1
Evolutionary rescue can be impeded by temporary environmental amelioration
2
3
Supporting Information
4
5
Methods
6
We examined whether progressive adaptation via accumulation of multiple beneficial
7
mutations was needed for the persistence of our phage populations in the increasingly
8
deteriorating environments.
9
stressful temperature cannot grow at even higher temperature, it is suggested that
If phage populations that evolved tolerance to a mildly
10
multiple beneficial mutations would be required for further adaptation.
Our ancestral
11
phage can grow poorly at 29.8 ºC, and not above 30 ºC (Zhang & Buckling 2011), thus
12
tolerance of temperature > 30 ºC would require at least one beneficial mutation.
13
our experiment, 45 out of 48 large-bottleneck phage populations survived at transfer 10
14
(30.2 ºC). Frozen stocks of these phage populations were reconditioned for 24 h at
15
30.2 ºC (grown with ~107 ancestral host bacteria cells); 1% of the phage extract was
16
transferred to new microcosms with ~107 ancestral host bacteria cells, and grown for
17
48 h at 31 ºC. None of these phage lines showed positive growth at 31 ºC.
18
suggests that adaptation to temperature above 31 ºC need accumulation of at least two
19
beneficial mutations.
In
This
20
We investigated whether adaptation to high temperature incurred fitness costs in
21
terms of growth performance at lower temperature in our experimental system.
22
Population growth profile at 29 ºC was assayed for the ancestral phage and eleven
23
populations from transfer 20 (the end of the experiment), of which two were from the
24
temperature-constant environment, three from the ‘amelioration’ treatment and six from
25
the ‘no amelioration’ treatment.
26
large-bottleneck lines.
27
at 29 ºC; 1% of the phage extract was transferred to new microcosms with ~10 7
28
stationary-phase ancestral bacteria cells, and grown at 29 ºC. For each population, 30
29
replicate microcosms were initiated; three microcosms were randomly chosen at each
30
of the following 10 points in time (0, 0.5, 1, 1.5, 2, 2.5, 3, 4, 8, and 48h), of which phage
All of the eleven evolved populations were from
Frozen stocks of these populations were first cultured for 24 h
1
31
densities were measured.
Mean density of the three replicates were used for assessing
32
the population growth profiles.
33
three-phase linear model for microbial growth (Buchanan et al. 1997; Novak et al.
34
2006).
35
population density, log phase with exponentially increasing density, and stationary
36
phase with constant density. Three parameters were obtained from this model: the
37
length of lag phase, the maximum growth rate, and final yield.
38
the model was very high: in all cases, the coefficient of determination was > 0.99.
39
Note that this was an assay of population-level growth performance, but not an assay
40
of ‘one-step’ growth curve. We examined phage populations, rather than individual
41
phage isolates.
42
population.
43
cells (consistent with our selection experiment), which may contain both active and less
44
active cells and thus cause further asynchrony in growth within each phage population.
The log-transformed measurements were used to fit a
The model assumes three distinct growth phases: lag phase with constant
The goodness of fit of
There may be heterogeneity in growth performance within each
Stationery-phase, but not exponential-phase, bacteria were used as host
45
46
2
47
Figure S1 A graphical illustration of the hypothesis that temporary environmental
48
amelioration during deterioration can slow the fixation of beneficial mutations.
49
asexual populations, as assumed here, favourable mutations need be established
50
sequentially and thus later steps of adaptation are contingent on the success of earlier
51
ones.
52
beneficial mutations may amplify through time.
53
scenario of relaxed, but not reversed, selection for beneficial mutations during the
54
amelioration phases.
In
Therefore the negative effect of environmental amelioration on the fixation of
55
56
3
Note that panel (d) assumes a
57
Figure S2 Population density of phage lines that evolved in the temperature-constant
58
environment (29 °C).
59
environment; and within each incubator, twelve populations (six with large bottleneck
60
population size and six with small bottleneck population size) were maintained.
61
Different symbols indicate phage lines from different individual incubators.
Two replicate incubators were used for this evolution
62
63
64
4
65
Figure S3 Growth profile at 29 ºC of the ancestral and eleven evolved populations,
66
either from the temperature-constant (control) or deteriorating environments.
67
line indicates an individual phage population.
68
69
5
Each
70
Figure S4 Phage population growth characteristics, length of lag phase (a), maximum
71
growth rate (b), and final yield (c), of the ancestral phage and eleven evolved phage
72
lines.
73
environments and those from the temperature-constant (control) environment in the
74
growth characteristics (Mann-Whitney U-test, P > 0.1). This suggests tolerance to
75
high temperature does not incur a cost in terms of growth performance at low
76
temperature, and thus in our evolutionary rescue experiment selection for beneficial
77
mutations was relaxed but not reversed during the amelioration periods.
78
absence of such fitness costs in the finally rescued populations of the ‘amelioration’
79
environment may result from either (i) that the high temperature-tolerant mutations did
80
not suffer a fitness costs, or (ii) that mutations with fitness costs existed but populations
81
with such mutations went extinct due to the antagonistic selection arising from the
82
environmental fluctuation.
83
amelioration’ evolution lines.
There is no difference between the rescued lines from the deteriorating
Note that the
But only the first explanation is relevant to the ‘no
84
85
86
87
6
88
REFERENCES
89
90
1.
91
Buchanan R.L., Whiting R.C. & Damert W.C. (1997). When is simple good enough: a
92
comparison of the Gompertz, Baranyi, and three-phase linear models for fitting
93
bacterial growth curves. Food Microbiol., 14, 313-326.
94
2.
95
Novak M., Pfeiffer T., Lenski R.E., Sauer U. & Bonhoeffer S. (2006). Experimental
96
tests for an evolutionary trade-off between growth rate and tield in E. coli. Am.
97
Nat., 168, 242-251.
98
3.
99
Zhang Q.-G. & Buckling A. (2011). Antagonistic coevolution limits population
100
persistence of a virus in a thermally deteriorating environment. Ecol. Lett., 14,
101
282-288.
102
103
7