ChapterPDF Available

Probing the Deep Genetic Basis of a Novel Trait in Escherichia coli*


Abstract and Figures

Evolution innovates by repurposing existing genetic elements to produce new functions. However, the range of new functions and traits this evolutionary tinkering can produce is limited to those that are supported and enabled by the rest of the genome. The full complement of genes in a genome required for a novel trait to manifest constitutes the trait’s “deep” genetic basis. The deep genetic basis of novel traits can be very difficult to determine under most circumstances, leaving it understudied despite its critical importance. Novel traits that arise during highly tractable microbial evolution experiments present opportunities to correct this deficit. One such novel trait is aerobic growth on citrate (Cit+ ), which evolved in one of twelve populations in the Long-Term Evolution Experiment with Escherichia coli (LTEE). We sought to uncover the deep genetic basis of this trait by transforming 3,985 single gene knockout mutants from the Keio collection with a plasmid that can confer aerobic growth on citrate. In our preliminary screen, we identified 111 genes putatively necessary for expression of the Cit+ trait. Of these, ∼ 32% are involved in core metabolic pathways, including the TCA and glycolysis pathways. Another ∼ 22% encode a variety of transporter proteins. The remaining genes are either of unknown function or uncertain involvement with citrate metabolism. Our work demonstrates how novel traits that are built upon pre-existing functions can depend on the activity of a large number of genes, hinting at an unappreciated level of complexity in the evolution of relatively simple new functions.
Content may be subject to copyright.
Chapter 10
Probing the Deep Genetic Basis of a Novel Trait
in Escherichia coli*
Tanush Jagdish, J. Jeffrey Morris, Brian D. Wade and Zachary D. Blount
Abstract Evolution innovates by repurposing existing genetic elements to produce
new functions. However, the range of new functions and traits this evolutionary tin-
kering can produce is limited to those that are supported and enabled by the rest of
the genome. The full complement of genes in a genome required for a novel trait to
manifest constitutes the trait’s “deep” genetic basis. The deep genetic basis of novel
traits can be very difficult to determine under most circumstances, leaving it under-
studied despite its critical importance. Novel traits that arise during highly tractable
microbial evolution experiments present opportunities to correct this deficit. One
such novel trait is aerobic growth on citrate (Cit+), which evolved in one of twelve
populations in the Long-Term Evolution Experiment with Escherichia coli (LTEE).
We sought to uncover the deep genetic basis of this trait by transforming 3,985 sin-
gle gene knockout mutants from the Keio collection with a plasmid that can confer
Tanush Jagdish
Department of Microbiology and Molecular Genetics, Michigan State University; BEACON Cen-
ter for the Study of Evolution in Action, East Lansing, MI 48824, USA; and Department of Physics
and of Biology, Kalamazoo College, Kalamazoo, MI 49006, USA. Currently at: Program for Sys-
tems Biology, Harvard University, Cambridge, MA 02138, USA e-mail: tanush@g.harvard.
J. Jeffrey Morris
Department of Microbiology and Molecular Genetics, Michigan State University; BEACON Cen-
ter for the Study of Evolution in Action, East Lansing, MI 48824, USA; and Department of Biol-
ogy, University of Alabama at Birmingham, Birmingham, AL 35924, USA
Brian D. Wade
Department of Plant, Soil and Microbial Sciences, Michigan State University, East Lansing, MI
48824, USA
Zachary D. Blount
Department of Microbiology and Molecular Genetics, Michigan State University; BEACON Cen-
ter for the Study of Evolution in Action, East Lansing, MI 48823, USA; and Department of Biol-
ogy, Gambier, OH 43022, USA e-mail:
* This paper was externally peer-reviewed.
© Springer Nature Switzerland AG 2020
Genetic and Evolutionary Computation,
W. Banzhaf et al. (eds.), Evolution in Action: Past, Present and Future,
108 Tanush Jagdish, J. Jeffrey Morris, Brian D. Wade and Zachary D. Blount
aerobic growth on citrate. In our preliminary screen, we identified 111 genes puta-
tively necessary for expression of the Cit+trait. Of these, 32% are involved in core
metabolic pathways, including the TCA and glycolysis pathways. Another 22%
encode a variety of transporter proteins. The remaining genes are either of unknown
function or uncertain involvement with citrate metabolism. Our work demonstrates
how novel traits that are built upon pre-existing functions can depend on the activity
of a large number of genes, hinting at an unappreciated level of complexity in the
evolution of relatively simple new functions.
Key words: Experimental Evolution, Microbial Evolution, Evolutionary Innova-
tion, Evolution Experiments, Cit+
10.1 Introduction
The living world is astonishingly diverse. Ecologically significant, qualitatively new
traits have played an important role in the origin of this diversity [1]. Evolutionary
innovations allow lineages to escape competitive pressures in their ancestral niches
by invading new niches. Adaptation to the new niches can then drive divergence,
speciation, and increased diversity [50]. Diversification driven by novel trait evo-
lution has likely been particularly consequential in microbes, where the origin of
novel traits and speciation are thought to be synonymous [12] and thus responsible
for the estimated billions to trillions of extant microbial clades [37,49].
Microbial lineages can evolve novel traits in two distinct ways: acquisition of
genes from other lineages through horizontal gene transfer (HGT) or the origination
of new genes through modification of existing sequences. HGT can rapidly dissem-
inate new traits among a diverse and distantly related community of microbes and
has been implicated in the spread of antibiotic resistance [32,55], virulence fac-
tors [19,21], and even entire metabolically related gene clusters [2]. However, truly
novel traits ultimately arise by the modification of genetic information that did not
originally encode them [23].
Such genetic modification, or ‘evolutionary tinkering’, occurs through four dis-
tinct mechanisms [1,16]. First, new genes can arise de novo from mutations that
induce expression of previously non-coding DNA that fortuitously yield functional
polypeptides [6,20]. Second, duplications can lead to neofunctionalization, in which
mutations of redundant gene copies confer novel functions [13,41,43]. Third, pre-
existing gene components recombine to yield new functions in a process called ‘do-
main shuffling’ [8,23,26]. And finally, via a mechanism directly relevant to our
work reported here, mutation or recombination can place existing genes under new
regulatory control, co-opting functionality by changing the physiological or devel-
opmental context within which the genes are expressed [8].
The mutational event that immediately causes the manifestation — or “actual-
ization” — of a novel trait has been the principal concern of most research into the
mechanistic bases of evolutionary innovations [1,16], but it is only one part of a
10 Probing the Deep Genetic Basis of a Novel Trait in Escherichia coli 109
larger evolutionary process [7,8]. A new trait can only be actualized if prior evo-
lution has “potentiated” its emergence, either by mechanistically increasing the rate
of actualizing mutations, or by epistatically enabling those mutations to produce the
trait upon their occurrence [8,9]. Moreover, new traits almost always first appear in
a weak form. The effectiveness of a new trait therefore requires a potentially open-
ended, selection-mediated accumulation of mutations that “refines” its functionality
and improves its fitness contribution [8,44,45]. In this process, potentiation makes
the innovation possible, actualization brings it into being, and refinement makes it
functionally effective [8].
Prior evolution crucially determines the potential for evolving a novel trait [10].
A new trait can only be actualized if it is mutationally reachable from an organism’s
existing genetic state [7,8]. This is to say, it must be possible for the genetic infor-
mation needed for the trait to arise from sequences that already exist in the genome.
One facet of this principle is the necessary interrelationships between genes within
a genome. All genes exist in a genome that includes both the set of regulatory ele-
ments that govern their expression and that of the other genes with which they in-
teract. Genes also function in a broader organismal context in which their expressed
products interact with those encoded by other genes. Typically, gene products only
produce a given function when working in tandem with many other gene products.
One consequence of this interdependence is that a new trait can only evolve in a lin-
eage in which it will be supported by the existing genomic and organismal context.
This broader context that is a necessary part of the potentiation of a novel trait’s
evolution may be called its “deep” genetic basis.
Consider, for example, an organism that grows on substrate A. However, the
organism’s environment also contains substrate B, which the organism cannot me-
tabolize. Suppose a mutation in a gene produces an enzyme for converting substrate
B to substrate A once it is in the cell, thus allowing the organism to survive on sub-
strate B. The capacity to grow on substrate B will only manifest if the organism has
the means to transport substrate B into the cell. Evolution of growth on substrate B
is hence contingent on and potentiated by not only the presence of the gene that can
be mutated to produce the new enzyme, but also the broader genetic context that
includes the requisite transporter gene.
In real organisms, the broader context that determines evolutionary potential is
exponentially more complicated [51,56]. Organisms possess a wide array of inte-
grated traits ranging from survival and stress tolerance to mating and reproduction
[52,54], and the genes and pathways coding for any given trait often also affect
other functions at different levels of causality [25,33,34]. These interactions be-
tween traits and their underlying genetic bases are highly complex and difficult to
understand [24,29,30]. Adding to this challenge is the fact that novel traits emerge
over evolutionary time, which can range from hundreds to thousands of generations
[10,39,40]. This system-wide, multi-level complexity coupled with the infeasible
timescales of novel trait evolution makes it inherently challenging to identify the
full set of interacting genetic elements underlying a new trait.
Novel traits sometimes arise during long-term evolution experiments with highly
tractable microbial model organisms, providing opportunities to examine their deep
110 Tanush Jagdish, J. Jeffrey Morris, Brian D. Wade and Zachary D. Blount
genetic bases [17]. One such instance arose during the Long-Term Evolution Experi-
ment with Escherichia coli (LTEE). The LTEE was begun in 1988 with the founding
of twelve populations of E. coli from a single clone. These populations have since
been evolved for more than 70,000 generations of daily 1:100 serial transfer in Davis
and Mingioli minimal medium supplemented with glucose (DM25) [15,34,35].
Throughout the experiment, population samples have been frozen every 500 gen-
erations, providing a complete, viable fossil record of the evolution in each popu-
lation [35]. DM25 also contains a high concentration of citrate (500 mg/L), which
is added as an iron chelating agent and constitutes a potential second carbon and
energy source. However, despite possessing a complete TCA cycle, E. coli is partly
defined as a species by its inability to grow aerobically on citrate (Cit-) [22,47].
This Cit-phenotype is due to an inability to transport citrate into the cell when oxy-
gen is present, but most E. coli strains can grow fermentatively on citrate using a
transporter expressed only under anoxia [42].
Despite having the cellular machinery to potentially evolve aerobic growth on cit-
rate (Cit+), spontaneous Cit+mutants of E. coli are extremely rare under most con-
ditions [22,53]. Nonetheless, a weak Cit+variant appeared in one LTEE population
after 31,000 generations [9]. Later genomic analysis showed that the trait was actu-
alized by a duplication that placed the previously silent citrate transporter gene, citT,
under the control of a promoter that directs expression under aerobic metabolism [8].
The Cit+subpopulation remained a minority until shortly after 33,000 generations,
when refined variants better able to exploit the citrate resource evolved and rose to
high frequency in the population, concurrently leading to a several-fold increase in
the size of the population. This process of refinement is ongoing and has involved
further adaptation to the idiosyncratic physiology of the Cit+trait [5].
The long-delayed and singular evolution of the Cit+trait in the LTEE was con-
tingent upon the particular history of the population in which it arose. The ongoing
research into this history has revealed the complexity of the interactions underlying
evolutionary potential. A series of “replay” experiments with clones isolated from
the population’s fossil record showed that later clones had a significantly higher rate
of mutation to Cit+, and thus a greater potential to evolve the trait than did earlier
ones [9]. This potentiation arose in part from adaptation to acetate-based ecological
interactions that evolved in the population. Quandt et al. [45] identified a series of
mutations that occurred in the citrate synthase gene, gltA, of the lineage in which
Cit+eventually arose. These mutations altered carbon flow into the TCA cycle, im-
proving growth on acetate. As citrate is also metabolized via the TCA cycle, they
also pre-adapted the lineage to growth on citrate, and rendered the Cit+actualiz-
ing mutation beneficial when it eventually arose. These mutations hence allowed
the weak initial Cit+variant lineage to remain in the population long enough to
accumulate refining mutations that improved growth on citrate [45]. Ironically, the
potentiating mutations in gltA compensated for the anti-potentiating effects of ear-
lier mutations that were beneficial to growth on glucose [36]. Moreover, even after
potentiation, the Cit+actualizing mutation’s fitness benefit prior to 30,000 genera-
tions was too low to permit it to outcompete other beneficial mutations available to
the population [36].
10 Probing the Deep Genetic Basis of a Novel Trait in Escherichia coli 111
Much has therefore been gleaned about the evolutionary potentiation of the Cit+
trait during the LTEE. However, the trait’s evolution also depended on the full
complement of genes necessary for citrate metabolism that existed in the ances-
tral genome prior to the start of the LTEE. These genes that interact and support
aerobic growth on citrate, the trait’s deep genetic basis, remain a mystery. Here we
describe an initial exploration of the deep genetic basis of the novel Cit+trait, in
which we sought to identify the non-essential genes required for aerobic growth on
citrate in the presence of the actualizing mutation.
10.2 Methods
10.2.1 Long-Term Evolution Experiment with E. coli
The LTEE has been described in greater detail elsewhere [35]. Briefly, twelve popu-
lations of E. coli B were founded in 1988 and have since been evolved for more than
70,000 generations of serial batch culture in Davis and Mingioli minimal medium
supplemented with 25 mg/L glucose (“DM25”; [15,34]. The populations are di-
luted 100-fold every 24 hours into fresh medium to a final volume of 10 mL and
maintained at 37 C with 120 rpm orbital aeration.
10.2.2 Screening for Genes Necessary for Aerobic Growth on
The pZBrnk-citT plasmid is a pUC19-based recombinant plasmid that contains a
copy of the novel rnk-citT module that actualized the Cit+trait in Ara-3 [8]. This
high copy number plasmid confers a Cit+phenotype in both E. coli B and E. coli
K12. The Keio collection is made up of 3985 strains of E. coli K12 in which a sin-
gle, non-essential gene has been deleted and replaced with a kanamycin resistance
cassette [4]. Genes required for aerobic growth on citrate were screened by chemi-
cally transforming each Keio strain with pZBrnk-citT and testing each transformant
for the capacity to grow aerobically on citrate using the procedure described below.
Each knock-out strain to be transformed was grown on LB plates supplemented
with kanamycin at a final working concentration of 50 µg/mL. Three to five colonies
of each strain were then resuspended in 300 µL of 50 mM CaCl2. The suspensions
were incubated on ice for 15 minutes, after which 100 ng of plasmid DNA was
added to the suspensions. Following 45 minutes of further incubation on ice, the
cells were heat-shocked at 42 C for 1 minute before being returned to ice for 5
minutes. A 500 µL volume of LB was added to the cells, shaken to mix, and the full
volume directly spread on Christensen’s citrate indicator plates supplemented with
50 µg/mL kanamycin and 100 µg/mL ampicillin [3,11]. The plates were then in-
112 Tanush Jagdish, J. Jeffrey Morris, Brian D. Wade and Zachary D. Blount
cubated for 10 days at 37 C. Plates were assessed after 5 and 10 days of incubation
for development of a hot pink coloration of the medium indicative of growth on cit-
rate. Those Keio strain transformants able to grow on citrate as indicated by a color
change were presumed to have deletions in genes that were not required for aero-
bic growth on citrate. Those that did not produce a color change were considered
to have deletions of candidate genes required for aerobic growth on citrate. Thirty
six mutants from the Keio collection were unused as viable transformants for those
mutants could not be obtained. This screen was carried out twice to rule out false
10.2.3 Curation and Analysis of Gene Functions
Functions of candidate genes were first manually curated using the EcoCyc database
[28]. All 111 functions were then grouped into five major categories: Metabolism,
Membrane-Related Proteins, Stress Response, Transcription, and Motility. Genes
that could not be classified into groups of three or more were labelled as “Other”,
and those with undefined functions were labelled “Unknown”.
10.2.4 Analysis of the Mutational History in Candidate Genes
Good et al. [18] conducted whole genome, whole population sequencing of all
frozen populations between 0 and 60,000 generations across all 12 LTEE lines. We
downloaded the annotated sequence data made publicly available by Good et al.
[18] and looked for mutations in any of the 111 candidate genes in all 12 LTEE
lines over the course of 60,000 generations. In order to ensure we only looked
at mutations that were nearing or had already reached fixation, we only consid-
ered mutations that had reached a frequency of 0.95. All mutational analysis was
carried out using R (version 3.5.1). Analysis scripts and raw data are deposited at
10.3 Results
We screened the entire Keio E. coli K-12 gene knockout collection for non-essential
genes required for aerobic growth on citrate by transforming each constituent knock-
out strain with a plasmid, pZBrnk-citT, which can confer a Cit+phenotype in a
wild type genetic background. We identified 111 candidate genes that are puta-
tively necessary for aerobic growth on citrate (Appendix). We manually categorized
each candidate gene by the primary function identified for it by the EcoCyc E. coli
database ([27], Fig. 10.1). Thirty-six percent of the candidates are involved in core
10 Probing the Deep Genetic Basis of a Novel Trait in Escherichia coli 113
Fig. 10.1: Functional Categories of Genes Putatively Necessary for Aerobic Growth on Citrate.
Genes grouped by functions assigned by EcoCyc [27]. Genes assigned to the ‘other’
category include those associated with fimbriae (2%), metalloproteases (2%),
lipoproteins (2%), cell division (1%), cell shape (1%), and curli formation (1%)
metabolic pathways, including sucA, sucB and sdhB, which encode TCA cycle en-
zymes, and others that encode glycolysis pathway enzymes (Fig. 10.1). Another
23% encode for membrane-associated proteins, such as tatB, macB, and garP,
which are involved in protein translocation, antibiotic export, and galactarate trans-
port, respectively (Fig. 10.1).
The remaining candidates are either of unknown function or uncertain involve-
ment with citrate metabolism. These genes include a substantial number that are
likely involved in bacterial stress response, such as multi-drug efflux pumps (emrD
and ybhF), and regulators of acid resistance and biofilm formation (ymgB). Other
candidates include transcription regulators (5), metalloproteases (3), lipoproteins
(3), and genes involved in regulating fimbriae (3), cell division (1), cell shape (1),
and curli formation (2).
Genes necessary for growth on citrate would be logical targets of selection in
the Cit+population. To determine if this has been the case for the 111 candidate
mutations we identified, we searched the whole-metagenome sequence data for the
citrate-using population from Good et al. [18]. We found that mutations had fixed or
nearly fixed in 8 of the candidate genes by 60,000 generations (Table 10.2). How-
ever, all mutations occurred in Ara-3 after 35,000 generations, at which point Ara-
3 had evolved a mutator phenotype [8].
All 12 LTEE populations have been evolving for over 70,000 generations. Ara-
3 remains the only population in which aerobic growth on citrate has evolved. An
ongoing question is that of whether it might evolve in any of the other 11 popula-
tions. Mutations that impair or eliminate the function of genes required for the Cit+
trait would presumably reduce the likelihood of evolving it. We therefore examined
114 Tanush Jagdish, J. Jeffrey Morris, Brian D. Wade and Zachary D. Blount
Fig. 10.2: Genes putatively identified as necessary for aerobic growth on citrate in which
mutations have fixed in LTEE populations by 60,000 generations. Ara-1, Ara-2, Ara-3,
Ara-4, Ara+3 and Ara+6 have all evolved heightened mutation rates over the course of
evolution. No fixed mutations in candidate genes were identified in populations Ara-6,
Ara+1, and Ara+5
the metagenome sequences for the other LTEE populations for mutations in the 111
genes our screen identified.
We found an array of mutations in genes putatively essential for aerobic growth
on citrate across the other LTEE populations. The vast majority of these mutations
occurred in the five populations, Ara-1, Ara-2, Ara-3, Ara-4, Ara+3, and Ara+6, in
which elevated mutation rates have also evolved over the course of the experiment
(Fig. 10.3; Fig. 10.2). However, we did identify mutations in three non-mutator pop-
ulations: a single mutation in a noncoding region of the transcription regulator gene,
iscR, fixed in Ara-5 by 45,000 generations; a missense mutation in a key oxoglu-
tarate dehydrogenase component gene, sucB, fixed in Ara+2 by 47,000 generations;
and in Ara+4 a missense mutation in the glycolysis regulating phosphofructokinase
gene, phoR, fixed by 18,000 generations, as well as the insertion of a mobile ge-
netic element in the aromatic amino acid biosynthesis gene aroG. We identified no
mutations in any candidate genes in Ara-6, Ara+1, or Ara+5.
10.4 Discussion
The Cit+trait that arose in the LTEE would seem to be a simple innovation, given
that the trait can be conferred in the ancestral genetic background by activating
expression of the citT gene that encodes a citrate-C4-dicarboxylate antiporter [8].
Despite this apparent simplicity, the trait’s manifestation required the activity of
other gene products, and thus the presence of other preexisting genes. These genes
10 Probing the Deep Genetic Basis of a Novel Trait in Escherichia coli 115
Fig. 10.3: Fixed mutations in candidate genes across 12 LTEE populations by 60,000 generations.
constitute the deep genetic basis of the Cit+trait, and we sought to identify them.
In total, our preliminary screen showed that 3%, or 111 out of 4288 non-essential
genes in E. coli were required for aerobic growth on citrate. Moreover, considering
the possible role of epistatic interactions and the conservative nature of our screen,
this number is likely to be an underestimate [14,46]. Our findings therefore suggest
that the deep genetic basis of the novel Cit+trait is quite broad, highlighting the
integrated nature of the organism.
The largest group of genes that we identified as putatively necessary for aerobic
growth on citrate belong to core metabolism. E. coli metabolizes exogenous citrate
via the TCA cycle, making the genes that encode the steps in the cycle necessary
[31,38,47]. Aerobic growth on citrate as a sole carbon source also creates problems
for biosynthesis as glycolysis is bypassed. Growth on citrate as a sole carbon source
thus requires the capacity to feed carbon into the gluconeogenesis pathways, as well
as to produce the TCA intermediates and key amino acid precursors 2-oxyglutarate
and succinyl-CoA. A further complication is caused by the physiology of the citT
transporter. Every molecule of citrate imported requires the simultaneous export of
a C4-dicarboxylate, and specifically the TCA intermediates of succinate, fumarate,
or malate [42]. Consequently, as few as 2 carbons per citrate molecule are available
for both catabolism and anabolism (Fig. 10.4). Given these considerations, several
genes might be predicted as necessary for aerobic growth on citrate.
Predicted genes include isocitrate dehydrogenase (icd), which is necessary to
yield 2-oxoglutarate because the 2-oxoglutarate decarboxylase (SucAB) reaction
is irreversible; isocitrate lyase (aceA), which is necessary to bypass the CO2-
producing SucAB and SucCD reactions via the glyoxylate shunt; and citrate syn-
thase (gltA), which is necessary to pass carbons harvested via glyoxylate back into
the TCA cycle to reach 2-oxoglutarate and succinyl-CoA. Interestingly, our screen
did not yield any of these genes. This odd result might be an artifact of the screen’s
design. These genes would be essential in minimal medium with citrate as a sole
carbon source. However, the Christensen’s citrate agar on which we conducted our
phenotypic screen likely contains sufficient amino acids and other metabolites to
compensate for anabolic deficiencies caused by loss of any one of these key en-
116 Tanush Jagdish, J. Jeffrey Morris, Brian D. Wade and Zachary D. Blount
Fig. 10.4: Citrate metabolic pathways in Ara-3 Cit+. Extracellular citrate is exchanged for the
intracellular C4-dicarboxylate TCA intermediates, succinate, fumarate, and malate by
the citT antiporter. This physiology means perhaps only 2 carbon atoms are available
per citrate molecule for both catabolism and anabolism. The citrate can be metabolized
in two ways. In the catabolic pathway (orange arrows), both carbons are lost as CO2
(leaving none for biosynthesis), but substantial energy is conserved in the form of
NAD(P)H. In the anabolic pathway (green arrows), no energy is gained, but the carbons
are harvested as glyoxylate, and can either be passed back into the TCA cycle or into
gluconeogenesis for biosynthesis of amino acids and other necessary metabolites.
Where two gene names are given (e.g. acnA, acnB) for a reaction, the genes are
redundant; where genes are listed as a complex (e.g. sucAB), all gene products are
needed to catalyze the reaction. Because of redundant genes and/or pathways, only 3
genes (gltA, icd, and aceA, shown in bold) may be predicted as essential for aerobic
growth on citrate as a sole carbon source. None of these genes were identified in our
screen, but others, shown in blue, were. Regulatory effects of 2-oxoglutarate are
indicated with dashed lines.
zymes. We plan to follow up on these preliminary findings with further screens
using Christensen’s citrate broth from which we will exclude yeast extract, which
should allow us to evaluate this hypothesis.
Some of the TCA genes we identified in our screen are not strictly essential, but
their absence could lead to the over-accumulation of intermediates, negatively af-
fecting regulation, and causing unbalanced growth. For instance, deletion of either
of the subunits of 2-oxoglutarate decarboxylase (sucA or sucB) eliminated the ca-
pacity to grow aerobically on citrate. This effect could be due to an accumulation
of 2-oxoglutarate levels that would inhibit citrate synthase, preventing the use of
citrate-derived carbon for amino acid biosynthesis. Similarly, loss of one of the four
subunits of succinate dehydrogenase, SdhB, eliminated the Cit+trait. SdhB is the
10 Probing the Deep Genetic Basis of a Novel Trait in Escherichia coli 117
cytoplasmic subunit responsible for passing electrons from SdhA, which binds both
succinate and FAD, to the electron transport chain [48]. It is possible that loss of
SdhB sequesters succinate in a form that cannot be accessed by citT, stopping the
flow of citrate into the cell.
The second largest subset of essential gene candidates were membrane proteins
and cell wall synthesis enzymes. The 22 membrane proteins include ABC trans-
porter families such as macB and yhhJ, and a diversity of importers, exporters,
and symporters. Several genes involved in peptidoglycan synthesis and remodeling
(ddlA, erfK, dacA) were also identified. This subset is more puzzling, as there is no
obvious direct relationship between these genes and citrate metabolism. While Cit+
requires an aerobically functioning citT membrane transporter, an rnk-citT module
(where rnk is an aerobic promoter) was available to the tested cells via a high-copy-
number plasmid [8]. Thus, the removal of the citT gene from the genome, or any
other transporter gene, should in principle not affect Cit+.
The role of most of the genes we have identified in aerobic growth on citrate is
unclear. This lack of obvious connection is perhaps most clearly seen in the cell
appendage biosynthesis genes our screen showed as putatively necessary to the Cit+
trait. These genes included those involved in flagellar biosynthesis (flgE, flgF, flgI,
and flgJ), fimbrial assembly (ycbS), and curli synthesis (csgABFG). Similarly, 18
candidate genes are involved in stress response functions in E. coli, including mul-
tidrug efflux pumps (emrD and ybhF), reactive oxygen defenses (sodB), acid resis-
tance regulators (ymgB), and response regulators for phosphate starvation (phoR).
At least some of these genes may be experimental artifacts of a screening procedure
that exposed the cells to multiple stressors, including exposure to two antibiotics,
treatment in high concentrations of calcium chloride, and cold.
Might the candidate genes we identified be targets of refining mutations that im-
proved aerobic growth on citrate after the evolution of the Cit+trait? To answer
this question, we examined the metagenome sequence data generated by Good et
al. [18] for the Ara-3 population through 60,000 generations. We found mutations
in 8 candidate genes, though none reached high frequency until well after the Cit+
clade evolved an elevated mutation rate around 35,000 generations [8]. The earli-
est mutation to rise to high frequency occurred at around 37,000 generations in the
gene csgG, which encodes an outer membrane lipoprotein involved in curli biosyn-
thesis. Due to the elevated mutation rate, however, it is unclear if these mutations
are beneficial or reached high frequency by hitchhiking with some other beneficial
Only 3 of the 8 mutated candidate genes in Ara-3, aroG, atpI, and ydiJ, are related
to central metabolism. These genes have a total of six mutations, all of which are
either synonymous or occur in noncoding regions. While it is possible that these
mutations might have beneficial fitness effects, it seems more likely that they are
hitchhikers. The five other candidate genes, csgG, flgF, ycbS, emrD, and ybhF, which
are involved in biofilm, flagellar, and fibrial biosynthesis, and encode multidrug
efflux pumps, respectively, have a total of 12 mutations. Five of these mutations are
synonymous or noncoding, but the remaining seven are missense or indel mutations.
The latter seven mutations would presumably affect gene function, and potentially
118 Tanush Jagdish, J. Jeffrey Morris, Brian D. Wade and Zachary D. Blount
conflict with our findings that they are necessary for the Cit+trait. Later work will
examine these mutations, their effects, and will determine if their identification in
the screen was perhaps an experimental artifact of some sort. Broadly, however,
the rarity of mutations in candidate genes in Ara-3 after the evolution of Cit+is
consistent with their being necessary for the trait and suggests that mutations in
them are generally detrimental.
If the 111 genes we identified are actually necessary for aerobic growth on citrate,
then mutations that impair or eliminate their function would seemingly reduce the
likelihood of evolving the Cit+trait. We identified nonsynonymous mutations and
indels in multiple candidate genes in 7 of the other 11 LTEE populations. We do not
yet know enough about the effects of these mutations on the function of the genes
in question. However, any that impair or eliminate function would likely reduce
or foreclose the possibility of the evolution of Cit+in the respective populations.
It will be interesting to examine how the evolvability of Cit+varies between the
populations in which mutations have occurred in candidate genes, and Ara-6, Ara+1,
and Ara+5, in which they have not.
Novel traits are not the result solely of the genetic changes that immediately un-
derlie them. Those genetic changes that actualize a trait must always occur in a
genetic background containing an integrated set of genes and gene products that
allow them to produce that new trait. Though the importance of this deep genetic
basis of novel traits is in a sense obvious, it has rarely been examined. The prelimi-
nary work we have described gives a glimpse into the deep genetic basis of the Cit+
trait that arose in the LTEE. The trait was actualized by a mutation that activated
the expression of a citrate transporter gene, citT, when oxygen was present via the
cooption of an alternate promoter, that of the aerobically expressed gene rnk. De-
spite this apparent simplicity, our results show that the manifestation of the Cit+trait
depends on the activity of more than 100 other genes. This finding shows how the
additive nature of novel traits in evolution means that even relatively minor, seem-
ingly simple novel traits nonetheless depend on a foundation of many pre-existing
elements that interact in complex and highly integrated ways.
Indeed, our findings suggest that what genes might be involved in the manifesta-
tion of a trait may be anything but obvious due to this complexity and integration.
Prior work has shown that the evolution of the Cit+trait was historically contingent
upon several potentiating mutations that arose over the course of the Ara-3 pop-
ulation’s history during the experiment, and the twisted paths of evolution during
that history. Our findings here show that the trait was contingent upon not simply
this history during the experiment. Indeed, it was contingent upon the much longer
history that preceded the experiment, over which evolution constructed an organism
with the full complement of genes that interacted in such a way as to support the
manifestation of the trait once the actualizing mutation took place. Such contingent
histories are necessary to provide the proper deep genetic basis that underlies the
evolution of all novel traits. Given the role of novel traits in evolution, our work
argues that it is time to take this deeper historical contingency seriously.
10 Probing the Deep Genetic Basis of a Novel Trait in Escherichia coli 119
Acknowledgements We thank Neerja Hajela for her outstanding technical expertise, logistical
aid, and general moral support. This project would not have been possible without her. We also
thank the undergraduate technicians who provided logistical support for our work, including
Camorrie Bradley, Michelle Mize, Maia Rowles, Kiyana Weatherspoon, Rafael Martinez, Jamie
Johnson, Devin Lake, and Jessica Baxter. We are grateful to Richard Lenski for helpful discus-
sions, advice, funding, lab space, and his continued mentorship. We are especially grateful to Erik
Goodman for his extensive and path-defining leadership of the BEACON Center for the Study
of Evolution in Action, and in whose honor this work is submitted. Mark Kauth, Kyle Card, and
Nkrumah Grant made critical contributions to the design of this project. This research was sup-
ported in part by grants from the National Science Foundation (currently DEB-1451740) and the
BEACON Center for the Study of Evolution in Action (DBI-0939454). T.J. acknowledges support
from the Internship Stipend Program at the Center for Career and Professional Development at
Kalamazoo College.
1. Andersson, D.I., Jerlstroem-Hultqvist, J., Naesvall, J.: Evolution of new functions de novo and
from preexisting genes. Cold Spring Harb Perspect Biol 7(6), a017996 (2015)
2. Araviad, L., Tatusov, R.L., Wolf, Y.I., Roland Walker, D., Koonin, E.: Evidence for massive
gene exchange between archaeal and bacterial hyperthermophiles. Trends in Genetics 14,
442–444 (1999)
3. Atlas, R.: Handbook of Microbiological Media for the Examination of Food. CRC, Boca
Raton, FL (2006)
4. Baba, T., Ara, T., Hasegawa, M., Takai, Y., Okumura, Y., Baba, M., Datsenko, K.A., Tomita,
M., Wanner, B.L., Mori, H.: Construction of Escherichia coli K-12 in-frame, single-gene
knockout mutants: The Keio collection. Molecular Systems Biology 2, 2006.0008 (2006)
5. Bajic, D., Vila, J.C.C., Blount, Z.D., Sanchez, A.: On the deformability of an empirical fitness
landscape by microbial evolution. Proceedings of the National Academy of Sciences of the
United States of America 115, 11,286–11,291 (2018)
6. Begun, D.J., Lindfors, H.A., Kern, A.D., Jones, C.D.: Evidence for de novo evolution of testis-
expressed genes in the Drosophila yakuba/Drosophila erecta clade. Genetics 176(2), 1131–
1137 (2007)
7. Blount, Z.D.: A case study in evolutionary contingency. Stud Hist Philos Sci Part C: Stud Hist
Philos Biol Biomed Sci 58, 82 – 92 (2016)
8. Blount, Z.D., Barrick, J.E., Davidson, C.J., Lenski, R.E.: Genomic analysis of a key innovation
in an experimental Escherichia coli population. Nature 489, 513–518 (2012)
9. Blount, Z.D., Borland, C.Z., Lenski, R.E.: Historical contingency and the evolution of a key
innovation in an experimental population of Escherichia coli. Proceedings of the National
Academy of Sciences of the United States of America 105(23), 7899–7906 (2008)
10. Blount, Z.D., Lenski, R.E., Losos, J.B.: Contingency and determinism in evolution: Replaying
life’s tape. Science 362, eaam5979 (2018)
11. Christensen, W.: Hydrogen sulfide production and citrate utilization in the differentiating of
enteric pathogens and coliform bacteria. Res Bull Weld County Health Dept Greeley Colo 1,
3–16 (1949)
12. Cohan, F.M.: Bacterial species and speciation. System Biology 50(4), 513–524 (2001)
13. Conant, G.C., Wolfe, K.H.: Turning a hobby into a job: How duplicated genes find new func-
tions. Nature Reviews Genetics 9, 938–950 (2008)
14. Dasmeh, P., Girard, E., Serohijos, A.W.R.: Highly expressed genes evolve under strong epis-
tasis from a proteome-wide scan in E. coli. Scientific Reports 7, 15844 (2017)
15. Davis, B.D., Mingioli, E.S.: Mutants of Escherichia coli requiring methionine or vitamin B12.
Journal of Bacteriology 60, 17–28 (1950)
120 Tanush Jagdish, J. Jeffrey Morris, Brian D. Wade and Zachary D. Blount
16. Ding, Y., Zhou, Q., Wang, W.: Origins of new genes and evolution of their novel functions.
Annual Review of Ecology, Evolution, and Systematics 43, 345–363 (2012)
17. Elena, S.F., Lenski, R.E.: Evolution experiments with microorganisms: The dynamics and ge-
netic bases of adaptation. Nature Reviews Genetics 4, 457–469 (2003)
18. Good, B.H., McDonald, M.J., Barrick, J.E., Lenski, R.E., Desai, M.M.: The dynamics of
molecular evolution over 60,000 generations. Nature 551, 45–50 (2017)
19. Groisman, E.A., Ochman, H.: Pathogenicity islands: Bacterial evolution in quantum leaps.
Cell 87, 791–794 (1996)
20. Guerzoni, D., McLysaght, A.: De novo origins of human genes. PLoS Genetics 7(11),
e1002381 (2011)
21. Hacker, J., Blum-Oehler, G., Muehldorfer, I., Tschaepe, H.: Pathogenicity islands of virulent
bacteria: Structure, function and impact on microbial evolution. Molecular Microbiology 23,
1089–1097 (1997)
22. Hall, B.G.: Chromosomal mutation for citrate utilization by Escherichia coli k-12. Journal of
Bacteriology 151, 269–273 (1982)
23. Jacob, F.: Evolution and tinkering. Science 196(4295), 1161–1166 (1977)
24. Jerison, E.R., Desai, M.M.: Genomic investigations of evolutionary dynamics and epistasis
in microbial evolution experiments. Current Opinion in Genetics & Development 35, 33–39
25. Jerison, E.R., Kryazhimskiy, S., Mitchell, J.K., Bloom, J.S., Kruglyak, L., Desai, M.M.: Ge-
netic variation in adaptability and pleiotropy in budding yeast. eLife 6, e27167 (2017)
26. Kawashima, T., Kawashima, S., Tanaka, C., Murai, M., Yoneda, M., Putnam, N.H., Rokhsar,
D.S., Kanehisa, M., Satoh, N., Wada, H.: Domain shuffling and the evolution of vertebrates.
Genome Research 19(8), 1393–1403 (2009)
27. Keseler, I.M., Collado-Vides, J., Gama-Castro, S., Ingraham, J., Paley, S., Paulsen, I.T.,
Peralta-Gil, M., Karp, P.D.: EcoCyc: A comprehensive database resource for Escherichia coli.
Nucleic Acids Research 33(suppl-1), D334–D337 (2005)
28. Keseler, I.M., Mackie, A., Santos-Zavaleta, A., Billington, R., Bonavides-Martnez, C., Caspi,
R., et al.: The EcoCyc database: Reflecting new knowledge about Escherichia coli K-12. Nu-
cleic Acids Research 45, D543–D550 (2016)
29. Khan, A.I., Dinh, D.M., Schneider, D., Lenski, R.E., Cooper, T.F.: Negative epistasis between
beneficial mutations in an evolving bacterial population. Science 332, 1193–1196 (2011)
30. Kryazhimskiy, S., Rice, D.P., Jerison, E.R., Desai, M.M.: Global epistasis makes adaptation
predictable despite sequence-level stochasticity. Science 344, 1519–1522 (2014)
31. Lara, F.J.S., Stokes, J.L.: Oxidation of citrate by Escherichia coli. Journal of Bacteriology
63(3), 415–420 (1952)
32. Lawrence, J.G., Ochman, H.: Reconciling the many faces of lateral gene transfer. Trends in
Microbiology 10, 1–4 (2002)
33. Lee, S.H., Yang, J., Goddard, M.E., Visscher, P.M., Wray, N.R.: Estimation of pleiotropy be-
tween complex diseases using single-nucleotide polymorphism-derived genomic relationships
and restricted maximum likelihood. Bioinformatics 28, 2540–2542 (2012)
34. Lenski, R.E.: Experimental studies of pleiotropy and epistasis in Escherichia coli. i. variation
in competitive fitness among mutants resistant to virus T4. Evolution 42, 425–432 (1988)
35. Lenski, R.E., Rose, M.R., Simpson, S.C., Tadler, S.C.: Long-term experimental evolution in
Escherichia coli. I. adaptation and divergence during 2,000 generations. American Naturalist
138(6), 1315–1341 (1991)
36. Leon, D., D’Alton, S., Quandt, E.M., Barrick, J.E.: Innovation in an E. coli evolution ex-
periment is contingent on maintaining adaptive potential until competition subsides. PLoS
Genetics 14, 1–22 (2018)
37. Locey, K.J., Lennon, J.T.: Scaling laws predict global microbial diversity. Proceedings of the
National Academy of Sciences of the United States of America 113(21), 5970–5975 (2016)
38. Luetgens, M., Gottschalk, G.: Why a co-substrate is required for anaerobic growth of Es-
cherichia coli on citrate. Journal of General Microbiology 119, 63–70 (1980)
39. Meyer, J.R., Dobias, D.T., Medina, S.J., Servilio, L., Gupta, A., Lenski, R.E.: Ecological spe-
ciation of bacteriophage lambda in allopatry and sympatry. Science 354, 1301–1304 (2016)
10 Probing the Deep Genetic Basis of a Novel Trait in Escherichia coli 121
40. Meyer, J.R., Dobias, D.T., Weitz, J.S., Barrick, J.E., Quick, R.T., Lenski, R.E.: Repeatability
and contingency in the evolution of a key innovation in phage lambda. Science 335, 428–432
41. Ohno, S.: Evolution by gene duplication. Springer-Verlag (1970)
42. Pos, K.M., Dimroth, P., Bott, M.: The Escherichia coli citrate carrier CitT: A member of a
novel eubacterial transporter family related to the 2-oxoglutarate/malate translocator from
spinach chloroplasts. Journal of Bacteriology 180, 4160–4165 (1998)
43. Qian, W., Zhang, J.: Genomic evidence for adaptation by gene duplication. Genome Research
24, 1356–1362 (2014)
44. Quandt, E.M., Deatherage, D.E., Ellington, A.D., Georgiou, G., Barrick, J.E.: Recursive
genomewide recombination and sequencing reveals a key refinement step in the evolution of a
metabolic innovation in Escherichia coli. Proceedings of the National Academy of Sciences
of the United States of America 111, 2217–2222 (2014)
45. Quandt, E.M., Gollihar, J., Blount, Z.D., Ellington, A.D., Georgiou, G., Barrick, J.E.: Fine-
tuning citrate synthase flux potentiates and refines metabolic innovation in the Lenski evolu-
tion experiment. eLife 4, e09696 (2015)
46. Remold, S.K., Lenski, R.E.: Pervasive joint influence of epistasis and plasticity on mutational
effects in Escherichia coli. Nature Genetics 36, 423–426 (2004)
47. Reynolds, C.H., Silver, S.: Citrate utilization by Escherichia coli: Plasmid- and chromosome-
encoded systems. Journal of Bacteriology 156, 1019–1024 (1983)
48. Ruprecht, J., Yankovskaya, V., Maklashina, E., Iwata, S., Cecchini, G.: Structure of Es-
cherichia coli succinate: Quinone oxidoreductase with an occupied and empty quinone-
binding site. Journal of Biological Chemistry 284, 29836–29846 (2009)
49. Schloss, P.D., Handelsman, J.: Status of the microbial census. Microbiol Mol Biol Rev 68(4),
686–691 (2004)
50. Schluter, D.: Evidence for ecological speciation and its alternative. Science 323(5915), 737–
741 (2009)
51. Segr`
e, D., DeLuna, A., Church, G.M., Kishony, R.: Modular epistasis in yeast metabolism.
Nature Genetics 37, 77–83 (2005)
52. Suding, K.N., Goldberg, D.E., Hartman, K.M.: Relationships among species traits: Separating
levels of response and identifying linkages to abundance. Ecology 84, 1–16 (2003)
53. Van Hofwegen, D.J., Hovde, C.J., Minnich, S.A.: Rapid evolution of citrate utilization by
Escherichia coli by direct selection requires citT and dctA. Journal of Bacteriology 198,
1022–1034 (2016)
54. Vasi, F., Travisano, M., Lenski, R.E.: Long-term experimental evolution in Escherichia coli.
ii. changes in life-history traits during adaptation to a seasonal environment. The American
Naturalist 144, 432–456 (1994)
55. von Wintersdorff, C.J.H., Penders, J., van Niekerk, J.M., Mills, N.D., Majumder, S., van
Alphen, L.B., Savelkoul, P.H.M., Wolffs, P.F.G.: Dissemination of antimicrobial resistance
in microbial ecosystems through horizontal gene transfer. Front Microbiol 7, 173 (2016)
56. Wang, Z., Liao, B.Y., Zhang, J.: Genomic patterns of pleiotropy and the evolution of complex-
ity. Proceedings of the National Academy of Sciences of the United States of America 107,
18034–18039 (2010)
122 Tanush Jagdish, J. Jeffrey Morris, Brian D. Wade and Zachary D. Blount
Genes identified as putatively necessary for aerobic growth on citrate and their annotations as
reported by EcoCyc.
ResearchGate has not been able to resolve any citations for this publication.
Full-text available
Key innovations are disruptive evolutionary events that enable a species to escape constraints and rapidly diversify. After 15 years of the Lenski long-term evolution experiment with Escherichia coli, cells in one of the twelve populations evolved the ability to utilize citrate, an abundant but previously untapped carbon source in the environment. Descendants of these cells became dominant in the population and subsequently diversified as a consequence of invading this vacant niche. Mutations responsible for the appearance of rudimentary citrate utilization and for refining this ability have been characterized. However, the complete nature of the genetic and/or ecological events that set the stage for this key innovation is unknown. In particular, it is unclear why it took so long for citrate utilization to evolve and why it still has evolved in only one of the twelve E. coli populations after 30 years of the Lenski experiment. In this study, we recapitulated the initial mutation needed to evolve citrate utilization in strains isolated from throughout the first 31,500 generations of the history of this population. We found that there was already a slight fitness benefit for this mutation in the original ancestor of the evolution experiment and in other early isolates. However, evolution of citrate utilization was blocked at this point due to competition with other mutations that improved fitness in the original niche. Subsequently, an anti-potentiated genetic background evolved in which it was deleterious to evolve rudimentary citrate utilization. Only later, after further mutations accumulated that restored the benefit of this first-step mutation and the overall rate of adaptation in the population slowed, was citrate utilization likely to evolve. Thus, intense competition and the types of mutations that it favors can lead to short-sighted evolutionary trajectories that hide a stepping stone needed to access a key innovation from many future generations.
Full-text available
Epistasis or the non-additivity of mutational effects is a major force in protein evolution, but it has not been systematically quantified at the level of a proteome. Here, we estimated the extent of epistasis for 2,382 genes in E. coli using several hundreds of orthologs for each gene within the class Gammaproteobacteria. We found that the average epistasis is ~41% across genes in the proteome and that epistasis is stronger among highly expressed genes. This trend is quantitatively explained by the prevailing model of sequence evolution based on minimizing the fitness cost of protein unfolding and aggregation. The genes with the highest epistasis are also functionally involved in the maintenance of proteostasis, translation and central metabolism. In contrast, genes evolving with low epistasis mainly encode for membrane proteins and are involved in transport activity. Our results highlight the coupling between selection and epistasis in the long-term evolution of a proteome.
Full-text available
The outcomes of evolution are determined by a stochastic dynamical process that governs how mutations arise and spread through a population. However, it is difficult to observe these dynamics directly over long periods and across entire genomes. Here we analyse the dynamics of molecular evolution in twelve experimental populations of Escherichia coli, using whole-genome metagenomic sequencing at five hundred-generation intervals through sixty thousand generations. Although the rate of fitness gain declines over time, molecular evolution is characterized by signatures of rapid adaptation throughout the duration of the experiment, with multiple beneficial variants simultaneously competing for dominance in each population. Interactions between ecological and evolutionary processes play an important role, as long-term quasi-stable coexistence arises spontaneously in most populations, and evolution continues within each clade. We also present evidence that the targets of natural selection change over time, as epistasis and historical contingency alter the strength of selection on different genes. Together, these results show that long-term adaptation to a constant environment can be a more complex and dynamic process than is often assumed.
Full-text available
Evolution can favor organisms that are more adaptable, provided that genetic variation in adaptability exists. Here, we quantify this variation among 230 offspring of a cross between diverged yeast strains. We measure the adaptability of each offspring genotype, defined as its average rate of adaptation in a specific environmental condition, and analyze the heritability, predictability, and genetic basis of this trait. We find that initial genotype strongly affects adaptability and can alter the genetic basis of future evolution. Initial genotype also affects the pleiotropic consequences of adaptation for fitness in a different environment. This genetic variation in adaptability and pleiotropy is largely determined by initial fitness, according to a rule of declining adaptability with increasing initial fitness, but several individual QTLs also have a significant idiosyncratic role. Our results demonstrate that both adaptability and pleiotropy are complex traits, with extensive heritable differences arising from naturally occurring variation.
Full-text available
EcoCyc ( is a freely accessible, comprehensive database that collects and summarizes experimental data for Escherichia coli K-12, the best-studied bacterial model organism. New experimental discoveries about gene products, their function and regulation, new metabolic pathways, enzymes and cofactors are regularly added to EcoCyc. New SmartTable tools allow users to browse collections of related EcoCyc content. SmartTables can also serve as repositories for user- or curator-generated lists. EcoCyc now supports running and modifying E. coli metabolic models directly on the EcoCyc website.
Historical processes display some degree of “contingency,” meaning their outcomes are sensitive to seemingly inconsequential events that can fundamentally change the future. Contingency is what makes historical outcomes unpredictable. Unlike many other natural phenomena, evolution is a historical process. Evolutionary change is often driven by the deterministic force of natural selection, but natural selection works upon variation that arises unpredictably through time by random mutation, and even beneficial mutations can be lost by chance through genetic drift. Moreover, evolution has taken place within a planetary environment with a particular history of its own. This tension between determinism and contingency makes evolutionary biology a kind of hybrid between science and history. While philosophers of science examine the nuances of contingency, biologists have performed many empirical studies of evolutionary repeatability and contingency. Here, we review the experimental and comparative evidence from these studies. Replicate populations in evolutionary “replay” experiments often show parallel changes, especially in overall performance, although idiosyncratic outcomes show that the particulars of a lineage’s history can affect which of several evolutionary paths is taken. Comparative biologists have found many notable examples of convergent adaptation to similar conditions, but quantification of how frequently such convergence occurs is difficult. On balance, the evidence indicates that evolution tends to be surprisingly repeatable among closely related lineages, but disparate outcomes become more likely as the footprint of history grows deeper. Ongoing research on the structure of adaptive landscapes is providing additional insight into the interplay of fate and chance in the evolutionary process.
A fitness landscape is a map between the genotype and its reproductive success in a given environment. The topography of fitness landscapes largely governs adaptive dynamics, constraining evolutionary trajectories and the predictability of evolution. Theory suggests that this topography can be deformed by mutations that produce substantial changes to the environment. Despite its importance, the deformability of fitness landscapes has not been systematically studied beyond abstract models, and little is known about its reach and consequences in empirical systems. Here we have systematically characterized the deformability of the genome-wide metabolic fitness landscape of the bacterium Escherichia coli. Deformability is quantified by the noncommutativity of epistatic interactions, which we experimentally demonstrate in mutant strains on the path to an evolutionary innovation. Our analysis shows that the deformation of fitness landscapes by metabolic mutations rarely affects evolutionary trajectories in the short range. However, mutations with large environmental effects produce long-range landscape deformations in distant regions of the genotype space that affect the fitness of later descendants. Our results therefore suggest that, even in situations in which mutations have strong environmental effects, fitness landscapes may retain their power to forecast evolution over small mutational distances despite the potential attenuation of that power over longer evolutionary trajectories. Our methods and results provide an avenue for integrating adaptive and eco-evolutionary dynamics with complex genetics and genomics.
Understanding the conditions that allow speciation to occur is difficult because most research has focused on either long-lived organisms or asexual microorganisms. We propagated bacteriophage λ, a virus with rapid generations and frequent recombination, on two Escherichia coli host genotypes that expressed either the LamB or OmpF receptor. When supplied with either single host (allopatry), λ improved its binding to the available receptor while losing its ability to use the alternative. When evolving on both hosts together (sympatry), the viruses split into two lineages with divergent receptor preferences. Although the level of divergence varied among replicates, some lineages evolved reproductive isolation via genetic incompatibilities. This outcome indicates that, under suitable conditions, allopatric and sympatric speciation can occur with similar ease.