Journal of Microbiological Methods 169 (2020) 105816

Contents lists available at ScienceDirect

Journal of Microbiological Methods

journal homepage: www.elsevier.com/locate/jmicmeth

A novel electroporation procedure for highly efficient transformation of T

Lipomyces starkeyi
⁎
Hiroaki Takakua, , Atsumi Miyajimaa, Haruka Kazamaa, Rikako Satoa, Satoshi Araa,
Tomohiko Matsuzawab, Katsuro Yaoib, Hideo Arakic, Yosuke Shidad, Wataru Ogasawarad,
Harutake Yamazakia
a
Department of Applied Life Sciences, Niigata University of Pharmacy and Applied Life Sciences, 265-1 Higashijima, Akiha-ku, Niigata 956-8603, Japan
b
Bioproduction Research Institute, National Institute of Advanced Industrial Science and Technology (AIST), Tsukuba Central 6, 1-1-1 Higashi, Tsukuba, Ibaraki 305-
8566, Japan
c
Research Institute for Creating the Future, FUJI OIL HOLDINGS INC, 4-3 Kinunodai, Tsukubamirai-shi, Ibaraki 300-2497, Japan
d
Department of Bioengineering, Nagaoka University of Technology, 1603-1 Kamitomioka, Nagaoka, Niigata 940-2188, Japan

A R T I C LE I N FO A B S T R A C T

Keywords: Microbial lipids produced by oleaginous microorganisms as raw materials for the production of oleochemicals
Electroporation and biodiesel are sustainable while avoiding competition with food products. The oleaginous yeast Lipomyces
Lipomyces starkeyi starkeyi is an excellent lipid producer with a great industrial potential that is suitable as a valuable host to
Oleaginous yeast improve lipid production through genetic engineering modifications. However, genetic tools, including effective
Transformation
transformation methods, for L. starkeyi are insufficient for improvement of lipid production and analysis of lipid
production mechanisms. We previously developed a polyethylene glycol (PEG)-mediated spheroplast transfor-
mation method that significantly improved the homologous recombination efficiency of L. starkeyi strain Δlslig4.
Although other transformation methods, including lithium acetate (LiAc)-mediated transformation and
Agrobacterium tumefaciens-mediated transformation, have been reported, a more efficient and convenient
transformation method for L. starkeyi is desired. In this study, we developed a novel electroporation transfor-
mation method that was first applied for integration of drug-resistance gene markers into the genome of L.
starkeyi strain Δlslig4 at the 18S ribosomal DNA locus of a multiple-copy gene, which yielded approximately 60
transformants/μg of DNA. Optimization of five parameters (i.e., cell growth phase, cell density, osmotic stabi-
lizers, pretreatment agents, and electric conditions) enhanced the efficiency of transformation to approximately
1.5 × 104 transformants/μg of DNA. As compared with those of LiAc-mediated transformation and PEG-
mediated spheroplast transformation, the efficiency of the proposed transformation method was increased by
about 111- and 7-fold, respectively. Additionally, the transformation efficiency of our proposed electroporation
method targeting a single-copy gene locus yielded 273 transformants/μg of DNA. To our knowledge, this is the
first report of a successful electroporation method to accelerate analysis of lipid production by L. starkeyi.

1. Introduction expensive than conventional diesel owing to the high cost of vegetable
oil. In addition, the biodiesel industry competes with the food industry
Because of the depletion of fossil fuel sources and increasing con- for oil crops (Sitepu et al., 2014). The fatty acid composition of mi-
cern over atmospheric CO2 levels, the sustainable production of the crobial lipids (triacylglycerols) produced by oleaginous yeasts is similar
next generation of renewable liquid transportation fuels is of great to that of vegetable oils (Beligon et al., 2016; Kosa and Ragauskas,
importance. Biodiesel, which is produced by transesterification of tri- 2011; Papanikolaou and Aggelis, 2011). Additionally, the production of
glycerides derived from vegetable oils with methanol, is a potentially triacylglycerols by oleaginous yeasts has many advantages over that
renewable fuel as a substitute for fossil fuels for use in compression with the use of vegetable oils, such as the short life cycle of oleaginous
ignition vehicles (Liang and Jiang, 2013). However, biodiesel is more yeasts, scaling-up facilitation of yeast culture, stable production

Abbreviations: DTT, dithiothreitol; LiAc, lithium acetate; OD, optical density; PEG, polyethylene glycol; PCR, polymerase chain reaction; YPD, yeast extract/
peptone/dextrose
⁎
Corresponding author.
E-mail address: htakaku@nupals.ac.jp (H. Takaku).

https://doi.org/10.1016/j.mimet.2019.105816
Received 6 November 2019; Received in revised form 20 December 2019; Accepted 20 December 2019
Available online 24 December 2019
0167-7012/ © 2019 Elsevier B.V. All rights reserved.
H. Takaku, et al. Journal of Microbiological Methods 169 (2020) 105816

regardless of seasonal and climatic factors, escape from competition 2. Materials and methods
with food production, and simplicity of enhanced microbial lipid pro-
duction by genetic engineering approaches (Li et al., 2008). 2.1. Microorganisms and culture conditions
Several oleaginous yeasts are known to accumulate intracellular
lipids by up to 60% of the cell dry weight (Ageitos et al., 2011). No- L. starkeyi strain CBS1807 was obtained from the Central Office for
tably, Lipomyces starkeyi is an excellent industrial oleaginous yeast be- Mold Cultures (Utrecht, The Netherlands). Homologous recombination
cause of the high lipid content of 75.2% of cell dry weight (Angerbauer with the deletion mutant L. starkeyi strain Δlslig4, generated from L.
et al., 2008). L. starkeyi can also accumulate lipids from various carbon starkeyi strain CBS1807, is highly efficient (Oguro et al., 2017). L.
sources, including glucose, xylose, mannose, arabinose, galactose, and starkeyi strains CBS1807 and Δlslig4 were grown in liquid yeast extract/
cellobiose (Oguri et al., 2012). In addition, L. starkeyi is able to si- peptone/dextrose (YPD) medium [1% yeast extract (Kyokuto-Sanki Co.,
multaneously utilize mixtures of sugars, such as glucose and mannose Ltd., Tatsuno, Japan), polypeptone (Nihon Pharmaceutical Co., Ltd.,
(Yang et al., 2014), cellobiose and xylose (Gong et al., 2012), and Tokyo, Japan), and 2% glucose] with or without 100 μg/mL of hy-
glucose and xylose (Zhao et al., 2008). Complete genome sequencing gromycin B (FUJIFILM Wako Pure Chemical, Osaka, Japan), 30 μg/mL
revealed that the genome of L. starkeyi strain NRRL Y-11557 (CBS1807) of nourseothricin (FUJIFILM Wako Pure Chemical), and/or 50 μg/mL of
consists of about 21.27 Mbp of DNA that encodes 8192 genes (Riley Zeocin™ (Invitrogen Corporation, Carlsbad, CA, USA). Solid media in-
et al., 2016). Elucidation of the complete genome sequence permits cluded 2% agar (FUJIFILM Wako Pure Chemical). Escherichia coli DH5α
comprehensive genome-wide analysis of gene expression patterns and cells were grown at 37 °C in liquid L-broth medium [1% Bacto™
functions. Genetic transformation technology is essential to both me- Tryptone (BD Biosciences, San Jose, CA, USA), 0.5% Bacto™ Yeast Ex-
tabolic engineering and physiological analysis of industrially important tract (BD Biosciences), and 1% NaCl] with 100 μg/mL of ampicillin
yeast, such as L. starkeyi. Consequently, the combination of complete (FUJIFILM Wako Pure Chemical).
genome sequencing information and genetic transformation techniques
will further improve lipid productivity by L. starkeyi.
2.2. General molecular biology techniques
Established genetic transformation methods for L. starkeyi include
lithium acetate (LiAc)-mediated transformation (Calvey et al., 2014), A.
Genomic DNA from L. starkeyi strains CBS1807 and Δlslig4 was
tumefaciens-mediated transformation (Dai et al., 2017), and poly-
prepared as described previously (Oguro et al., 2017). Plasmid DNA
ethylene glycol (PEG)-mediated spheroplast transformation (Oguro
was extracted using the FastGene® Plasmid Mini Kit (Nippon Genetics
et al., 2015). We previously reported the use of spheroplast–PEG
Co., Ltd., Tokyo, Japan). Polymerase chain reaction (PCR) was per-
transformation for targeted ribosomal DNA (rDNA) integration of
formed using KOD FX Neo DNA polymerase in accordance with the
multiple copies of a gene into the L. starkeyi genome (Oguro et al.,
manufacturer's instructions (Toyobo Co., Ltd., Osaka, Japan). Amplified
2015). This feature is especially useful for metabolic engineering-
PCR products were extracted using the FastGene® Gel/PCR extraction
mediated overexpression of genes involved in lipid production. We also
kit (Nippon Genetics Co., Ltd.).
developed a gene targeting system for application in genetic studies
using mutant strains of L. starkeyi (Δlsku70, Δlsku80, Δlslig4,
Δlsku70Δlsku80, Δlsku70Δlslig4, Δlsku80Δlslig4, and Δlsku70Δlsku80Δl- 2.3. Preparation of DNA fragments for transformation
slig4), as the gene targeting frequency of the wild-type strain is very low
(Oguro et al., 2017). For genetic engineering of L. starkeyi, a simple, Two DNA fragments were used for L. starkeyi transformation in this
quick, and efficient transformation method to facilitate the deletion of study. One was obtained by digesting the plasmid pKS-18S-hph, which
target genes is essential for the functional study of various genes at the contained a portion of 18S rDNA gene of L. starkeyi and the hygromycin
molecular level. However, A. tumefaciens-mediated transformation and B resistance gene, with the restriction enzyme ApaI (Takara Bio, Inc.,
PEG-mediated spheroplast transformation are relatively laborious and Shiga, Japan), as previously described (Oguro et al., 2015), and used for
time-consuming. Thus, a more high-efficiency and convenient trans- targeted integration into the 18S rDNA locus. The plasmid pKS-LsLIG4-
formation method is desired. Electroporation, which is one of the most sNAT1, containing the LsTDH3 promoter/sNAT1/LsTDH3 terminator
common transformation methods for DNA delivery, is simple, quick, DNA fragment for expression in L. starkeyi and the LsLIG4 5′ and 3′ non-
and efficient but requires appropriate adjustment of physical para- coding regions, was constructed by assembling the following four PCR
meters and specific treatments for each species of yeast (Delorme, 1989; fragments using NEBuilder HiFi DNA Assembly Master Mix (New
Faber et al., 1994; Hood and Stachow, 1990; Liu et al., 2017; Wang England Biolabs, Ipswich, MA, USA). Two PCR fragments of the 5′ and
et al., 2011). However, the transformation of L. starkeyi by electro- 3′ non-coding regions of the LsLIG4 gene were amplified with the
poration has not been previously reported. Therefore, we attempted to primer sets 5'nonLsLIG4Fw (5′-AACAAAAGCTGGGTACCGGGCCCCTT
develop an electroporation procedure as an alternative method for CGCGGATGCCAAAGAC-3′) and 5'nonLsLIG4Rw (5′-AGCAAATTAAAG
highly efficient transformation of L. starkeyi. TTACCACAATTATATGCACATGGAGTC-3′), and 3'nonLsLIG4Fw (
In this study, an electroporation procedure optimized for the olea- 5′-CCCGCTACATGCCTGTTATAGAAGTCAAGTTCGC-3′) and 3'non-
ginous yeast L. starkeyi was developed. With this optimized electro- LsLIG4Rw (5′-TCGACCTCGAGGGGGGGCCCCCGCATCCGTTCGCTGA
poration procedure, a genomic integrative cassette containing a hy- AAG-3′), respectively, using the genomic DNA of L. starkeyi strain
gromycin-resistance gene was integrated into the rDNA locus of a multi- Δlslig4 as a template. Another PCR fragment containing the LsTDH3
copy gene, which produced about 1.5 × 104 transformants/μg of DNA. promoter region, nourseothricin resistance marker (sNAT1), and
Further, when electroporation was performed for genome integration of LsTDH3 terminator region was amplified with the primer set
a single-copy gene locus, the transformation efficiency was approxi- 5′TDH3NATFw (5′-TGTGGTAACTTTAATTTGCTGAAGCGGTTTGCC-3′)
mately 300 transformants/μg of DNA. As compared with those of non- and 3′TDH3NATRw (5′-TATAACAGGCATGTAGCGGGTGGTGATG-3′)
electroporation transformation methods, the transformation efficiency using the plasmid pKS-sNAT1-LsURA3 (Oguro et al., 2017) as a tem-
of electroporation was increased by 22- to 136-fold. Therefore, the plate. The other PCR fragment of the vector region was amplified with
proposed electroporation transformation method is useful for metabolic the primer set 5'pKSnonLsLIG4Fw (5′-GGGCCCGGTACCCAGCTT
engineering and comprehensive investigations of the functional genome TTG-3′) and 3'pKSnonLsLIG4Rw (5′-GGGCCCCCCCTCGAGGTC-3′)
of L. starkeyi. using the plasmid vector pBluescript KS(+) as a template. The DNA
fragment used for targeted integration into a single-copy gene was
obtained by digesting the plasmid pKS-LsLIG4-sNAT1 with ApaI (Ta-
kara Bio).

2
H. Takaku, et al. Journal of Microbiological Methods 169 (2020) 105816

2.4. LiAc-mediated transformation method with 50 mL of ice-cold sterilized distilled water. After centrifugation,
the cells were washed with 3 mL of ice-cold sorbitol (0.5 M) or sucrose
LiAc-mediated transformation of L. starkeyi was performed as de- (0.25, 0.5, or 1 M). Finally, the cells were resuspended in ice-cold
scribed previously (Calvey et al., 2014). Briefly, L. starkeyi cells at an sorbitol (0.5 M) or sucrose (0.25, 0.5, or 1 M) at a final concentration of
optical density at 600 nm (OD600) of 0.625 in 50 mL of YPD medium 0.5, 1.0, or 3.0 × 109 cells/mL. One microliter of the DNA fragments
were cultured overnight to an OD600 of about 10 and then harvested, (1 μg/μL) was gently mixed with 40 μL of the cell suspension in a
washed twice with sterilized distilled water, and resuspended in 500 μL chilled 0.2 cm electroporation cuvette (Bio-Rad Laboratories, Hercules,
of 0.1 M LiAc. A 100 μL aliquot of the cell suspension was centrifuged in CA, USA) and pulsed with the conditions of capacitance at 25 μF; field
a microtube at 5000 ×g for 1 min to separate the supernatant from the strength at 2.5, 3.75, 5.0, 6.25, or 7.5 kV/cm; and resistance at 200,
pellet. The pellet was suspended in 240 μL of 50% (w/v) PEG (BioUltra 400, or 800 Ω using the Gene Pulser X-Cell (Bio-Rad Laboratories).
3350; Sigma-Aldrich Corporation, St. Louis, MO, USA), 30 μL of 1.0 M After pulsing, 1 mL of ice-cold sorbitol (0.5 M) or sucrose (0.25, 0.5, or
LiAc, 3 μL of 10 mg/mL single-stranded salmon testes carrier DNA 1 M) was immediately added to the cuvette. Subsequently, the cell
D9156 (Sigma-Aldrich), and sterilized distilled water to a final volume suspension was gently mixed, transferred into 5 mL of YPD medium
of 350 μL. After vortexing thoroughly, the cell suspension was in- with sorbitol (0.5 M) or sucrose (0.25, 0.5, or 1 M), and incubated at
cubated at 30 °C for 3 h. After heat shock at 40 °C for 5 min, the cell 30 °C for 10 h while slowly shaking. After the culture was centrifugated
suspension was quickly centrifugated and the resultant pellet was re- at 4000 ×g for 5 min, the resultant pellet was resuspended in sterilized
suspended in 1 mL of YPD medium and cultured for 4 h at 30 °C. After distilled water and plated on selective solid medium.
the culture was further centrifuged at 5000 ×g for 1 min, the resultant
pellet was resuspended in 100 μL of sterilized distilled water and plated 2.7. Confirmation of target DNA integration in transformants by colony
on selective solid medium. PCR

2.5. PEG-mediated spheroplast transformation method Transformants selected on YPD solid medium with 100 μg/mL of
hygromycin B, 30 μg/mL of nourseothricin, and/or 50 μg/mL of zeocin
PEG-mediated spheroplast transformation of L. starkeyi was per- were streaked on fresh YPD solid medium containing hygromycin B,
formed as described previously (Oguro et al., 2015). Briefly, L. starkeyi nourseothricin, and/or zeocin. A single colony was suspended in 100 μL
cells at the log phase (OD600 = 1.0) of culture growth in YPD medium of lysis buffer [10 mM Tris–Cl (pH 8.0), 1 mM ethylenediaminete-
were collected, washed twice with sterilized distilled water, re- traacetic acid, 100 mM NaCl, 2% Triton X-100 (Alfa Aesar, Ward Hill,
suspended in McIlvaine buffer (pH 6.0) containing 0.4 M Na-tartrate, MA, USA), and 1% sodium dodecyl sulfate (FUJIFILM Wako Pure
5 mg/mL of Westase™ Enzyme (Takara Bio) to 3.0 × 106 cells/mL, and Chemical)] and incubated at 70 °C for 30 min. After the addition of
then incubated at 30 °C for 90 min to promote spheroplast formation. 100 μL of phenol/chloroform, the cell suspension was centrifugated at
The spheroplasts were gently centrifugated at 1000 ×g for 5 min and 15,000 × g for 5 min. The resultant supernatant was used as templates
subsequently washed twice with 3 mL of STC buffer (1.2 M sorbitol, for the following PCRs to confirm that the DNA fragment used in the
50 mM Tris–HCl (pH 8.0), and 50 mM CaCl2·2H2O). After centrifuga- transformation was correctly integrated into the target genomic locus.
tion at 1000 ×g for 5 min, the final pellet was resuspended in 0.5 mL of To confirm integration of heterogeneous genes at the 18S rDNA locus,
the same modified McIlvaine buffer. The DNA fragment for transfor- the primer sets 18SrDNAFw (5′-CTACTTGGATAACCGTGG-3′) and
mation of L. starkeyi was added to the spheroplast suspension (500 μL), HphRv (5′-TATCGGCGAGTACTTCTACAC-3′), and 18SrDNARw (
which was then incubated at room temperature for 20 min. Then, 1 mL 5′-ACGGTATCTGCGTTAACC-3′) and HphFw (5′-ATCTCGTGCTTTCAG
of 40% PEG (w/v) (molecular weight, 3000; FUJIFILM Wako Pure CTTGC-3′) were used to amplify the plasmid pKS-18S-hph-derived DNA
Chemical) in STC buffer was added and mixed thoroughly. After further fragments (a portion of 18S rDNA from L. starkeyi and the hygromycin B
incubation at room temperature for 20 min, the spheroplast mixture resistance gene expression cassette). To confirm the replacement of the
was incubated at 30 °C for 16 h in 10 mL of TB3 solution [0.3% yeast zeocin-resistance gene with the nourseothricin resistance gene, the
extract (Kyokuto-Sanki Co., Ltd.), 0.3% Difco Casamino Acids–Vitamin primer pair 5'nonLsLIG4Fw (5′-ATAATTGGACGCCAACGAGG-3′) and
Assay (Difco Laboratories, Franklin Lakes, NJ, USA), and 20% sucrose] 3'nonLsLIG4 Rv (5′-TTGCGCGTTTAGAACAGCTG-3′) was used to am-
for spheroplast regeneration. After centrifugation of the culture, the plify the plasmid pKS-LsLIG4-sNAT1-derived DNA fragments (LsLIG4 5′
resultant pellet was resuspended in sterilized distilled water and plated and 3′ non-coding regions and the nourseothricin resistance gene ex-
on selective solid medium. pression cassette).

2.6. Electroporation transformation method 2.8. Statistical analysis

Electroporation transformation of L. starkeyi was performed as Statistical analysis was performed using Tukey's test with GraphPad
previously described (Thompson et al., 1998) but with a slight mod- Prism software, version 8 (GraphPad Software Inc., San Diego, CA,
ification. Briefly, L. starkeyi Δlslig4 cells were grown in YPD medium at USA). Differences were considered to be significant when the P-value
30 °C for 2 days. Part of the culture described above was added to was < 0.05.
50 mL of YPD medium to yield a concentration of 2.5 × 107 cells/mL,
and the mixture was further cultured at 30 °C for 6 h (early log phase), 3. Results and discussion
12 h (mid-log phase), 18 h (late log phase), or 24 h (stationary phase).
The cultures were put on ice for 15 min and centrifugated at 4000 ×g 3.1. Transformation of L. starkeyi strain Δlslig4 by non-electroporation
for 5 min at 4 °C to collect the cells. After removing the supernatant, the methods
cells were suspended in 8 mL of sterilized distilled water. Following the
addition of 1 mL of TE [10 mM Tris–Cl (pH 8.0), 1 mM ethylenedia- L. starkeyi strain Δlslig4 is an efficient recipient for gene targeting
minetetraacetic acid] and 1 mL of LiAc (0, 1, 2, or 4 M), the culture was and provides a useful tool for high-throughput genome-wide functional
incubated at 30 °C for 45 min while slowly shaking. Then, 500 μL of gene analysis (Oguro et al., 2017). The Δlslig4 cells were transformed
dithiothreitol (DTT) (0, 200, 500, or 1000 mM) was added, and the cell with the ApaI-digested pKS-18S-hph linear DNA fragment containing a
suspension was incubated at 30 °C for 15 min while slowly shaking. The portion of 18S rDNA from L. starkeyi and the hygromycin B resistance
treated cell suspension was diluted until 50 mL with ice-cold sterilized gene using either LiAc-mediated transformation (Calvey et al., 2014) or
distilled water, centrifugated at 4000 ×g for 5 min at 4 °C, and washed PEG-mediated spheroplast transformation (Oguro et al., 2015). The 18S

3
H. Takaku, et al. Journal of Microbiological Methods 169 (2020) 105816

Table 1
Efficiencies of two different transformation methods for L. starkeyi Δlslig4 using either pKS-18S-hph or pKS-LsLIG4-nat. Data are presented as the mean ± standard
error of the mean (SEM) of the results of three independent experiments.
Construct name Target site Transformation method Transformation efficiency (transformants/μg DNA)

pKS-18S-hph 18S rDNA region LiAc-mediated transformation 131 ± 10

18S rDNA region PEG-mediated spheroplast transformation 2136 ± 275
pKS-LsLIG4-nat LsLIG4 locus LiAc-mediated transformation 2 ± 1
LsLIG4 locus PEG-mediated spheroplast transformation 12 ± 2

rDNA recombination site is suitable for targeted integration because of sorbitol as an osmotic stabilizer. One microgram of ApaI-digested pKS-
its existence as multiple repeated copies in the genomes of all eu- 18S-hph linear DNA fragment was transformed into Δlslig4 competent
karyotes. The transformation efficiencies of LiAc-mediated transfor- cells by electroporation using the Gene Pulser Xcell™ Electroporation
mation and PEG-mediated spheroplast transformation are shown in System (Bio-Rad Laboratories) with the following instrument settings:
Table 1. The transformation efficiency of LiAc-mediated transformation field strength, 7.5 kV/cm; capacitance, 25 μF; and resistance, 200 Ω.
was 131 transformants/μg of DNA with pKS-18S-hph digested with After electronic shock, cold 0.5 M sorbitol was added immediately to
ApaI, whereas the efficiency of PEG-mediated spheroplast transforma- the culture and the cells were transferred into YPD medium containing
tion was much higher at 2136 transformants/μg of DNA with pKS-18S- 0.5 M sorbitol and incubated at 30 °C for 10 h. Finally, the cells were
hph digested with ApaI. Contrary to expectations, the efficiency of LiAc- spread on the selected medium and incubated at 30 °C for 3 days.
mediated transformation was low. Transformation by the electroporation method resulted in an average of
The integration of targeted DNA at the rDNA locus has been known 60 transformants/μg of DNA of pKS-18S-hph digested with ApaI. Al-
to lead to a higher transformation efficiency in comparison with that at though this is the first report of the successful electroporation trans-
a single-copy gene locus (Fujii et al., 1990; Szostak and Wu, 1979). formation of L. starkeyi, the transformation efficiency of electroporation
Since the 18S rDNA locus is a multi-copy gene for targeted integration, was low. Hence, we attempted to increase the transformation efficiency
we further examined the transformation efficiency for targeted in- of electroporation.
tegration into the genome of L. starkeyi strain Δlslig4 at a single-copy
gene locus. A construct containing the LsTDH3 promoter/sNAT1/
LsTDH3 terminator DNA fragment for expression in L. starkeyi and 3.3. Optimization of transformation efficiency by electroporation
LsLIG4 5′ and 3′ untranslated regions was created. A scheme of the
transformation method is depicted in Supplementary Materials Fig. S1. We tested the following five parameters that appeared to contribute
Transformation of the LsTDH3 promoter/sNAT1/LsTDH3 terminator to transformation efficiency using the pKS-18S-hph digested with ApaI
DNA fragment into Δlslig4 cells enabled marker switching through for hph gene integration into the genome of L. starkeyi strain Δlslig4 at
homologous recombination at the disrupted chromosomal LsLIG4 locus the 18S rDNA locus of a multiple-copy gene: cell growth phase, cell
of the Δlslig4 genome. The transformants were selected in the presence density, osmotic stabilizers (sorbitol and sucrose), pretreatment agents
of nourseothricin. Successful marker switching was tested by assessing (LiAc and DTT), and electric conditions (voltage and resistance).
nourseothricin resistance and zeocin sensitivity. The correct replace- At four different phases of growth [early log phase (4.5–5.2 × 107
ment of drug markers was subsequently confirmed by the colony PCR cells/mL), mid-log phase (0.8–1.0 × 108 cells/mL), late log phase
method. We examined the transformation efficiencies of the LiAc- (1.3–1.5 × 108 cells/mL), and stationary phase (2.0 × 108 cells/mL)],
mediated transformation method and PEG-mediated spheroplast Δlslig4 cells were tested for suitability for electroporation transforma-
transformation method for targeted integration into the genome of L. tion. The collected cells were resuspended in electroporation buffer at a
starkeyi strain Δlslig4 at a single-copy gene locus using the above marker concentration of approximately 3.0 × 109 cells/mL for further elec-
switching strategy (Table 1). The efficiencies of LiAc-mediated trans- troporation procedures. The most efficient transformation frequency
formation and PEG-mediated spheroplast transformation were ex- was achieved with cells in the mid-log phase (Tukey's test, P < .05;
tremely low at 2 and 12 transformants/μg of DNA, respectively. Be- Fig. 1A). This observation is in agreement with previously reported
cause the extremely low efficiencies of these transformation methods transformation electroporation methods with other yeasts (Faber et al.,
interfere with the analysis of gene function by target gene deletion in L. 1994; Voronovsky et al., 2002).
starkeyi, development of a high-efficiency transformation method in L. It is known that cell density influences transformation efficiency by
starkeyi other than LiAc-mediated and PEG-mediated spheroplast electroporation (Delorme, 1989; Faber et al., 1994; Sánchez et al.,
transformation methods is essential for the functional study of various 1993; Wu and Letchworth, 2004). To investigate the effect of cell
genes at the molecular level. density on transformation efficiency by electroporation, Δlslig4 cells at
the mid-log phase were harvested, resuspended in electroporation
buffer at concentrations between 5.0 × 108 and 3.0 × 109 cells/mL,
3.2. Transformation of L. starkeyi strain Δlslig4 by the electroporation and transformed into the plasmid pKS-18S-hph digested with ApaI. It
method was not possible to prepare transformation samples up to a cell density
of 3.0 × 109 cells/mL because it was physically impossible to obtain
Because of its ease, speed, and efficiency in comparison with al- transformation samples at concentrations of > 3.0 × 109 cells/mL.
ternative transformation techniques, electroporation is the most pop- However, as shown in Fig. 1B, the number of transformants increased
ular method for fungal transformation (Liu et al., 2017; Manivasakam with increasing cell density (Tukey's test, P < .05). Since it was phy-
and Schiestl, 1993; Miklenić et al., 2015; Suga and Hatakeyama, 2001; sically impossible to prepare a cell suspension at a concentration of >
Wang et al., 2011; Watanabe et al., 2010; Wu and Letchworth, 2004). 3.0 × 109 cells/mL due to the low mobility, we concluded that a cell
Transformation of L. starkeyi by electroporation was based on the density of 3.0 × 109 cells/mL was optimum for electroporation trans-
electroporation procedure previously described for S. cerevisiae formation.
(Thompson et al., 1998). At the log phase, Δlslig4 cells were collected by An osmotic stabilizer with a low ionic strength, such as sorbitol or
centrifugation and then treated with 0.1 M LiAc and 10 mM DTT. The sucrose, stabilizes the host cell membrane against osmotic pressure and
treated cells were collected by centrifugation, washed with sterile increases the efficiency of biochemical reactions within the cell mem-
water, and resuspended in ionic electroporation buffer containing 0.5 M brane. Osmotic stabilization is needed during electroporation and the

4
H. Takaku, et al. Journal of Microbiological Methods 169 (2020) 105816

A) Table 3
70 Effect of LiAc and DTT on electroporation transformation efficiency. L. starkeyi
Δlslig4 cells were transformed with pKS-18S-hph digested with ApaI at the 18S
Transformants/µ g DNA

60 ribosomal DNA locus of a multiple-copy gene. Mid-log phase L. starkeyi Δlslig4

50 cells pretreated with LiAc and/or DTT were resuspended in 0.5 M ice-cold su-
crose at a final concentration of 3.0 × 109 cells/mL. One microgram of pKS-
40 18S-hph digested with ApaI was used for electroporation with the settings
30 25 μF, 6.25 V/cm, and 200 Ω. Data are presented as the mean ± SEM of the
results of three independent experiments.
20
LiAc (M) DTT (mM) Transformation efficiency (transformants/μg DNA)
10
0 10 –
0 0.1 10 12 ± 2
early log middle log late log stationary 0.2 10 1348 ± 208
0.4 10 42 ± 15
Growth phase 0.2 0 –
B) 0.2 25 707 ± 54
0.2 50 736 ± 214
120
Transformants/µ g DNA

100

80

60

40

20

0
0.5 1 3

Cell density (cells/mL 109)

Fig. 1. Effect of the cell growth phase and cell density on the electroporation
transformation efficiency of L. starkeyi. L. starkeyi Δlslig4 cells pretreated with
0.1 M LiAc and 10 mM DTT were resuspended in ice-cold sorbitol (0.5 M). One
microgram of pKS-18S-hph digested with ApaI was used for electroporation
with the settings 25 μF, 7.5 kV/cm, and 200 Ω. Data of three independent
Fig. 2. Effect of electric conditions (voltage, resistance, and capacitance) on
experiments in the bar graph are presented as the mean ± standard error of
electroporation transformation efficiency of L. starkeyi. Mid-log phase L. starkeyi
the mean (SEM). (A) Effect of cell growth phase on electroporation transfor-
Δlslig4 cells pretreated with 0.2 M LiAc and 10 mM DTT were resuspended in
mation efficiency. (B) Effect of cell density on electroporation transformation
0.5 M ice-cold sucrose at a final concentration of 3.0 × 109 cells/mL. Data in
efficiency. L. starkeyi Δlslig4 cells at the mid-log phase were harvested and re-
the plots are presented as the mean ± SEM of three independent experiments.
suspended in ice-cold sorbitol (0.5 M) at a cell density of 5.0 × 108, 1.0 × 109,
or 3.0 × 109 cells/mL.
and the best concentration of sucrose for the transformation was 0.5 M
(Tukey's test, P < .05); thus, 0.5 M sucrose was considered optimum.
Table 2
Effect of osmotic stabilizer on transformation efficiency of electroporation. L. A pretreatment procedure with LiAc and DTT can improve elec-
starkeyi Δlslig4 cells were transformed with pKS-18S-hph digested with ApaI at troporation transformation efficiency in some yeast species (Liu et al.,
the 18S ribosomal DNA locus of a multiple-copy gene. Mid-log phase L. starkeyi 2017; Markham et al., 2018; Suga and Hatakeyama, 2001; Thompson
Δlslig4 cells pretreated with 0.1 M LiAc and 10 mM DTT were resuspended in et al., 1998; Wang et al., 2011; Watanabe et al., 2010; Wu and
ice-cold sorbitol (0.5 M) or sucrose (0.25, 0.5, or 1 M) at a final concentration of Letchworth, 2004). LiAc can modify the cell wall structure and increase
3.0 × 109 cells/mL. One microgram of pKS-18S-hph digested with ApaI was the permeability of intact cells to nucleic acid (Brzobohatý and KovÁČ,
used for electroporation with the settings 25 μF, 7.5 kV/cm, and 200 Ω. Data 1986; Pham et al., 2011). DTT can reduce the tightness and increase the
are presented as the mean ± SEM of the results of three independent experi- porosity of the cell wall (Brzobohatý and KovÁČ, 1986; Ganeva et al.,
ments.
1995). To determine whether treatment with LiAc and/or DTT can
Osmotic Concentration (M) Transformation efficiency enhance the transformation efficiency of L. starkeyi, Δlslig4 cells were
stabilizer (transformants/μg DNA) incubated in various concentrations of LiAc and/or DTT. Notably, no
transformant colonies were observed without pretreatment with LiAc or
Sorbitol 0.5 40 ± 15
Sucrose 0.5 97 ± 10 DTT, suggesting that pretreatment with both LiAc and DTT was es-
Sucrose 0.25 7 ± 2 sential for successful electroporation transformation of L. starkeyi
Sucrose 1.0 4 ± 1 (Table 3). To determine the optimum LiAc concentration for high
transformation efficiency, the concentration of LiAc was varied from
0.1 to 0.4 while keeping the concentration of DTT at 10 mM. The
subsequent cell liquid culture. Therefore, we investigated the effect of highest number of transformants was obtained by treatment of cells
the concentration of sorbitol or sucrose as an osmotic stabilizer on the with 0.2 mM LiAc and 10 mM DTT (Tukey's test, P < .05; Table 3). The
efficiency of electroporation transformation. The concentration of su- addition of 0.2 mM LiAc improved the electroporation efficiency by
crose was varied from 0.25 to 1.0 M to determine the optimum con- approximately 110-fold, as compared with that of 0.1 mM LiAc. In re-
centration for high transformation efficiency. As shown in Table 2, the gard to DTT, concentrations of 25 and 50 with 0.2 M LiAc were also
use of 0.5 M sucrose instead of 0.5 M sorbitol as an osmotic stabilizer tested. The results showed that DTT at a concentration of > 25 mM had
increased transformation efficiency by 2.5-fold (Tukey's test, P < .05), a mild negative effect on transformation efficiency as compared with

5
H. Takaku, et al. Journal of Microbiological Methods 169 (2020) 105816

Table 4
Efficiency of the electroporation transformation method for L. starkeyi Δlslig4 using either pKS-18S-hph or pKS-LsLIG4-nat. Mid-log phase L. starkeyi Δlslig4 cells
pretreated with 0.2 M LiAc and 10 mM DTT were resuspended in 0.5 M ice-cold sucrose at a final concentration of 3.0 × 109 cells/mL. One microgram of DNA was
used for electroporation with the settings 25 μF, 3.75 kV/cm, and 800 Ω. Data are presented as the mean ± SEM of the results of three independent experiments.
Targeting efficiency was calculated by dividing the number of homologous integrants by the number of screened transformants. The number of homologous
integrants was confirmed by PCR.
Construct name Target site Transformation efficiency (transformants/μg DNA) Targeting efficiency (Homologous integrants/Transformants)

pKS-18S-hph 18S rDNA region 14,607 ± 3572 95% (95/100)

pKS-LsLIG4-nat LsLIG4 locus 273 ± 73 84.2% (16/19)

that of 10 mM DTT (Tukey's test, P < .05; Table 3), suggesting that the required for the PEG-mediated spheroplast transformation method. The
optimum concentration of DTT was 10 mM. proposed electroporation method is expected to advance molecular
A relatively high electric field is needed to form reversible pores in genetics studies of industrially useful oleaginous yeast, such as L. star-
the cell membrane to provide a temporary pathway for incorporation of keyi.
exogenous DNA. After cessation of the electric field, the pores gradually
shrink and reseal, and most cells recover. However, if the electric field Ethical approval
is above a certain level, most pores are unable to reseal (Kotnik et al.,
2015). Previous studies have shown that the efficiency of electropora- No human or animal studies were performed by any of the authors.
tion transformation of various yeast species can be improved by the
optimization of electric conditions (Faber et al., 1994; Liu et al., 2017; Informed consent
Miklenić et al., 2015; Sánchez et al., 1993; Takahashi et al., 2014;
Voronovsky et al., 2002; Wu and Letchworth, 2004). To optimize the Informed consent was obtained from all individual participants in-
electric conditions for L. starkeyi, various field strengths (2.5, 3.75, 5.0, cluded in the study.
6.25, and 7.5 kV/cm) and resistance values (200, 400, and 800 Ω) at a
constant capacitance (25 μF) were tested. As shown in Fig. 2, an electric Declaration of Competing Interest
field of 3.75 kV/cm, resistance of 800 Ω, and capacitance of 25 μF or
5 kV/cm, resistance of 400 Ω, and capacitance of 25 μ provided the The authors have no conflict of interest to declare in regard to the
highest transformation efficiency (Tukey's test, P < .05), in which the contents of this article.
former was tend to be superior than the latter. The number of trans-
formants was up to 16-fold higher under the optimum electric condi- Acknowledgments
tions (3.75 kV/cm, 800 Ω, and 25 μF) than that of the initial values
(6.25 kV/cm, 200 Ω, and 25 μF). This study was supported in part by a grant for the Development of
Production Techniques for Highly Functional Biomaterials Using Smart
Cells of Plants and Other Organisms from JSPS KAKENHI (grant
3.4. Comparison of transformation efficiency of L. starkeyi strain Δlslig4 by
number 18 K05401).
three different transformation methods

Appendix A. Supplementary data

We compared the efficiencies of the three most frequently used
methods for yeast transformation (i.e., LiAc-mediated transformation,
Supplementary data to this article can be found online at https://
PEG-mediated spheroplast transformation, and electroporation trans-
doi.org/10.1016/j.mimet.2019.105816.
formation). The electroporation transformation efficiency for integra-
tion into the genome of L. starkeyi strain Δlslig4 at the 18S rDNA locus of
References
a multi-copy gene was increased by approximately 111- and 7-fold as
compared with those of the LiAc-mediated transformation and PEG-
Ageitos, J.M., Vallejo, J.A., Veiga-Crespo, P., Villa, T.G., 2011. Oily yeasts as oleaginous
mediated spheroplast transformation, respectively (Tukey's test, cell factories. Appl. Microbiol. Biotechnol. 90, 1219–1227. https://doi.org/10.1007/
P < .05; Tables 1 and 4). Furthermore, the transformation efficiency s00253-011-3200-z.
for integration into the genome of L. starkeyi strain Δlslig4 at a single- Angerbauer, C., Siebenhofer, M., Mittelbach, M., Guebitz, G.M., 2008. Conversion of
sewage sludge into lipids by Lipomyces starkeyi for biodiesel production. Bioresour.
copy gene locus was increased by approximately 136- and 22-fold as Technol. 99, 3051–3056. https://doi.org/10.1016/j.biortech.2007.06.045.
compared with those of the LiAc-mediated transformation and PEG- Beligon, V., Christophe, G., Fontanille, P., Larroche, C., 2016. Microbial lipids as potential
mediated spheroplast transformation, respectively (Tukey's test, source to food supplements. Curr. Opin. Food Sci. 7, 35–42. https://doi.org/10.1016/
j.cofs.2015.10.002.
P < .05; Tables 1 and 4). We also confirmed highly efficient gene
Brzobohatý, B., KovÁČ, L., 1986. Factors enhancing genetic transformation of intact yeast
targeting of the electroporation transformation method for L. starkeyi cells modify cell wall porosity. Microbiology. 132, 3089–3093. https://doi.org/10.
Δlslig4 (Table 4). These results suggest that the electroporation proce- 1099/00221287-132-11-3089.
dure developed in this study has advantages for the functional study of Calvey, C.H., Willis, L.B., Jeffries, T.W., 2014. An optimized transformation protocol for
Lipomyces starkeyi. Curr. Genet. 60, 223–230. https://doi.org/10.1007/s00294-014-
various genes at the molecular level when high transformation effi- 0427-0.
ciency is needed. Dai, Z., Deng, S., Culley, D.E., Bruno, K.S., Magnuson, J.K., 2017. Agrobacterium tume-
faciens–mediated transformation of oleaginous yeast Lipomyces species. Appl.
Microbiol. Biotechnol. 101, 6099–6110. https://doi.org/10.1007/s00253-017-
4. Conclusions 8357-7.
Delorme, E., 1989. Transformation of Saccharomyces cerevisiae by electroporation. Appl.
Environ. Microbiol. 55, 2242–2246.
In this study, we successfully developed a highly efficient and Faber, K.N., Haima, P., Harder, W., Veenhuis, M., Geert, A., 1994. Highly–efficient
simple method for transformation of the oleaginous yeast L. starkeyi. electrotransformation of the yeast Hansenula polymorpha. Curr. Genet. 25, 305–310.
https://doi.org/10.1007/bf00351482.
The transformation efficiency of the developed electroporation proce-
Fujii, T., Kondo, K., Shimizu, F., Sone, H., Tanaka, J., Inoue, T., 1990. Application of a
dure was significantly increased by > 100-fold as compared with that ribosomal DNA integration vector in the construction of a brewer's yeast having al-
of the LiAc-mediated transformation method. This method does not pha–acetolactate decarboxylase activity. Appl. Environ. Microbiol. 56, 997–1003.
Ganeva, V., Galutzov, B., Teissie, J., 1995. Electric field mediated loading of
require some time-consuming steps, such as spheroplast preparation,

6
H. Takaku, et al. Journal of Microbiological Methods 169 (2020) 105816

macromolecules in intact yeast cells is critically controlled at the wall level. Biochim. Curr. Genet. 57, 233–239. https://doi.org/10.1007/s00294-011-0341-7.
Biophys. Acta 1240, 229–236. https://doi.org/10.1016/0005-2736(95)00181-6. Riley, R., Haridas, S., Wolfe, K.H., Lopes, M.R., Hittinger, C.T., Göker, M., Salamov, A.A.,
Gong, Z., Wang, Q., Shen, H., Hu, C., Jin, G., Zhao, Z.K., 2012. Co-fermentation of cel- Wisecaver, J.H., Long, T.M., Calvey, C.H., 2016. Comparative genomics of bio-
lobiose and xylose by Lipomyces starkeyi for lipid production. Bioresour. Technol. 117, technologically important yeasts. Proc. Natl. Acad. Sci. U. S. A. 113, 9882–9887.
20–24. https://doi.org/10.1016/j.biortech.2012.04.063. https://doi.org/10.1073/pnas.1603941113.
Hood, M.T., Stachow, C., 1990. Transformation of Schizosaccharomyces pombe by elec- Sánchez, M., Iglesias, F.J., Santamaría, C., Domínguez, A., 1993. Transformation of
troporation. Nucleic Acids Res. 18, 688. https://doi.org/10.1093/nar/18.3.688. Kluyveromyces lactis by electroporation. Appl. Environ. Microbiol. 59, 2087–2092.
Kosa, M., Ragauskas, A.J., 2011. Lipids from heterotrophic microbes: advances in meta- Sitepu, I.R., Garay, L.A., Sestric, R., Levin, D., Block, D.E., German, J.B., Boundy–Mills,
bolism research. Trends Biotechnol. 29, 53–61. https://doi.org/10.1016/j.tibtech. K.L., 2014. Oleaginous yeasts for biodiesel: current and future trends in biology and
2010.11.002. production. Biotechnol. Adv. 32, 1336–1360. https://doi.org/10.1016/j.biotechadv.
Kotnik, T., Frey, W., Sack, M., Meglič, S.H., Peterka, M., Miklavčič, D., 2015. 2014.08.003.
Electroporation–based applications in biotechnology. Trends Biotechnol. 33, Suga, M., Hatakeyama, T., 2001. High efficiency transformation of Schizosaccharomyces
480–488. https://doi.org/10.1016/j.tibtech.2015.06.002. pombe pretreated with thiol compounds by electroporation. Yeast. 18, 1015–1021.
Li, Q., Du, W., Liu, D., 2008. Perspectives of microbial oils for biodiesel production. Appl. https://doi.org/10.1002/yea.753.
Microbiol. Biotechnol. 80, 749–756. https://doi.org/10.1007/s00253-008-1625-9. Szostak, J.W., Wu, R., 1979. Insertion of a genetic marker into the ribosomal DNA of
Liang, M.H., Jiang, J.G., 2013. Advancing oleaginous microorganisms to produce lipid via yeast. Plasmid. 2, 536–554. https://doi.org/10.1016/0147-619x(79)90053-2.
metabolic engineering technology. Prog. Lipid Res. 52, 395–408. https://doi.org/10. Takahashi, S., Okada, H., Abe, K., Kera, Y., 2014. Genetic transformation of the yeast
1016/j.plipres.2013.05.002. Rhodotorula gracilis ATCC 26217 by electroporation. Appl. Biochem. Microbiol. 50,
Liu, H., Jiao, X., Wang, Y., Yang, X., Sun, W., Wang, J., Zhang, S., Zhao, Z.K., 2017. Fast 624–628. https://doi.org/10.1134/S0003683814110040.
and efficient genetic transformation of oleaginous yeast Rhodosporidium toruloides by Thompson, J.R., Register, E., Curotto, J., Kurtz, M., Kelly, R., 1998. An improved protocol
using electroporation. FEMS Yeast Res. 17https://doi.org/10.1093/femsyr/fox017. for the preparation of yeast cells for transformation by electroporation. Yeast. 14,
fox017. 565–571. https://doi.org/10.1002/(SICI)1097-0061(19980430)14:6<565::AID-
Manivasakam, P., Schiestl, R.H., 1993. High efficiency transformation of Saccharomyces YEA251>3.0.CO;2-B.
cerevisiae by electroporation. Nucleic Acids Res. 21, 4414. https://doi.org/10.1093/ Voronovsky, A.A., Abbas, C.A., Fayura, L.R., Kshanovska, B.V., Dmytruk, K.V., Sybirna,
nar/21.18.4414. K.A., Sibirny, A.A., 2002. Development of a transformation system for the flavino-
Markham, K.A., Vazquez, S., Alper, H.S., 2018. High-efficiency transformation of genic yeast Candida famata. FEMS Yeast Res. 2, 381–388. https://doi.org/10.1016/
Yarrowia lipolytica using electroporation. FEMS Yeast Res. 18https://doi.org/10. S1567-1356(02)00112-5.
1093/femsyr/foy081. foy081. Wang, J.-H., Hung, W., Tsai, S.-H., 2011. High efficiency transformation by electro-
Miklenić, M., Žunar, B., Štafa, A., Svetec, I.-K., 2015. Improved electroporation procedure poration of Yarrowia lipolytica. J. Microbiol. 49, 469–472. https://doi.org/10.1007/
for genetic transformation of Dekkera/Brettanomyces bruxellensis. FEMS Yeast Res. 15, s12275-011-0433-6.
fov096. https://doi.org/10.1093/femsyr/fov096. Watanabe, J., Uehara, K., Mogi, Y., Suzuki, K., Watanabe, T., Yamazaki, T., 2010.
Oguri, E., Masaki, K., Naganuma, T., Iefuji, H., 2012. Phylogenetic and biochemical Improved transformation of the halo-tolerant yeast Zygosaccharomyces rouxii by
characterization of the oil–producing yeast Lipomyces starkeyi. Antonie Van electroporation. Biosci. Biotechnol. Biochem. 74, 1092–1094. https://doi.org/10.
Leeuwenhoek 101, 359–368. https://doi.org/10.1007/s10482-011-9641-7. 1271/bbb.90865.
Oguro, Y., Yamazaki, H., Shida, Y., Ogasawara, W., Takagi, M., Takaku, H., 2015. Wu, S., Letchworth, G.J., 2004. High efficiency transformation by electroporation of
Multicopy integration and expression of heterologous genes in the oleaginous yeast, Pichia pastoris pretreated with lithium acetate and dithiothreitol. BioTechniques. 36,
Lipomyces starkeyi. Biosci. Biotechnol. Biochem. 79, 512–515. https://doi.org/10. 152–154. https://doi.org/10.2144/04361DD02.
1080/09168451.2014.982504. Yang, X., Jin, G., Gong, Z., Shen, H., Song, Y., Bai, F., Zhao, Z.K., 2014. Simultaneous
Oguro, Y., Yamazaki, H., Ara, S., Shida, Y., Ogasawara, W., Takagi, M., Takaku, H., 2017. utilization of glucose and mannose from spent yeast cell mass for lipid production by
Efficient gene targeting in non-homologous end-joining-deficient Lipomyces starkeyi Lipomyces starkeyi. Bioresour. Technol. 158, 383–387. https://doi.org/10.1016/j.
strains. Curr. Genet. 63, 751–763. https://doi.org/10.1007/s00294-017-0679-6. biortech.2014.02.121.
Papanikolaou, S., Aggelis, G., 2011. Lipids of oleaginous yeasts. Part I: biochemistry of Zhao, X., Kong, X., Hua, Y., Feng, B., Zhao, Z., 2008. Medium optimization for lipid
single cell oil production. Eur. J. Lipid Sci. Technol. 113, 1031–1051. https://doi. production through co-fermentation of glucose and xylose by the oleaginous yeast
org/10.1002/ejlt.201100014. Lipomyces starkeyi. Eur. J. Lipid Sci. Technol. 110, 405–412. https://doi.org/10.
Pham, T.A., Kawai, S., Murata, K., 2011. Visualization of the synergistic effect of lithium 1002/ejlt.200700224.
acetate and single–stranded carrier DNA on Saccharomyces cerevisiae transformation.