Accepted Manuscript Title: Codon optimization, promoter and expression system selection that achieved high-level production of Yarrowia lipolytica lipase in Pichia pastoris Author: Wen-Jing Zhou Jiang-Ke Yang Lin Mao Li-Hong Miao PII: DOI: Reference:
S0141-0229(14)00194-X http://dx.doi.org/doi:10.1016/j.enzmictec.2014.10.007 EMT 8697
To appear in:
Enzyme and Microbial Technology
Received date: Revised date: Accepted date:
17-7-2014 19-10-2014 24-10-2014
Please cite this article as: Zhou W-J, Yang J-K, Mao L, Miao L-H, Codon optimization, promoter and expression system selection that achieved high-level production of Yarrowia lipolytica lipase in Pichia pastoris, Enzyme and Microbial Technology (2014), http://dx.doi.org/10.1016/j.enzmictec.2014.10.007 This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
1
Title: Codon optimization, promoter and expression system selection that achieved
2
high-level production of Yarrowia lipolytica lipase in Pichia pastoris
4
Authors: Wen-Jing Zhou, Jiang-Ke Yang, Lin Mao, Li-Hong Miao
cr
5
ip t
3
Address: College of Biological and Pharmaceutical Engineering, Wuhan Polytechnic
7
University, Wuhan 430023, China
us
6
12 13 14 15 16 17 18
M
86-27-83943875
d
11
Jiang-Ke Yang, Email: [email protected], Tel: (+) 86-27-83943875; Fax: (+)
te
10
Corresponding author:
Ac ce p
9
an
8
19 20 21 22
1
Page 1 of 31
23
Abstract Lipase (EC 3.1.1.3) stands amongst the most important and promising biocatalysts for
25
industrial applications. In this study, in order to realize a high-level expression of the Y.
26
lipolytica lipase gene in Pichia pastoris, we optimized the codon of LIP2 by de novo gene
27
design and synthesis, which significantly improved the lipase expression when compared to
28
the native lip2 gene. We also comparatively analyzed the effects of the promoter types
29
(PAOX1 and PFLD1) and the Pichia expression systems, including the newly developed
30
PichiaPink system, on lipase production and obtained the optimal recombinants. Bench-top
31
scale fermentation studies indicated that the recombinant carrying the codon-optimized
32
lipase gene syn-lip under the control of promoter PAOX1 has a significantly higher lipase
33
production capacity in the fermenter than other types of recombinants. After undergoing
34
methanol inducible expression for 96 h, the wet cell weight of Pichia, the lipase activity and
35
the protein content in the fermentation broth reached their highest values of 262 g/L, 38500
36
U/mL and 2.82 g/L, respectively. This study has not only greatly facilitated the
37
bioapplication of lipase in industrial fields but the strategies utilized, such as de novo gene
38
design and synthesis, the comparative analysis among promoters and different generations of
39
Pichia expression systems will also be useful as references for future work in this field.
40
Keywords: lipase, de novo design, synthesis, PichiaPink, fermentation
Ac ce p
te
d
M
an
us
cr
ip t
24
41 42 43 44
2
Page 2 of 31
45
1. Introduction Lipases (EC 3.1.1.3) from Yarrowia lipolytica are prolific biocatalyzers that are widely
47
used to catalyze the synthesis of functional esters and biofuels by esterification and
48
transesterification [1,2], as well as with the enantio- and regio-selective synthesis of
49
chemical compounds [3], to name a few uses. However, the high cost of Y. lipolytica lipase
50
remains a barrier for its utilization in industrial applications. In order to realize high-level
51
expression of Y. lipolytica lipases and facilitate their use in industrial applications, high-level
52
expression in industrial-grade hosts is generally the best choice. The Pichia expression
53
system is the most commonly used eukaryotic system for heterologous gene expression. It
54
not only shares similar advantages as the prokaryote systems, such as having a genetic
55
operability, an efficient expression capacity, fast reproduction, simple nutrient utilization and
56
a suitability for high-density fermentation, but also has eukaryotic cell characteristics, such
57
as correct post-translation modifications [4,5]. Since the first P. pastoris expression system
58
was commercially developed by the Phillips Petroleum Company, it was broadly used for
59
foreign gene expression, and now has almost become one of the most popular eukaryotic
60
expression systems in the world [6,4,7].
cr
us
an
M
d
te
Ac ce p
61
ip t
46
The efficient expression of heterologous genes, or the flow of genetic information from
62
DNA via mRNA to protein, can be affected by many factors. The promoter types
63
(constitutive or inducible) certainly affect the transcription efficiency. The translation of
64
genetic information from mRNA to protein is also affected by the GC content, intragenic
65
repeats, and complexity of the mRNA secondary structure. Complex mRNA secondary
66
structure and the uneven distribution of nucleotides could act as roadblocks that dramatically
3
Page 3 of 31
influence the rhythm of protein synthesis [8,9,10]. As with most organisms, Pichia displays a
68
non-random pattern of synonymous codon usage and shows a general bias towards a subset
69
of codons. This non-random pattern of synonymous codon usage and bias on a subset of
70
codons affects the heterogeneous gene expression in Pichia [11,12].
ip t
67
de novo gene design is a powerful measure for the codon optimization of the gene. Far
72
superior to other methods, such as point mutation for molecular rebuilding, de novo gene
73
design could substitute the most frequently used codons for less used codons while
74
considering multiple elements connected with gene expression in parallel. It can efficiently
75
enhance codon usage frequency, adjust the GC content, artificially add or delete specific
76
restriction sites or protease sensitive sites, and decrease the complexity of mRNA secondary
77
structures. Currently, this gene design and synthesis method has gradually been adopted as
78
an efficient measure to improve the expression level [13,14,15]. In this study, in order to
79
facilitate lipase industrial bioapplications, we optimized the codon of Y. lipolytica lipase
80
gene lip2 by de novo design and synthesis, comparatively analyzed the effects of the
81
promoter types (PAOX1 and PFLD1) and different Pichia systems on its expression, and
82
realized a high level expression of lip2 gene in Pichia.
us
an
M
d
te
Ac ce p
83
cr
71
84
2. Materials and Methods
85
2.1 Y. lipolytica lipase gene lip2 cloning and Pichia recombinant construction
86
Y. lipolytica strain CICC32187 was purchased from the China Center of Industrial
87
Culture Collection, CICC. Primers YLF1 (5’-GAATTCGTGTACACCTCTACCGAGAC-3’,
88
EcoRI) and YLR1 (5’-GCGGCCGCTTAGATACCACAGACACCCTC-3’, NotI) were used
4
Page 4 of 31
89
to amplify the lipase gene lip2 with genomic DNA of strain CICC32187 as the template. The
90
amplified lip2 gene was verified by DNA sequencing, and the sequence of lip2 was
91
submitted into GenBank with accession number KJ603521. PCR product was double digested by EcoRI and NotI enzymes and then ligated into
93
Pichia expression vector pPICZA, which carries the methanol inducible expression
94
promoter PAOX1, to form the recombinant plasmid pPIC-lip2. pPIC-lip2 was then transferred
95
into Pichia X-33 cells to obtain the Pichia transformants X-33 (PAOX1-LIP2). The steps
96
detailing the Pichia competent cell preparation, electroporation, positive colony screening
97
and verification were conducted mainly according to the descriptions of Cregg [16].
an
us
cr
ip t
92
99
M
98
2.2 de novo lipase gene design, synthesis and recombinant construction Based on the native amino acid sequence of Y. lipolytica LIP2 lipase (GenBank:
101
AIB08846), we de novo designed the nucleotide sequence of syn-lip. With the assistance of
102
DNA2.0 software (http://www.dna20.com), the more frequently used codons in Pichia were
103
selected from synonymous codons to replace the less frequently used codons. The secondary
104
structure and the minimal free energy of the mRNA was calculated by RNAfold software
105
(http://rna.tbi.univie.ac.at/cgi-bin/RNAfold.cgi). The codon usage frequency in the native
106
lip2 gene and the designed gene were analyzed online by graphical codon usage analyzer
107
software 2.0 (http://gcua.schoedl.de/). The codon usage frequency, the codon adaptation
108
index (CAI), the frequency of optimal codons (FOP) and the GC content of the original lip2
109
gene and designed syn-lip gene are shown in Supplementary data Fig. S1 through S6.
110
Ac ce p
te
d
100
To synthesize the designed lipase gene, a batch of oligonucleotides (see Supplementary
5
Page 5 of 31
data Table S1, Fig. S3) was designed by Gene2Oligo software [17] to make the
112
thermodynamic properties of each oligonucleotide consistent. A two-step gene synthesis
113
method was used to assemble these oligonucleotides into the full length lipase gene. This
114
process was conducted according to the descriptions of Yang et al. [18]. The gene synthesis
115
flowchart is shown in Supplementary data Fig. S7.
cr
ip t
111
The full-length synthesized lipase gene syn-lip gene was ligated into a pMD-18T vector
117
and the insertion was verified by sequencing. A fragment of syn-lip was digested from the
118
pMD-18T vector by EcoRI and NotI and then ligated into a pPICZA vector to form the
119
recombinant pPIC-syn-lip. The recombinant was then transferred into the Pichia strain X-33
120
to obtain the yeast transformant carrying methanol inducible expression syn-lip gene.
M
an
us
116
121
d
123
2.3 Promoter PFLD1 cloning and recombinant construction The promoter region (PFLD1) of formaldehyde dehydrogenase gene (FLD) was amplified
te
122
by using the primer pair FLD5BZ (5’-GCGAGATCTGCATGCAGGAATCTCTGG-3’, Bgl II)
125
and FLD3HZ (5’- GCGAAGCTTTGTGAATATCAAGAATTGTATGAACAAGC-3’, Hind
126
III) with the genome DNA of P. pastoris X-33 as the template. After verification by DNA
127
sequencing, the PFLD1 fragment was inserted into a pPICZA vector by using restriction
128
enzymes BglII and HindIII to replace the original AOX1 promoter in pPICZA to form the
129
PFLD1 initiated vector pPICFLD. The codon optimized lipase gene syn-lip was then inserted
130
into this vector by the EcoRI and NotI sites to form the recombinant PFLD1-syn-lip. After the
131
transfer into the strain X-33, a Pichia transformant carrying a PFLD1 controlled lipase gene
132
was obtained.
Ac ce p
124
6
Page 6 of 31
133 134
2.4 Recombinant construction by using the PichiaPink system The Invitrogen newly developed yeast expression system, PichiaPink, was used in this
136
study to construct the yeast recombinants suitable for the large-scale expression and
137
secretion of lipase. Vector pPink-HC, containing a Saccharomyces cerevisiae -mating
138
factor prepro-sequence and a selection marker gene ADE2, and PichiaPink Strain 1 (ade2)
139
were
140
(5’-CTAGGCCTGTTTACACTTCTACTGAGAC-3’,
141
(5’-CTGCCGGCGATACCACAAACACCCTC-3’, NaeI) were used to amplify the syn-lip
142
gene and then were inserted into the pPink-HC vector by StuI and NaeI sites to construct
143
recombinant pPink-syn-lip.
the
secreted
expression
of
lipase.
Primer
us
for
StuI)
and
pairs
SLPF SLPR
M
an
selected
cr
ip t
135
The operations on the Pichia competent cell preparation, electroporation and positive
145
colony screenings were conducted mainly according to the manufacturer’s instructions
146
(Invitrogen). Briefly, approximately 5 g of plasmid was linearized by Spe I, which cut at
147
the TRP2 region of the PichiaPink vectors, and then was transferred into Strain 1 (ade2) by
148
electroporation with the conditions set at 2 kV, 25 , and 200 F for a 0.2 cm
149
electroporation cuvette (Electroporator 2510, Eppendorf). PAD (Pichia Adenine Dropout)
150
agar was used for the screening of positive PichiaPink transformants by ADE2
151
complementation. The insertion of the syn-lip gene into the PichiaPink genome was verified
152
by PCR with the primer pair SLPF/SLPR.
Ac ce p
te
d
144
153 154
2.5 Methanol inducible expression of lipase
7
Page 7 of 31
A single colony of Pichia transformants PAOX1-lip2, PAXO1-syn-lip, PFLD1-syn-lip and
156
PichiaPinK (PAOX1-syn-lip) was deposited into 20 mL of YPD medium (yeast extract 1%,
157
peptone 2%, dextrose 2%) and cultured at 28°C overnight. The Pichia culture was then
158
inoculated into 50 mL of BMGY medium (yeast extract 1%, peptone 2%, 100 mM potassium
159
phosphate buffer with pH 6.0, yeast nitrogen base 1.34%, glycerol 1%) at a 1:50 ratio and
160
incubated at 28°C with a shaking speed of 200 rpm for approximately 20 h to reach an
161
OD600=4.0. The cells were harvested and transferred into 50 mL of BMMY medium (yeast
162
extract 1%, peptone 2%, 100 mM potassium phosphate buffer with pH 6.0, yeast nitrogen
163
base 1.34%, 0.5% methanol) for the methanol inducible expression of lipase. Methanol was
164
added into the medium every 24 h to reach a final concentration of 0.5% to induce the
165
expression of lipase. Lipase activity, protein content and wet cell weight in the medium were
166
checked at predetermined intervals.
d
M
an
us
cr
ip t
155
2.6 Glycosylation site prediction and PNGase digestion of PichiaPink expression product
Ac ce p
168
te
167
The NetNGlyc 1.0 server (www.cbs.dtu.dk/services/NetNGlyc/) was used to analysis
169 170
the
N-glycosylation
sites,
and
the
GlycoEP
server
171
(http://www.imtech.res.in/raghava/glycoep/) was used to predict the O-glycosylation site in
172
the LIP2 protein.
173
N-Glycosidase F (PNGase F, New England Biolabs) was used to hydrolyze the
174
N-glycan chain from the PichiaPink expression product of LIP2. The PichiaPink expressed
175
protein in the fermentation broth was first concentrated by using an Amicon Ultra-15 ml
176
centrifugal filter unit with 10 kDa filter (Millipore), and then twice dialyzed in a protein
8
Page 8 of 31
buffer (0.05 M sodium phosphate 50 mM, DTT 0.05 mM). Approximately 20 l aliquots of
178
the protein solution were mixed with a glycoprotein denaturing buffer (5% SDS, 0.4 M DTT)
179
and then incubated in 100 °C bath for 10 min. The denatured protein solution was then added
180
into the G7 reaction buffer (0.5 M sodium phosphate, pH 7.5), 10% NP-40, and 2 l PNGase
181
F to make a 20 l volume deglycosidation mixture. The reaction mixture was incubated at
182
37 °C for 1 h and then loaded into the gel for SDS-PAGE analysis.
2.7 Recombinant fermentation in a bench-top fermenter
an
184
us
183
cr
ip t
177
Pichia transformants X-33 (PAOX1-lip2) and X-33 (PAOX1-syn-lip) were selected for the
186
bench-top scale lipase production assay. The fermentation was conducted in a 14 L
187
bioreactor (BioFlo 415, New Brunswick, Canada) with a medium of basal salts (H2PO4 26.7
188
mL, CaSO4 0.93 g, K2SO4 18.2 g, MgSO4•7H2O 14.9 g, KOH 4.13 g, glycerol 40.0 g, per
189
liter). The fermentation parameters were monitored and controlled throughout the whole
190
process. In the biomass accumulation phase, the fermentation parameters were maintained as:
191
temperature (28.0oC), dissolved oxygen (DO, >40%), pH=6.0, agitation ratio (rpm, 400-650)
192
and aeration (0.1-1.0 vvm). In the methanol inducible expression phase, the fermentation
193
parameters were maintained as: temperature (25.0oC), dissolved oxygen (DO, >35%),
194
pH=6.0, agitation ratio (rpm, 500-850) and aeration (0.1-1.0 vvm). The amount of methanol
195
fed into the system was changed according to the set time intervals during this phase. In the
196
first 24 h, methanol was fed into the broth with the ratio of 2 mL/h per liter of broth. In the
197
next 24 h, the amount of methanol fed into the system was set to 4 mL/h per liter of broth,
198
and then increased to 5.5 mL/h per liter of broth, which was maintained until the end of the
Ac ce p
te
d
M
185
9
Page 9 of 31
199
fermentation. Samples were collected at intervals, and the wet cell weight, lipase activity and
200
protein content in the broth were analyzed for each sample.
202
2.8 Wet cell weight, lipase activity and protein content assays
ip t
201
To measure the yeast wet cell weight, an aliquot of 50 mL broth and the cell pellet were
204
harvested by centrifuge. The wet cell weight was calculated as the mass of the cells in grams
205
per liter. The lipase activity in the broth was determined at pH 7.5 by free fatty acid titration
206
using 50 mM NaOH after incubation in a thermostated vessel for 10 min. The assay mixture
207
consisted of 5.9 mL of Tris-HCl buffer (50 mM, pH 7.5), 50 mM NaCl, 4 mL of emulsified
208
tributyrin and 0.1 mL of a diluted enzyme solution. One unit (U) of activity was defined as
209
the amount of enzyme liberating 1 micromole of fatty acid per min at 40°C. The protein
210
content in the fermentation broth was determined by the Bradford method and the expression
211
profiles were checked by SDS-PAGE.
us
an
M
d
te
Ac ce p
212
cr
203
213
3. Results and discussions
214
3.1 de novo design and synthesis of the lipase gene
215
Considering the codon usage bias between P. pastoris and Y. lipolytica, the codon of Y.
216
lipolytica lipase gene lip2 was optimized based on the amino acid sequence of LIP2
217
(AIB08846) to realize a high level expression of lipase in Pichia. We first replaced the less
218
frequently used codons with more frequently used ones (Fig. 1, Supplementary data Fig. S1
219
and S2). As reported before, the codon order, such as being either AT-rich or GC-rich in the
220
region, and the complexity of the mRNA secondary structures can act as roadblocks that
10
Page 10 of 31
might influence the rhythm of protein synthesis, decrease the translation efficiency, or even
222
cause the premature termination of translation [19]. To make the nucleotides A, T, G and C
223
evenly distribute in the sequence, which could also be used to deduce the complexity of the
224
secondary structure of mRNA, more frequently used codons may also be selected to make
225
different types of nucleotides distribute evenly, as well as to keep the GC content of the gene
226
within the range of 45-60%. After codon optimization, the codon adaptation index (CAI)
227
changed from 0.74 of the native lip2 gene to the optimized 0.87, which means the codons
228
were well adapted to the Pichia expression system (Supplementary data Fig. S4). As
229
indicated by the frequency of the optimal codons (FOP) parameter, in the original gene,
230
22.5% of the codons have the usage frequency 90%. While in optimized syn-lip, approximately 60% of codons have the
232
usage frequency >90%. (Supplementary data Fig. S5). The GC content changed from the
233
original 47.5% to the optimized 50.0%, and the GC-rich regions in the native lip2 with peaks
234
higher than 60% were eliminated (Supplementary data Fig. S6). As shown by Figure 2, after
235
re-designing, the ensemble diversity of the first 200 nucleotides decreased from 38.7 to 25.6,
236
and the minimal free energy of the original gene significantly changed from the original
237
value of -45.00 kcal/mol to -34.5 kcal/mol (Fig. 2).
239
cr
us
an
M
d
te
Ac ce p
238
ip t
221
3.2 Comparative assay of lipase expression between native and codon optimized genes
240
The lipase production capacities of native lip2 and the codon optimized syn-lip gene
241
were evaluated in flasks in this study, and the wet cell weight, lipase activity and protein
242
content in the fermentation broth were calculated. As shown in figure 3, the recombinant
11
Page 11 of 31
carrying the codon-optimized lipase gene (syn-lip) demonstrated a much stronger lipase
244
secretion capacity than the cells with the native gene (lip2). After methanol-inducible
245
expression was allowed to occur for 96 h, the lipase activity and protein content of the
246
syn-lip gene reached 5800 U/mL and 0.55 g/L in the fermentation broth, respectively. In
247
contrast, recombinants (pPIC-lip) carrying the native lip2 gene reached levels of only 4700
248
U/mL and 0.46 g/L, respectively (Fig. 3). Furthermore, as reflected by figure 3A, the
249
biomass of the recombinant carrying syn-lip was approximately 102 g/L at 96 h, which is
250
lower than the value of the native lip2 gene of 120 g/L. Thus the lipase production efficiency
251
of syn-lip was pronouncedly higher than the native lip2 gene if the production was calculated
252
on a per unit of yeast cell biomass basis.
3.3 Comparative assay of promoter types on lipase expression
d
254
M
253
an
us
cr
ip t
243
In the metabolism process of methanol in the Pichia cell, alcohol oxidases first oxidize
256
methanol into formaldehyde. The alcohol oxidases possess a very low affinity for molecular
257
oxygen. In order to compensate for their poor capacity for carrying oxygen, P. pastoris
258
produces large amounts of the enzymes by the promoters PAOXI and PAOXII [20]. These
259
methanol inducible promoters could be used to drive the heterologous gene expression in
260
Pichia, and generally it yields a higher level of expression. Glutathione-dependent
261
formaldehyde dehydrogenase (FLD) is another key enzyme required for the metabolism of
262
methanol, as a carbon source, and certain alkylated amines, as nitrogen sources. The
263
promoter of formaldehyde dehydrogenase 1 (PFLD1) was strongly and independently induced
264
by either methanol as a sole carbon source or methylamine as a sole nitrogen source [21].
Ac ce p
te
255
12
Page 12 of 31
Under the control of PFLD1, the expression level of the foreign gene was generally
266
comparable to those obtained with the commonly used alcohol oxidase I gene promoter
267
(PAOX1), which is now becoming an attractive alternative to PAOX1 in P. pastoris [22,23,24].
268
In this study, we comparatively analyzed the lipase production capacity of the Pichia
269
recombinants PFLD1-syn-lip and PAOX1-syn-lip in flasks under the 0.5% methanol induction.
270
As reflected by figure 4, the PFLD1 controlled recombinant (PFLD1-syn-lip) showed a
271
considerable lipase production capacity. After methanol inducible expression was allowed to
272
occur for 96 h, the lipase activity and protein content in the broth reached 4950 U/mL and
273
0.48 g/L, respectively. For the recombinant PAOX1-syn-lip, the lipase activity and the protein
274
content reached the highest values at 5650 U/mL and 0.54 g/L, respectively, at the 96 h time
275
point (Fig. 4). Although this study further demonstrated that PFLD1 is an efficient promoter
276
for the expression of heterologous proteins in P. pastoris and that the expression levels of the
277
lipase of the PFLD1-controlled system may be comparable to the PAOX1-controlled system, we
278
also noticed that over the timeframe of the whole experiment, the lipase production capacity
279
of PAOX1-syn-lip was always higher than that of PFLD1-syn-lip. So in this study, the
280
recombinant under the control of PAOX1 (PAOX1-syn-lip) was still selected for its stronger
281
lipase production capacity.
283
cr
us
an
M
d
te
Ac ce p
282
ip t
265
3.4 Comparative assay of lipase expression between different Pichia expression systems
284
To facilitate a larger scale expression of exogenous protein, Invitrogen has developed a
285
novel Pichia eukaryotic expression system, PichiaPink, by using ADE2 as selectable marker
286
gene. In this study, we compared the lipase production capacity between recombinants of
13
Page 13 of 31
X-33 (PAOX1-syn-LIP) and PichiaPink (pPink-syn-lip). As shown in figure 5, the
288
lipase-producing capacity of these two types of recombinants was almost equal to each other
289
and there was no distinct difference in the lipase activity and protein content between them
290
throughout the entire time of the experiment (Fig. 5B and 5C). The total protein profiles in
291
PichiaPink (pPink-syn-lip) fermentation broth detected by SDS-PAGE showed that the
292
specific band of lipase with a size of approximately 37 kDa was apparently much weaker
293
than the X-33 (PAOX1-syn-lip) recombinants, although the detected total protein and lipase
294
activities were equivalent with each other (Fig. 5D). We speculated that this might be caused
295
by the excess glycosylation of the partial Lip2 protein in PichiaPink system.
an
us
cr
ip t
287
We further checked the whole SDS-PAGE gels of the fermentation broth and found that,
297
when compared with the protein profile of the X-33 (PAOX1-syn-lip) fermentation broth,
298
which had a specific lipase band (approximately 37 kDa) and a clear background, PichiaPink
299
(pPink-syn-lip) had a smeared background and a blurred band with a size of approximately
300
97 kDa (Fig. 6A and 6B), which might be the excess glycosylated LIP2 band. N-Glycosidase
301
F was used to digest the lip2 gene produced from PichiaPink. As shown in Fig. 6C, PNGase
302
F could efficiently digest the PichiaPink expression product, including the 97 KDa smear
303
band into two bands, where the small band possessed a size of approximately 32 kDa, which
304
coincided with the speculated size of the Lip2 without glycosylation, and a bigger band with
305
a size of approximately 40 kDa.
Ac ce p
te
d
M
296
306
We further analyzed the N- and O-glycosylation site by the NetNGlyc 1.0 server and
307
GlycoEP server and found that there are two N-glycosylation sites, a strong N-glycosylation
308
site N113 and a weak site N134, and two O-glycosylation sites (T159 and T295) that exist in
14
Page 14 of 31
the Lip2 protein (Fig. 6D). The strong N-glycosylation site (N113) might be generally
310
slightly glycosylated in a host such as X-33. Considering PNGase F can efficiently cleave
311
between the innermost GlcNAc and asparagine residues and can hydrolyze nearly all types
312
of N-glycan chains from glycopeptides and proteins, we believe that the bigger band
313
(approximately 40 kDa) in PNGase F digested products comes from the O-glycosylation of
314
Lip2 in the PichiaPink system.
cr
ip t
309
3.5 Lipase production of the recombinants in the bench-top fermenter
an
316
us
315
Scale-up experiments with a bench-top fermenter were performed on the recombinants
318
carrying the native lip2 and codon-optimized syn-lip gene in this study (Fig. 7). With better
319
parameter control and a better feed of methanol into the fermenter, the wet cell weights,
320
lipase activities and protein contents in the fermentation broth from the fermenter were
321
significantly higher than those in the flasks. Both the recombinants X-33(PAOX1-lip2) and
322
X-33(PAOX1-syn-lip) had wet cell weights higher than 250 g/L and lipase activities higher
323
than 15000 U/mL. When compared with the native lip2 gene, the lipase activity and protein
324
content in the fermentation broth of the codon-optimized syn-lip X-33(PAOX1-syn-lip) their
325
highest values at 38,500 U/mL and 2.82 g/L, which is significantly higher than the native
326
lip2 values, where the lipase activity and the protein content was 17,500 U/mL and 1.68 g/L,
327
respectively (Fig. 7).
Ac ce p
te
d
M
317
328
Currently, several studies have been conducted on the expression of the Y. lipolytica
329
lipase gene. By using methods such as multiple copy integration, Pignède et al. [25]
330
achieved a level of 0.5 g of active Y. lipolytica lipase per liter in the supernatant. Through the
15
Page 15 of 31
optimization of fermentation parameters, Yu et al. [26] achieved a lipase activity of 12,500
332
U/mL, and by using the constitutive expression promoter PGAP, Wang et al. [27] has realized
333
lip2 expression in P. pastoris to the level of 13,500 U/mL in a 10-L fermenter. The
334
expression levels achieved by these studies were almost equivalent to the level seen in our
335
Pichia recombinants carrying the native lip2 gene. Through de novo design and synthesis of
336
the lipase gene, a comparative selection between the promoter types and the Pichia
337
expression system, the expression level in X-33 (PAOX1-syn-lip) reached levels far higher
338
than in previous works.
an
us
cr
ip t
331
339
4. Conclusions
M
340
Through de novo gene design and synthesis of syn-lip, a comparative analysis of the
342
lipase production capacity between the promoters PAOX1 and PFLD1, and the different Pichia
343
expression system, we have successfully obtained the highest level of expression in the
344
Pichia recombinants of X-33(PAOX1-syn-lip). After a fermentation assay in a bench-top scale
345
fermenter, the lipase production capacity reached the highest value of 38,500 U/mL in the
346
fermentation broth. This study has greatly facilitated the bioapplication of lipase in industrial
347
fields. In addition, the strategies utilized in this study, such as the use of de novo gene design
348
and synthesis, a comparative analysis among promoters and different generation of Pichia
349
expression systems, will also be utilized as references for future work in this field.
Ac ce p
te
d
341
350 351 352
Acknowledgments We thank J. S. Cheng and C. Hu for their help in the conduction of the experiments.
16
Page 16 of 31
This work was financially supported by the Science and Technology Supporting Program of
354
Hubei Province (2014), the National High Technology Research and Development Program
355
of China (no. 2013AA102805-4), the Key Project of Science and Technology Research
356
Program of the Education Department of Hubei Province (no. D20131703), and the Key
357
Project of Chinese Ministry of Education (no. 212118), where Dr. JK Yang is an incumbent
358
member of the Chutian Scholar Program.
cr
ip t
353
us
359
References
361
[1] Meng Y, Wang G, Yang N, Zhou Z, Li Y, Liang X, Chen J, Li Y, Li J. Two-step
362
synthesis of fatty acid ethyl ester from soybean oil catalyzed by Yarrowia lipolytica
363
lipase. Biotechnol Biofuels. 2011;4(1):6.
M
an
360
[2] Casas-Godoy L, Meunchan M, Cot M, Duquesne S, Bordes F, Marty A. Yarrowia
365
lipolytica lipase Lip2: an efficient enzyme for the production of concentrates of
366
docosahexaenoic acid ethyl ester. J Biotechnol. 2014;180:30−36.
Ac ce p
te
d
364
367
[3] Cambon E, Piamtongkam R, Bordes F, Duquesne S, André I, Marty A. Rationally
368
engineered double substituted variants of Yarrowia lipolytica lipase with enhanced
369 370 371 372 373 374
activity coupled with highly inverted enantioselectivity towards 2-bromo phenyl acetic acid esters. Biotechnol Bioeng. 2010;106(6):852−859.
[4] Cregg JM, Vedvick TS, Raschke WC. Recent advances in the expression of foreign genes in Pichia pastoris. Biotechnol (N Y). 1993;11:905−910. [5] Cregg JM. Introduction: distinctions between Pichia pastoris and other expression systems. Methods Mol Biol 2007a;389:1−10.
17
Page 17 of 31
375 376
[6] Vedvick TS. Gene expression in yeast: Pichia pastoris. Curr Opin Biotechnol 1991;2:742−745. [7] Gasser B, Prielhofer R, Marx H, Maurer M, Nocon J, Steiger M, Puxbaum V, Sauer M,
378
Mattanovich D. Pichia pastoris: protein production host and model organism for
379
biomedical research. Future Microbiol 2013;8:191−208.
cr
381
[8] Zur, H. Tuller T. Strong association between mRNA folding strength and protein abundance in S. cerevisiae. EMBO Rep. 2012;13,272−277.
us
380
ip t
377
[9] Park C, Chen X, Yang JR, Zhang J. Differential requirements for mRNA folding
383
partially explain why highly expressed proteins evolved slowly. Proc Natl Acad Sci
384
USA 2013;110,E678−E686.
M
an
382
[10] Mao Y, Liu H, Liu Y, Tao S. Deciphering the rules by which dynamics of mRNA
386
secondary structure affect translation efficiency in Saccharomyces cerevisiae. Nucleic
387
Acids Res. 2014, 42(8):4813−4822.
te
d
385
[11] Sinclair G, Choy FY. Synonymous codon usage bias and the expression of human
389
glucocerebrosidase in the methylotrophic yeast, Pichia pastoris. Protein Expr Purif
390
Ac ce p
388
2002;26(1):96−105.
391
[12] Mirzaei SA, Yazdi MT, Sepehrizadeh Z. Secretory expression and purification of a
392
soluble NADH cytochrome b5 reductase enzyme from Mucor racemosus in Pichia
393
pastoris based on codon usage adaptation. Biotechnol Lett 2010;32(11):1705−1711.
394
[13] Öberg F, Sjöhamn J, Conner MT, Bill RM, Hedfalk K. Improving recombinant
395
eukaryotic membrane protein yields in Pichia pastoris: the importance of codon
396
optimization and clone selection. Mol Membr Biol 2011;28(6):398−411.
397
[14] Quan J, Saaem I, Tang N, Ma S, Negre N, Gong H, White KP, Tian J. Parallel on-chip
398
gene synthesis and application to optimization of protein expression. Nat Biotechnol 18
Page 18 of 31
399
2011;29:449–452. [15] Yang JK, Liu LY, Dai JH, Li Q. de novo design and synthesis of Candida antarctica
401
lipase B gene and α-factor leads to high-level expression in Pichia pastoris. PLoS One
402
2013;8(1):e53939.
ip t
400
[16] Cregg JM. Pichia ptotocols. 2nd 2007b, The Humana Press, Totowa, NJ.
404
[17] Rouillard JM, Lee W, Truan G, et al. Gene2Oligo: oligonucleotide design for in vitro
406
gene synthesis. Nucleic Acids Res, 2004;32:W176−180.
us
405
cr
403
[18] Yang JK, Chen FY, Yan XX, Miao LH, Dai JH. A simple and accurate two-step long DNA sequences synthesis strategy to improve heterologous gene expression in Pichia.
408
PLoS ONE 2012;7(5):e36607.
410
[19] Richard P, Manley JL. Transcription termination by nuclear RNA polymerases. Genes Dev 2009, 23(11): 1247−1269.
M
409
an
407
[20] Cregg JM, Madden KR, Barringer KJ, Thill GP, Stillman CA. Functional
412
characterization of the two alcohol oxidase genes from the yeast Pichia pastoris. Mol
413
Cell Biol 1989;9(3):1316–1323.
Ac ce p
te
d
411
414
[21] Shen S, Sulter G, Jeffries TW, Cregg JM. A strong nitrogen source-regulated promoter
415
for controlled expression of foreign genes in the yeast Pichia pastoris. Gene
416
1998;216(1):93−102.
417
[22] Resina D, Serrano A, Valero F, Ferrer P. Expression of a Rhizopus oryzae lipase in
418
Pichia pastoris under control of the nitrogen source-regulated formaldehyde
419
dehydrogenase promoter. J Biotechnol 2004;109(1-2):103−113.
420
[23] Daly R, Hearn MT. Expression of heterologous proteins in Pichia pastoris: a useful
421
experimental tool in protein engineering and production. J Mol Recognit
19
Page 19 of 31
422
2005;18:119−138. [24] Resina D, Maurer M, Cos O, Arnau C, Carnicer M, Marx H, Gasser B, Valero F,
424
Mattanovich D, Ferrer P. Engineering of bottlenecks in Rhizopus oryzae lipase
425
production in Pichia pastoris using the nitrogen source-regulated FLD1 promoter. N
426
Biotechnol 2009;25(6):396−403.
cr
[25] Pignède G, Wang HJ, Fudalej F, Seman M, Gaillardin C, Nicaud JM. Autocloning and amplification
of
LIP2
429
2000;66(8):3283−3289.
in
Yarrowia
lipolytica.
Appl
Environ
Microbiol
an
428
us
427
ip t
423
[26] Yu M, Lange S, Richter S, Tan T, Schmid RD. High-level expression of extracellular
431
lipase Lip2 from Yarrowia lipolytica in Pichia pastoris and its purification and
432
characterization. Protein Expr Purif 2007;53(2): 255−263.
M
430
[27] Wang X, Sun Y, Ke F, Zhao H, Liu T, Xu L, Liu Y, Yan Y. Constitutive expression of
434
Yarrowia lipolytica lipase LIP2 in Pichia pastoris using GAP as promoter. Appl
435
Biochem Biotechnol 2012;166(5):1355−1367.
437 438 439 440
te
Ac ce p
436
d
433
441 442 443 444 445 20
Page 20 of 31
Fig. 1. Sequence comparison between the native Y. lipolytica lipase gene lip2 and the
447
designed lipase gene syn-lip. Bars in same color represent the same nucleotides between the
448
native and codon optimized genes.
449
Fig. 2. Secondary structure of the first 200 bp of the original lip2 and codon-optimized
450
syn-lip mRNA generated by the software RNAfolder. A. Structure of the original lip2 mRNA
451
where the MFE is -45.00 kcal/mol and the ensemble diversity is 38.73. (B) Structure of the
452
codon optimized syn-lip mRNA where the MFE is -34.5 kcal/mol and the ensemble diversity
453
is 25.69.
454
Fig. 3. Expression profiles of the yeast recombinants carrying the original lip2 and codon
455
optimized syn-lip gene. Pichia wet cell weight (A), lipase activity (B), and protein content
456
(C) in the fermentation broth of the lip2 and syn-lip recombinants were calculated in this
457
study. Protein profiles in the broth were checked by SDS-PAGE, and approximately 10 L of
458
the protein solution per sample was loaded into the gels (D). * indicates the significant
459
difference (P
S0141-0229(14)00194-X http://dx.doi.org/doi:10.1016/j.enzmictec.2014.10.007 EMT 8697
To appear in:
Enzyme and Microbial Technology
Received date: Revised date: Accepted date:
17-7-2014 19-10-2014 24-10-2014
Please cite this article as: Zhou W-J, Yang J-K, Mao L, Miao L-H, Codon optimization, promoter and expression system selection that achieved high-level production of Yarrowia lipolytica lipase in Pichia pastoris, Enzyme and Microbial Technology (2014), http://dx.doi.org/10.1016/j.enzmictec.2014.10.007 This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
1
Title: Codon optimization, promoter and expression system selection that achieved
2
high-level production of Yarrowia lipolytica lipase in Pichia pastoris
4
Authors: Wen-Jing Zhou, Jiang-Ke Yang, Lin Mao, Li-Hong Miao
cr
5
ip t
3
Address: College of Biological and Pharmaceutical Engineering, Wuhan Polytechnic
7
University, Wuhan 430023, China
us
6
12 13 14 15 16 17 18
M
86-27-83943875
d
11
Jiang-Ke Yang, Email: [email protected], Tel: (+) 86-27-83943875; Fax: (+)
te
10
Corresponding author:
Ac ce p
9
an
8
19 20 21 22
1
Page 1 of 31
23
Abstract Lipase (EC 3.1.1.3) stands amongst the most important and promising biocatalysts for
25
industrial applications. In this study, in order to realize a high-level expression of the Y.
26
lipolytica lipase gene in Pichia pastoris, we optimized the codon of LIP2 by de novo gene
27
design and synthesis, which significantly improved the lipase expression when compared to
28
the native lip2 gene. We also comparatively analyzed the effects of the promoter types
29
(PAOX1 and PFLD1) and the Pichia expression systems, including the newly developed
30
PichiaPink system, on lipase production and obtained the optimal recombinants. Bench-top
31
scale fermentation studies indicated that the recombinant carrying the codon-optimized
32
lipase gene syn-lip under the control of promoter PAOX1 has a significantly higher lipase
33
production capacity in the fermenter than other types of recombinants. After undergoing
34
methanol inducible expression for 96 h, the wet cell weight of Pichia, the lipase activity and
35
the protein content in the fermentation broth reached their highest values of 262 g/L, 38500
36
U/mL and 2.82 g/L, respectively. This study has not only greatly facilitated the
37
bioapplication of lipase in industrial fields but the strategies utilized, such as de novo gene
38
design and synthesis, the comparative analysis among promoters and different generations of
39
Pichia expression systems will also be useful as references for future work in this field.
40
Keywords: lipase, de novo design, synthesis, PichiaPink, fermentation
Ac ce p
te
d
M
an
us
cr
ip t
24
41 42 43 44
2
Page 2 of 31
45
1. Introduction Lipases (EC 3.1.1.3) from Yarrowia lipolytica are prolific biocatalyzers that are widely
47
used to catalyze the synthesis of functional esters and biofuels by esterification and
48
transesterification [1,2], as well as with the enantio- and regio-selective synthesis of
49
chemical compounds [3], to name a few uses. However, the high cost of Y. lipolytica lipase
50
remains a barrier for its utilization in industrial applications. In order to realize high-level
51
expression of Y. lipolytica lipases and facilitate their use in industrial applications, high-level
52
expression in industrial-grade hosts is generally the best choice. The Pichia expression
53
system is the most commonly used eukaryotic system for heterologous gene expression. It
54
not only shares similar advantages as the prokaryote systems, such as having a genetic
55
operability, an efficient expression capacity, fast reproduction, simple nutrient utilization and
56
a suitability for high-density fermentation, but also has eukaryotic cell characteristics, such
57
as correct post-translation modifications [4,5]. Since the first P. pastoris expression system
58
was commercially developed by the Phillips Petroleum Company, it was broadly used for
59
foreign gene expression, and now has almost become one of the most popular eukaryotic
60
expression systems in the world [6,4,7].
cr
us
an
M
d
te
Ac ce p
61
ip t
46
The efficient expression of heterologous genes, or the flow of genetic information from
62
DNA via mRNA to protein, can be affected by many factors. The promoter types
63
(constitutive or inducible) certainly affect the transcription efficiency. The translation of
64
genetic information from mRNA to protein is also affected by the GC content, intragenic
65
repeats, and complexity of the mRNA secondary structure. Complex mRNA secondary
66
structure and the uneven distribution of nucleotides could act as roadblocks that dramatically
3
Page 3 of 31
influence the rhythm of protein synthesis [8,9,10]. As with most organisms, Pichia displays a
68
non-random pattern of synonymous codon usage and shows a general bias towards a subset
69
of codons. This non-random pattern of synonymous codon usage and bias on a subset of
70
codons affects the heterogeneous gene expression in Pichia [11,12].
ip t
67
de novo gene design is a powerful measure for the codon optimization of the gene. Far
72
superior to other methods, such as point mutation for molecular rebuilding, de novo gene
73
design could substitute the most frequently used codons for less used codons while
74
considering multiple elements connected with gene expression in parallel. It can efficiently
75
enhance codon usage frequency, adjust the GC content, artificially add or delete specific
76
restriction sites or protease sensitive sites, and decrease the complexity of mRNA secondary
77
structures. Currently, this gene design and synthesis method has gradually been adopted as
78
an efficient measure to improve the expression level [13,14,15]. In this study, in order to
79
facilitate lipase industrial bioapplications, we optimized the codon of Y. lipolytica lipase
80
gene lip2 by de novo design and synthesis, comparatively analyzed the effects of the
81
promoter types (PAOX1 and PFLD1) and different Pichia systems on its expression, and
82
realized a high level expression of lip2 gene in Pichia.
us
an
M
d
te
Ac ce p
83
cr
71
84
2. Materials and Methods
85
2.1 Y. lipolytica lipase gene lip2 cloning and Pichia recombinant construction
86
Y. lipolytica strain CICC32187 was purchased from the China Center of Industrial
87
Culture Collection, CICC. Primers YLF1 (5’-GAATTCGTGTACACCTCTACCGAGAC-3’,
88
EcoRI) and YLR1 (5’-GCGGCCGCTTAGATACCACAGACACCCTC-3’, NotI) were used
4
Page 4 of 31
89
to amplify the lipase gene lip2 with genomic DNA of strain CICC32187 as the template. The
90
amplified lip2 gene was verified by DNA sequencing, and the sequence of lip2 was
91
submitted into GenBank with accession number KJ603521. PCR product was double digested by EcoRI and NotI enzymes and then ligated into
93
Pichia expression vector pPICZA, which carries the methanol inducible expression
94
promoter PAOX1, to form the recombinant plasmid pPIC-lip2. pPIC-lip2 was then transferred
95
into Pichia X-33 cells to obtain the Pichia transformants X-33 (PAOX1-LIP2). The steps
96
detailing the Pichia competent cell preparation, electroporation, positive colony screening
97
and verification were conducted mainly according to the descriptions of Cregg [16].
an
us
cr
ip t
92
99
M
98
2.2 de novo lipase gene design, synthesis and recombinant construction Based on the native amino acid sequence of Y. lipolytica LIP2 lipase (GenBank:
101
AIB08846), we de novo designed the nucleotide sequence of syn-lip. With the assistance of
102
DNA2.0 software (http://www.dna20.com), the more frequently used codons in Pichia were
103
selected from synonymous codons to replace the less frequently used codons. The secondary
104
structure and the minimal free energy of the mRNA was calculated by RNAfold software
105
(http://rna.tbi.univie.ac.at/cgi-bin/RNAfold.cgi). The codon usage frequency in the native
106
lip2 gene and the designed gene were analyzed online by graphical codon usage analyzer
107
software 2.0 (http://gcua.schoedl.de/). The codon usage frequency, the codon adaptation
108
index (CAI), the frequency of optimal codons (FOP) and the GC content of the original lip2
109
gene and designed syn-lip gene are shown in Supplementary data Fig. S1 through S6.
110
Ac ce p
te
d
100
To synthesize the designed lipase gene, a batch of oligonucleotides (see Supplementary
5
Page 5 of 31
data Table S1, Fig. S3) was designed by Gene2Oligo software [17] to make the
112
thermodynamic properties of each oligonucleotide consistent. A two-step gene synthesis
113
method was used to assemble these oligonucleotides into the full length lipase gene. This
114
process was conducted according to the descriptions of Yang et al. [18]. The gene synthesis
115
flowchart is shown in Supplementary data Fig. S7.
cr
ip t
111
The full-length synthesized lipase gene syn-lip gene was ligated into a pMD-18T vector
117
and the insertion was verified by sequencing. A fragment of syn-lip was digested from the
118
pMD-18T vector by EcoRI and NotI and then ligated into a pPICZA vector to form the
119
recombinant pPIC-syn-lip. The recombinant was then transferred into the Pichia strain X-33
120
to obtain the yeast transformant carrying methanol inducible expression syn-lip gene.
M
an
us
116
121
d
123
2.3 Promoter PFLD1 cloning and recombinant construction The promoter region (PFLD1) of formaldehyde dehydrogenase gene (FLD) was amplified
te
122
by using the primer pair FLD5BZ (5’-GCGAGATCTGCATGCAGGAATCTCTGG-3’, Bgl II)
125
and FLD3HZ (5’- GCGAAGCTTTGTGAATATCAAGAATTGTATGAACAAGC-3’, Hind
126
III) with the genome DNA of P. pastoris X-33 as the template. After verification by DNA
127
sequencing, the PFLD1 fragment was inserted into a pPICZA vector by using restriction
128
enzymes BglII and HindIII to replace the original AOX1 promoter in pPICZA to form the
129
PFLD1 initiated vector pPICFLD. The codon optimized lipase gene syn-lip was then inserted
130
into this vector by the EcoRI and NotI sites to form the recombinant PFLD1-syn-lip. After the
131
transfer into the strain X-33, a Pichia transformant carrying a PFLD1 controlled lipase gene
132
was obtained.
Ac ce p
124
6
Page 6 of 31
133 134
2.4 Recombinant construction by using the PichiaPink system The Invitrogen newly developed yeast expression system, PichiaPink, was used in this
136
study to construct the yeast recombinants suitable for the large-scale expression and
137
secretion of lipase. Vector pPink-HC, containing a Saccharomyces cerevisiae -mating
138
factor prepro-sequence and a selection marker gene ADE2, and PichiaPink Strain 1 (ade2)
139
were
140
(5’-CTAGGCCTGTTTACACTTCTACTGAGAC-3’,
141
(5’-CTGCCGGCGATACCACAAACACCCTC-3’, NaeI) were used to amplify the syn-lip
142
gene and then were inserted into the pPink-HC vector by StuI and NaeI sites to construct
143
recombinant pPink-syn-lip.
the
secreted
expression
of
lipase.
Primer
us
for
StuI)
and
pairs
SLPF SLPR
M
an
selected
cr
ip t
135
The operations on the Pichia competent cell preparation, electroporation and positive
145
colony screenings were conducted mainly according to the manufacturer’s instructions
146
(Invitrogen). Briefly, approximately 5 g of plasmid was linearized by Spe I, which cut at
147
the TRP2 region of the PichiaPink vectors, and then was transferred into Strain 1 (ade2) by
148
electroporation with the conditions set at 2 kV, 25 , and 200 F for a 0.2 cm
149
electroporation cuvette (Electroporator 2510, Eppendorf). PAD (Pichia Adenine Dropout)
150
agar was used for the screening of positive PichiaPink transformants by ADE2
151
complementation. The insertion of the syn-lip gene into the PichiaPink genome was verified
152
by PCR with the primer pair SLPF/SLPR.
Ac ce p
te
d
144
153 154
2.5 Methanol inducible expression of lipase
7
Page 7 of 31
A single colony of Pichia transformants PAOX1-lip2, PAXO1-syn-lip, PFLD1-syn-lip and
156
PichiaPinK (PAOX1-syn-lip) was deposited into 20 mL of YPD medium (yeast extract 1%,
157
peptone 2%, dextrose 2%) and cultured at 28°C overnight. The Pichia culture was then
158
inoculated into 50 mL of BMGY medium (yeast extract 1%, peptone 2%, 100 mM potassium
159
phosphate buffer with pH 6.0, yeast nitrogen base 1.34%, glycerol 1%) at a 1:50 ratio and
160
incubated at 28°C with a shaking speed of 200 rpm for approximately 20 h to reach an
161
OD600=4.0. The cells were harvested and transferred into 50 mL of BMMY medium (yeast
162
extract 1%, peptone 2%, 100 mM potassium phosphate buffer with pH 6.0, yeast nitrogen
163
base 1.34%, 0.5% methanol) for the methanol inducible expression of lipase. Methanol was
164
added into the medium every 24 h to reach a final concentration of 0.5% to induce the
165
expression of lipase. Lipase activity, protein content and wet cell weight in the medium were
166
checked at predetermined intervals.
d
M
an
us
cr
ip t
155
2.6 Glycosylation site prediction and PNGase digestion of PichiaPink expression product
Ac ce p
168
te
167
The NetNGlyc 1.0 server (www.cbs.dtu.dk/services/NetNGlyc/) was used to analysis
169 170
the
N-glycosylation
sites,
and
the
GlycoEP
server
171
(http://www.imtech.res.in/raghava/glycoep/) was used to predict the O-glycosylation site in
172
the LIP2 protein.
173
N-Glycosidase F (PNGase F, New England Biolabs) was used to hydrolyze the
174
N-glycan chain from the PichiaPink expression product of LIP2. The PichiaPink expressed
175
protein in the fermentation broth was first concentrated by using an Amicon Ultra-15 ml
176
centrifugal filter unit with 10 kDa filter (Millipore), and then twice dialyzed in a protein
8
Page 8 of 31
buffer (0.05 M sodium phosphate 50 mM, DTT 0.05 mM). Approximately 20 l aliquots of
178
the protein solution were mixed with a glycoprotein denaturing buffer (5% SDS, 0.4 M DTT)
179
and then incubated in 100 °C bath for 10 min. The denatured protein solution was then added
180
into the G7 reaction buffer (0.5 M sodium phosphate, pH 7.5), 10% NP-40, and 2 l PNGase
181
F to make a 20 l volume deglycosidation mixture. The reaction mixture was incubated at
182
37 °C for 1 h and then loaded into the gel for SDS-PAGE analysis.
2.7 Recombinant fermentation in a bench-top fermenter
an
184
us
183
cr
ip t
177
Pichia transformants X-33 (PAOX1-lip2) and X-33 (PAOX1-syn-lip) were selected for the
186
bench-top scale lipase production assay. The fermentation was conducted in a 14 L
187
bioreactor (BioFlo 415, New Brunswick, Canada) with a medium of basal salts (H2PO4 26.7
188
mL, CaSO4 0.93 g, K2SO4 18.2 g, MgSO4•7H2O 14.9 g, KOH 4.13 g, glycerol 40.0 g, per
189
liter). The fermentation parameters were monitored and controlled throughout the whole
190
process. In the biomass accumulation phase, the fermentation parameters were maintained as:
191
temperature (28.0oC), dissolved oxygen (DO, >40%), pH=6.0, agitation ratio (rpm, 400-650)
192
and aeration (0.1-1.0 vvm). In the methanol inducible expression phase, the fermentation
193
parameters were maintained as: temperature (25.0oC), dissolved oxygen (DO, >35%),
194
pH=6.0, agitation ratio (rpm, 500-850) and aeration (0.1-1.0 vvm). The amount of methanol
195
fed into the system was changed according to the set time intervals during this phase. In the
196
first 24 h, methanol was fed into the broth with the ratio of 2 mL/h per liter of broth. In the
197
next 24 h, the amount of methanol fed into the system was set to 4 mL/h per liter of broth,
198
and then increased to 5.5 mL/h per liter of broth, which was maintained until the end of the
Ac ce p
te
d
M
185
9
Page 9 of 31
199
fermentation. Samples were collected at intervals, and the wet cell weight, lipase activity and
200
protein content in the broth were analyzed for each sample.
202
2.8 Wet cell weight, lipase activity and protein content assays
ip t
201
To measure the yeast wet cell weight, an aliquot of 50 mL broth and the cell pellet were
204
harvested by centrifuge. The wet cell weight was calculated as the mass of the cells in grams
205
per liter. The lipase activity in the broth was determined at pH 7.5 by free fatty acid titration
206
using 50 mM NaOH after incubation in a thermostated vessel for 10 min. The assay mixture
207
consisted of 5.9 mL of Tris-HCl buffer (50 mM, pH 7.5), 50 mM NaCl, 4 mL of emulsified
208
tributyrin and 0.1 mL of a diluted enzyme solution. One unit (U) of activity was defined as
209
the amount of enzyme liberating 1 micromole of fatty acid per min at 40°C. The protein
210
content in the fermentation broth was determined by the Bradford method and the expression
211
profiles were checked by SDS-PAGE.
us
an
M
d
te
Ac ce p
212
cr
203
213
3. Results and discussions
214
3.1 de novo design and synthesis of the lipase gene
215
Considering the codon usage bias between P. pastoris and Y. lipolytica, the codon of Y.
216
lipolytica lipase gene lip2 was optimized based on the amino acid sequence of LIP2
217
(AIB08846) to realize a high level expression of lipase in Pichia. We first replaced the less
218
frequently used codons with more frequently used ones (Fig. 1, Supplementary data Fig. S1
219
and S2). As reported before, the codon order, such as being either AT-rich or GC-rich in the
220
region, and the complexity of the mRNA secondary structures can act as roadblocks that
10
Page 10 of 31
might influence the rhythm of protein synthesis, decrease the translation efficiency, or even
222
cause the premature termination of translation [19]. To make the nucleotides A, T, G and C
223
evenly distribute in the sequence, which could also be used to deduce the complexity of the
224
secondary structure of mRNA, more frequently used codons may also be selected to make
225
different types of nucleotides distribute evenly, as well as to keep the GC content of the gene
226
within the range of 45-60%. After codon optimization, the codon adaptation index (CAI)
227
changed from 0.74 of the native lip2 gene to the optimized 0.87, which means the codons
228
were well adapted to the Pichia expression system (Supplementary data Fig. S4). As
229
indicated by the frequency of the optimal codons (FOP) parameter, in the original gene,
230
22.5% of the codons have the usage frequency 90%. While in optimized syn-lip, approximately 60% of codons have the
232
usage frequency >90%. (Supplementary data Fig. S5). The GC content changed from the
233
original 47.5% to the optimized 50.0%, and the GC-rich regions in the native lip2 with peaks
234
higher than 60% were eliminated (Supplementary data Fig. S6). As shown by Figure 2, after
235
re-designing, the ensemble diversity of the first 200 nucleotides decreased from 38.7 to 25.6,
236
and the minimal free energy of the original gene significantly changed from the original
237
value of -45.00 kcal/mol to -34.5 kcal/mol (Fig. 2).
239
cr
us
an
M
d
te
Ac ce p
238
ip t
221
3.2 Comparative assay of lipase expression between native and codon optimized genes
240
The lipase production capacities of native lip2 and the codon optimized syn-lip gene
241
were evaluated in flasks in this study, and the wet cell weight, lipase activity and protein
242
content in the fermentation broth were calculated. As shown in figure 3, the recombinant
11
Page 11 of 31
carrying the codon-optimized lipase gene (syn-lip) demonstrated a much stronger lipase
244
secretion capacity than the cells with the native gene (lip2). After methanol-inducible
245
expression was allowed to occur for 96 h, the lipase activity and protein content of the
246
syn-lip gene reached 5800 U/mL and 0.55 g/L in the fermentation broth, respectively. In
247
contrast, recombinants (pPIC-lip) carrying the native lip2 gene reached levels of only 4700
248
U/mL and 0.46 g/L, respectively (Fig. 3). Furthermore, as reflected by figure 3A, the
249
biomass of the recombinant carrying syn-lip was approximately 102 g/L at 96 h, which is
250
lower than the value of the native lip2 gene of 120 g/L. Thus the lipase production efficiency
251
of syn-lip was pronouncedly higher than the native lip2 gene if the production was calculated
252
on a per unit of yeast cell biomass basis.
3.3 Comparative assay of promoter types on lipase expression
d
254
M
253
an
us
cr
ip t
243
In the metabolism process of methanol in the Pichia cell, alcohol oxidases first oxidize
256
methanol into formaldehyde. The alcohol oxidases possess a very low affinity for molecular
257
oxygen. In order to compensate for their poor capacity for carrying oxygen, P. pastoris
258
produces large amounts of the enzymes by the promoters PAOXI and PAOXII [20]. These
259
methanol inducible promoters could be used to drive the heterologous gene expression in
260
Pichia, and generally it yields a higher level of expression. Glutathione-dependent
261
formaldehyde dehydrogenase (FLD) is another key enzyme required for the metabolism of
262
methanol, as a carbon source, and certain alkylated amines, as nitrogen sources. The
263
promoter of formaldehyde dehydrogenase 1 (PFLD1) was strongly and independently induced
264
by either methanol as a sole carbon source or methylamine as a sole nitrogen source [21].
Ac ce p
te
255
12
Page 12 of 31
Under the control of PFLD1, the expression level of the foreign gene was generally
266
comparable to those obtained with the commonly used alcohol oxidase I gene promoter
267
(PAOX1), which is now becoming an attractive alternative to PAOX1 in P. pastoris [22,23,24].
268
In this study, we comparatively analyzed the lipase production capacity of the Pichia
269
recombinants PFLD1-syn-lip and PAOX1-syn-lip in flasks under the 0.5% methanol induction.
270
As reflected by figure 4, the PFLD1 controlled recombinant (PFLD1-syn-lip) showed a
271
considerable lipase production capacity. After methanol inducible expression was allowed to
272
occur for 96 h, the lipase activity and protein content in the broth reached 4950 U/mL and
273
0.48 g/L, respectively. For the recombinant PAOX1-syn-lip, the lipase activity and the protein
274
content reached the highest values at 5650 U/mL and 0.54 g/L, respectively, at the 96 h time
275
point (Fig. 4). Although this study further demonstrated that PFLD1 is an efficient promoter
276
for the expression of heterologous proteins in P. pastoris and that the expression levels of the
277
lipase of the PFLD1-controlled system may be comparable to the PAOX1-controlled system, we
278
also noticed that over the timeframe of the whole experiment, the lipase production capacity
279
of PAOX1-syn-lip was always higher than that of PFLD1-syn-lip. So in this study, the
280
recombinant under the control of PAOX1 (PAOX1-syn-lip) was still selected for its stronger
281
lipase production capacity.
283
cr
us
an
M
d
te
Ac ce p
282
ip t
265
3.4 Comparative assay of lipase expression between different Pichia expression systems
284
To facilitate a larger scale expression of exogenous protein, Invitrogen has developed a
285
novel Pichia eukaryotic expression system, PichiaPink, by using ADE2 as selectable marker
286
gene. In this study, we compared the lipase production capacity between recombinants of
13
Page 13 of 31
X-33 (PAOX1-syn-LIP) and PichiaPink (pPink-syn-lip). As shown in figure 5, the
288
lipase-producing capacity of these two types of recombinants was almost equal to each other
289
and there was no distinct difference in the lipase activity and protein content between them
290
throughout the entire time of the experiment (Fig. 5B and 5C). The total protein profiles in
291
PichiaPink (pPink-syn-lip) fermentation broth detected by SDS-PAGE showed that the
292
specific band of lipase with a size of approximately 37 kDa was apparently much weaker
293
than the X-33 (PAOX1-syn-lip) recombinants, although the detected total protein and lipase
294
activities were equivalent with each other (Fig. 5D). We speculated that this might be caused
295
by the excess glycosylation of the partial Lip2 protein in PichiaPink system.
an
us
cr
ip t
287
We further checked the whole SDS-PAGE gels of the fermentation broth and found that,
297
when compared with the protein profile of the X-33 (PAOX1-syn-lip) fermentation broth,
298
which had a specific lipase band (approximately 37 kDa) and a clear background, PichiaPink
299
(pPink-syn-lip) had a smeared background and a blurred band with a size of approximately
300
97 kDa (Fig. 6A and 6B), which might be the excess glycosylated LIP2 band. N-Glycosidase
301
F was used to digest the lip2 gene produced from PichiaPink. As shown in Fig. 6C, PNGase
302
F could efficiently digest the PichiaPink expression product, including the 97 KDa smear
303
band into two bands, where the small band possessed a size of approximately 32 kDa, which
304
coincided with the speculated size of the Lip2 without glycosylation, and a bigger band with
305
a size of approximately 40 kDa.
Ac ce p
te
d
M
296
306
We further analyzed the N- and O-glycosylation site by the NetNGlyc 1.0 server and
307
GlycoEP server and found that there are two N-glycosylation sites, a strong N-glycosylation
308
site N113 and a weak site N134, and two O-glycosylation sites (T159 and T295) that exist in
14
Page 14 of 31
the Lip2 protein (Fig. 6D). The strong N-glycosylation site (N113) might be generally
310
slightly glycosylated in a host such as X-33. Considering PNGase F can efficiently cleave
311
between the innermost GlcNAc and asparagine residues and can hydrolyze nearly all types
312
of N-glycan chains from glycopeptides and proteins, we believe that the bigger band
313
(approximately 40 kDa) in PNGase F digested products comes from the O-glycosylation of
314
Lip2 in the PichiaPink system.
cr
ip t
309
3.5 Lipase production of the recombinants in the bench-top fermenter
an
316
us
315
Scale-up experiments with a bench-top fermenter were performed on the recombinants
318
carrying the native lip2 and codon-optimized syn-lip gene in this study (Fig. 7). With better
319
parameter control and a better feed of methanol into the fermenter, the wet cell weights,
320
lipase activities and protein contents in the fermentation broth from the fermenter were
321
significantly higher than those in the flasks. Both the recombinants X-33(PAOX1-lip2) and
322
X-33(PAOX1-syn-lip) had wet cell weights higher than 250 g/L and lipase activities higher
323
than 15000 U/mL. When compared with the native lip2 gene, the lipase activity and protein
324
content in the fermentation broth of the codon-optimized syn-lip X-33(PAOX1-syn-lip) their
325
highest values at 38,500 U/mL and 2.82 g/L, which is significantly higher than the native
326
lip2 values, where the lipase activity and the protein content was 17,500 U/mL and 1.68 g/L,
327
respectively (Fig. 7).
Ac ce p
te
d
M
317
328
Currently, several studies have been conducted on the expression of the Y. lipolytica
329
lipase gene. By using methods such as multiple copy integration, Pignède et al. [25]
330
achieved a level of 0.5 g of active Y. lipolytica lipase per liter in the supernatant. Through the
15
Page 15 of 31
optimization of fermentation parameters, Yu et al. [26] achieved a lipase activity of 12,500
332
U/mL, and by using the constitutive expression promoter PGAP, Wang et al. [27] has realized
333
lip2 expression in P. pastoris to the level of 13,500 U/mL in a 10-L fermenter. The
334
expression levels achieved by these studies were almost equivalent to the level seen in our
335
Pichia recombinants carrying the native lip2 gene. Through de novo design and synthesis of
336
the lipase gene, a comparative selection between the promoter types and the Pichia
337
expression system, the expression level in X-33 (PAOX1-syn-lip) reached levels far higher
338
than in previous works.
an
us
cr
ip t
331
339
4. Conclusions
M
340
Through de novo gene design and synthesis of syn-lip, a comparative analysis of the
342
lipase production capacity between the promoters PAOX1 and PFLD1, and the different Pichia
343
expression system, we have successfully obtained the highest level of expression in the
344
Pichia recombinants of X-33(PAOX1-syn-lip). After a fermentation assay in a bench-top scale
345
fermenter, the lipase production capacity reached the highest value of 38,500 U/mL in the
346
fermentation broth. This study has greatly facilitated the bioapplication of lipase in industrial
347
fields. In addition, the strategies utilized in this study, such as the use of de novo gene design
348
and synthesis, a comparative analysis among promoters and different generation of Pichia
349
expression systems, will also be utilized as references for future work in this field.
Ac ce p
te
d
341
350 351 352
Acknowledgments We thank J. S. Cheng and C. Hu for their help in the conduction of the experiments.
16
Page 16 of 31
This work was financially supported by the Science and Technology Supporting Program of
354
Hubei Province (2014), the National High Technology Research and Development Program
355
of China (no. 2013AA102805-4), the Key Project of Science and Technology Research
356
Program of the Education Department of Hubei Province (no. D20131703), and the Key
357
Project of Chinese Ministry of Education (no. 212118), where Dr. JK Yang is an incumbent
358
member of the Chutian Scholar Program.
cr
ip t
353
us
359
References
361
[1] Meng Y, Wang G, Yang N, Zhou Z, Li Y, Liang X, Chen J, Li Y, Li J. Two-step
362
synthesis of fatty acid ethyl ester from soybean oil catalyzed by Yarrowia lipolytica
363
lipase. Biotechnol Biofuels. 2011;4(1):6.
M
an
360
[2] Casas-Godoy L, Meunchan M, Cot M, Duquesne S, Bordes F, Marty A. Yarrowia
365
lipolytica lipase Lip2: an efficient enzyme for the production of concentrates of
366
docosahexaenoic acid ethyl ester. J Biotechnol. 2014;180:30−36.
Ac ce p
te
d
364
367
[3] Cambon E, Piamtongkam R, Bordes F, Duquesne S, André I, Marty A. Rationally
368
engineered double substituted variants of Yarrowia lipolytica lipase with enhanced
369 370 371 372 373 374
activity coupled with highly inverted enantioselectivity towards 2-bromo phenyl acetic acid esters. Biotechnol Bioeng. 2010;106(6):852−859.
[4] Cregg JM, Vedvick TS, Raschke WC. Recent advances in the expression of foreign genes in Pichia pastoris. Biotechnol (N Y). 1993;11:905−910. [5] Cregg JM. Introduction: distinctions between Pichia pastoris and other expression systems. Methods Mol Biol 2007a;389:1−10.
17
Page 17 of 31
375 376
[6] Vedvick TS. Gene expression in yeast: Pichia pastoris. Curr Opin Biotechnol 1991;2:742−745. [7] Gasser B, Prielhofer R, Marx H, Maurer M, Nocon J, Steiger M, Puxbaum V, Sauer M,
378
Mattanovich D. Pichia pastoris: protein production host and model organism for
379
biomedical research. Future Microbiol 2013;8:191−208.
cr
381
[8] Zur, H. Tuller T. Strong association between mRNA folding strength and protein abundance in S. cerevisiae. EMBO Rep. 2012;13,272−277.
us
380
ip t
377
[9] Park C, Chen X, Yang JR, Zhang J. Differential requirements for mRNA folding
383
partially explain why highly expressed proteins evolved slowly. Proc Natl Acad Sci
384
USA 2013;110,E678−E686.
M
an
382
[10] Mao Y, Liu H, Liu Y, Tao S. Deciphering the rules by which dynamics of mRNA
386
secondary structure affect translation efficiency in Saccharomyces cerevisiae. Nucleic
387
Acids Res. 2014, 42(8):4813−4822.
te
d
385
[11] Sinclair G, Choy FY. Synonymous codon usage bias and the expression of human
389
glucocerebrosidase in the methylotrophic yeast, Pichia pastoris. Protein Expr Purif
390
Ac ce p
388
2002;26(1):96−105.
391
[12] Mirzaei SA, Yazdi MT, Sepehrizadeh Z. Secretory expression and purification of a
392
soluble NADH cytochrome b5 reductase enzyme from Mucor racemosus in Pichia
393
pastoris based on codon usage adaptation. Biotechnol Lett 2010;32(11):1705−1711.
394
[13] Öberg F, Sjöhamn J, Conner MT, Bill RM, Hedfalk K. Improving recombinant
395
eukaryotic membrane protein yields in Pichia pastoris: the importance of codon
396
optimization and clone selection. Mol Membr Biol 2011;28(6):398−411.
397
[14] Quan J, Saaem I, Tang N, Ma S, Negre N, Gong H, White KP, Tian J. Parallel on-chip
398
gene synthesis and application to optimization of protein expression. Nat Biotechnol 18
Page 18 of 31
399
2011;29:449–452. [15] Yang JK, Liu LY, Dai JH, Li Q. de novo design and synthesis of Candida antarctica
401
lipase B gene and α-factor leads to high-level expression in Pichia pastoris. PLoS One
402
2013;8(1):e53939.
ip t
400
[16] Cregg JM. Pichia ptotocols. 2nd 2007b, The Humana Press, Totowa, NJ.
404
[17] Rouillard JM, Lee W, Truan G, et al. Gene2Oligo: oligonucleotide design for in vitro
406
gene synthesis. Nucleic Acids Res, 2004;32:W176−180.
us
405
cr
403
[18] Yang JK, Chen FY, Yan XX, Miao LH, Dai JH. A simple and accurate two-step long DNA sequences synthesis strategy to improve heterologous gene expression in Pichia.
408
PLoS ONE 2012;7(5):e36607.
410
[19] Richard P, Manley JL. Transcription termination by nuclear RNA polymerases. Genes Dev 2009, 23(11): 1247−1269.
M
409
an
407
[20] Cregg JM, Madden KR, Barringer KJ, Thill GP, Stillman CA. Functional
412
characterization of the two alcohol oxidase genes from the yeast Pichia pastoris. Mol
413
Cell Biol 1989;9(3):1316–1323.
Ac ce p
te
d
411
414
[21] Shen S, Sulter G, Jeffries TW, Cregg JM. A strong nitrogen source-regulated promoter
415
for controlled expression of foreign genes in the yeast Pichia pastoris. Gene
416
1998;216(1):93−102.
417
[22] Resina D, Serrano A, Valero F, Ferrer P. Expression of a Rhizopus oryzae lipase in
418
Pichia pastoris under control of the nitrogen source-regulated formaldehyde
419
dehydrogenase promoter. J Biotechnol 2004;109(1-2):103−113.
420
[23] Daly R, Hearn MT. Expression of heterologous proteins in Pichia pastoris: a useful
421
experimental tool in protein engineering and production. J Mol Recognit
19
Page 19 of 31
422
2005;18:119−138. [24] Resina D, Maurer M, Cos O, Arnau C, Carnicer M, Marx H, Gasser B, Valero F,
424
Mattanovich D, Ferrer P. Engineering of bottlenecks in Rhizopus oryzae lipase
425
production in Pichia pastoris using the nitrogen source-regulated FLD1 promoter. N
426
Biotechnol 2009;25(6):396−403.
cr
[25] Pignède G, Wang HJ, Fudalej F, Seman M, Gaillardin C, Nicaud JM. Autocloning and amplification
of
LIP2
429
2000;66(8):3283−3289.
in
Yarrowia
lipolytica.
Appl
Environ
Microbiol
an
428
us
427
ip t
423
[26] Yu M, Lange S, Richter S, Tan T, Schmid RD. High-level expression of extracellular
431
lipase Lip2 from Yarrowia lipolytica in Pichia pastoris and its purification and
432
characterization. Protein Expr Purif 2007;53(2): 255−263.
M
430
[27] Wang X, Sun Y, Ke F, Zhao H, Liu T, Xu L, Liu Y, Yan Y. Constitutive expression of
434
Yarrowia lipolytica lipase LIP2 in Pichia pastoris using GAP as promoter. Appl
435
Biochem Biotechnol 2012;166(5):1355−1367.
437 438 439 440
te
Ac ce p
436
d
433
441 442 443 444 445 20
Page 20 of 31
Fig. 1. Sequence comparison between the native Y. lipolytica lipase gene lip2 and the
447
designed lipase gene syn-lip. Bars in same color represent the same nucleotides between the
448
native and codon optimized genes.
449
Fig. 2. Secondary structure of the first 200 bp of the original lip2 and codon-optimized
450
syn-lip mRNA generated by the software RNAfolder. A. Structure of the original lip2 mRNA
451
where the MFE is -45.00 kcal/mol and the ensemble diversity is 38.73. (B) Structure of the
452
codon optimized syn-lip mRNA where the MFE is -34.5 kcal/mol and the ensemble diversity
453
is 25.69.
454
Fig. 3. Expression profiles of the yeast recombinants carrying the original lip2 and codon
455
optimized syn-lip gene. Pichia wet cell weight (A), lipase activity (B), and protein content
456
(C) in the fermentation broth of the lip2 and syn-lip recombinants were calculated in this
457
study. Protein profiles in the broth were checked by SDS-PAGE, and approximately 10 L of
458
the protein solution per sample was loaded into the gels (D). * indicates the significant
459
difference (P
