duplications, leading to reduced copy numbers or ENB1 302
absence in some W/S-clade yeasts. Disproportionate secretion and uptake of enterobactin could 303
also result in its extracellular accumulation, potentially reaching levels detrimental to the cells 304
that produce it. We hypothesize that altering the copy number of the ENB1 gene through 305
duplications and losses might be an evolutionary strategy taken by the W/S-clade yeasts to 306
balance the expression levels of the biosynthetic enzymes and the transporter, while adapting to 307
new environments with different iron availability. 308
Copy number variation aside, the presence and absence of ENB1 almost perfectly mirrors 309
that of the ent genes encoding enterobactin biosynthesis pathway. Specifically, ENB1 is present 310
in all the W/S-clade yeasts that retained the ent genes, while it is absent in nearly all species that 311
lost the ent genes with St. stellata as the only exception. The correlation between the presence of 312
the ENB1 and ent genes is likely driven by the fact that the loss of ENB1 would be deleterious to 313
yeasts actively producing enterobactin because they would sequester iron that they could not 314
import, a phenotype we observed in the St. bombicola enb1
D
mutant (Figure 1B). The sole 315
exception is St. stellata, which is the only W/S yeast that harbors the ENB1 gene in the absence 316
of ent genes, a genetic configuration known to create public goods cheaters in other siderophore 317
systems (Wang et al. 2015; Krause et al. 2018). Regardless of whether St. stellata reacquired 318
.CC-BY-NC 4.0 International licenseavailable under a
(which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made
The copyright holder for this preprintthis version posted November 21, 2023. ; https://doi.org/10.1101/2023.11.21.568075doi: bioRxiv preprint
15
ENB1 from an ancient Wickerhamiella species via HGT or not (Figure S4), it likely represents a 319
reversion to the ancestral cheater state prior to the HOT event, either through simple loss of the 320
ent operon or also through loss and subsequent regain of ENB1. 321
322
Conclusions 323
Siderophores can readily bind iron, even in an iron-limited environment, thereby both 324
reserving iron for producers and depriving iron from species that lack the matching transporter 325
for uptake (Niehus et al. 2017; Schiessl et al. 2017). However, it has been repeatedly shown that 326
non-producers possess conditional competitive advantages over the producers because they act 327
as cheaters and exploit foreign siderophores without the cost of synthesis (Griffin et al. 2004; 328
Jiricny et al. 2010; Dumas and Kümmerli 2012). The fact that the MRCA of the W/S clade was 329
likely converted from a cheater for enterobactin into an enterobactin producer through HOT 330
suggests that there were ecological and evolutionary circumstances where being a producer had 331
an advantage over being a cheater. The HOT was proposed to have occurred in insect guts, 332
which are believed to foster intense competition for iron among bacteria, yeasts, and the host 333
itself (Barber and Elde 2015; Kominek et al. 2019). Yeasts with the capacity to produce their 334
own enterobactin, which has exceptionally high affinity for Fe3+ (Kf = 1051) (Carrano and 335
Raymond 1979), may have been better positioned to outcompete nonproducers and species that 336
produce different siderophores, especially in communities lacking enterobactin producers. Thus, 337
the cooperative integration of the fungal ENB1 gene and the bacterial ent genes as a functional 338
unit for sequestering iron would have substantially contributed to the fitness of yeasts in highly 339
competitive, iron-limited environments. After enterobactin-producing W/S-clade yeasts had 340
begun to radiate, ENB1 was horizontally transferred into an ancient lineage of 341
.CC-BY-NC 4.0 International licenseavailable under a
(which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made
The copyright holder for this preprintthis version posted November 21, 2023. ; https://doi.org/10.1101/2023.11.21.568075doi: bioRxiv preprint
16
Saccharomycetales (Figure 3), which allowed many additional yeasts, including S. cerevisiae 342
(Petra Heymann et al. 2000), to exploit the enterobactin produced by others. Several secondary 343
losses of ENB1 then led to its patchy distribution in the order Saccharomycetales. 344
Siderophore uptake has rarely been studied in nonconventional yeasts, but this study 345
suggests it is subject to surprisingly complex eco-evolutionary dynamics. Here, we identified 346
Enb1 as the key transporter mediating the uptake of enterobactin-bound iron in St. bombicola, 347
one of the extant W/S-clade yeasts that acquired a complete bacterial enterobactin biosynthesis 348
pathway via an ancient HOT. Through phylogenomic analyses, we showed that the ENB1 gene is 349
of fungal origin and has a broad, but patchy, distribution in the Dikarya. Along with the 350
horizontally acquired bacterial operon, this fungal gene formed a mosaic functional unit that 351
operates coordinately in many W/S-clade yeasts to scavenge iron from the environment. We also 352
propose how cheaters arose through the secondary loss of the ent operon and/or HGT of ENB1. 353
Multiple gene duplication and loss events punctuate the ENB1 phylogeny and further suggest a 354
dynamic eco-evolutionary history of enterobactin cheaters in yeasts. This story shows how 355
fungal and bacterial genes can be functionally integrated in the same species for the production 356
of a resource, as well as how eco-evolutionary forces can then separate them as cheaters arise to 357
exploit that resource. 358
359
Materials and Methods 360
Construction of St. bombicola deletion mutants 361
Genetic manipulations were performed in the St. bombicola PYCC 5882 strain obtained 362
from the Portuguese Yeast Culture Collection. Strains used in this study are listed in Dataset S5. 363
.CC-BY-NC 4.0 International licenseavailable under a
(which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made
The copyright holder for this preprintthis version posted November 21, 2023. ; https://doi.org/10.1101/2023.11.21.568075doi: bioRxiv preprint
17
All the oligonucleotides used for construction of St. bombicola mutants are listed in Dataset S6. 364
The hygromycin-resistance cassette with a St. bombicola GPD promoter and S. cerevisiae CYC1 365
terminator was amplified from the previously constructed plasmid PJET-ADH1-HYG 366
(Gonçalves et al. 2018). Genomic DNA of St. bombicola PYCC 5882 was isolated and purified 367
using a modified phenol:chloroform extraction method as previously described (Hittinger et al. 368
2010). For disruption of the ENB1 and FRE-TM genes, two sets of primers were used to amplify 369
~1 kb upstream and ~1 kb downstream of their coding sequences with 30 bp overlaps to the 370
hygromycin-resistance cassette. The ENB1 and FRE-TM deletion cassettes were obtained by 371
assembling the upstream and downstream fragments of each gene with the hygromycin-372
resistance cassette using the NEBuilder® HiFi DNA Assembly Master Mix (Catalog# E2621), 373
followed by PCR amplifications and cleanup using the QIAquick® PCR Purification Kit. 374
Phusion® High-Fidelity DNA Polymerase (Catalog# M0530) was used for all the PCR 375
amplifications. 376
St. bombicola was transformed with each deletion cassette using a previously described 377
electroporation protocol (Saerens et al. 2011). The transformants were screened on YPD (10 g/L 378
yeast extract, 20 g/L peptone, 20 g/L dextrose, pH 7.2) plates supplemented with 650 µg/mL 379
hygromycin B (US Biological) and verified using colony PCR and Sanger sequencing. 380
381
Growth and iron-binding assays 382
St. bombicola strains were inoculated from glycerol stocks, which were stored at -80 °C, 383
to YPD medium and precultured for 3 days at 30 °C in culture tubes. 2% dextrose was used as 384
the carbon source for all cell cultivations. For low-iron cultivations, 1.7g/L Yeast Nitrogen Base 385
without Amino Acids, Carbohydrates, Ammonium Sulfate, Ferric Chloride, or Cupric Sulfate 386
.CC-BY-NC 4.0 International licenseavailable under a
(which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made
The copyright holder for this preprintthis version posted November 21, 2023. ; https://doi.org/10.1101/2023.11.21.568075doi: bioRxiv preprint
18
(US Biological) was used to make the low-iron synthetic complete (SC) medium (“low-iron 387
medium” in brief) without direct addition of any iron source, although it may inadvertently 388
acquire ambient iron during the medium preparation and cultivation processes. The low-iron 389
medium also consisted of 5 g/L ammonium sulfate, 2 g/L Complete Dropout Mix (US 390
Biological), and 200 nM cupric sulfate. The regular SC medium (“regular medium” in brief) 391
consisted of 200 µg/L (1.2 µM) ferric chloride with all other components kept identical. YP 392
medium contains about 25 µM of iron (Du et al. 2012), which is approximately 20 times of the 393
iron concentration in the regular-iron SC medium. To avoid substantial iron carryover from the 394
YPD precultures, cells were washed twice in deionized water and reinoculated at an initial 395
optical density (OD600) of 0.2 in culture tubes with regular- and low-iron media for a second 396
preculture. After 4 days, the cells were again washed and reinoculated at an OD600 of 0.2 into the 397
corresponding media for the growth experiments, which were conducted in 250 mL shake flasks 398
with 50 mL of regular- or low-iron medium at 30 °C. 399
The chromeazurol S overlay (O-CAS) solution was made with chromeazurol S, ferric 400
chloride hexahydrate, PIPES (free acid), and agarose as described previously (Kominek et al. 401
2019). Cells of the St. bombicola wildtype and enb1
D
strains were precultured in 5 mL YPD 402
medium for 3 days, harvested, and resuspended in 5 mL deionized water. 5 µL of the resulting 403
cell suspension was spotted on the center of 60-mm diameter petri dishes consisting of low-iron 404
SC agarose (1% w/v) and incubated at 30 °C. After 7 days, 6 mL of the O-CAS solution were 405
overlaid onto low-iron SC plates. The plates were sealed with Parafilm and put in the dark at 406
room temperature for another 7 days before pictures were taken on an LED light box. 407
408
Enterobactin quantification 409
.CC-BY-NC 4.0 International licenseavailable under a
(which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made
The copyright holder for this preprintthis version posted November 21, 2023. ; https://doi.org/10.1101/2023.11.21.568075doi: bioRxiv preprint
19
Supernatants of yeast cultures were harvested by centrifugation at the end of the growth 410
experiments. The amount of enterobactin was quantified using a Shimadzu LCMS8040 by 411
injecting 10 µL of the filtered supernatants onto a Phemonenex Kinetex XB-C18 column (P/N: 412
00G-4605-E0, 100Å, 250 ´ 4.6 mm, 5 µ particle size) held at 50 °C. The mobile phase was a 413
binary gradient of A) water and B) acetonitrile, pumped at 0.7 mL/min. The gradient program 414
was set according to %B using linear gradients between set time points, starting by holding at 415
5%B for 1 min; then ramping up to be 10%B at 5 min and 95%B at 6.5 min; then held at 95%B 416
until 8 min, at which point it was returned to 5%B at 11 min and held at 5%B until the end of the 417
program at 15 min. The column eluent flowed through a photo diode array detector scanning 418
from 250–600 nm and into a 2-way switch valve to send it to waste or the mass spectrometer. 419
The mobile phase was diverted to waste from time 0.1 – 3.9 min to aid in keeping the 420
electrospray ionization (ESI) source clean by preventing the buffer salts present in the media 421
from entering the mass spectrometer. The column eluent was sent to the ESI source from time 422
3.9 – 15 min. 423
The dual ion source was set to operate in ESI negative mode, with 2.5 L/min nitrogen 424
nebulizing gas, 15 L/min nitrogen drying gas, desolvation line temperature set to 250 °C, and the 425
heating block temperature set to 400 °C. The mass spectrometer was programed to collect Q3-426
scans negative ions from 200 – 1000 m/z and, for the quantification of enterobactin, three 427
multiple-reaction monitoring (MRM) transitions were acquired: 668®178, CE:53, relative 428
intensity: 100; 668®222, CE:33, relative intensity: 53; and 668®445, CE:21, relative intensity: 429
22. Argon gas at 230 kPa was used for the MRM fragmentations. The retention time (10.2 min), 430
relative intensities of the three MRM transitions, and a 7-point calibration curve from 125–10 431
µg/mL were determined using an enterobactin standard (Sigma Aldrich #E3910). To minimize 432
.CC-BY-NC 4.0 International licenseavailable under a
(which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made
The copyright holder for this preprintthis version posted November 21, 2023. ; https://doi.org/10.1101/2023.11.21.568075doi: bioRxiv preprint
20
carryover the needle was washed for 10 sec before and after each injection. Two blank methanol 433
injections were made between each sample to ensure any carryover enterobactin was below the 434
threshold of detection (~7 µg/mL). As the cell growth varied substantially for the 3 strains grown 435
in different conditions, the concentrations of enterobactin were normalized by the final cell 436
optical density of yeast cultures for fair comparisons. 437
438
Screening for the presence of ENB1 homologs 439
Basic Local Alignment Search Tool protein (BLASTp) analyses in a publicly available 440
genomic assembly of the St. bombicola PYCC 5882 strain (GeneBank ID: GCA_003033785.1) 441
(Gonçalves et al. 2018) were performed using the Enb1 protein sequences from Saccharomyces 442
cerevisiae (ScEnb1) as the query. The identified ENB1 ortholog in St. bombicola was 443
functionally validated using targeted gene replacement as described above and designated as 444
StbENB1. To ascertain whether Enb1-like transporters could be found outside the fungal 445
kingdom, the ScEnb1 amino acid sequence was used as BLASTp query against the NCBI non-446
redundant protein sequence (nr) database (Datasets S1-1 & S1-2). The hits with E-values < 2e-60 447
were filtered by removing redundant sequences with > 90% similarity using the clustering 448
function of the MMseq2 (Steinegger and Söding 2017). The resulting 1165 sequences were used 449
to construct a maximum likelihood (ML) phylogeny. We visualized and plotted phylogenetic 450
trees using iTOL v5 (Letunic and Bork 2021). To evaluate the prevalence of ENB1 homologs 451
among fungal genomes, a hidden Markov model (HMMER) sequence similarity search (v3.3, 452
http://hmmer.org) was conducted against all the protein-coding sequences pulled out from 1644 453
published fungal genome assemblies (Li et al. 2021) using the aligned sequences of ScEnb1 and 454
.CC-BY-NC 4.0 International licenseavailable under a
(which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made
The copyright holder for this preprintthis version posted November 21, 2023. ; https://doi.org/10.1101/2023.11.21.568075doi: bioRxiv preprint
21
StbEnb1 as the query. HMMER hits with E-values < 0.05 are listed in Dataset S3, and protein 455
hits with bit-scores > 300 were retrieved for phylogenetic analyses. 456
We also identified homologs of Enb1 across the 345 published yeast genome assemblies 457
(Shen et al. 2018)(Kominek et al. 2019) using a separate HMMER search. Protein annotations 458
used for the HMMER search were generated previously by the MAKER genome annotation 459
pipeline v2.31.8 (Holt and Yandell 2011; Shen et al. 2018; Kominek et al. 2019), again using the 460
alignment of ScEnb1 and StbEnb1 sequences as the query. HMMER hits with E-values < 0.05 461
were listed in Dataset S3, and the 435 hits with bit scores > 300 were retrieved for constructing a 462
maximum likelihood (ML) phylogeny (Figure 2). We similarly retrieved Enb1 homologs from 463
the recently published genomes of W. shivajii and St. stellata (Opulente et al. 2023). 464
465
Phylogenetic analyses and topology tests 466
Protein sequences of hits from the aforementioned HMMER and BLASTp analyses were 467
retrieved to construct maximum likelihood trees to infer the phylogenetic relationships of ENB1 468
homologs. MAFFT (Katoh and Standley 2013) v7.508 was used to align the sequence with the 469
strategy L-INS-i. Initial inferences of the phylogenies were performed with IQ-TREE (Nguyen et 470
al. 2015) v2.2.0.3 using ModelFinder (Kalyaanamoorthy et al. 2017) with automated model 471
selection based on 1000 bootstrap pseudoreplicates. The LG+R10, LG+G4, LG+R7, and 472
LG+R10 amino-acid substitution models were selected because they had the lowest Bayesian 473
Information Criterion scores to construct the trees in Figure 2, Figure 3A, Figure S1, and Figure 474
S2, respectively. The seven closest relatives of the Enb1 orthologs in filamentous fungi were 475
chosen as outgroups for the Enb1 phylogeny (Figure 3). To investigate the likelihood of HGT 476
events between the W/S clade and order Saccharomycetales, phylogenetic analyses of Enb1 477
.CC-BY-NC 4.0 International licenseavailable under a
(which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made
The copyright holder for this preprintthis version posted November 21, 2023. ; https://doi.org/10.1101/2023.11.21.568075doi: bioRxiv preprint
22
proteins were repeated with specific constraints, and significance was assessed with AU and SH 478
topology tests. The constrained tree in Figure 3B enforced the reciprocal monophyly of the 479
Saccharomycetales and the W/S clade, while the constrained trees in Figure S4 enforced the 480
repositioning of St. stellata and W. versatilis. The substitution model selected for all the 481
constrained trees was LG+G4. 482
483
Supplementary Materials 484
Supplementary data are available at Molecular Biology and Evolution online. 485
486
Acknowledgments 487
We thank David J. Krause, Trey K. Sato, and members of the Hittinger and Rokas groups for 488
helpful discussions. This material is based upon work supported in part by the Great Lakes 489
Bioenergy Research Center, U.S. Department of Energy, Office of Science, Office of Biological 490
and Environmental Research under Award Number DE-SC0018409; the National Science 491
Foundation (under grant Nos. DBI-1906759 to K.T.D., DEB-2110403 to C.T.H., and DEB-492
2110404 to A.R.); and the National Institute of Food and Agriculture, United States Department 493
of Agriculture, Hatch project 7005101. C.T.H. is an H. I. Romnes Faculty Fellow, supported by 494
the Vice Chancellor for Research and Graduate Education with funding from the Wisconsin 495
Alumni Research Foundation. Research in the Rokas lab is also supported by the National 496
Institutes of Health/National Institute of Allergy and Infectious Diseases (R01 AI153356), and 497
the Burroughs Welcome Fund. 498
499
Data Availability 500
.CC-BY-NC 4.0 International licenseavailable under a
(which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made
The copyright holder for this preprintthis version posted November 21, 2023. ; https://doi.org/10.1101/2023.11.21.568075doi: bioRxiv preprint
23
The data underlying this article are available in the article and in its online supplementary 501
material. 502
503
Competing Interest Statement 504
AR is a scientific consultant for LifeMine Therapeutics, Inc. 505
506
Author Contributions 507
LS performed all experiments and analyses, except where otherwise noted, and drafted and 508
edited the manuscript. KTD assisted with topology tests. JFW set up the HMMER pipeline and 509
assisted with interpretation. SDK performed enterobactin detection assays. CG and MG provided 510
reagents and advised on their use. DAO, ALL, XZ, and XXS provided genome sequences and 511
annotations. LS, AR, and CTH designed and conceived the study. AR and CTH secured funding 512
and edited the manuscript. All authors approved the manuscript. 513
514
References 515
Ajayi-Oyetunde OO, Bradley CA. 2018. Rhizoctonia solani: taxonomy, population biology and 516
management of rhizoctonia seedling disease of soybean. Plant Pathology 67:3–17. 517
Ali SS, Asman A, Shao J, Firmansyah AP, Susilo AW, Rosmana A, McMahon P, Junaid M, 518
Guest D, Kheng TY, et al. 2019. Draft genome sequence of fastidious pathogen 519
Ceratobasidium theobromae, which causes vascular-streak dieback in Theobroma cacao. 520
Fungal Biology and Biotechnology 6:14. 521
Carrano C, Raymond K. Ferric ion sequestering agents. 2. Kinetics and mechanism of iron 522
removal from transferrin by enterobactin and synthetic tricatechols. Journal of the 523
American Chemical Society 101:18. 524
.CC-BY-NC 4.0 International licenseavailable under a
(which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made
The copyright holder for this preprintthis version posted November 21, 2023. ; https://doi.org/10.1101/2023.11.21.568075doi: bioRxiv preprint
24
Ardon O, Bussey H, Philpott C, Ward DM, Davis-Kaplan S, Verroneau S, Jiang B, Kaplan J. 525
2001. Identification of a Candida albicans Ferrichrome Transporter and Its 526
Characterization by Expression in Saccharomyces cerevisiae. Journal of Biological 527
Chemistry 276:43049–43055. 528
Barber MF, Elde NC. 2015. Buried Treasure: Evolutionary Perspectives on Microbial Iron 529
Piracy. Trends in Genetics 31:627–636. 530
Butler G, Kenny C, Fagan A, Kurischko C, Gaillardin C, Wolfe KH. 2004. Evolution of the 531
MAT locus and its Ho endonuclease in yeast species. Proceedings of the National 532
Academy of Sciences 101:1632–1637. 533
Dhaoui M, Auchère F, Blaiseau P-L, Lesuisse E, Landoulsi A, Camadro J-M, Haguenauer-534
Tsapis R, Belgareh-Touzé N. 2011. Gex1 is a yeast glutathione exchanger that interferes 535
with pH and redox homeostasis. MBoC 22:2054–2067. 536
Dias PJ, Sá-Correia I. 2013. The drug:H+ antiporters of family 2 (DHA2), siderophore 537
transporters (ARN) and glutathione:H+antiporters (GEX) have a common evolutionary 538
origin in hemiascomycete yeasts. BMC Genomics 14:901. 539
Diffels JF, Seret M-L, Goffeau A, Baret PV. 2006. Heavy metal transporters in Hemiascomycete 540
yeasts. Biochimie 88:1639–1649. 541
Du Y, Cheng W, Li W-F. 2012. Expression Profiling Reveals an Unexpected Growth-542
Stimulating Effect of Surplus Iron on the Yeast Saccharomyces cerevisiae. Molecular 543
Cells 34:127–132. 544
Dumas Z, Kümmerli R. 2012. Cost of cooperation rules selection for cheats in bacterial 545
metapopulations. Journal of Evolutionary Biology 25:473–484. 546
Furrer JL, Sanders DN, Hook-Barnard IG, McIntosh MA. 2002. Export of the siderophore 547
enterobactin in Escherichia coli: involvement of a 43 kDa membrane exporter. Molecular 548
Microbiology 44:1225–1234. 549
Gonçalves C, Gonçalves P. 2019. Multilayered horizontal operon transfers from bacteria 550
reconstruct a thiamine salvage pathway in yeasts. Proceedings of the National Academy 551
of Sciences 116:22219–22228. 552
.CC-BY-NC 4.0 International licenseavailable under a
(which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made
The copyright holder for this preprintthis version posted November 21, 2023. ; https://doi.org/10.1101/2023.11.21.568075doi: bioRxiv preprint
25
Gonçalves C, Wisecaver JH, Kominek J, Oom MS, Leandro MJ, Shen X-X, Opulente DA, Zhou 553
X, Peris D, Kurtzman CP, et al. Evidence for loss and reacquisition of alcoholic 554
fermentation in a fructophilic yeast lineage. eLife 7:e33034. 555
Griffin AS, West SA, Buckling A. 2004. Cooperation and competition in pathogenic bacteria. 556
Nature 430:1024–1027. 557
Groenewald M, Hittinger CT, Bensch K, Opulente DA, Shen X-X, Li Y, Liu C, LaBella AL, 558
Zhou X, Limtong S, et al. 2023. A genome-informed higher rank classification of the 559
biotechnologically important fungal subphylum Saccharomycotina. Studies in Mycology 560
105:1–22. 561
Haas H, Eisendle M, Turgeon BG. 2008. Siderophores in Fungal Physiology and Virulence. 562
Annual Review of Phytopathology 46:149–187. 563
Heymann P, Ernst JF, Winkelmann G. 1999. Identification of a fungal triacetylfusarinine C 564
siderophore transport gene (TAF1) in Saccharomyces cerevisiae as a member of the 565
major facilitator superfamily. Biometals 12:301–306. 566
Heymann P., Ernst JF, Winkelmann G. 2000. Identification and substrate specificity of a 567
ferrichrome-type siderophore transporter (Arn1p) in Saccharomyces cerevisiae. FEMS 568
Microbiology Letters 186:221–227. 569
Heymann Petra, Ernst JF, Winkelmann G. 2000. A gene of the major facilitator superfamily 570
encodes a transporter for enterobactin (Enb1p) in Saccharomyces cerevisiae. Biometals 571
13:65–72. 572
Heymann P, Gerads M, Schaller M, Dromer F, Winkelmann G, Ernst JF. 2002. The Siderophore 573
Iron Transporter of Candida albicans (Sit1p/Arn1p) Mediates Uptake of Ferrichrome-574
Type Siderophores and Is Required for Epithelial Invasion. Infection and Immunity 575
70:5246–5255. 576
Hittinger CT, Carroll SB. 2007. Gene duplication and the adaptive evolution of a classic genetic 577
switch. Nature 449:677–681. 578
Hittinger CT, Gonçalves P, Sampaio JP, Dover J, Johnston M, Rokas A. 2010. Remarkably 579
ancient balanced polymorphisms in a multi-locus gene network. Nature 464:54–58. 580
Holt C, Yandell M. 2011. MAKER2: An annotation pipeline and genome-database management 581
tool for second-generation genome projects. BMC Bioinformatics 12. 582
.CC-BY-NC 4.0 International licenseavailable under a
(which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made
The copyright holder for this preprintthis version posted November 21, 2023. ; https://doi.org/10.1101/2023.11.21.568075doi: bioRxiv preprint
26
Hurst LD, Pál C, Lercher MJ. 2004. The evolutionary dynamics of eukaryotic gene order. Nature 583
Review Genetics 5:299–310. 584
James SA, Bond CJ, Stanley R, Ravella SR, Péter G, Dlauchy D, Roberts IN. 2016. Apiotrichum 585
terrigenum sp. nov., a soil-associated yeast found in both the UK and mainland Europe. 586
International Journal of Systematic and Evolutionary Microbiology 66:5046–5050. 587
Jiricny N, Diggle SP, West SA, Evans BA, Ballantyne G, Ross-Gillespie A, Griffin AS. 2010. 588
Fitness correlates with the extent of cheating in a bacterium. Journal of Evolutionary 589
Biology 23:738–747. 590
Julou T, Mora T, Guillon L, Croquette V, Schalk IJ, Bensimon D, Desprat N. 2013. Cell–cell 591
contacts confine public goods diffusion inside Pseudomonas aeruginosa clonal 592
microcolonies. Proceedings of the National Academy of Sciences 110:12577–12582. 593
Kalyaanamoorthy S, Minh BQ, Wong TK, von Haeseler A, Jermiin LS. 2017. ModelFinder: fast 594
model selection for accurate phylogenetic estimates. Nature methods 14:587. 595
Katoh K, Standley DM. 2013. MAFFT multiple sequence alignment software version 7: 596
improvements in performance and usability. Molecular biology and evolution 30:772–597
780. 598
Kellis M, Patterson N, Endrizzi M, Birren B, Lander ES. 2003. Sequencing and comparison of 599
yeast species to identify genes and regulatory elements. Nature 423:241–254. 600
Kominek J, Doering DT, Opulente DA, Shen X-X, Zhou X, DeVirgilio J, Hulfachor AB, 601
Groenewald M, Mcgee MA, Karlen SD, et al. 2019. Eukaryotic Acquisition of a Bacterial 602
Operon. Cell 176:1356-1366.e10. 603
Kramer J, Özkaya Ö, Kümmerli R. 2020. Bacterial siderophores in community and host 604
interactions. Nature Review Microbiology 18:152–163. 605
Krause DJ, Hittinger CT. 2022. Functional Divergence in a Multi-gene Family Is a Key 606
Evolutionary Innovation for Anaerobic Growth in Saccharomyces cerevisiae. Molecular 607
Biology and Evolution 39:202. 608
Krause DJ, Kominek J, Opulente DA, Shen X-X, Zhou X, Langdon QK, DeVirgilio J, Hulfachor 609
AB, Kurtzman CP, Rokas A, et al. 2018. Functional and evolutionary characterization of 610
.CC-BY-NC 4.0 International licenseavailable under a
(which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made
The copyright holder for this preprintthis version posted November 21, 2023. ; https://doi.org/10.1101/2023.11.21.568075doi: bioRxiv preprint
27
a secondary metabolite gene cluster in budding yeasts. Proceedings of the National 611
Academy of Sciences 115:11030–11035. 612
Lesuisse E, Blaiseau P-L, Dancis A, Camadro J-M. 2001. Siderophore uptake and use by the 613
yeast Saccharomyces cerevisiae. Microbiology 147:289–298. 614
Lesuisse E, Simon-Casteras M, Labbe P. 1998. Siderophore-mediated iron uptake in 615
Saccharomyces cerevisiae: the SIT1 gene encodes a ferrioxamine B permease that 616
belongs to the major facilitator superfamily. Microbiology 144:3455–3462. 617
Letunic I, Bork P. 2021. Interactive Tree Of Life (iTOL) v5: an online tool for phylogenetic tree 618
display and annotation. Nucleic Acids Research 49:W293–W296. 619
Li Y, Steenwyk JL, Chang Y, Wang Y, James TY, Stajich JE, Spatafora JW, Groenewald M, 620
Dunn CW, Hittinger CT, et al. 2021. A genome-scale phylogeny of the kingdom Fungi. 621
Current Biology 31:1653-1665.e5. 622
Lolas MA, Castro A, Polanco R, Gainza-Cortés F, Ferrada E, Sosnowski MR, Díaz GA. 2020. 623
First Report of Eutypa lata Causing Dieback of Grapevines (Vitis vinifera) in Chile. Plant 624
Disease 104:2024. 625
Marcet-Houben M, Gabaldón T. 2015. Beyond the Whole-Genome Duplication: Phylogenetic 626
Evidence for an Ancient Interspecies Hybridization in the Baker’s Yeast Lineage. PLOS 627
Biology 13:e1002220. 628
Nguyen L-T, Schmidt HA, Von Haeseler A, Minh BQ. 2015. IQ-TREE: a fast and effective 629
stochastic algorithm for estimating maximum-likelihood phylogenies. Molecular biology 630
and evolution 32:268–274. 631
Niehus R, Picot A, Oliveira NM, Mitri S, Foster KR. 2017. The evolution of siderophore 632
production as a competitive trait. Evolution 71:1443–1455. 633
Novo M, Bigey F, Beyne E, Galeote V, Gavory F, Mallet S, Cambon B, Legras J-L, Wincker P, 634
Casaregola S, et al. 2009. Eukaryote-to-eukaryote gene transfer events revealed by the 635
genome sequence of the wine yeast Saccharomyces cerevisiae EC1118. Proceedings of 636
the National Academy of Sciences 106:16333–16338. 637
Opulente DA, LaBella AL, Harrison M-C, Wolters JF, Liu C, Li Yonglin, Kominek J, Steenwyk 638
JL, Stoneman HR, VanDenAvond J, et al. 2023. Genomic and ecological factors shaping 639
.CC-BY-NC 4.0 International licenseavailable under a
(which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made
The copyright holder for this preprintthis version posted November 21, 2023. ; https://doi.org/10.1101/2023.11.21.568075doi: bioRxiv preprint
28
specialism and generalism across an entire subphylum. unpublished data, 640
https://doi.org/10.1101/2023.06.19.545611, bioRxiv, last accessed November 13, 2023 641
Quaedvlieg W, Kema GHJ, Groenewald JZ, Verkley GJM, Seifbarghi S, Razavi M, Mirzadi 642
Gohari A, Mehrabi R, Crous PW. 2011. Zymoseptoria gen. nov.: a new genus to 643
accommodate Septoria-like species occurring on graminicolous hosts. Persoonia 26:57–644
69. 645
Raguzzi F, Lesuisse E, Crichton RR. 1988. Iron storage in Saccharomyces cerevisiae. FEBS 646
Letters 231:253–258. 647
Raymond KN, Dertz EA, Kim SS. 2003. Enterobactin: An archetype for microbial iron transport. 648
Proceedings of the National Academy of Sciences 100:3584–3588. 649
Saerens KMJ, Saey L, Soetaert W. 2011. One-step production of unacetylated sophorolipids by 650
an acetyltransferase negative Candida bombicola. Biotechnology and Bioengineering 651
108:2923–2931. 652
Schiessl KT, Janssen EM-L, Kraemer SM, McNeill K, Ackermann M. 2017. Magnitude and 653
Mechanism of Siderophore-Mediated Competition at Low Iron Solubility in the 654
Pseudomonas aeruginosa Pyochelin System. Frontiers in Microbiology 8: 1964. 655
Shen X-X, Opulente DA, Kominek J, Zhou X, Steenwyk JL, Buh KV, Haase MAB, Wisecaver 656
JH, Wang M, Doering DT, et al. 2018. Tempo and Mode of Genome Evolution in the 657
Budding Yeast Subphylum. Cell 175:1533-1545.e20. 658
Stearman R, Yuan DS, Yamaguchi-Iwai Y, Klausner RD, Dancis A. 1996. A Permease-Oxidase 659
Complex Involved in High-Affinity Iron Uptake in Yeast. Science 271:1552–1557. 660
Steinegger M, Söding J. 2017. MMseqs2 enables sensitive protein sequence searching for the 661
analysis of massive data sets. Nature Biotechnology 35:1026–1028. 662
Takashima M, Manabe R, Nishimura Y, Endoh R, Ohkuma M, Sriswasdi S, Sugita T, Iwasaki 663
W. 2019. Recognition and delineation of yeast genera based on genomic data: Lessons 664
from Trichosporonales. Fungal Genetics and Biology 130:31–42. 665
Thomma BPHJ, Van Esse HP, Crous PW, De Wit PJGM. 2005. Cladosporium fulvum (syn. 666
Passalora fulva), a highly specialized plant pathogen as a model for functional studies on 667
plant pathogenic Mycosphaerellaceae. Molecular Plant Pathology 6:379–393. 668
.CC-BY-NC 4.0 International licenseavailable under a
(which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made
The copyright holder for this preprintthis version posted November 21, 2023. ; https://doi.org/10.1101/2023.11.21.568075doi: bioRxiv preprint
29
Usher KC, Özkan E, Gardner KH, Deisenhofer J. 2001. The plug domain of FepA, a TonB-669
dependent transport protein from Escherichia coli, binds its siderophore in the absence of 670
the transmembrane barrel domain. Proceedings of the National Academy of Sciences 671
98:10676–10681. 672
Vakirlis N, Sarilar V, Drillon G, Fleiss A, Agier N, Meyniel J-P, Blanpain L, Carbone A, 673
Devillers H, Dubois K, et al. 2016. Reconstruction of ancestral chromosome architecture 674
and gene repertoire reveals principles of genome evolution in a model yeast genus. 675
Genome Research 26:918–932. 676
Wang M, Schaefer AL, Dandekar AA, Greenberg EP. 2015. Quorum sensing and policing of 677
Pseudomonas aeruginosa social cheaters. Proceedings of the National Academy of 678
Sciences 112:2187–2191. 679
Weigert M, Kümmerli R. 2017. The physical boundaries of public goods cooperation between 680
surface-attached bacterial cells. Proceedings of the Royal Society B 284:20170631. 681
Yun C-W, Bauler M, Moore RE, Klebba PE, Philpott CC. 2001. The Role of the FRE Family of 682
Plasma Membrane Reductases in the Uptake of Siderophore-Iron in Saccharomyces 683
cerevisiae. Journal of Biological Chemistry 276:10218–10223. 684
Yun C-W, Tiedeman JS, Moore RE, Philpott CC. 2000. Siderophore-Iron Uptake in 685
Saccharomyces cerevisiae: Identification of Ferrichrome and Fusarinine Transporters. 686
Journal of Biological Chemistry 275:16354–16359. 687
688
.CC-BY-NC 4.0 International licenseavailable under a
(which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made
The copyright holder for this preprintthis version posted November 21, 2023. ; https://doi.org/10.1101/2023.11.21.568075doi: bioRxiv preprint
30
Figures 689
Figure 1 690
691
692
Figure 1. Enb1 is the key transporter mediating the uptake of enterobactin-bound iron in St. 693
bombicola. Genes related to enterobactin biosynthesis and candidate genes for transport are 694
native fungal genes (green) or underwent horizontal gene transfer (HGT) from a bacterial operon 695
(salmon) (A). Growth of St. bombicola wildtype (WT) and mutants in regular- and low-iron 696
conditions (B). The minimal growth of the enb1
D
strain in low-iron medium could be attributed 697
to intracellular iron storage in vacuoles and other compartments during preculturing (Raguzzi et 698
al. 1988) or reductive transport of enterobactin-bound iron with low affinity (Lesuisse et al. 699
2001). Chromeazurol S overlay (O-CAS) assay of the WT and enb1
D
mutants; iron sequestration 700
converts the blue color to orange (C). Extracellular enterobactin accumulation of St. bombicola 701
mutants (D). Data are presented as mean values and standard deviations of three independent 702
biological replicates. Statistical significance of the differences between the enb1
D
, fre-tm
D
, and 703
WT strains was evaluated by Student’s t-Test. *, p-value < 0.001. 704
705
A
C
B
D
.CC-BY-NC 4.0 International licenseavailable under a
(which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made
The copyright holder for this preprintthis version posted November 21, 2023. ; https://doi.org/10.1101/2023.11.21.568075doi: bioRxiv preprint
31
Figure 2 706
707
708
Figure 2. Maximum likelihood phylogeny of Enb1 homologs, including paralogs, from the yeast 709
subphylum Saccharomycotina using an amino acid sequence alignment. Branch values shown 710
are the percentages of branch support out of 1000 bootstrap replicates. The four Arn family 711
siderophore transporters (Arn1, Arn2, Sit1, and Enb1) and two glutathione exchangers (Gex1 712
and Gex2) from S. cerevisiae are labeled in the tree. Yeast major clades are colored according to 713
(Shen et al. 2018) and are now circumscribed as orders (Groenewald et al. 2023). Figure S3 714
shows a midpoint-rooted version of the same ML phylogeny; this rooting supports the 715
conclusion that Enb1 is monophyletic. 716
717
.CC-BY-NC 4.0 International licenseavailable under a
(which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made
The copyright holder for this preprintthis version posted November 21, 2023. ; https://doi.org/10.1101/2023.11.21.568075doi: bioRxiv preprint
32
Figure 3 718
719
720
721
Figure 3. Evidence for HGT of ENB1 in yeasts. Maximum likelihood phylogeny of the fungal 722
Enb1 protein sequences most closely related to Saccharomycotina Enb1 (A), which are a clade 723
from the full analyses shown in Figure S1. The 7 closest relatives of Enb1 proteins from 724
filamentous fungi were chosen as outgroups. Branch values shown are the percentages of branch 725
support out of 1000 bootstrap replicates. Branches are colored in light blue, dark blue, red, and 726
black to represent species from the genera Wickerhamiella, Starmerella, order 727
Saccharomycetales, and outgroup lineages, respectively. Branches with the same species name 728
occur because some genomes encode multiple copies of Enb1. The simplified topologies of the 729
unconstrained and constrained gene trees are shown in (B). Branches repositioned in the 730
constrained topology are highlighted with wider lines and denoted with (*). The SH and AU 731
topology test results are shown in (C). p-values of weighted Shimodaira-Hasegawa (wSH) test 732
and approximately unbiased (AU) test were both lower than the statistical significance threshold 733
of 0.05. 734
735
A
C
B
.CC-BY-NC 4.0 International licenseavailable under a
(which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made
The copyright holder for this preprintthis version posted November 21, 2023. ; https://doi.org/10.1101/2023.11.21.568075doi: bioRxiv preprint
33
Figure 4 736
737
738
739
Figure 4. Evolution of the fungal enterobactin transporter encoded by ENB1 and its functional 740
integration with the bacterial ent operon in yeasts. The genome-scale species phylogeny was 741
adopted from a recent publication (Opulente et al. 2023) by pruning irrelevant branches of the 742
phylogeny of 1,154 yeasts using iTOL v5 (Letunic and Bork 2021). Branches colored in orange 743
represent the presence of ent genes. Dashed arrows in red denote horizontal gene transfer (HGT) 744
events. The brown square represents genes encoding the Enterobacteriales FepC/D/G ABC-type 745
enterobactin transporter. Green stars represent the number of genes encoding fungal Enb1 746
transporters. Branch names in green represent enterobactin producers harboring both ENB1 and 747
ent genes. Branch names in blue represent enterobactin cheaters harboring only genes encoding 748
the Enb1 transporter. Species with names in black have neither ENB1 nor ent genes. Gene icons 749
are colored as in Figure 1. 750
751
.CC-BY-NC 4.0 International licenseavailable under a
(which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made
The copyright holder for this preprintthis version posted November 21, 2023. ; https://doi.org/10.1101/2023.11.21.568075doi: bioRxiv preprint
34
Figure S1 752
753
754
755
Figure S1. Midpoint-rooted maximum likelihood phylogeny of top Enb1 BLASTp hits from 756
NCBI nr database. The maximum likelihood (ML) phylogeny was constructed with 1,165 757
sequences obtained from the 2,626 top BLASTp hits with E-values < e-60 by removing 758
sequences with > 90% similarity using the clustering function of the mmseq2 (Steinegger and 759
Söding 2017). The branches and outer strip were colored based on the key on the bottom right. 760
Nodes with bootstrap values > 90% were highlighted using (•). 761
762
.CC-BY-NC 4.0 International licenseavailable under a
(which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made
The copyright holder for this preprintthis version posted November 21, 2023. ; https://doi.org/10.1101/2023.11.21.568075doi: bioRxiv preprint
35
Figure S2 763
764
765
766
767
Figure S2. Midpoint-rooted maximum likelihood amino acid phylogeny of Enb1 homologs, 768
including paralogs, from a database of 1644 fungal genomes. The branches were colored based 769
on the key on the bottom right. Nodes with bootstrap values > 90% were highlighted using (•). 770
Other Arn family proteins from S. cerevisiae and St. bombicola are labeled. The monophyletic 771
clade consisting of Enb1 proteins from yeasts is zoomed in with the pruned tree. 772
773
.CC-BY-NC 4.0 International licenseavailable under a
(which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made
The copyright holder for this preprintthis version posted November 21, 2023. ; https://doi.org/10.1101/2023.11.21.568075doi: bioRxiv preprint
36
Figure S3 774
775
776
Figure S3. Midpoint-rooted maximum likelihood phylogeny of Enb1 homologs, including 777
paralogs, from the yeast subphylum Saccharomycotina using an amino acid sequence alignment. 778
Enb1 orthologs are marked in black. Nodes with bootstrap values > 90% out of 1000 bootstrap 779
replicates were highlighted using (•). Yeast major clades are colored according to (Shen et al. 780
2018) and are now circumscribed as orders (Groenewald et al. 2023). 781
782
.CC-BY-NC 4.0 International licenseavailable under a
(which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made
The copyright holder for this preprintthis version posted November 21, 2023. ; https://doi.org/10.1101/2023.11.21.568075doi: bioRxiv preprint
37
Figure S4 783
784
785
786
Figure S4. Topology tests of possible HGTs within the W/S clade yeasts. Maximum likelihood 787
phylogeny of all W/S clade Enb1 protein sequences (A). The 7 closest relatives of Enb1 proteins 788
from filamentous fungi were chosen as outgroups. Branch values shown are the percentages of 789
branch support out of 1000 bootstrap replicates. Branches are colored in light blue, dark blue, 790
and black to represent species from the Wickerhamiella, Starmerella, and outgroup lineages, 791
respectively. Branches with the same species name occur because some genomes encode 792
multiple copies of Enb1. The simplified topologies of the unconstrained and constrained gene 793
trees are shown in (B). Branches repositioned in the constrained topology are denoted with (*). 794
Constrained Topology 1 enforced the Enb1 homologs of W. versatilis to be outgroups to the 795
Enb1 homologs of all the Starmerella species, except for St. stellata. Constrained Topology 2 796
enforced the Enb1 homologs of St. stellata to be outgroups to the Enb1 homologs of all other 797
Starmerella species. Constrained Topology 3 enforced the clustering of the Enb1 homologs from 798
St. stellata with those from the St. ratchasimensis, St. apicola, St. riodocensis, St. kuoi, and St. 799
starmerella to conform to the species phylogeny. The delta log-likelihoods of constrained 800
topologies are compared to the unconstrained topology, and the SH and AU topology test results 801
are shown in (C). 802
803
A
B
C
.CC-BY-NC 4.0 International licenseavailable under a
(which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made
The copyright holder for this preprintthis version posted November 21, 2023. ; https://doi.org/10.1101/2023.11.21.568075doi: bioRxiv preprint
38
Supplementary Datasets 804
Dataset S1-1. Top hits of BLASTp against the NCBI non-redundant protein sequences database 805
using St. bombicola Enb1 protein sequence as query. 806
Dataset S1-2. Top hits of BLASTp against the NCBI non-redundant protein sequences database, 807
excluding fungi, using St. bombicola Enb1 protein sequence as query. 808
Dataset S2. Top hits of HMMER scan against the 1644-fungi genome database using the aligned 809
sequences of S. cerevisiae and St. bombicola Enb1 proteins as queries. 810
Dataset S3. Top hits of HMMER scan against the 345-yeast genome database using the aligned 811
sequences of S. cerevisiae and St. bombicola Enb1 proteins as queries. 812
Dataset S4. Newick format strings of phylogenetic trees constructed in this work. 813
Dataset S5. Strains used in this study. 814
Dataset S6. Oligonucleotides for construction of St. bombicola deletion mutants. 815
816
.CC-BY-NC 4.0 International licenseavailable under a
(which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made
The copyright holder for this preprintthis version posted November 21, 2023. ; https://doi.org/10.1101/2023.11.21.568075doi: bioRxiv preprint
