Distributing a metabolic pathway among a microbial
consortium enhances production of natural products
The MIT Faculty has made this article openly available. Please share
how this access benefits you. Your story matters.
Citation
Zhou, Kang, Kangjian Qiao, Steven Edgar, and Gregory
Stephanopoulos. “Distributing a Metabolic Pathway Among a
Microbial Consortium Enhances Production of Natural Products.”
Nature Biotechnology 33, no. 4 (January 5, 2015): 377–383.
As Published
http://dx.doi.org/10.1038/nbt.3095
Publisher
Nature Publishing Group
Version
Author's final manuscript
Accessed
Thu May 26 00:57:33 EDT 2016
Citable Link
http://hdl.handle.net/1721.1/102311
Terms of Use
Creative Commons Attribution-Noncommercial-Share Alike
Detailed Terms
http://creativecommons.org/licenses/by-nc-sa/4.0/
1
Editor’s summary
2
Stable co-culture of yeast and Escherichia coli enables the synthesis of an acetylated diol from a
3
taxadiene precursor
4
6
Distributing a metabolic pathway among a microbial consortium
enhances production of natural products
7
Kang Zhou, Kangjian Qiao, Steven Edgar, Gregory Stephanopoulos *
8
Department of Chemical Engineering, Massachusetts Institute of Technology,
9
Cambridge, MA, USA
5
10
11
Article type: Article
12
* To whom correspondence should be addressed. E-mail: gregstep@mit.edu
13
1
14
Abstract
15
Metabolic engineering of microorganisms such as Escherichia coli and Saccharomyces
16
cerevisiae to produce high-value natural metabolites is often done through functional
17
reconstitution of long metabolic pathways. Problems arise when parts of pathways require
18
specialized environments or compartments for optimal function. Here we solve this problem
19
through co-culture of multiple engineered organisms, each of which contains the part of the
20
pathway that it is best suited to hosting. In one example, we divided the synthetic pathway for
21
paclitaxel precursors into two modules, expressed in either S. cerevisiae or E. coli, and cultured
22
the microorganisms in the same bioreactor. Stable co-culture was achieved by designing a
23
mutualistic relationship between the two species in which a metabolic intermediate produced
24
by E. coli was used and functionalized by yeast. This synthetic consortium efficiently produced
25
oxygenated taxanes, tanshinone precursors and functionalized sesquiterpenes.
26
2
27
Introduction
28
Plants synthesize numerous structurally complex compounds that have important
29
therapeutic properties1-6, e.g. paclitaxel, a potent antitumor agent1. Heterologous production of
30
these molecules in industrial microbes—mainly bacteria and yeasts—could provide a robust and
31
sustainable production process. However, in bacteria it has been challenging to functionally
32
express sophisticated eukaryotic enzymes that are often required in the synthesis of complex
33
compounds7; on the other hand, it has been equally difficult to engineer yeasts for high-yield
34
production of building blocks of natural products, e.g. the isoprenoid biosynthetic pathway of
35
bacteria has higher theoretical yield than that of yeasts1.
36
In nature, microbes can form interacting communities to accomplish chemically difficult
37
tasks through division of labor among different species8. These natural microbial consortia have
38
been used in food and other industries for decades9. Furthermore, interactions of microbial
39
species in mixed microbial cultures were studied extensively in the 60s and 70s10, 11, aiming to
40
establish operating diagrams for maintaining synthetic co-culture, which has been challenging
41
due to difference in their doubling time and secretion of toxic metabolites11. Recently, a few
42
synthetic consortia comprising genetically engineered microbes have been reported for
43
production of biofuels and chemicals12-14. However, these prior studies were mostly concerned
44
with the stability of microbial consortia while the more recent work focused on utilizing non-
45
conventional biomass, e.g. cellulose12, 13. In these examples, which both involved two different
46
species, the first species only provided the carbon source for the second, which harbored the
47
essential pathway for the final product in its entirety and was able to make the final product on
48
its own. Strictly speaking, none of this prior work examined the potential to use more than one
3
49
species for the purpose of constructing a long synthetic pathway, which enables production of
50
structurally complex compounds.
51
In this study, we demonstrate the concept of reconstituting a heterologous metabolic
52
pathway in a microbial partnership in which one microbe is engineered to synthesize a
53
metabolic intermediate that is translocated to another microbe, in which it is further
54
functionalized. In principle, it could be attractive to use synthetic microbial consortia for
55
production of valuable metabolites, especially those with complex structures. One major
56
advantage of this design is that each expression system and pathway module can be constructed
57
and optimized in parallel, so that the time required would be significantly reduced. Other
58
advantages of using synthetic consortia include, (i) taking advantage of unique properties and
59
functions of different microbes, (ii) exploring beneficial interactions among consortium
60
members to enhance productivity, and, (iii) minimizing problems arising from feedback
61
inhibition through spatial pathway module segregation.
62
We report the use of two model laboratory and industrial microbes, E. coli and S.
63
cerevisiae in a consortium to produce precursors of the anti-cancer drug paclitaxel. E. coli is a
64
fast growing bacterium that can be engineered to overproduce taxadiene, the scaffold molecule
65
of paclitaxel1. S. cerevisiae, having advanced protein expression machinery and abundant
66
intracellular membranes, has been suggested as a preferable host for expressing cytochrome
67
P450s (CYPs), which functionalize taxadiene by catalyzing multiple oxygenation reactions15-17.
68
We find that integration of parts of the whole pathway in separate species cultured together
69
combines dual properties of rapid production of taxadiene in E. coli with efficient oxygenation of
70
taxadiene by S. cerevisiae. This novel approach has overcome the challenges of using E. coli
4
71
alone—perturbation of the fine-tuned taxadiene production by introducing CYPs and functional
72
expression of these enzymes in E. coli1.
73
RESULTS
74
75
Co-culture design to produce paclitaxel precursors
76
We first engineered S. cerevisiae BY4700 to express taxadiene 5α-hydroxylase and its
77
reductase (5αCYP-CPR, Supplementary Fig.1a), which catalyze the first oxygenation reaction in
78
the pathway of paclitaxel biosynthesis17. Taxadiene was efficiently oxygenated by this yeast
79
(named TaxS1) when taxadiene was externally fed into the culture medium (Supplementary
80
Fig.1b), confirming that the 5aCYP was functional in S. cerevisiae BY4700. Next, we co-cultured
81
this 5αCYP-CPR-expressing yeast with a taxadiene-producing E. coli (named TaxE1) in a fed-
82
batch bioreactor with glucose as the sole carbon and energy source (Figure 1a). The mixed
83
culture produced 2 mg/L of oxygenated taxanes in 72 h (Figure 1b, identification of the
84
oxygenated taxanes is described in online methods, quantification of isoprenoids), whereas in
85
control experiments in which only E. coli TaxE1 (Figure 1b) or S. cerevisiae TaxS1 (data not
86
shown) was cultured, no oxygenated taxanes were produced. These results showed that
87
taxadiene produced by E. coli can diffuse into S. cerevisiae and be subsequently oxygenated.
88
However, the cell density of E. coli (Figure 1c) and the total titer of taxanes (Figure 1d) were
89
significantly reduced in the presence of S. cerevisiae. The cause could be inhibition of E. coli by
90
accumulated ethanol produced by yeast when grown on glucose (Figure 1e). This hypothesis
91
was validated by the fact that ethanol, at the highest concentration observed (50 g/L, Figure 1e),
92
completely inhibited E. coli cell growth and taxadiene production (Supplementary Fig. 2). Similar
5
93
instances of inhibition have been observed before in natural systems when microbes compete
94
for common resources18.
95
To overcome this problem we designed a mutualistic interaction between the two
96
microorganisms 18. When E. coli metabolizes xylose it excretes acetate, which is inhibitory to its
97
own growth19. S. cerevisiae, on the other hand, cannot metabolize xylose but can use acetate as
98
the sole carbon source for growth without producing ethanol (Figure 2a, Supplementary Table
99
1). We therefore switched the co-culture carbon source from glucose to xylose. Under these
100
conditions S. cerevisiae only grew in the xylose medium in the presence of E. coli (Figure 2b),
101
and the concentration of extracellular acetate in the co-culture was significantly reduced
102
compared with that observed when E. coli was grown on xylose on its own (Figure 2c). More
103
importantly, this stable co-culture minimised the ethanol concentration to below the limit of
104
detection (0.1 g/L) throughout the experiment. In addition, the titer of total taxanes produced
105
by E. coli was not substantially affected by the presence of S. cerevisiae (Figure 2d), suggesting
106
that ethanol inhibition of E. coli was successfully eliminated and taxadiene production
107
proceeded unabated by the presence of yeast. However, although more oxygenated taxanes
108
were produced in this co-culture (4 mg/L in 72 h, Figure 2e) compared with the previous co-
109
culture (2 mg/L in 72 h, Figure 1b), the taxadiene oxygenation efficiency was still low (only 8% of
110
total taxadiene produced, Figure 2).
111
112
Optimization to Improve taxadiene oxygenation
113
To increase taxadiene oxygenation, we first focused on optimizing the growth of S.
114
cerevisiae, using the rationale that more yeast cells would express more 5αCYP and therefore
115
functionalize more taxadiene. We noted that acetate accumulated in the co-culture during the
6
116
first 24 h (Figure 2c), indicating that the initial yeast population was insufficient to convert all
117
available substrate in the medium. This was corrected by increasing the initial inoculum of yeast
118
and also periodically feeding additional carbon (xylose), nitrogen (ammonium) and phosphorous
119
(phosphate) sources to ensure that these major nutrients were not limiting yeast growth. After
120
these modifications, no acetate was detected throughout the entire fermentation and the
121
oxygenated taxane titer was improved ~3 fold (16 mg/L in 90 h, Figure 3a). Under these
122
conditions, as growth of S. cerevisiae was strictly limited by the amount of acetate secreted by E.
123
coli, further increase of the relative amount of yeast in the culture relied on engineering the
124
acetate pathway in E. coli (see below). We opted not to feed exogenous acetate in order to
125
preserve the autonomous nature of the co-culture (Supplementary Fig. 3).
126
We next improved the specific oxygenation activity of yeast TaxS1. 5αCYP-CPR (fused as
127
a single polypeptide, Supplementary Fig. 1a) was previously expressed under a strong
128
constitutive promoter (TEFp). We replaced TEFp by GPDp (a widely used strong promoter20),
129
UAS-GPDp (an enhanced version of GPDp21) and ACSp (a promoter from the acetate assimilation
130
pathway22, 23 that we hypothesized to be strong in our study since yeast TaxS1 grew on acetate
131
here) and tested taxadiene oxygenation efficiency by the corresponding strains. To this end,
132
yeast strains (TaxS1, TaxS2, TaxS3 and TaxS4) were cultured without E. coli and the oxygenation
133
rate of exogenously supplied taxadiene was measured (Figure 3b). Based on the results of this
134
assay, UAS-GPDp was selected as strongest promoter. Yeast strain TaxS4 was then co-cultured
135
using xylose as a substrate with E. coli TaxE1, and this co-culture produced significantly higher
136
concentrations of oxygenated taxanes (25 mg/L in 90 h) compared with a co-culture in which the
137
TEFp promoter was used (16 mg/L in 90 h, Figure 3c). GPDp and ACSp were also tested in co-
138
culture (Figure 3c), and the results were fairly consistent with those of the feeding experiments
7
139
(Figure 3b), e.g. ACSp, the promoter characterized to be weaker than TEFp in the feeding
140
experiment, also led to lower production of oxygenate taxanes compared with TEFp in co-cuture
141
(Figure 3c).
142
After increasing oxygenation efficiency in yeast, we engineered E. coli to overproduce
143
acetate and thereby further potentially improve the growth rate of S. cerevisiae by increasing
144
the concentration of available substrate. Production of acetate by E. coli is auto-regulated: when
145
acetate accumulates, E. coli growth is inhibited, resulting in lower acetate production rate. First,
146
we overexpressed the genes in the E. coli acetate production pathway (phosphate
147
acetyltransferase, pta, and acetate kinase, ackA), but this neither increased the S. cerevisiae
148
population density nor the oxygenation efficiency substantially (Supplementary Fig. 4). To
149
overcome this problem, we inactivated oxidative phosphorylation by knocking out atpFH24,
150
which is the primary means of ATP production under aerobic conditions. The rationale for this
151
modification was that the atpFH knock-out would force E. coli to produce more acetate, because
152
acetate production would, under these conditions, become the primary ATP generation
153
pathway (Figure 4a). Indeed, this E. coli mutant (named TaxE4) produced up to 5.0±0.1 g/L
154
acetate in test tube, while the parental strain (E. coli TaxE1) only produced 2.3±0.2 g/L acetate.
155
The relative S. cerevisiae population was also much larger when yeast TaxS4 was co-cultured
156
with E. coli TaxE4 compared to that with E. coli TaxE1 (Figure 4b). More importantly, the titer of
157
the oxygenated taxanes was further improved (33 mg/L in 120 h), and the percentage of the
158
taxadiene oxygenated was significantly increased (up to 75%, Figure 4c). Another strategy that
159
could be tested in future to further improve acetate production is the knockout of E. coli ACS,
160
which assimilates extracellular acetate under certain conditions. Such a knockout might make
161
more of the produced acetate available to the yeast strain.
8
162
163
Co-culture to produce monoacetylated dioxygenated taxane
164
We further engineered the co-culture to produce more complex paclitaxel precursors. A
165
prevailing theory of paclitaxel early-synthesis suggests taxadien-5α-ol is acetylated at its C-5α
166
position, followed by oxygenation at the C-10β position15 (Figure 5a). Because of the modular
167
nature of our microbial consortium, the ability to functionalize taxadien-5α-ol could be achieved
168
by engineering of only the yeast module. Taxadien-5α-ol acetyl-transferase (TAT25) and taxane
169
10β-hydroxylase (10βCYP26, fused with a CYP reductase1) were co-expressed in yeast TaxS4.
170
When the resulting yeast (named as TaxS6) was co-cultured with E. coli TaxE4, the co-culture
171
produced a monoacetylated dioxygenated taxane (molecular weight 346), which was identified
172
as a single peak on the extracted ion chromatography (346 m/z, GCMS) and was absent from the
173
control co-culture not expressing the TAT and 10βCYP (Figure 5b). A 13C labelling experiment
174
confirmed that the oxygenated diol was indeed derived from taxadiene (Supplementary Fig. 5).
175
The identified compound could be taxadien-5α-acetate-10β-ol, an important intermediate in the
176
paclitaxel synthesis15, because its spectrum contained many of its fragment ions (346, 303, 286,
177
271 and 243 m/z27, Supplementary Fig. 5). To improve the titer and yield of this compound, we
178
used a stronger promoter for expressing TAT (strain TaxS7), and the change of promoter
179
improved the titer from 0.6 mg/L to 1 mg/L (Figure 5c), confirming the hypothesis that this step
180
was limiting. We then operated the bioreactor under a xylose limited condition, which further
181
increased the titer and also substantially improved the yield, by reducing the xylose
182
consumption (from ~120 g/L to 80 g/L, Figure 5c, Supplementary Fig. 6). This is the first report
183
of producing a monoacetylated dioxygenated taxane from a simple substrate (xylose) in
9
184
microbes, and it reveals the usefulness of the modularity of a microbial partnership for synthesis
185
of complex metabolites.
186
187
188
Production of other oxygenated isoprenoids by co-culture
189
The E. coli-S. cerevisiae co-culture developed in this study could be used for production
190
of any compound if one of the pathway precursors can cross cell membranes. The method
191
should be applicable to most isoprenoids, the largest class of natural products, because their
192
scaffold molecules are generally membrane-permeable. To test this hypothesis, we examined
193
the synthesis of another diterpene, ferruginol, the precursor of tanshinone, which is in clinical
194
trials for treating heart disease28, 29. We replaced taxadiene synthase in E. coli TaxE4 with two
195
enzymes (KSL and CPS, resulting in strain TaxE7) that are required for synthesizing miltiradiene29,
196
a membrane-crossing molecule. At the same time, in S. cerevisiae BY4700 we overexpressed a
197
specific CYP and its reductase (SmCYP and SmCPR, resulting in strain TaxS8), which were
198
reported to oxygenate miltiradiene into ferruginol28 (Figure 6a). When E. coli TaxE5 and yeast
199
TaxS8 were co-cultured in the xylose medium, the co-culture successfully produced 18 mg/L
200
ferruginol (Figure 6b), which exceeds the highest titer reported in the literature (10 mg/L by S.
201
cerevisiae28). This shows that the co-culture concept is generally applicable to diterpenes, and
202
demonstrates the advantages of co-culture over mono-culture, that is, being able to construct
203
parts of the pathway in parallel and achieve higher titers owing to microbial cooperation.
204
Finally, we synthesized a sesquiterpene—nootkatone, which is a high-end fragrance
205
molecule30. Similarly, we replaced the taxadiene synthase and geranylgeranyl diphosphate
10
206
synthase in E. coli TaxE4 with a sesquiterpene synthase (VALC, resulting in strain TaxE8) to
207
produce valencene, and in yeast BY4700 we overexpressed a specific CYP and its reductase
208
(HmCYP and AtCPR, resulting in strain TaxS9) that can oxygenate valencene30 (Figure 6a). When
209
these strains (TaxE8 and TaxS9) were co-cultured, they produced 30 mg/L nootkatol and a small
210
quantity of nootkatone (0.8 mg/L, Figure 6c). Recently, a Pichia alcohol dehydrogenase
211
(PpADH3C) was shown to oxidize nootkatol in its native host30. We introduced this enzyme to
212
yeast TaxS9, yielding strain TaxS10 which upon co-culture with E. coli TaxE8, increased the
213
nootkatone titer by a factor of 5 (4 mg/L, Figure 6c). Again, these results supported the
214
hypothesis that the co-culture concept should be widely applicable to production of oxygenated
215
isoprenoids.
216
217
DISCUSSION
218
Our major motivation for using a stable co-culture is the introduction of modularity to
219
the design of pathways for microbial metabolite production by assigning a different part of the
220
metabolic pathway to each member of a partnership or synthetic consortium. In such an
221
experimental set-up pathway modules can be separately optimized and assembled to enable
222
optimal functioning of the complete pathway. The examples in this report demonstrate this
223
modularity: the screening of a better promoter for CYP expression in yeast could be carried out
224
independent of E. coli (Figure 3), and producing the acetylated diol in the co-culture also only
225
required modification of one of its modules (Figure 5). Such modularity should significantly
226
expedite the reconstruction of long biosynthetic pathways in microorganisms as the
227
construction of the cells carrying the pathway modules can be carried out in parallel and the
228
number of genetic modifications per cell is substantially reduced. To achieve this modularity,
11
229
pathway modules in different cells should not directly interact with each other to minimize
230
possible regulation. For example, CYPs and their reductase involved in taxane oxygenation
231
generate reactive oxygen species31, 32, which inhibit two enzymes (ISPG and ISPH) in the
232
taxadiene biosynthetic pathway containing iron-sulfur clusters that are hyper-sensitive to ROS33.
233
Spatial segregation, in two different microbes, of the pathway of taxadiene production from its
234
oxygenation pathway prevents inactivation of ISPG/ISPH by ROS generated by CYPs.
235
Because of modularity of a co-culture approach, we were able to exploit advantages of
236
the different species. Before this study, taxadiene could only be overproduced in E. coli1 while
237
most biochemical characterizations of the taxadiene-functionalizing enzymes were carried out in
238
S. cerevisiae16, 26, 34. By using E. coli to synthesize taxadiene and S. cerevisiae to functionalize it,
239
we combined the advantages of the two species for taxane production (fast growth of E. coli and
240
complete protein expression system of S. cerevisiae). Using co-culture, we were able to
241
synthesize a complex taxane (putative taxadiene-5α-acetate-10β-ol) (Figure 5) that has never
242
been produced by microorganisms growing on a simple carbon source in the past, and achieve
243
higher titers of isoprenoid production than has been reported previously (Figure 6b).
244
As most synthetic microbial consortia are competitive12, 13 (Supplementary Fig. 7), a
245
primary challenge in their design is to avoid the dominance of one species over another, due to
246
a shorter doubling time11,
247
species13. Conventionally, titration of the inoculum ratio13 and optimization of growth conditions
248
(such as pH and temperature11) can be exploited to maintain coexistence. However, these
249
strategies require time-consuming experimental trials or construction of sophisticated
250
mathematical models13, whose parameters also need to be estimated experimentally. In
251
addition, batch-to-batch variability can be high in these competitive co-cultures (data not
12
12
or production of substances that are inhibitory to the other
252
shown). In this study, we avoided these complications by building a mutualistic co-culture in
253
which S. cerevisiae used as its sole carbon source acetate, which was provided by and inhibitory
254
to E. coli, which in turn grew better in the presence of yeast compared to without the yeast
255
(Supplementary Fig. 8). We applied additional genetic and growth constraints to enforce this
256
cooperation, for instance, the respiration-deficient E. coli was forced to produce acetate as this
257
was its primary way to generate cellular ATP (Figure 4a), and the yeast also had to consume
258
acetate because it cannot utilize xylose (Figure 2a). Under such interdependency, the inoculum
259
ratio of our co-culture can be simply set to over-inoculation of yeast (the inoculum ratio of yeast
260
to E. coli was approximately 40:1, online methods, bioreactor experiments for the E. coli – S.
261
cerevisiae co-culture). This eliminated the inhibitory acetate levels, but did not result in yeast
262
overpopulation, because yeast growth was strictly limited by the concentration of acetate
263
produced by E. coli, leading to a balanced ratio of the two species (the ratio of yeast to E. coli
264
was 1:2 at 41 h, Figure 4b). Furthermore, this ratio was controllable through altering the specific
265
acetate productivity (Figure 4b). Because of this ability to alter the consortium composition by
266
increasing the relative yeast population, we managed to minimize accumulation of the pathway
267
intermediate (taxadiene) and increase the titer of oxygenated taxanes (Figure 4c and
268
Supplementary Fig. 9).
269
In addition to the mutualistic design, we also explored other strategies to avoid
270
microbial competition. The first was a two-stage culture, in which E. coli was cultured separately
271
for a few days before mixing with an active S. cerevisiae culture. This approach allowed both
272
microbes to grow at their preferred conditions and taxadiene to be efficiently oxygenated
273
(Supplementary Fig. 10). However, this process required a longer cultivation time (180 h) and,
274
additionally, it is more complicated than that of the mutualistic co-culture. We also explored a
13
275
two-carbon-source strategy, in which xylose can only be utilized by E. coli and ethanol (manually
276
added at low concentration, <2 g/L) was exclusively used by yeast (Supplementary Fig. 11). A
277
stable co-culture could be maintained under these conditions by controlling the ethanol
278
addition, and oxygenated taxanes were also produced at a relatively high titer (8 mg/L in 130 h,
279
Supplementary Fig. 12). However, both E. coli and S. cerevisiae produced acetate under this
280
scheme leading to microbial inhibition (Supplementary Fig. 12), which was eliminated in the
281
mutualistic design.
282
The co-culture concept is not restricted to the pairing E. coli with S. cerevisiae. We have
283
briefly explored the use of two different E. coli strains for production of oxygenated taxanes
284
(Supplementary Fig. 13), which worked, although the titer was low, mainly due to lack of the
285
mutualistic interactions present in the E. coli-S. cerevisiae co-culture. As a general guideline, a
286
target pathway should be divided into modules, each of which should be assigned to a specific
287
host strain so that the combined genetic traits of the consortium strains are favorable for
288
pathway completion. These microorganisms should rely on each other for supply of an essential
289
nutrient or detoxification of an inhibitory substance, ensuring a stable and controllable microbial
290
composition.
291
A necessary condition for co-culture is that the pathway intermediate (taxadiene) can
292
cross cell membranes and is secreted to the extracellular medium. This property was first
293
confirmed for taxadiene in prior studies where organic solvent mixed with E. coli cell culture was
294
found to efficiently extract taxadiene (C20) from the cells in a bioreactor1. We also measured
295
distribution of taxadiene in E. coli, medium and yeast in this study, which confirmed that
296
taxadiene can cross cell membranes efficiently even in absence of an organic solvent
297
(Supplementary Fig. 14). This physiochemical property is shared by many isoprenoids ranging
14
298
from C5 to C40, including isoprene35, limonene3, amorphadiene36 and canthaxanthin37. Hence,
299
the co-culture concept should be generally applicable to the production of most isoprenoids (in
300
this study, we have experimentally validated production of sesquiterpene and diterpene, Figure
301
6).
302
The experiments reported here provide evidence that a secondary metabolite pathway
303
can be reconstructed in a microbial consortium, paving the way for engineering the microbial
304
synthesis of natural compounds with complex structures that currently cannot be efficiently
305
synthesized in a single microbe such as alkaloids and flavonoids (including >10,000 molecules),
306
which all derived from aromatic amino acids that can be high-titer produced and excreted by E.
307
coli38 and functionalized by S. cerevisiae39. The co-culture can also benefit producing short chain
308
dicarboxylic acids (C6-C10), whose precursors are short chain fatty acids that can be easily
309
produced in engineered E. coli40, 41 and efficiently oxidized in the yeast expressing CYPs42.
310
311
312
313
314
315
15
316
References
317
318
319
320
321
322
323
324
325
326
327
328
329
330
331
332
333
334
335
336
337
338
339
340
341
342
343
344
345
346
347
348
349
350
351
352
353
354
355
356
357
358
359
360
1.
2.
3.
4.
5.
6.
7.
8.
9.
10.
11.
12.
13.
14.
15.
16.
17.
18.
Ajikumar, P.K. et al. Isoprenoid pathway optimization for Taxol precursor
overproduction in Escherichia coli. Science 330, 70-74 (2010).
Paddon, C.J. et al. High-level semi-synthetic production of the potent antimalarial
artemisinin. Nature 496, 528-532 (2013).
Alonso-Gutierrez, J. et al. Metabolic Engineering of Escherichia coli for Limonene
and Perillyl Alcohol Production. Metabolic engineering (2013).
Ajikumar, P.K. et al. Terpenoids: opportunities for biosynthesis of natural product
drugs using engineered microorganisms. Mol Pharm 5, 167-190 (2008).
Hefferon, K. Plant-derived pharmaceuticals for the developing world. Biotechnol J
8, 1193-1202 (2013).
Melnik, S. & Stoger, E. Green factories for biopharmaceuticals. Current medicinal
chemistry 20, 1038-1046 (2013).
Chang, M.C., Eachus, R.A., Trieu, W., Ro, D.K. & Keasling, J.D. Engineering
Escherichia coli for production of functionalized terpenoids using plant P450s.
Nat Chem Biol 3, 274-277 (2007).
Agapakis, C.M., Boyle, P.M. & Silver, P.A. Natural strategies for the spatial
optimization of metabolism in synthetic biology. Nat Chem Biol 8, 527-535 (2012).
Smid, E.J. & Lacroix, C. Microbe-microbe interactions in mixed culture food
fermentations. Current opinion in biotechnology 24, 148-154 (2013).
Fredrickson, A.G. & Stephanopoulos, G. Microbial competition. Science 213,
972-979 (1981).
Davison, B.H. & Stephanopoulos, G. Effect of pH oscillations on a competing
mixed culture. Biotechnology and bioengineering 28, 1127-1137 (1986).
Bayer, T.S. et al. Synthesis of methyl halides from biomass using engineered
microbes. Journal of the American Chemical Society 131, 6508-6515 (2009).
Minty, J.J. et al. Design and characterization of synthetic fungal-bacterial
consortia for direct production of isobutanol from cellulosic biomass. Proceedings
of the National Academy of Sciences of the United States of America 110,
14592-14597 (2013).
Xia, T., Eiteman, M.A. & Altman, E. Simultaneous utilization of glucose, xylose
and arabinose in the presence of acetate by a consortium of Escherichia coli
strains. Microbial cell factories 11, 77 (2012).
Guerra-Bubb, J., Croteau, R. & Williams, R.M. The early stages of taxol
biosynthesis: an interim report on the synthesis and identification of early
pathway metabolites. Natural product reports 29, 683-696 (2012).
Jennewein, S., Long, R.M., Williams, R.M. & Croteau, R. Cytochrome p450
taxadiene 5alpha-hydroxylase, a mechanistically unusual monooxygenase
catalyzing the first oxygenation step of taxol biosynthesis. Chemistry & biology 11,
379-387 (2004).
Hefner, J. et al. Cytochrome P450-catalyzed hydroxylation of taxa-4(5),11(12)diene to taxa-4(20),11(12)-dien-5alpha-ol: the first oxygenation step in taxol
biosynthesis. Chemistry & biology 3, 479-489 (1996).
Nowak, M.A. Five rules for the evolution of cooperation. Science 314, 1560-1563
(2006).
16
361
362
363
364
365
366
367
368
369
370
371
372
373
374
375
376
377
378
379
380
381
382
383
384
385
386
387
388
389
390
391
392
393
394
395
396
397
398
399
400
401
402
403
404
405
406
407
408
19.
20.
21.
22.
23.
24.
25.
26.
27.
28.
29.
30.
31.
32.
33.
Eiteman, M.A. & Altman, E. Overcoming acetate in Escherichia coli recombinant
protein fermentations. Trends in biotechnology 24, 530-536 (2006).
Sun, J. et al. Cloning and characterization of a panel of constitutive promoters for
applications in pathway engineering in Saccharomyces cerevisiae. Biotechnology
and bioengineering 109, 2082-2092 (2012).
Blazeck, J., Garg, R., Reed, B. & Alper, H.S. Controlling promoter strength and
regulation in Saccharomyces cerevisiae using synthetic hybrid promoters.
Biotechnology and bioengineering 109, 2884-2895 (2012).
De Virgilio, C. et al. Cloning and disruption of a gene required for growth on
acetate but not on ethanol: the acetyl-coenzyme A synthetase gene of
Saccharomyces cerevisiae. Yeast 8, 1043-1051 (1992).
Kratzer, S. & Schuller, H.J. Transcriptional control of the yeast acetyl-CoA
synthetase gene, ACS1, by the positive regulators CAT8 and ADR1 and the
pleiotropic repressor UME6. Molecular microbiology 26, 631-641 (1997).
Causey, T.B., Zhou, S., Shanmugam, K.T. & Ingram, L.O. Engineering the
metabolism of Escherichia coli W3110 for the conversion of sugar to redoxneutral and oxidized products: homoacetate production. Proceedings of the
National Academy of Sciences of the United States of America 100, 825-832
(2003).
Walker, K., Schoendorf, A. & Croteau, R. Molecular cloning of a taxa4(20),11(12)-dien-5alpha-ol-O-acetyl transferase cDNA from Taxus and
functional expression in Escherichia coli. Archives of biochemistry and biophysics
374, 371-380 (2000).
Schoendorf, A., Rithner, C.D., Williams, R.M. & Croteau, R.B. Molecular cloning
of a cytochrome P450 taxane 10 beta-hydroxylase cDNA from Taxus and
functional expression in yeast. Proceedings of the National Academy of Sciences
of the United States of America 98, 1501-1506 (2001).
Wheeler, A.L. et al. Taxol biosynthesis: differential transformations of taxadien-5
alpha-ol and its acetate ester by cytochrome P450 hydroxylases from Taxus
suspension cells. Archives of biochemistry and biophysics 390, 265-278 (2001).
Guo, J. et al. CYP76AH1 catalyzes turnover of miltiradiene in tanshinones
biosynthesis and enables heterologous production of ferruginol in yeasts.
Proceedings of the National Academy of Sciences of the United States of
America 110, 12108-12113 (2013).
Zhou, Y.J. et al. Modular pathway engineering of diterpenoid synthases and the
mevalonic acid pathway for miltiradiene production. Journal of the American
Chemical Society 134, 3234-3241 (2012).
Hom, E.F. & Murray, A.W. Niche engineering demonstrates a latent capacity for
fungal-algal mutualism. Science 345, 94-98 (2014).
Pillai, V.C., Snyder, R.O., Gumaste, U., Thekkumkara, T.J. & Mehvar, R. Effects
of transient overexpression or knockdown of cytochrome P450 reductase on
reactive oxygen species generation and hypoxia reoxygenation injury in liver
cells. Clin Exp Pharmacol Physiol 38, 846-853 (2011).
Reed, J.R., Cawley, G.F. & Backes, W.L. Inhibition of cytochrome P450 1A2mediated metabolism and production of reactive oxygen species by heme
oxygenase-1 in rat liver microsomes. Drug Metab Lett 5, 6-16 (2011).
Artsatbanov, V.Y. et al. Influence of Oxidative and Nitrosative Stress on
Accumulation of Diphosphate Intermediates of the Non-mevalonate Pathway of
17
409
410
411
412
413
414
415
416
417
418
419
420
421
422
423
424
425
426
427
428
429
430
431
432
433
434
435
436
437
438
439
34.
35.
36.
37.
38.
39.
40.
41.
42.
Isoprenoid Biosynthesis in Corynebacteria and Mycobacteria. Biochemistry.
Biokhimiia 77, 362-371 (2012).
Rontein, D. et al. CYP725A4 from yew catalyzes complex structural
rearrangement of taxa-4(5),11(12)-diene into the cyclic ether 5(12)-oxa-3(11)cyclotaxane. The Journal of biological chemistry 283, 6067-6075 (2008).
Xue, J. & Ahring, B.K. Enhancing Isoprene Production by Genetic Modification of
the 1-Deoxy-D-Xylulose-5-Phosphate Pathway in Bacillus subtilis. Applied and
environmental microbiology 77, 2399-2405 (2011).
Zhou, K., Zou, R., Zhang, C., Stephanopoulos, G. & Too, H.P. Optimization of
amorphadiene synthesis in Bacillus subtilis via transcriptional, translational, and
media modulation. Biotechnology and bioengineering (2013).
Doshi, R., Nguyen, T. & Chang, G. Transporter-mediated biofuel secretion.
Proceedings of the National Academy of Sciences of the United States of
America 110, 7642-7647 (2013).
Santos, C.N., Xiao, W. & Stephanopoulos, G. Rational, combinatorial, and
genomic approaches for engineering L-tyrosine production in Escherichia coli.
Proceedings of the National Academy of Sciences of the United States of
America 109, 13538-13543 (2012).
Minami, H. et al. Microbial production of plant benzylisoquinoline alkaloids.
Proceedings of the National Academy of Sciences of the United States of
America 105, 7393-7398 (2008).
Choi, Y.J. & Lee, S.Y. Microbial production of short-chain alkanes. Nature 502,
571-574 (2013).
Leber, C. & Da Silva, N.A. Engineering of Saccharomyces cerevisiae for the
synthesis of short chain fatty acids. Biotechnology and bioengineering 111, 347358 (2014).
Craft, D.L., Madduri, K.M., Eshoo, M. & Wilson, C.R. Identification and
characterization of the CYP52 family of Candida tropicalis ATCC 20336,
important for the conversion of fatty acids and alkanes to alpha,omegadicarboxylic acids. Applied and environmental microbiology 69, 5983-5991
(2003).
440
441
442
443
18
Figures and tables
a
b
Glucose
E. coli
Taxadiene
Oxygenated
taxanes
S. cerevisiae
Ethanol
1.4E+12
d
8E+11
6E+11
4E+11
2E+11
E. coli
1
culture
445
446
447
448
449
450
451
452
453
454
455
35
30
25
20
15
10
5
0
0
Co-culture
2
2.0
1.5
1.0
0.5
0.0
0
24
48
Time (h)
72
0
24
48
Time (h)
72
60
Extracellular ethanol (g/L)
1E+12
2.5
e
40
1.2E+12
Total taxanes (mg/L)
44h E. coli CFU (cell/mL)
c
Oxygenated taxanes (mg/L)
444
50
40
30
20
10
0
0
24
48
Time (h)
E. coli culture
72
Co-culture
Figure 1 A competitive E. coli – S. cerevisiae consortium for production of oxygenated taxanes.
(a) Both E. coli TaxE1 and the yeast TaxS1 grew on glucose; E. coli TaxE1 produced taxadiene
which can diffuse to the yeast, where it is oxygenated. (b) Only the co-culture produced the
oxygenated taxanes. (c) Growth of E. coli TaxE1 was inhibited by the presence of the yeast. (d)
The taxane productivity of E. coli TaxE1 was compromised by the presence of the yeast. Total
taxanes = Taxadiene + Oxygenated taxanes. (e) These inhibitions could be due to the ethanol
produced by the yeast, which was confirmed by follow-up experiments (Supplementary Fig. 2).
Error bars, s.e. in all graphs (some error bars are smaller than the plot symbols). All replicates
have also been plotted in all graphs (open circle), which indicates the number of replicates for
each experiment.
456
19
b
Xylose
Taxadiene
E. coli
Oxygenated
taxanes
S. cerevisiae
Acetate
S. cerevisiae density (OD600)
a
16
14
12
10
8
6
4
2
0
0
d 70
Total taxanes (mg/L)
Extracellular acetate (g/L)
2.5
2
1.5
1
0.5
60
50
40
30
20
10
0
0
0
457
458
459
460
461
462
463
464
465
466
467
e
Oxygenated taxanes (mg/L)
c
24
48
Time (h)
72
E. coli culture
0
24
48
Time (h)
72
S. cerevisiae culture
24
48
Time (h)
72
4.5
4.0
3.5
3.0
2.5
2.0
1.5
1.0
0.5
0.0
0
24
48
Time (h)
72
Co-culture
Figure 2 A mutualistic E. coli – S. cerevisiae consortium for production of oxygenated taxanes. (a)
E. coli TaxE1 grew on xylose and produced acetate that served as sole carbon source for the
yeast to grow. The taxadiene produced by E. coli TaxE1 was oxygenated in yeast TaxS1. (b) Yeast
TaxS1 could only grow in presence of the E. coli TaxE1. (c) Yeast TaxS1 removed the acetate
produced by E. coli TaxE1. (d) The presence of yeast TaxS1 did not compromise taxane
production of E. coli TaxE1. (e) Yeast TaxS1 can only produce oxygenated taxanes when E. coli
TaxE1 supplied taxadiene. The taxadiene oxygenation efficiency of this co-culture was 8% (4
mg/L out of 50 mg/L taxadiene was oxygenated). Error bars, s.e. in all graphs (some error bars
are smaller than the plot symbols). All replicates have also been plotted in all graphs (open
circle), which indicates the number of replicates for each experiment.
468
20
15
12
9
6
3
0
0
48
72
96
Time (h)
After growth optimization
469
470
471
472
473
474
475
476
477
478
c
0.9
24
Before growth optimization
0.6
0.3
0.0
ACSp
UAS-GPDp
0 TEFp
0 GPDp
0
0
Oxygenated taxanes (mg/L)
b
18
5αCYP activity (mg/L/h)
Oxygenated taxanes
(mg/L)
a
30
25
20
15
10
5
0
0
24
48
72
Time (h)
96
TEFp
GPDp
ACSp
UAS-GPDp
Figure 3 Optimizing the yeast growth and engineering the yeast promoters improved production
of the oxygenated taxanes. (a) Growth optimization (increasing the yeast inoculum and feeding
additional nutrients) improved production of the oxygenated taxanes by more than two-fold. (b)
A stronger promoter (UAS-GPDp), compared to the previously used TEFp, was found in the
promoter screening in terms of taxadiene oxygenation. (c) The co-culture using UAS-GPDp also
produced significantly (p<0.01, based on Student’s t-test) more oxygenated taxanes than that
using TEFp. Error bars, s.e. in all graphs (some error bars are smaller than the plot symbols). All
replicates have also been plotted in all graphs (open circle), which indicates the number of
replicates for each experiment.
479
480
21
a
3.0
S. cerevisiae / E. coli (OD600)
b
Xylose
Acetyl-CoA
ATP + Acetate
Oxidative phosphorylation
ATP + CO2
2.5
2.0
1.5
1.0
0.5
0.0
0 41h 0
Control Control
KnockoutKnockout
c
Knockout
50
40
40
Taxanes (mg/L)
Taxanes (mg/L)
Control
50
30
20
10
0
20
10
0
0
481
482
483
484
485
486
487
488
489
490
491
492
493
30
24
48 72
Time (h)
96 120
0
24
48 72
Time (h)
96 120
Oxygenated taxanes
Taxadiene
Figure 4 Inactivating oxidative phosphorylation of the E. coli improved yeast growth and
production of the oxygenated taxanes. (a) Inactivation of the E. coli oxidative phosphorylation
forces the production of acetate, which became the major pathway of generating ATP in the E.
coli. (b) The acetate-overproducing E. coli (TaxE4) improved the yeast growth in the co-culture.
Control: TaxE1-TaxS4 co-culture; Knockout: TaxE4-TaxS4 co-culture. (c) The taxadiene
oxygenation efficiency was greatly improved when the S. cerevisiae was co-cultured with the
acetate-overproducing E. coli. Oxygenation efficiency of the TaxE1-TaxS4 co-culture was ~50%
(20 mg/L oxygenated taxanes per 40 mg/L total taxanes), and that of the TaxE4-TaxS4 co-culture
was ~75% (30 mg/L oxygenated taxanes per 40 mg/L total taxanes). Error bars, s.e. in all graphs
(some error bars are smaller than the plot symbols). All replicates have also been plotted in all
graphs (open circle, except b, in which N=4), which indicates the number of replicates for each
experiment.
494
22
5αCYP
CPR
Taxadiene
b
10βCYP
CPR
TAT
Taxadien-5α-ol
Taxadien-5α-acetate
c
300
Signal @ 346 m/z
200
5αCYP
100
0
14.0
15.0
16.0
17.0
300
5αCYP-TAT-10βCYP
200
Monoacetylated dioxygenated
taxane (mg/L)
a
Taxadien-5α-acetate-10β-ol
1.2
1
0.8
0.6
0.4
0.2
0
0
24 48 72 96 120
Time (h)
100
0
14.0
495
496
TEFp-TAT
15.0
16.0
UASGPDp-TAT
17.0
UASGPDp-TAT CL
Time (min)
497
498
499
500
501
502
503
504
505
506
507
508
509
510
511
Figure 5 Production of a monoacetylated dioxygenated taxane by the E. coli – S. cerevisiae coculture. (a) Early paclitaxel biosynthetic pathway. (b) The yeast co-expressing 5αCYP-CPR, TAT
and 10βCYP-CPR (TaxS6) produced putative taxadien-5α-acetate-10β-ol when co-cultured with a
taxadiene-producing E. coli. Extracted ion chromatograms (346 m/z, molecular weight of
monoacetylated dioxygenated taxane) are shown here. 5αCYP: TaxE4/TaxS4 co-culture; 5αCYPTAT-10βCYP: TaxE4/TaxS6 co-culture. (c) Using a stronger promoter (UASGPDp) to express TAT
improved titer of the monoacetylated dioxygenated taxane. Operating the bioreactor at a
carbon-limited (CL) condition further improved the production titer and yield (xylose
consumption was reduced by 30%). TEFp-TAT: TaxE4/TaxS6 co-culture, where expression of TAT
was driven by TEFp; UASGPDp-TAT: TaxE4/TaxS7 co-culture, where UASGPDp was used to
express TAT; UASGPDp-TAT CL: TaxE4/TaxS7 co-culture at a carbon limited condition. Error bars,
s.e. in all graphs (some error bars are smaller than the plot symbols). All replicates have also
been plotted in all graphs (open circle), which indicates the number of replicates for each
experiment.
512
23
a
MEP Pathway
ISPA
VALC
FPP (C15)
Valencene
HmCYP
AtCPR
ADHC3
Nootkatol
Nootkatone
GGPPS
KSL
SmCYP
CPS
SmCPR
GGPP (C20)
Miltiradiene
b
c
Ferruginol
Nootkatone
10
5
6
30
Nootkatone (mg/L)
15
0
25
20
15
10
5
0
0
513
514
515
516
517
518
519
520
521
522
523
524
525
526
Nootkatol
35
Nootkatol (mg/L)
Ferruginol (mg/L)
20
Ferruginol
24
TaxE5
48 72
Time (h)
96 120
TaxE5 + TaxS8
5
4
3
2
1
0
0
24
48 72
Time (h)
TaxE6
96 120
0
TaxE6 + TaxS9
24
48 72
Time (h)
96
120
TaxE6 + TaxS10
Figure 6 Use of the E. coli-S. cerevisiae co-culture for production of other oxygenated
Isoprenoids. (a) Illustration of biosynthetic pathways of ferruginol and nootkatone. (b) An E. coli
was engineered to produce miltiradiene from xylose (TaxE7), which cannot produce ferruginol
on its own. When this E. coli was co-cultured with a yeast expressing a specific CYP and its
reductase (TaxS8), the co-culture can produce 18 mg/L ferruginol. Mass spectrum of the
produced ferruginol was identical to the one in the literature (data not shown). (c) Similarly, an
E. coli was engineered to produce valencene (TaxE8); itself cannot produce any oxygenated
valencene. When it was co-cultured with a yeast expressing a specific CYP and its reductase
(TaxS9), the co-culture can produce 30 mg/L nootkatol and low quantity of nootkatone. When
an alcohol dehydrogenase was introduced to TaxS9, the resulting strain TaxS10 can produce 4
mg/L nootkatone in presence of TaxE8. Error bars, s.e. in all graphs. (some error bars are smaller
than the plot symbols). All replicates have also been plotted in all graphs (open circle), which
indicates the number of replicates for each experiment.
527
24
528
ONLINE METHODS
529
E. coli strains. E. coli TaxE1 was previously constructed by Chin Giaw Lim in our lab
530
(unpublished works). In brief, the MEP operon1 (dxs-idi-ispDF controlled by T7 promoter) and
531
the TG operon1 (ts-ggpps controlled by T7 promoter) were integrated into locus araA and locus
532
lacY of E. coli MG1655_ΔrecA_ΔendA_DE31 respectively. Strains used in this study are
533
summarized in Supplementary Table 2.
534
To engineer E. coli TaxE1 to overproduce acetate, we overexpressed pta or pta-ackA
535
operon by using a pSC101 based plasmid containing trc promoter (p5trc1). pta or ackA amplified
536
from E. coli MG1655 chromosome was assembled with part of p5trc by using the recently
537
developed Cross-Lapping In Vitro Assembly (CLIVA) method43 (primer P1-P6 used), yielding
538
plasmid p5trc-pta and p5trc-ackA respectively. Primers used in this study are summarized in
539
Supplementary Table 3. All the plasmids constructed in this study were validated via sequencing.
540
Plasmid p5trc-pta was transformed into E. coli TaxE1, yielding E. coli TaxE2. ackA with trc
541
promoter and terminator was amplified from p5trc-ackA and cloned into p5trc-pta via CLIVA
542
(primer P7-P10 used), yielding plasmid p5trc-pta-trc-ackA. This plasmid was transformed into E.
543
coli TaxE1, yielding E. coli TaxE3. After overexpression of pta and pta-ackA, we inactivated
544
oxidative phosphorylation of E. coli TaxE1 by knocking out atpFH as described previously24
545
(primer P11 and P12 used), yielding E. coli TaxE4.
546
To construct E. coli to produce miltiradiene, we knocked out atpFH of E. coli TaxE5 (a
547
strain previously constructed by Chin Giaw Lim in our lab, unpublished works) as described
548
previously24 (primer P11 and P12 used), resulting in strain TaxE6. Then we transformed plasmid
549
p5T7-KSL-CPS-GGPPS into E. coli TaxE6, resulting in strain TaxE7. To obtain plasmid p5T7-KSL-
550
CPS-GGPPS, KSL and CPS amplified from synthetic DNA were assembled with part of p5T7TG1 via
25
551
CLIVA (primer P13-P18 used). To construct E. coli to produce valencene, ispA amplified from E.
552
coli genome and valC amplified from synthetic DNA were assembled with part of p5T7TG via
553
CLIVA (primer P18 – P23 used), yielding plasmid p5T7-ISPA-VALC, which was transformed into E.
554
coli TaxE6, resulting in strain TaxE8.
555
To engineer E. coli to express taxadiene 5a-hydroxylase with its reductase (5αCYP-CPR,
556
as
a
fusion
protein),
plasmid
p5trc-5αCYP-CPR
was
transformed
into
E.
coli
557
MG1655_ΔrecA_ΔendA_DE3, yielding E. coli TaxE9. Plasmid p5trc-5αCYP-CPR was previously
558
constructed by Chin Giaw Lim in our lab (unpublished works). To obtain this plasmid, coding
559
sequence of 5αCYP-CPR amplified from p10At24T5αOH-tTCPR1 was cloned into p5trc. To be
560
compatible with E. coli TaxE9, E. coli EDE3Ch1TrcMEPp5T7TG1 (named as TaxE10 in this study)
561
was used to produce taxadiene in the E. coli – E. coli co-culture, as both strains were resistant to
562
spectinomycin. An E. coli carrying unbalanced taxadiene synthetic pathway was also constructed
563
in this study: plasmid p5T7TG was transformed into E. coli TaxE4, resulting in strain TaxE11.
564
S. cerevisiae strains. S. cerevisiae BY4700 (ATCC 200866, MATa ura3Δ0) was used to
565
express the 5αCYP-CPR. Its coding gene amplified from plasmid p10At24T5αOH-tTCPR1 was
566
cloned into plasmid p416-TEF (ATCC 87368) by using the restriction enzyme cloning (XbaI and
567
HindIII, primer P24 and P25 used), yielding plasmid p416-TEFp-5αCYP-CPR. The auxotrophic
568
marker and expression cassette of the new plasmid (URA-TEFp-5αCYP-CPR-CYC1t) was via CLIVA
569
cloned into the integration shuttle vector pUC-YPRC15 (primer P26-P29 used), which was
570
constructed by cloning PCR fragment of BY4700 YPRC locus into plasmid pUC19 (New England
571
Biology) via the restriction enzyme cloning (NotI and EcoRI, primer P30-P33 used). The resulting
572
plasmid (pUC-YPRC15-URA-TEFp-17α5αCYP-CPR-CYCt) was linearized by using NotI and
26
573
transformed into BY4700 (YPRC15 locus44), yielding yeast TaxS1. This construction was
574
illustrated in Supplementary Fig. 15.
575
To replace the TEFp with GPDp and ACSp, GPDp amplified from plasmid p414-GPD
576
(ATCC 87356) or ACSp amplified from BY4700 chromosome was combined with part of pUC-
577
YPRC15-URA-TEFp-5αCYP-CPR via CLIVA (primer P34-P41 used), yielding plasmid pUC-YPRC15-
578
URA-GPDp-5αCYP-CPR-CYCt and pUC-YPRC15-URA-ACSp-5αCYP-CPR-CYCt respectively. These
579
two plasmids were linearized by using NotI and transformed into BY4700 (YPRC15 locus),
580
yielding yeast TaxS2 and TaxS3 respectively. To add upstream activation sequence (UAS) to
581
GPDp, the UASTEF-UASCIT1-UASCLB221 was synthesized (as gblock gene fragment, Integrated DNA
582
Technologies) and cloned into pUC-YPRC15-URA-GPDp-5αCYP-CPR-CYCt via CLIVA (primer P42-
583
P45 used), yielding pUC-YPRC15-URA-UAS-GPDp-5αCYP-CPR-CYCt. This plasmid was linearized
584
by using NotI and transformed into BY4700 (YPRC15 locus), yielding yeast TaxS4. Sequences of
585
all the synthetic genes used in this study are summarized in Supplementary Table 4.
586
S. cerevisiae BY4719 (ATCC 200882, MATa trp1Δ63 ura3Δ0) was used to co-express
587
5aCYP-CPR, taxadien-5α-ol acetyl-transferase (TAT) and taxane 10β-hydroxylase with its
588
reductase (10βCYP-CPR, as a fusion protein). Plasmid pUC-YPRC15-URA-GPDp-5αCYP-CPR-CYCt
589
was linearized by using NotI and first transformed into BY4719 (YPRC15 locus), yielding yeast
590
TaxS5. To further express TAT and 10βCYP-CPR in TaxS5, we constructed an integration vector
591
(pUC-PDC6-TRP) that targeted locus PDC6 and contained TRP marker. First, plasmid pUC19 was
592
combined with PCR fragment of BY4700 PDC6 locus via CLIVA (primer P46-P49 used), yielding
593
integration plasmid pUC-PDC6. The auxotrophic marker (TRP) of plasmid p414-GPD was then
594
cloned into pUC-PDC6 via CLIVA (primer P50-P53 used), yielding integration plasmid pUC-PDC6-
595
TRP. After the construction of the integration vector, coding gene of Taxus cuspidata TAT was
27
596
synthesized (Genscript) and cloned into plasmid pJA11545 via CLIVA (primer P54-P57 used),
597
yielding p426-TEFp-TAT-ACTt. Coding gene of Taxus cuspidate 10βCYP was synthesized (as
598
gblocks gene fragments, Integrated DNA Technologies) and cloned into pUC-YPRC15-URA-GPDp-
599
5αCYP-CPR to replace the 5αCYP via CLIVA (primer P58-P63 used), yielding pUC-YPRC15-URA-
600
GPDp-10βCYP-CPR-CYCt. The expression cassettes of these two plasmids (TEFp-TAT-ACTt and
601
GPDp-10βCYP-CPR-CYCt) were assembled with part of the integration vector pUC-PDC6-TRP via
602
CLIVA (primer P64-P69 used), yielding pUC-PDC6-TRP-(GPDp-10βCYP-CPR-CYCt)-(TEFp-TAT-ACTt).
603
This plasmid was linearized by using NotI and transformed into TaxS5 (PDC6 locus44), yielding
604
yeast TaxS6.
605
To replace the promoter of TAT (TEFp) with a stronger promoter (UASGPDp), coding
606
gene of TAT amplified from plasmid p426-TEFp-TAT-ACTt was assembled with part of pUC-
607
YPRC15-URA-UAS-GPDp-5αCYP-CPR-CYCt via CLIVA (primer P59 and P70-72 used), resulting in
608
plasmid pUC-YPRC15-URA-UAS-GPDp-TAT-CYCt; coding gene of 10βCYP-CPR amplified from
609
pUC-YPRC15-URA-GPDp-10βCYP-CPR-CYCt was assembled with part of p426-TEFp-TAT-ACTt via
610
CLIVA (primer P56, P57, P73 and P74 used), resulting in plasmid p426-TEFp-10βCYP-CPR-ACTt.
611
Then expression operon of plasmid pUC-YPRC15-URA-UAS-GPDp-TAT-CYCt and plasmid p426-
612
TEFp-10βCYP-CPR-ACTt was assembled with part of integration plasmid pUC-PDC6-TRP via CLIVA
613
(primer P65-P68, P75 and P76 used), resulting in plasmid pUC-PDC6-TRP-(TEFp-10βCYP-CPR-
614
ACTt)-(UAS-GPDp-TAT-CYCt), which was linearized by NotI and transformed into TaxS5, resulting
615
in strain TaxS7.
616
To construct the yeast that can oxygenate miltiradiene, coding gene of SmCYP-SmCPR
617
were synthesized and assembled with part of plasmid pUC-YPRC15-URA-UAS-GPDp-5αCYP-CPR-
618
CYCt via CLIVA (primer P77-P82 used), resulting in plasmid pUC-YPRC15-URA-UAS-GPDp-SmCYP28
619
SmCPR-CYCt, which was transformed into S. cerevisiae BY4700, resulting in strain TaxS8. To
620
construct the yeast that can produce nootkatone from valencene, coding gene of HmCYP-AtCPR
621
was synthesized and assembled with part of plasmid pUC-YPRC15-URA-UAS-GPDp-5αCYP-CPR-
622
CYCt via CLIVA (primer P81-P86 used), resulting in plasmid pUC-YPRC15-URA-UAS-GPDp-HmCYP-
623
AtCPR-CYCt, which was linearized by NotI and transformed into S. cerevisiae BY4700, resulting in
624
strain TaxS9. To improve the nootkatone production, coding gene of PpADHC3 was amplified
625
from Pichia pastoris genomic DNA and assembled with part of plasmid p426-TEFp-TAT-ACTt via
626
CLIVA (primer), resulting in plasmid p426-TEFp-PpADHC3-ACTt; expression operon of this
627
plasmid was further assembled with plasmid pUC-YPRC15-URA-UAS-GPDp-HmCYP-AtCPR-CYCt
628
via CLIVA (primer P56, P57, P87 and P88 used), resulting in plasmid pUC-YPRC15-URA-(UAS-
629
GPDp-HmCYP-AtCPR-CYCt)-(TEFp-PpADHC3-ACTt),
630
transformed into S. cerevisiae BY4700, resulting in strain TaxS10.
which
was
linearized
by
NotI
and
631
Characterization of the yeast cultures by feeding taxadiene. All S. cerevisiae strains
632
were characterized in absence of E. coli prior to co-culture experiment. We used 14 mL glass
633
tubes (Pyrex) for this type of characterizations. A colony of the S. cerevisiae was inoculated into
634
1 mL YPD medium (10 g/L yeast extract, 20 g/L peptone, 20 g/L glucose) and grown at 30 ᵒC/250
635
rpm until cell density OD600 reached 2. 10 μL of 6 g/L synthetic taxadiene stock solution (in
636
DMSO) was added to start the experiments, and the cultures were then incubated at 22 ᵒC/250
637
rpm. To compare yeast growth and activity when growing on glucose or acetate, the same
638
procedure was used except the medium was the one used in bioreactor experiments with
639
indicated carbon source.
640
Bioreactor experiments for the E. coli – S. cerevisiae co-culture. A 1 L Bioflo bioreactor
641
(New Brunswick) was used for all the bioreactor works in this study. In initial experiments, seed
29
642
cultures of E. coli and S. cerevisiae were inoculated into 500 mL of defined medium (13.3 g/L
643
KH2PO4, 4 g/L (NH4)2HPO4, 1.7 g/L citric acid, 0.0084 g/L EDTA, 0.0025 g/L CoCl2, 0.015 g/L MnCl2,
644
0.0015 g/L CuCl2, 0.003 g/L H3BO3, 0.0025 g/L Na2MoO4, 0.008 g/L Zn(CH3COO)2), 0.06 g/L Fe(III)
645
citrate, 0.0045 g/L thiamine, 1.3 g/L MgSO4, pH 7.0) containing 5 g/L yeast extract and 40 g/L
646
glucose (or 20 g/L xylose). To prepare seed culture of E. coli, a colony of the E. coli was
647
inoculated into Luria-Bertani (LB) medium (10 g/L tryptone, 5 g/L yeast extract, 10 g/L NaCl,
648
pH=7) and grown at 37 ᵒC/250 rpm overnight. 5 mL of the grown cell suspension (OD of ~6) was
649
inoculated into the bioreactor. To prepare seed culture of S. cerevisiae, a colony of the S.
650
cerevisiae was inoculated into YPD medium and grown at 30 ᵒC/250 rpm until cell density
651
OD600 reached 20. 10 mL of the grown cell suspension were centrifuged at 3,000 g for 2 min,
652
and pellets were resuspended in phosphate buffered saline (PBS) and inoculated into the
653
bioreactor. In the control experiments, only E. coli or S. cerevisiae was inoculated into the
654
bioreactor. To improve growth of the microbes (refer to Figure 3a), ammonium phosphate was
655
co-fed with xylose (1 g (NH4)2HPO4 per 5 g xylose) and more seed culture of the S. cerevisiae
656
were inoculated (pellets of 50 mL of grown cell suspension, OD600=20).
657
During the fermentation, oxygen was supplied by filtered air at 0.5 L/min and agitation
658
was adjusted to maintain dissolved oxygen levels at 30% (280-800 rpm). The pH of the culture
659
was controlled at 7.0 using 10% NaOH and 0.5 M HCl. The temperature of the culture in the
660
bioreactor was controlled at 30 ᵒC until the dissolved oxygen level dropped below 40%. The
661
temperature of the bioreactor was then reduced to 22 ᵒC and the E. coli was induced with 0.1
662
mM IPTG. During the course of the fermentation, the concentration of glucose (or xylose),
663
acetate and ethanol was monitored at constant time intervals. As the glucose concentration
30
664
dropped below 20 g/L, 20 g/L of glucose was introduced into the bioreactor. As the xylose
665
concentration dropped below 10 g/L, 50 g/L of xylose was introduced into the bioreactor.
666
Bioreactor experiments for the E. coli – E. coli co-culture. Half liter of rich medium (5
667
g/L yeast extract, 10 g/L tryptone, 10 g/L NaCl, 5 g/L K2HPO4, 8 g/L glycerol, pH7) containing 50
668
mg/L spectinomycin, was inoculated with 5 mL of grown culture (OD of 4) of E. coli TaxE5 and 5
669
mL of grown culture (OD of 4) of E. coli TaxE6.
670
During the fermentation, oxygen was supplied by filtered air at 0.5 L/min and agitation
671
was adjusted to maintain dissolved oxygen levels at 30% (280-800 rpm). The pH of the culture
672
was controlled at 7.0 using 10% NaOH. The temperature of the culture in the bioreactor was
673
controlled at 30 ᵒC until the dissolved oxygen level dropped below 40%. The temperature of the
674
bioreactor was then reduced to 22 ᵒC and the E. coli was induced with 0.1 mM IPTG. During the
675
course of the fermentation, the concentration of glycerol and acetate was monitored at
676
constant time intervals. Glycerol was fed into the bioreactor at the rate of 0.65 g/h.
677
Test tube experiments for characterizing acetate production of E. coli. A colony of E.
678
coli was inoculated into LB medium, and incubated at 37 ᵒC/250 rpm overnight. 10 μL of grown
679
cells were inoculated into the same medium as the one used in E. coli – S. cerevisiae bioreactors.
680
The cell suspension was incubated at 22 ᵒC/250 rpm for 96 h and samples were taken for
681
extracellular acetate measurement.
682
Quantification of isoprenoids. At indicated time points, 200 μL of cell suspension was
683
sampled and mixed with 200 μL ethyl acetate and 100 uL 0.5mm glass beads. The mixture was
684
vortexed at room temperature for 20min, and clarified by centrifugation at 18,000 g for 2 min. 1
685
μL of the ethyl acetate phase was analyzed by GCMS (Varian saturn 3800 GC attached to a
31
686
Varian 2000 MS). The samples were injected into a HP-5ms column (Agilent Technologies USA).
687
Helium at flow rate 1.0 mL/min was used as the carrier gas. The oven temperature was kept at
688
100 °C for 1 min, then increased to 175 °C at the increment of 15 °C/min, then increased to
689
220 °C at the increment of 4 °C/min, then increased to 290 °C at the increment of 50 °C/min and
690
finally held at this temperature for 1 min. The injector and transfer line temperatures were both
691
set at 250 °C. The MS was operated under scan mode (40-400 m/z) and total ion count of
692
taxanes was used for the quantification. Taxadiene, nootkatol and nootkatone were quantified
693
by using the calibration curve (total ion count vs. concentration) constructed with authentic
694
standard.
695
The 5αCYP was reported to produce multiple oxygenated taxanes in S. cerevisiae34. After
696
analyzing co-culture samples, we also observed many peaks on total ion chromatography (40-
697
400 m/z, GCMS) between 11-18.5 min, where we did not observe any peak when sample of the
698
single cultures was analyzed (Supplementary Fig. 16a). Five of the major peaks contained
699
significant amount of 288 m/z signal (characteristic mass of mono-oxygenated taxane, 272
700
(taxadiene) + 16 (oxygen), Supplementary Fig. 16b). Among them, two were previously
701
identified as oxa-cyclotaxane (OCT) and taxadien-5α-ol34 (Supplementary Fig. 17), but the other
702
three taxanes have not been identified before (Supplementary Fig. 18). As a conservative
703
estimate, we only considered these five oxygenated taxanes for calculating titer of total
704
oxygenated taxanes. As standards of these five monooxygenated taxanes, the monoacetylated
705
dioxygenated taxane and ferruginol were not available, they were quantified by using the
706
taxadiene calibration curve.
707
Quantification of extracellular metabolites. At indicated time points, 1.1 mL of cell
708
suspension was sampled and centrifuged at 18,000 g for 1 min. The supernatant was sterilized
32
709
by using 0.2 μm filter. 1mL of filtered supernatant was analyzed by a HPLC (Waters 2695
710
separation module coupled to Waters 410 differential refractometer) to measure concentration
711
of extracellular glucose, xylose, acetate and ethanol. Bio-rad HPX-87H column was used and 14
712
mM sulfuric acid was used as mobile phase at the flow rate of 0.7 mL/min.
713
Quantification of E. coli and S. cerevisiae cell number. To measure cell number of E. coli
714
in the E. coli – S. cerevisiae co-cultures, 2 μL of cell suspension was diluted in 200 μL sterile PBS,
715
and 2 μL of the diluted cell suspension was further diluted in 200 μL sterile PBS. 50 μL of the
716
repeatedly diluted cell suspension was plated on LB agar plate (1.5% agar) and incubated at 37
717
ᵒC for 20 h. After the incubation, only E. coli colonies were visible on the plate (the yeast
718
colonies were only visible after at least 48 h at this condition). As such method of measuring
719
colony forming unit was time consuming and low throughput. We later developed a sucrose
720
gradient centrifugation method to quantify cell number of both E. coli and S. cerevisiae in the
721
co-culture. At indicated time points, 0.5 mL of cell suspension was sampled and loaded onto 1
722
mL of 45% sucrose solution in a 14 mL falcon tube, which was then centrifuged at 2,100 g for 2
723
min. Microbes in the supernatant were exclusively E. coli and those in the pellets were mostly S.
724
cerevisiae (Supplementary Fig. 19). After this separation, cell number of the two microbes could
725
be quantified by measuring optical density at 600 nm.
726
727
728
729
730
731
732
733
734
735
43.
44.
45.
Zou, R., Zhou, K., Stephanopoulos, G. & Too, H.P. Combinatorial Engineering of
1-Deoxy-D-Xylulose 5-Phosphate Pathway Using Cross-Lapping In Vitro
Assembly (CLIVA) Method. PloS one 8, e79557 (2013).
Flagfeldt, D.B., Siewers, V., Huang, L. & Nielsen, J. Characterization of
chromosomal integration sites for heterologous gene expression in
Saccharomyces cerevisiae. Yeast 26, 545-551 (2009).
Avalos, J.L., Fink, G.R. & Stephanopoulos, G. Compartmentalization of metabolic
pathways in yeast mitochondria improves the production of branched-chain
alcohols. Nature biotechnology 31, 335-341 (2013).
736
33
737
738
34
739
740
ACKNOWLEDGEMENT
741
Zhang, Jose Avalos and Wen Wang. This work was supported by National Institutes of Health
742
grant 1-R01-GM085323-01A1 and the Singapore MIT Alliance.
We acknowledge useful discussions and input from Adel Ghaderi, Felix Lam, Haoran
743
744
745
AUTHOR CONTRIBUTIONS
746
analyzed the results and wrote the manuscript. K.Z., K.Q. and S.E. executed all the experiments.
K.Z. and G.S. conceived the project. K.Z., K.Q. S.E. and G.S. designed the experiments,
747
748
749
COMPETING FINANCIAL INTERESTS
750
concept has been filed.
The authors declare no competing financial interests. A patent on the co-culture
751
35
1
Supplementary information for
2
3
4
5
Synthetic Microbial Consortia for Overproducing
Oxygenated Paclitaxel Precursors
6
7
Kang Zhou, Kangjian Qiao, Steven Edgar, Gregory Stephanopoulos
8
9
1
a
b
14
Taxanes (mg/L)
12
Taxadiene
TEFp
10
8
6
4
2
5αCYP reductase
0
0
12
24
Time (h)
36
48
Taxadien-5α-ol
Taxadiene
10
11
Oxygenated taxanes
S. cerevisiae
12
13
14
15
16
17
Supplementary Fig. 1 Taxadiene 5α-hydroxylase was confirmed to be functional in the S.
cerevisiae. (a) The 5αCYP and its reductase was expressed as a fusion protein, and their
transcription was controlled by the TEF promoter. (b) Taxadiene can be efficiently oxygenated
when fed into culture of the 5αCYP-expressing yeast. Error bars, s.e. in all graphs (some error
bars are smaller than the plot symbols, N=2).
18
2
b 2.5
Taxadiene (mg/L)
Cell density (OD600)
a8
6
4
2
0
1.5
1.0
0.5
0.0
0
19
20
2.0
8
16
24
Time (h)
32
0
No ethanol
8
16
24
Time (h)
32
50g/L ethanol
21
22
23
24
Supplementary Fig. 2 50 g/L exogenously added ethanol completely stopped growth and
taxadiene production of the E. coli in shake flask. The ethanol was added at 0 h. Error bars, s.e.
in all graphs (some error bars are smaller than the plot symbols, N=2).
25
26
27
3
40
3.00
35
Taxanes (mg/L)
Acetate (g/L)
Acetate feeding rate (a.u.)
3.50
2.50
2.00
1.50
1.00
25
20
15
10
0.50
5
0.00
0
0
28
29
30
24 48 72 96 120
Time (h)
0
24
48 72
Time (h)
96 120
Acetate
Taxadiene
Acetate feeding rate
Oxygenated taxanes
30
31
32
33
34
Supplementary Fig. 3 Feeding exogenous acetate did not improve production of the oxygenated
taxanes. (a) The feeding led to acetate accumulation. (b) Production of oxygenated taxanes was
not improved by the feeding as compared to the control (Figure 3a). Error bars, s.e. in all graphs
(some error bars are smaller than the plot symbols, N=2).
35
4
b
S. cerevisiae / E. coli(OD600)
a
Acetyl-CoA
pta
Acetyl-Phosphate
ackA
ATP + Acetate
0.7
0.6
0.5
0.4
0.3
0.2
0.1
0
41h
65h
Control
Pta pta
Control
c
Control
Pta
50
Taxanes (mg/L)
Taxanes (mg/L)
50
40
30
20
10
0
40
30
20
10
0
0
24
48 72
Time (h)
96
120
0
24
48 72
Time (h)
96
120
Oxygenated taxanes
36
37
38
39
40
41
42
43
44
Supplementary Fig. 4 Overexpression of pta neither improved the yeast growth nor the
taxadiene oxygenation. (a) Schematic illustration of the major acetate production pathway in E.
coli. (b) Effect of the overexpression on the yeast growth. Control: TaxE1 – TaxS4 co-culture, Pta:
TaxE2 – TaxS4 co-culture. (c) Effect of the overexpression on the taxane production. The E. coli
overexpressing both pta and ackA did not grow in LB medium at 22 ᵒC. Error bars, s.e. in all
graphs (some error bars are smaller than the plot symbols, N=2).
45
5
10000
9000
Signal (a.u.)
8000
7000
105
147
119 133
6000
5000
171
159
243
189
215 228
4000
251
3000
271
331
287
303
2000
346
1000
0
b
m/z
12000
125
Signal (a.u.)
10000
8000
6000
244
158
202
113
143
171
184
261
219
4000
290
231
306
322
2000
350
366
0
m/z
46
47
48
49
50
51
52
53
Supplementary Fig. 5 Mass spectrum of the monoacetylated dioxygenated taxane produced by
the co-culture. (a) Spectrum of the compound that was derived from non-labeled taxadiene. (b)
Spectrum of the compound that was derived from uniformly 13C-labeled taxadiene. In the latter
case, molecular weight of the compound was increased to 366 from 346, consistent with the
fact that twenty 12C atoms were substituted by 13C atoms.
54
6
Monoacetylated dioxygenated
taxane (mg/L)
1.4
1.2
1
0.8
0.6
0.4
0.2
0
0
55
56
57
58
59
60
61
62
5g/day
24
48 72 96 120
Time (h)
10g/day
15g/day
Supplementary Fig. 6 Optimization of xylose feeding rate for improving titer of the production
of the monoacetylated dioxygenated taxane in co-culture of TaxE4 and TaxS7. Linear feeding of
xylose started at the beginning of day 3, and volume of the culture was maintained at 500 mL
through the experiments. Rate 10 g/day was found to be optimal. In this case, xylose
concentration in medium was always below its detection limit (0.1 g/L) after day 3, and total
amount of consumed xylose was 80 g/L. Error bars, s.e. in all graphs (some error bars are smaller
than the plot symbols, N=2).
63
7
CH3I
Glucose
Cellulase
Acetate,
ethanol
Cellulose
A. fermentans
64
65
66
67
68
S. cerevisiae
Supplementary Fig. 7 Illustration of the interactions between the two microbes of the Bayer et
al.’s study (Ref. 12 in the manuscript). A. fermentans secreted cellulase that in medium broke
down cellulose into glucose, which can be utilized by both A. fermentans and S. cerevisiae (these
details were not shown in the Bayer et al’s paper). Thus, this co-culture was competitive.
69
8
a
b
c
Acetate
Cell density
Acetate (g/L)
5
4
3
2
1
0
0
70
71
72
73
74
75
76
77
24
48 72
Time (h)
96
120
Consumed xylose
40
35
30
25
20
15
10
5
0
Consumed xylose (g/L)
E. coli density (OD600)
6
0
24
48 72
Time (h)
96 120
140
120
100
80
60
40
20
0
0
24
48 72 96 120
Time (h)
Supplementary Fig. 8 Effect of S. cerevisiae on E. coli growth and its xylose consumption. (a) E.
coli TaxE4 accumulated high concentration of acetate in absence of S. cerevisiae TaxS7, which
can eliminate the acetate in co-culture. (b) After acetate concentration reached 5 g/L, the E. coli
mono-culture stopped to grow, and the E. coli grew to much higher cell density in the co-culture.
(c) After reaching 5 g/L acetate concentration, the E. coli mono-culture also stopped to consume
xylose while E. coli kept consuming xylose in presence of the yeast. Error bars, s.e. in all graphs
(some error bars are smaller than the plot symbols, N=2).
78
79
80
81
9
Putative Taxadien-5αacetate-10β-ol (mg/L)
0.8
0.7
0.6
0.5
0.4
0.3
0.2
0.1
0
0
24
48 72 96 120
Time (h)
Control
ΔatpFH
82
83
84
85
86
87
88
Supplementary Fig. 9 Production of the putative taxadien-5α-acetate-10β-ol by the co-culture
was also improved by inactivation of the oxidative phosphorylation. Control: TaxE1-TaxS6 coculture; Knockout: TaxE4-TaxS6 co-culture. Error bars, s.e. in all graphs (some error bars are
smaller than the plot symbols, N=2).
89
90
10
Mixing the E. coli culture with
an active S. cerevisiae culture
(volumetric ratio 4:1)
Taxanes (mg/L)
E. coli culture
phase
E. coli- S. cerevisiae co-culture
phase
45
40
35
30
25
20
15
10
5
0
0
50
Taxadiene
100
Time (h)
150
200
Oxygenated taxanes
91
92
93
94
95
96
97
98
99
Supplementary Fig. 10 Production of oxygenated taxanes by using a two-stage culture. The
taxadiene-producing E. coli and 5αCYP-expressing yeast were cultured separately in the glucose
medium for three days, and then mixed to produce oxygenated taxanes. This allowed
accumulation of taxadiene in the first phase and efficient oxygenation of the taxadiene in the
second phase. Error bars, s.e. in all graphs (some error bars are smaller than the plot symbols,
N=2).
100
101
11
a
b
20
Ethanol (g/L)
Xylose (g/L)
25
15
10
5
0
0
100 150
Time (h)
200
0
d
7
6
5
4
3
2
1
0
Oxygenated taxanes
(mg/L)
Taxadiene (mg/L)
c
50
1.4
1.2
1
0.8
0.6
0.4
0.2
0
0
50
100
150
Time (h)
200
50
100 150
Time (h)
200
50
100 150
Time (h)
200
5
4
3
2
1
0
0
102
103
104
105
106
107
108
109
Supplementary Fig. 11 Characterization of the E. coli culture, the S. cerevisiae culture and the
co-culture in the xylose/ethanol medium. (a) The S. cerevisiae could not utilize xylose. (b) The E.
coli could not utilize ethanol. (c) Only E. coli can produce taxadiene. (d) Only the co-culture can
produce oxygenated taxanes. Error bars, s.e. in all graphs (some error bars are smaller than the
plot symbols, N=2).
110
12
a
b
E. coli
Ethanol
Taxadiene
0
111
112
Oxygenated
taxanes
d
24 48 72 96 120 144
Time (h)
10
9
8
7
6
5
4
3
2
1
0
0
24 48 72 96 120 144
Time (h)
0
24 48 72 96 120 144
Time (h)
e
1.8
1.6
1.4
1.2
1
0.8
0.6
0.4
0.2
0
Acetate (g/L)
20
18
16
14
12
10
8
6
4
2
0
Ethanol (g/L)
Xylose (g/L)
c
S. cerevisiae
Taxanes (mg/L)
Xylose
0
24 48 72 96 120 144
Time (h)
4.5
4
3.5
3
2.5
2
1.5
1
0.5
0
113
114
115
116
117
118
119
120
121
Supplementary Fig. 12 Maintaining a stable co-culture of E. coli and S. cerevisiae for production
of oxygenated taxanes by applying two carbon sources. (a) In this co-culture, xylose can only be
utilized by the E. coli and ethanol can only be utilized by the S. cerevisiae. Taxadiene produced
by the E. coli can be oxygenated when it gets into the yeast. Both cells may produce acetate. (b)
Production of taxadiene and oxygenated taxanes in the co-culture. (c) Xylose consumption in the
co-culture. (d) Ethanol consumption in the co-culture. Ethanol was periodically added. (e)
Acetate accumulation in the co-culture. Error bars, s.e. in all graphs (some error bars are smaller
than the plot symbols, N=2).
122
13
a
b8
Taxadiene
E. coli
E. coli
A
B
Oxygenated
taxanes
Taxanes (mg/L)
7
Rich medium
6
5
4
3
2
1
0
0
Taxadiene
123
124
125
126
127
128
129
24
48 72 96 120
Time (h)
Oxygenated taxanes
Supplementary Fig. 13 An E. coli-E. coli consortium for production of oxygenated taxanes. (a) A
synthetic microbial consortium comprising two E. coli strains that cooperatively synthesize
oxygenated taxanes. (b) 0.8 mg/L oxygenated taxanes were produced by the E. coli – E. coli
consortium in fed-batch bioreactor, whereas no oxygenated taxanes was produced by culture of
any single E. coli strain (data not shown). Error bars, s.e. in all graphs (some error bars are
smaller than the plot symbols, N=2).
130
14
a
b
Control
5
4
3
2
1
0
40
35
30
25
20
15
10
5
0
50h
131
132
133
134
135
136
137
138
139
140
141
142
p5-T7TG
Taxadiene (mg/L)
6
Taxadiene (mg/L)
3d taxadiene (mg/L)
7
E. coli
69h
90h
Medium
Yeast
40
35
30
25
20
15
10
5
0
50h
E. coli
69h
90h
Medium
Yeast
Supplementary Fig. 14 Distribution of taxadiene in E. coli, medium and yeast, and effect of
taxadiene productivity of E. coli on it. (a) an E. coli carrying an unbalanced taxadiene synthetic
pathway (TaxE11) was confirmed to produce less taxadiene. Control: TaxE4 mono-culture in
shake flask. p5T7TG: TaxE11 mono-culture in shake flask. (b) the taxadiene distribution in the E.
coli and S. cerevisiae co-culture. Control: TaxE4/TaxS7 co-culture; p5T7TG: TaxE11/TaxS11 coculture. Taxadiene concentration in the co-culture was significantly reduced when a poor
taxadiene producer (E. coli TaxE11) was used. Nevertheless, at all conditions, more than 50% of
taxadiene was found to be outside E. coli (in medium or yeast), indicating that taxadiene can
cross cell membranes efficiently (E. coli has two cell membranes), and thus its mass transfer
should not be a limiting step in the isoprenoid production processes. Error bars, s.e. in all graphs
(some error bars are smaller than the plot symbols, N=2).
143
15
NotI
PCR fragment of YPRC15 locus
Up
YPRC15
PCR fragment of p416-TEFp-5αCYP-CPR
Down
pUC19
TEFp 5αCYP CYCt
pUC-YPRC15
Cloning
Cloning
URA
pUC-YPRC15URA-TEFp17α5αCYPCPR-CYCt
NotI linearization
Up
TEFp 5αCYP CYCt
URA
Down
The linearized plasmid
BY4700 YPRC15 locus
Up
YPRC15
Down
Recombination
TaxS1 YPRC15 locus
144
Up
TEFp 5αCYP CYCt
URA
Down
145
146
147
148
149
150
Supplementary Fig. 15 Schematic illustration of the yeast genome modification method used in
this study. Construction of yeast TaxS1 was demonstrated here, and other yeast strains were
constructed similarly. Up: upstream homologous sequence of YPRC 15. Down: downstream
homologous sequence of YPRC15.
151
16
Signal of total ions
a
150000
120000
X2
OCT
90000
X1
60000
X3
5α-ol
30000
0
11
Signal of extracted ion
(288m/z)
b
12
13
14
3500
15
Time (h)
16
17
18
OCT
3000
2500
X2
2000
X1
1500
X3
5α-ol
1000
500
0
11
12
13
14
15
Time (h)
16
17
18
152
153
154
155
156
157
158
159
Supplementary Fig. 16 The five oxygenated taxanes quantified in this study. (a) Samples of E.
coli co-culture, yeast culture and co-culture were analyzed by GC-MS. Multiple new peaks were
identified in the co-culture sample as compared to other samples (total ion chromatography). (b)
Five of them should be monooxygenated taxane as they also appeared on extracted ion
chromatography - 288m/z (272 (taxadiene)+16 (oxygen)).
160
17
a
8000
OCT
191
7000
Signal (a.u.)
6000
5000
4000
288
3000
2000
105 119
245
217
147
259
273
1000
b
0
1600
5α-ol
105
1400
Signal (a.u.)
1200
120
1000
800
600
133 145
255
171 185
227
400
273 288
200
0
m/z
161
162
163
164
165
Supplementary Fig. 17 Mass spectra of the known oxygenated taxanes produced by the coculture. (a) Mass spectrum of oxa-cyclotaxane. (b) Mass spectrum of taxadien-5α-ol.
166
167
18
a
1400
109
X1
270
1200
122
Signal (a.u.)
1000
273
201
149
800
288
255
219
600
400
200
0
b
3500
105
3000
270
186
X2
123
Signal (a.u.)
2500
131
2000
171
288
1500
255
1000
500
0
c
1200
Signal (a.u.)
1000
800
600
X3
201
109
273
121 133
255
219
161
288
191
245
400
200
0
m/z
168
169
170
171
172
Supplementary Fig. 18 Mass spectra of novel oxygenated taxanes produced by the co-culture. (a)
Mass spectrum of taxane X1. (b) Mass spectrum of taxane X2. (c) Mass spectrum of taxane X3.
173
174
19
Before centrifugation
The supernatant
The pellet
175
176
177
178
179
180
Supplementary Fig. 19 Separation of E. coli from S. cerevisiae by using a sucrose-gradient based
centrifugation method. The supernatant after the centrifugation mostly contained E. coli, and
the pellets mostly contained S. cerevisiae.
181
182
20
183
Supplementary Table 1 Characterization of Yeast TaxS7 on two carbon sources
Biomass yield (OD600/(g/L carbon source))
184
Glucose
2.35 ± 0.05
Acetate
1.81 ± 0.08
‘±’ indicates standard error.
185
21
Specific productivity of the
monoacetylated dioxygenated
taxane (μg/L/h/OD600)
0.281 ± 0.037
0.190 ± 0.025
186
Supplementary Table 2 Microbial strains used in this study
Strain
TaxE1
TaxE2
TaxE3
TaxE4
TaxE5
TaxE6
TaxE7
TaxE8
TaxE9
TaxE10
TaxE11
TaxS1
TaxS2
TaxS3
TaxS4
TaxS5
TaxS6
TaxS7
TaxS8
TaxS9
TaxS10
Genotype
MG1655_ΔrecA_ΔendA_DE3 ΔaraA::T7-MEP ΔlacY::T7-TG
MG1655_ΔrecA_ΔendA_DE3 ΔaraA::T7-MEP ΔlacY::T7-TG p5trc-pta
MG1655_ΔrecA_ΔendA_DE3 ΔaraA::T7-MEP ΔlacY::T7-TG p5trc-pta-trc-ackA
MG1655_ΔrecA_ΔendA_DE3 ΔaraA::T7-MEP ΔlacY::T7-TG ΔatpFH::FRT-KanR-FRT
MG1655_ΔrecA_ΔendA_DE3 ΔaraA::T7-MEP
MG1655_ΔrecA_ΔendA_DE3 ΔaraA::T7-MEP ΔatpFH::FRT-KanR-FRT
MG1655_ΔrecA_ΔendA_DE3 ΔaraA::T7-MEP ΔatpFH::FRT-KanR-FRT p5T7-KSLCPS-GGPPS
MG1655_ΔrecA_ΔendA_DE3 ΔaraA::T7-MEP ΔatpFH::FRT-KanR-FRT p5T7-ISPAVALC
MG1655_ΔrecA_ΔendA_DE3 p5trc-5αCYP-CPR
MG1655_ΔrecA_ΔendA_DE3 p5T7TG
MG1655_ΔrecA_ΔendA_DE3 ΔaraA::T7-MEP ΔlacY::T7-TG ΔatpFH::FRT-KanR-FRT
p5T7TG
MATa ura3Δ0::URA-TEFp-5αCYP-CPR-CYCt
MATa ura3Δ0::URA-GPDp-5αCYP-CPR-CYCt
MATa ura3Δ0::URA-ACSp-5αCYP-CPR-CYCt
MATa ura3Δ0::URA-UAS-GPDp-5αCYP-CPR-CYCt
MATa ura3Δ0::URA-GPDp-5αCYP-CPR-CYCt trp1Δ63
MATa ura3Δ0::URA-UAS-GPDp-5αCYP-CPR-CYCt trp1Δ63::TRP-(GPDp-10βCYPCPR-CYCt)-(TEFp-TAT-ACTt)
MATa ura3Δ0::URA-UAS-GPDp-5αCYP-CPR-CYCt trp1Δ63::TRP-(TEFp-10βCYPCPR-ACTt)-(UAS-GPDp-TAT-CYCt)
MATa ura3Δ0::URA-UAS-GPDp-SmCYP-SmCPR-CYCt
MATa ura3Δ0::URA-UAS-GPDp-HmCYP-AtCPR-CYCt
MATa ura3Δ0::URA-(UAS-GPDp-HmCYP-AtCPR-CYCt)-(TEFp-PpADHC3-ACTt)
187
188
22
189
190
Supplementary Table 3 Primers used in this study
No.
P1
P2
P3
P4
P5
P6
P7
P8
P9
P10
P11
Oligo
p5 LIC F
p5 LIC R
p5 - Ec.pta F
Ec.pta - p5 R
p5 - Ec.ackA F
Ec.ackA - p5 R
p5 site 1 LIC F
p5 site1 LIC R
p5 site1 - operon F
operon - p5 site1 R
atpF(KO)-FRT F
P12 FRT2-atpH(KO) R
P13
P14
P15
P16
P17
P18
P19
P20
P21
P22
P23
P24
P25
P26
P27
P28
P29
P30
P31
P32
P33
P34
P35
P36
P37
T7 RBS 1 - ksl F
ksl - GGGS_cps R
GGGS_cps F
cps - tail of tds R
tail of tds F
T7 RBS 1 LIC R
T7 RBS 1 - valC_3-1 F
ValC_3-1 - GS.LK
GS.LK_Ec.ispA F
Ec. ispA - tail of ggpps R
tail of ggpps F
XbaI-bovine17a F
CPR-his-HindIII R
YPRC-p4xx F
p4xx-YPRC R
p4xx-YPRC F
YPRC-p4xx R
NotI-YPRC15 F
YPRC15-EcoRI R
EcoRI-pUC19 F
pUC19-NotI R
GPDp-17a F
p4xx-GPDp R
p4xx-GPDp F
GPDp-17a R
Sequence
GTCGA*CCATC*ATCATCATC
ATGTT*ATTCC*TCCTTATTT
GGAAT*AACAT*GTGTCCCGTATTATT
GATGG*TCGAC*TTACTGCTGCTGTGC
GGAAT*AACAT*ATGTCGAGTAAGTTA
GATGG*TCGAC*TCAGGCAGTCAGGCG
ACTGG*GCCTT*GGCGTTTAAGGGCAC
GGAGT*CGCAT*TGGTGCTACGCCTGT
ATGCG*ACTCC*TGCATTAGG
AAGGC*CCAGT*CTTTCGACT
TAGTAAGCGTTGCTTTTATTTAAAGAGCAATATCAGAACGTTAA
CGGCTGGAGCTGCTTC
AGCGCGCCAGAGAGAAGCTCTGCCATTTGTTCGTTTTTGGTTAC
CAATTAGCCATGGTCC
TACCATGG*GCATGATG*AGCCTGGCATTTAAT
TTGCAGAA*CCACCACC*TTTACCGCGAACATT
GGTGGTGG*TTCTGCAA*GCCTGAGCAGC
AGACCTGG*ATTGGATC*TTAGGCAACCGGCTC
GATCCAAT*CCAGGTCT*AA
CATCATGC*CCATGGTA*TA
TACCATGG*GCATGATG*GCCGAGATGTTCAACGGC
GATCCGG*TGCTGCC*GGGGATGATGGGCTCGAC
GGCAGCA*CCGGATC*CGACTTTCCGCAGCAA
AGTTTTGA*CGAAAGGC*TTATTTATTACGCTGGATG
GCCTTTCG*TCAAAACT*AA
GCtctagaAAAATGGCTCTGTTATTAGCAGTT
GCaagcttTTAgtgatggtgatgatgatgCCAAATATCCCGTAAGTAGC
TCGCAAA*ACTAAAG*GGAACAAAAGCTG
TTCCATT*ATCAGAG*CAGATTGTACTGAGA
CTCTGAT*AATGGAA*GGTCGGGATGA
CTTTAGT*TTTGCGA*AACCCTATGCT
GCgcggccgcTTTATATCATATAATTAAGACACAAAAG
GCgaattcATAAAGCAGCCGCTACCA
GCgaattcAAAGCCTGGGGTGCCTAA
GCgcggccgcAGGTGGCACTTTTCGGGG
AACAAA*ATGGCT*CTGTTATTAGCAG
TTTTTTAT*CAGCTT*TTGTTCCCTTT
AAGCTG*ATAAAAAA*CACGCTTTTTC
AGCCAT*TTTGTT*TGTTTATGTGTGT
23
P38
P39
P40
P41
P42
P43
P44
P45
P46
P47
P48
P49
P50
P51
P52
P53
P54
P55
P56
P57
P58
P59
P60
P61
P62
P63
P64
P65
P66
P67
P68
P69
P70
P71
P72
P73
P74
P75
P76
P77
P78
P79
P80
P81
ACS1p-17a F
p4xx-ACS1p R
p4xx-ACS1p F
ACS1p-17a
p4xx - UAS(TEF) F
UAS(CLB) - GPD R
GPD LIC F
p4xx (Seq-F) R
pUC-PDC6 F
pDC6-pUC R
pDC6-pUC F
pUC-PDC6 R
Inter - Express F
Marker - Inter R
Marker - Inter F
Inter - Express R
p4xx RE - TAT F
TAT - Myc LIC R
Myc LIC F
p4xx RE LIC R
Linker CPR LIC F
GPD LIC R
GPD - CO.10bCYP F
Co.10bCYP part1 - part2 R
Co.10bCYP part1 - part2 F
CO.10bCYP - Linker CPR R
Inter - GPDp F
CYC1t - ACT1t rc R
CYC1t - ACT1t rc F
TEFp rc - Inter R
TEFp rc - Inter F
Inter - GPDp
GPD - TAT F
TAT - CYCt R
CYCt LIC F
p4xx RE - Co.10CYP F
CPR - Myc LIC R
Inter - UAS (45-F)
Inter - UAS (45-R)
GPD - Sm.CYP F
SmCYP - LK R
LK_SmCPR1 F
Sm. CPR - His R
6xHis LIC F
TGTGCT*ATGGCT*CTGTTATTAGCAG
ATTGTT*CAGCTT*TTGTTCCCTTTAG
AAGCTG*AACAAT*CTGTTTATTACCC
AGCCAT*AGCACA*GTGGGCAATG
ACAAA*AGCTG*AATGTTTCTACTCCT
TGTTT*TTTAT*GGGACAGGCACCGAA
ATAAA*AAACA*CGCTTTTTC
CAGCT*TTTGT*TCCCTTTAG
GCGG*CCGC*CTTT*CAAG*GGTGGGGG
CAGG*CTTT*GGCT*GAAC*AACAGTCTCTCC
GTTC*AGCC*AAAG*CCTG*GGGTGCCT
CTTG*AAAG*GCGG*CCGC*AGGTGGCA
TAAAGGGA*ACAAAAGC*TG
GCAGATTG*TACTGAGA*GT
TCTCAGTA*CAATCTGC*TC
GCTTTTGT*TCCCTTTA*GT
TAGAA*CTATG*GAAAAAACTGATCTG
TCAAT*TTTTG*AACTTTGGCCACGTA
CAAAA*ATTGA*TTTCTGAAG
CATAG*TTCTA*GAGCTAGC
GGCAG*CACCG*GATCC
CATTT*TGTTT*GTTTATGTG
AAACA*AAATG*GACTCCTTCATCTTC
GACAAGAT*TTCGTCCA*ACTTCAATC
TGGACGAA*ATCTTGTC*CTCCTTGAT
CGGTG*CTGCC*AGATCTTGGGAACAA
AAAGCT*AGTTTATC*ATTATCAATAC
TATCATAT*CAAATTAA*AGCCTTCGA
TTAATTTG*ATATGATA*CACGGTCCA
ATCGGC*ATAGCTTC*AAAATGTTTCT
GAAGCTAT*GCCGAT*TTCGGCCTATT
GATAAACT*AGCTTT*TGTTCCCTTT
AAACA*AAATG*GAAAAAACTGATCTG
TAAGC*TTTTA*AACTTTGGCCACGTA
TAAAA*GCTTA*TCGAT
TAGAA*CTATG*GACTCCTTCATCTTC
TCAAT*TTTTG*CCAAATATCCCGTAA
CAAAAGCT*AATGTTTC*TACTCCTTT
GAAACATT*AGCTTTTG*TTCCCTTTA
ATAAACAA*ACAAAATG*GACTCTTTTCCATTATTG
GATCCGG*TGCTGCC*GGACTTGACGATTGG
GGCAGCA*CCGGATC*CGAACCATCCTCTAAAAA
GATGGTGA*TGATGATG*CCAGACATCTCTCAAGTA
CATCATCA*TCACCATC*AC
24
191
P82 GPD LIC R2
CATTTTGT*TTGTTTAT*GTG
P83 GPD - Hm.CYP F
ATAAACAA*ACAAAATG*CAATTCTTCTCCTTGG
P84 Hm.CYP - LK R2
GATCCGG*TGCTGCC*TTCTCTGGATGGTTG
P85 LK_At.CPR F
GGCAGCA*CCGGATC*CACTTCTGCCTTGTAT
P86 At. CPR - His R
GATGGTGA*TGATGATG*CCAGACATCTCTCAA
P87 p4xx RE - Pp.ADH_C3 F
TAGAA*CTATG*ACCACAGTTTTCGCT
P88 Pp.ADH_C3 - Myc LIC R
TCAAT*TTTTG*CCCCCTGACTTTACT
* indicates a phosphorothioate bond
192
25
193
194
Supplementary Table 4 Sequence of synthetic genes used in this study
Synthetic gene
UASTEF-UASCIT1-UASCLB2
TAT
10βCYP
Sequence
aatgtttctactccttttttactcttccagattttctcggactccgcgcatcgccgtaccacttcaaaac
acccaagcacagcatactaaatttcccctctttcttcctctagggtgtcgttaattacccgtactaaag
gtttggaaaagaaaaaagagaccgcctcgtttctttttcttcgtcgaaaaaggcaataaaaatttttt
agagattactacatattccaacaagaccttcgcaggaaagtatacctaaactaattaaagaaatctc
cgaagttcgcatttcattgaacggctcaattaatctttgtaaatatgagcgtttttacgttcacattgcc
tttttttttatgtatttaccttgcatttttgtgctaaaaggcgtcacgtttttttccgccgcagccgcccgg
aaatgaaaagtatgacccccgctagaccaaaaatacttttgtgttattggaggatcgcaatccctttc
agtggaattattagaatgaccactactccttctaatcaaacacgcggaaatagccgccaaaagaca
gattttattccaaatgcgggtaactatttgtataatatgtttacatattgagcccgtttaggaaagtgc
aagttcaaggcactaatcaaaaaaggagatttgtaaatatagcgaccgaatcaggaaaaggtcaa
caacgaagttcgcgatatggatgaacttcggtgcctgtccc
atggaaaaaactgatctgcatgtcaatctgattgaaaaagttatggttggtcctagccctccactgcc
gaaaactaccctgcaactgtccagcattgataacctgccgggtgtacgcggtagcattttcaacgca
ctgctgatctataacgcaagcccgagcccgaccatgatctctgcagatccagcaaaaccgatccgc
gaagcgctggcgaaaattctggtttactacccgccatttgcaggccgcctgcgtgaaaccgaaaac
ggtgatctggaagttgagtgcaccggtgaaggtgcgatgttcctggaagcgatggcggacaacgaa
ctgagcgttctgggtgactttgacgattccaacccgtcttttcagcaactgctgttttccctgcctctgg
ataccaactttaaagatctgtccctgctggttgttcaggtgacccgctttacctgtggcggtttcgtcgt
cggcgttagcttccaccacggcgtttgcgacggccgtggcgccgcacagtttctgaaaggcctggct
gaaatggcacgtggtgaggtgaagctgtctctggaaccgatttggaaccgcgaactggttaaactg
gatgacccgaaatacctgcagttctttcacttcgagttcctgcgtgcgccgagcatcgtagagaaaa
tcgttcagacctatttcatcattgatttcgagacgatcaactatatcaaacagtctgtaatggaagag
tgcaaggagttctgctcttccttcgaagtggcttccgctatgacttggatcgcgcgtactcgtgcgttc
cagattccggaaagcgaatacgtgaagatcctgttcggcatggacatgcgtaactctttcaacccgc
cgctgccgtccggctattacggtaacagcatcggcacggcgtgtgctgtggacaacgttcaggatct
gctgtctggttctctgctgcgtgctatcatgattatcaagaaatccaaagtctccctgaacgacaattt
caaatctcgcgcggtggtaaaaccgtccgaactggacgtgaacatgaatcacgaaaacgtagtcgc
cttcgctgactggtcccgtctgggtttcgacgaagtagacttcggctggggtaatgctgtatctgtttc
tccggtgcagcagcagtctgctctggccatgcaaaactacttcctgttcctgaaaccgtctaaaaac
aaaccagatggtatcaaaatcctgatgtttctgccgctgtctaagatgaaatctttcaaaatcgaaat
ggaagccatgatgaagaaatacgtggccaaagtttaa
atggactccttcatcttcttgagatccattggtactaagttcggtcaattggaatcttccccagctattt
tgtctttgactttggctccaattttggccatcatcttgttgttgttattcagatacaaccacagatcctct
gttaagttgccaccaggtaaattgggttttccattgattggtgaaaccatccaattattgagaaccttg
agatctgaaaccccacaaaagttcttcgatgacagattgaaaaagtttggtccagtctacatgacct
cattgataggtcatccaactgttgttttgtgtggtccagctggtaacaaattggttttgtctaacgaag
ataagttggttgaaatggaaggtccaaagtctttcatgaagttgatcggtgaagattctatcgttgct
aagagaggtgaagatcacagaattttgagaactgctttggctagatttttgggtgctcaagctttaca
aaactacttgggtagaatgtcctccgaaattggtcatcattttaacgaaaagtggaagggtaaggat
gaagttaaggttttgccattggtcagaggtttgattttctctattgcttccaccttgttcttcgacgttaa
26
KSL
CPS
tgatggtcatcaacaaaagcaattgcaccacttgttggaaaccattttggttggttctttgtccgttcc
attggattttccaggtactagatacagaaagggtttacaagctagattgaagttggacgaaatcttgt
cctccttgattaagcgtagaagaagagatttgagatccggtattgcttccgatgatcaagatttgttg
tctgtcttgttgaccttcagagatgaaaagggtaactctttgaccgatcaaggtatcttggataacttc
tctgctatgttccatgcttcttacgatacaactgttgctccaatggctttgatcttcaagttgttgtactc
taacccagaataccacgaaaaggttttccaagaacaattggaaatcatcggtaacaagaaagaag
gtgaagaaatctcctggaaggacttgaaatctatgaagtacacttggcaagccgtccaagaatcttt
gagaatgtatccaccagttttcggtattttcagaaaggccattaccgatatccattacgatggttaca
ctattccaaagggttggagagttttgtgttctccatataccactcacttgagagaagaatactttcca
gaaccagaagaattcagaccatccagatttgaagatgaaggtagacatgttactccatacacttac
gttccatttggtggtggtttgagaacttgtccaggttgggaattttctaagatcgaaatcttgttgttcg
tccaccacttcgttaagaacttctcatcttacattccagtcgacccaaacgaaaaagttttgtctgatc
cattgccaccattaccagctaatggtttctctattaagttgttcccaagatcttaa
atgagcctggcatttaatccggcagcaaccgcatttagcggtaatggtgcacgtagccgtcgtgaaa
actttccggttaaacatgttaccgttcgtggttttccgatgattaccaataaaagcagctttgccgtta
aatgcaatctgaccaccaccgatctgatgggcaaaattgcagaaaaattcaaaggcgaggatagc
aattttccggcagccgcagcagttcagcctgcagcagatatgccgagcaatctgtgtattattgatac
cctgcagcgtctgggtgttgatcgttattttcgtagcgaaattgataccatcctggaagatacctatcg
tctgtggcagcgtaaagaacgtgcaatttttagcgataccgcaattcatgcaatggcatttcgtctgc
tgcgtgttaaaggttatgaagttagcagcgaagaactggcaccgtatgcagatcaagaacatgtgg
atctgcagaccattgaagttgcaaccgttattgaactgtatcgtgcagcacaagaacgtaccggtga
agatgaaagcagcctgaaaaaactgcatgcatggaccaccacatttctgaaacagaaactgctga
ccaatagcatcccggataaaaaactgcacaaactggtggaatactatctgaaaaactttcacggca
ttctggatcgtatgggtgttcgtcagaatctggatctgtacgatattagttattaccgtaccagcaaag
cagccaatcgttttagtaatctgtgctccgaagattttctggcatttgcacgtcaggattttaacatttg
tcaggcacagcatcagaaagaactgcagcagctgcaacgttggtatgccgattgtaaactggatac
cctgaaatatggtcgtgatgttgttcgtgttgcaaattttctgaccagcgcaattattggtgatccgga
actgagtgatgttcgtattgtttttgcacagcatattgttctggtgacccgcatcgatgatttttttgatc
atcgtggtagccgtgaagagagctacaaaattctggaactgatcaaagaatggaaagaaaaaccg
gcagcagaatatggtagcgaagaagttgaaattctgttcaccgcagtgtataataccgtgaatgaa
ctggcagaacgtgcccatgttgaacagggtcgtagcgttaaagatttcctgattaaactgtgggtgc
agatcctgagcatctttaaacgtgagctggatacctggtcagatgataccgcactgaccctggatga
ttatctgagcgcaagctgggttagcattggttgtcgtatttgtattctgatgtccatgcagttcattggc
atcaaactgtcagatgaaatgctgctgagcgaagaatgtattgatctgtgtcgtcatgttagcatggt
ggatcgcctgctgaatgatgttcagacctttgaaaaagaacgcaaagagaataccggtaatagcgt
taccctgctgctggcagcaaataaagatgatagcagttttaccgaagaagaggcaattcgtattgca
aaagaaatggccgaatgtaatcgtcgtcagctgatgcagattgtgtataaaaccggtacaatttttc
cgcgtcagtgcaaagatatgtttctgaaagtttgccgcattgggtgttatctgtatgcaagcggtgat
gaatttaccagtccgcagcagatgatggaagatatgaaaagcctggtttatgaaccgctgaccattc
atccgctggttgcaaataatgttcgcggtaaataa
atggcaagcctgagcagcaccattctgagccgtagtccggcagcacgtcgtcgtattacaccggca
agcgcaaaactgcatcgtccggaatgttttgcaaccagcgcatggatgggtagcagcagcaaaaat
ctgagcctgagctatcagctgaaccacaaaaaaatcagcgttgcaaccgttgatgcaccgcaggtt
catgatcacgatggcaccaccgttcatcagggtcatgatgcagttaaaaacattgaagatccgatcg
aatatatccgtaccctgctgcgtaccaccggtgatggtcgtattagcgttagcccgtatgataccgca
27
SmCYP
tgggttgcaatgattaaagatgttgaaggtcgtgatggtccgcagtttccgagcagcctggaatgga
ttgttcagaatcagctggaagatggtagctggggtgatcagaaactgttttgtgtttatgatcgtctgg
tgaataccattgcatgtgttgttgcactgcgtagctggaatgttcatgcacataaagttaaacgtggc
gtgacctatatcaaagaaaacgtggataaactgatggaaggcaacgaagaacatatgacctgtgg
ttttgaagttgtttttccggcactgctgcagaaagcaaaaagcctgggtatcgaagatctgccgtatg
attcaccggcagttcaagaggtttatcatgttcgtgaacaaaaactgaaacgcattccgctggaaat
catgcataaaattccgaccagtctgctgtttagcctggaaggtctggaaaatctggattgggacaaa
ctgctgaaactgcagagcgcagatggtagttttctgaccagcccgagcagtaccgcatttgcatttat
gcagaccaaagatgagaaatgctatcagttcatcaaaaacaccatcgacacctttaatggtggtgc
accgcatacctatccggttgatgtttttggtcgtctgtgggcaattgatcgcctgcagcgtctgggtat
tagccgtttttttgaaccggaaattgcagattgtctgagccacattcacaaattctggaccgataaag
gtgtttttagcggtcgtgaaagcgaattttgcgatattgatgataccagtatgggtatgcgtctgatgc
gtatgcatggttatgatgttgatccgaatgttctgcgcaactttaaacagaaagatggcaaatttagc
tgctatggtggtcagatgattgaaagcccgagcccgatttataacctgtatcgtgcaagccagctgc
gttttccgggtgaagaaattctggaagatgccaaacgttttgcctatgatttcctgaaagaaaaactg
gccaataaccagatcctggataaatgggttattagcaaacatctgccggatgaaattaaactgggc
ctggaaatgccgtggctggcaaccctgcctcgtgttgaagcaaaatactatattcagtattatgccgg
tagcggtgatgtgtggattggtaaaaccctgtatcgcatgccggaaattagcaatgatacctatcat
gatctggccaaaaccgattttaaacgttgtcaggcaaaacaccagtttgagtggctgtatatgcaag
aatggtatgaaagctgcggcattgaagaatttggcattagccgtaaagatctgctgctgagctatttt
ctggcaaccgcaagcatctttgaactggaacgtaccaatgaacgtattgcatgggccaaaagccag
attattgcaaaaatgatcaccagcttttttaacaaagaaaccacgagcgaagaagataaacgcgca
ctgctgaatgaactgggtaacattaatggtctgaatgataccaatggtgcaggtcgtgaaggtggtg
cgggtagcattgcactggccaccctgacccagtttctggaaggttttgatcgttatacccgtcaccag
ctgaaaaatgcatggtcagtttggctgacccagctgcagcatggtgaagcagatgatgcagaactg
ctgaccaataccctgaatatttgtgccggtcatattgcctttcgcgaagaaatcctggcacacaatga
atataaagcactgagcaatctgacgagcaaaatttgtcgccagctgagctttattcagagcgaaaa
agaaatgggtgtggaaggtgaaattgccgcaaaaagcagcattaaaaacaaagaactggaagag
gatatgcagatgctggttaaactggttctggagaaatatggtggtattgaccgcaatatcaaaaaag
catttctggcagttgccaaaacctattattaccgtgcatatcatgcagccgataccattgatacccat
atgttcaaagttctgtttgagccggttgcctaa
atggactcttttccattattggctgccttgtttttcattgctgctactattaccttcttgtccttccgtaga
agaagaaatttgccaccaggtccatttccatatccaatcgttggtaatatgttgcaattgggtgctaa
cccacatcaagtttttgctaagttgtctaagagatacggtccattgatgtccattcatttgggttccttg
tacaccgttatagtctcttcaccagaaatggccaaagaaatcttgcatagacatggtcaagttttctc
cggtagaactattgctcaagctgttcatgcttgtgatcacgataagatttctatgggttttttgccagtt
gcctctgaatggagagatatgagaaagatctgcaaagaacaaatgttctccaatcaatccatggaa
gcttctcaaggtttgagaagacaaaagttgcaacaattattggaccacgtccaaaagtgttctgatt
ctggtagagctgttgatattagagaagctgctttcattaccaccttgaatttgatgtctgctaccttgtt
ttcttcacaagctaccgaatttgattccaaggctaccatggaattcaaagaaattattgaaggtgttg
ccaccatcgttggtgttccaaattttgctgattacttcccaatcttaagaccattcgatccacaaggtg
ttaagagaagagctgatgtttttttcggtaagttgttggccaagatcgaaggttatttgaacgaaaga
ttggaatccaagagagctaatccaaacgctccaaagaaggatgatttcttggaaatcgttgtcgata
tcatccaagccaacgaattcaagttgaaaacccatcatttcacccacttgatgttggatttgtttgttg
gtggttctgataccaacaccacttctattgaatgggctatgtctgaattggttatgaacccagataag
28
SmCPR
VALC
atggctagattgaaggctgaattgaaatctgttgctggtgacgaaaagatcgttgatgaatctgctat
gccaaagttgccatacttgcaagctgttatcaaagaagtcatgagaattcatccacctggtcctttgt
tgttaccaagaaaagctgaatccgatcaagaagtcaacggttacttaattccaaagggtactcaaat
cttgattaacgcttacgccattggtagagatccatctatttggactgatccagaaacttttgacccag
aaagattcttggacaacaagatcgatttcaagggtcaagactacgaattattgccatttggttcaggt
agaagagtttgtccaggtatgccattggctactagaatattgcatatggctactgctactttggttcac
aatttcgattggaagttggaagatgattctactgctgctgctgatcatgctggtgaattatttggtgtt
gctgttagaagagcagtcccattgagaattattccaatcgtcaagtcctaa
atggaaccatcctctaaaaagttgtccccattggatttcattaccgccattttgaagggtgatattga
aggtgttgctccaagaggtgttgcagctatgttgatggaaaacagagatttggctatggttttgacta
cctctgttgctgttttgattggttgcgttgttgttttggcttggagaagaactgctggttctgctggtaaa
aaacaattgcaaccaccaaagttggttgttccaaaaccagctgctgaacctgaagaagctgaagac
gaaaaaactaaggtcagtgttttcttcggtactcaaactggtactgctgaaggttttgctaaagctttt
gccgaagaagctaaagctagatatccacaagctaagttcaaggttatcgatttggatgattacgctg
ccgatgatgatgaatacgaagaaaagttgaagaaagaatccttggccttcttcttcttggcttcttat
ggtgatggtgaacctactgataatgctgctagattttacaagtggttcaccgaaggtaaggatagag
aagattggttgaagaacttgcaatacggtgtttttggtttgggtaacagacaatacgaacacttcaa
caagattgccatcgttgtcgatgatttgattactgaacaaggtggtaagaagttggttccagttggttt
aggtgatgatgatcaatgcatcgaagatgatttttccgcttggagagaattggtttggccagaattgg
ataagttgttgagaaatgaagatgatgctactgttgctactccatataccgctgttgttttacaataca
gagttgtcttgcacgatcaaactgatggtttgatcacagaaaatggttctccaaatggtcatgctaac
ggtaacactatctatgatgctcaacatccatgtagagctaacgttgctgttagaagagaattgcata
ctccagcttcagatagatcttgtacccatttggaattcgatacttcaggtactggtttggtttacgaaa
ctggtgatcatgttggtgtttactgcgaaaacttgttggaaaatgtcgaagaagccgaaaagttattg
aacttgtctccacaaacctacttctccgttcatactgataacgaagatggtactccattgtctggttctt
cattgccaccaccatttccaccatgtactttgagaactgctttgactaagtacgccgatttgatttctat
gccaaagaagtctgttttggttgctttggctgaatacgcctctaatcaatcagaagctgatagattga
gatacttggcttcaccagatggtaaagaagaatacgcccaatatatcgttgcctcccaaagatcatt
attggaagttatggctgaattcccatctgctaaaccaccattgggtgttttttttgctgctattgctcct
agattgcaacctagattctactccatttcttcctccccaaaaattgctccaactagagttcatgttacc
tgtgctttggtttatgataagactccaactggtagaatccataagggtatttgttctacctggattaag
aacgctgttccattggaagaatcttcagattgctcttgggctccaattttcatcagaaactctaacttt
aagttgccagccgatccaaaggttccaattatcatggttggtccaggtacaggtttagctccttttag
aggtttcttacaagaaagattggccttgaaagaatctggtgctgaattgggtccagctattttgttttt
tggttgtagaaacagaaagatggacttcatatacgaagatgaattgaactccttcgttaaggttggt
gccatttctgaattgatcgttgctttttctagagaaggtccagccaaagaatacgttcaacataagat
gtctcaaagagcctccgatatttggaagatgatatctgatggtggttacatgtacgtttgtggtgatg
ctaaaggtatggctagagatgttcatagaaccttgcataccattgctcaagaacaaggttctttgtca
tcttctgaagcagaaggtatggtcaaaaacttgcaaactactggtagatacttgagagatgtctggt
aa
atggccgagatgttcaacggcaactcttctaacgacggatcttcttgcatgcccgtgaaggacgccc
tgcgacgaaccggcaaccaccaccccaacctgtggaccgacgacttcatccagtctctgaactctcc
ctactctgactcttcttaccacaagcaccgagagatcctgatcgacgagatccgagacatgttctcta
acggcgagggcgacgagttcggcgtgctcgagaacatctggttcgtggacgtggtgcagcgactgg
gcatcgaccgacacttccaggaggagatcaagaccgccctggactacatctacaagttctggaacc
29
HmCYP
acgactctatcttcggcgacctgaacatggtggccctgggcttccgaatcctgcgactgaaccgata
cgtggcctcttctgacgtgttcaagaagttcaagggcgaggagggccagttctctggcttcgagtcct
ctgaccaggacgctaagctcgaaatgatgctgaacctgtacaaggcctctgagctggacttccccga
cgaggacatcctgaaggaggcccgagccttcgcctctatgtacctgaagcacgtgatcaaggagta
cggcgacatccaggagtctaagaaccccctgctgatggagatcgagtacaccttcaagtacccctg
gcgatgccgactgccccgactcgaggcctggaacttcatccacatcatgcgacagcaggactgcaa
catctctctggccaacaacctctacaagatccccaagatctacatgaagaagatcctcgagctggcc
atcctggacttcaacatcctgcagtctcagcaccagcacgagatgaagctgatctctacctggtgga
agaactcttctgctatccagctggacttcttccgacaccgacacatcgagtcttacttttggtgggcct
cgcccctgttcgagcccgagttctctacctgccgaatcaactgcaccaagctgtctaccaagatgttc
ctgctggacgacatctacgacacctacggcaccgtcgaggagctgaagcccttcaccaccaccctg
acccgatgggacgtgtctaccgtggacaaccaccccgactacatgaagatcgccttcaacttctctt
acgagatctacaaggagatcgcctctgaggccgagcgaaagcacggccccttcgtgtacaagtacc
tgcagtcttgctggaagtcttacatcgaggcctacatgcaggaggccgagtggatcgcctctaacca
catccccggcttcgacgagtacctgatgaacggcgtgaagtcctctggcatgcgaatcctgatgatc
cacgccctgatcctgatggacacccccctgtctgacgagattctcgagcagctggacatcccctcgtc
taagtctcaggccctgctgtctctgatcacccgactggtggacgacgtgaaggacttcgaggacgag
caggcccacggcgagatggcctcttctatcgagtgctacatgaaggacaaccacggctctacccga
gaggacgccctgaactacctgaagatccgaatcgagtcttgcgtgcaggagctgaacaaggagctg
ctcgagccctctaacatgcacggatctttccgaaacctgtacctgaacgtgggaatgcgagtgatttt
cttcatgctgaacgacggcgacctgttcacccactctaaccgaaaggagatccaggacgccatcac
caagttcttcgtcgagcccatcatcccctga
atgcaattcttctccttggtttccatcttcttgttcttgtcctttttgtttttgttgagaaagtggaagaac
tccaactcccaatctaaaaagttgccaccaggtccatggaaattgccattattgggttctatgttgca
tatggttggtggtttgccacatcatgttttgagagatttggctaaaaagtacggtccattgatgcactt
gcaattgggtgaagtttctgctgttgttgttacttctccagatatggccaaagaagttttgaaaaccca
cgatattgctttcgcttctagaccaaagttgttggctccagaaatcgtttgttacaacagatctgatat
tgccttctgtccatacggtgattattggagacaaatgagaaagatctgcgtcttggaagttttgtctgc
taagaacgtcagatccttcagttccattagaagagatgaagtcttgagattggtcaacttcgttagat
cttctacttccgaaccagttaacttcaccgaaagattattcttgttcacctcttctatgacctgtagatc
tgcttttggtaaggttttcaaagaacaagaaaccttcatccaattgatcaaagaagtcattggtttgg
ctggtggttttgatgttgctgatattttcccatccttgaagttcttgcatgtcttgactggtatggaaggt
aagattatgaaggcccatcataaggttgatgccatcgttgaagatgttatcaacgaacacaaaaag
aacttggctatgggtaagactaatggtgctttgggtggtgaagatttgatcgatgttttgttaagattg
atgaacgatggtggtttacaattcccaatcaccaacgataacattaaggccatcatcttcgatatgtt
tgctgctggtactgaaacttcttcctctactttggtttgggctatggttcaaatgatgagaaacccaac
tattttggctaaggctcaagctgaagttagagaagcttttaagggtaaagaaactttcgacgaaaac
gacgtcgaagaattgaagtacttgaagttggttatcaaagaaacattgagattgcacccaccagttc
ctttgttggttccaagagaatgtagagaagaaaccgaaatcaacggttacaccattccagttaagac
caaggttatggttaatgtttgggctttgggtagagatccaaagtattgggatgatgctgataacttca
agccagaaagattcgaacaatgctccgttgactttatcggtaacaacttcgaatacttgccatttggt
ggtggtagaagaatttgtccaggtatttctttcggtttggccaatgtttatttgccattggctcaattgt
tgtaccacttcgattggaaattaccaacaggtatggaacctaaggatttggatttgactgaattggtc
ggtattaccattgccagaaagtccgatttgatgttagttgctactccataccaaccatccagagaata
a
30
AtCPR
atgacttctgccttgtatgcctctgatttgttcaagcaattgaagtccattatgggtactgactcattgt
ccgatgatgttgttttggttattgctactacctccttggctttggttgctggttttgttgttttattgtgga
aaaagaccaccgccgatagatctggtgaattgaaaccattgatgatcccaaagtctttgatggcca
aagatgaagatgatgatttggatttgggttccggtaagactagagtttctattttcttcggtactcaaa
ccggtactgctgaaggttttgctaaagctttgtctgaagaaatcaaggccagatacgaaaaagctgc
cgttaaggttatagatttggatgattatgctgccgatgacgaccaatacgaagaaaagttgaagaa
agaaaccttggccttcttctgtgttgctacttatggtgatggtgaacctactgataatgctgctagattt
tacaagtggttcactgaagaaaacgaaagagacatcaagttgcaacaattggcttacggtgtttttg
ctttgggtaacagacaatacgaacacttcaacaagatcggtatcgttttggatgaagaattgtgtaa
gaagggtgccaagagattgattgaagttggtttgggtgatgatgaccaatccatcgaagatgatttt
aacgcctggaaagaatccttgtggtctgaattggataagttgttgaaggacgaagatgacaaatct
gttgctacaccatacactgctgttatcccagaatatagagttgttacccatgatccaagattcaccac
tcaaaagtctatggaatctaacgttgctaacggtaacaccaccatcgatattcatcatccatgtaga
gttgatgtcgccgtccaaaaagaattgcatactcatgaatctgacagatcctgcatccatttggaatt
cgatatttccagaaccggtattacttacgaaaccggtgatcatgttggtgtttacgctgaaaatcacg
ttgaaatcgttgaagaagccggtaagttgttaggtcattccttagatttggttttctccatccatgccg
acaaagaagatggttctccattggaatctgctgttccaccaccatttccaggtccatgtactttgggta
ctggtttggctagatatgctgacttgttgaatccaccaagaaagtctgctttagttgctttggctgctta
tgctactgaaccatctgaagccgaaaaattgaaacatttgacttccccagatggtaaggacgaatat
tctcaatggatagttgcctcccaaagatccttgttggaagttatggctgcttttccatctgctaaacca
ccattgggtgttttttttgctgctattgctccaagattgcaacctagatattactccatttcctccagtcc
aagattagctccatcaagagttcatgttacatccgctttggtttatggtccaactccaactggtagaat
tcataagggtgtttgttctacctggatgaagaacgctgttccagctgaaaaatctcatgaatgttctg
gtgccccaattttcattagagcttctaatttcaagttgccatccaacccatctactccaatagttatggt
tggtccaggtacaggtttagctccttttagaggtttcttacaagaaagaatggccttgaaagaagat
ggtgaagaattgggttcctccttgttgttttttggttgcagaaacagacaaatggatttcatctatgaa
gatgaattgaacaacttcgtcgaccaaggtgttatctccgaattgattatggccttttctagagaagg
tgctcaaaaagaatacgtccaacacaagatgatggaaaaagccgctcaagtttgggacttgatcaa
agaagaaggttacttgtacgtttgcggtgatgctaaaggtatggctagagatgttcatagaacattg
cataccatcgttcaagaacaagaaggtgtctcatcttctgaagctgaagctatcgttaagaagttgc
aaactgaaggtagatacttgagagatgtctggtaa
195
31