Yeast strains, media and culturing conditions
All mutant strains described in this study were created in the K. lactis
GG799 expression strain background (Colussi & Taron, 2005
). Kluyveromyces lactis
strains were routinely grown in YPGal medium (1% yeast extract, 2% peptone, 2% galactose, optionally containing 2% agar for solid medium), YPGlu medium (1% yeast extract, 2% peptone and 2% glucose, ±2% agar) or YPGly medium (1% yeast extract, 2% peptone and 2% glycerol, ±2% agar) at 30 °C for 2–3 days. Nitrogen-free yeast carbon base (YCB) medium and acetamide were from New England Biolabs (Ipswich, MA). G418 was from Sigma-Aldrich (St. Louis, MO) and used in YPGal medium at a final concentration of 200 μg mL−1
. In all experiments, samples of spent culture medium (SCM) were prepared by clearing cells from aliquots of liquid cultures grown to saturation by centrifugation at 4000 g
for 10 min.
All oligonucleotide primers used for PCR-based assembly of gene disruption DNA fragments and for whole-cell PCR identification of chromosomal integration events are presented in Supporting Information, Table S1
DNA constructs for disruption of K. lactis
genes encoding putative pfam00026 proteases were assembled using a multistep ‘PCR-knitting’ strategy shown in . The PCR template vector pCT468 contained an expression cassette consisting of the Aspergillus nidulans
acetamidase gene (amdS
) cloned downstream of the S. cerevisiae
alcohol dehydrogenase (ADH1
) promoter. This cassette was flanked on both sides by 300 bp directly repeating DNA sequences comprising the amdS
gene's native 3′ untranslated region (UTR). The entire cassette was assembled by gene synthesis and cloned into the KpnI and HindIII sites of pUC57 (GenScript USA, Piscataway, NJ) and its sequence is available from GenBank (#HM015509). In the PCR knitting strategy, two halves of a gene disruption fragment were amplified by PCR using primer pairs KO1/KO2 and KO3/KO4. Primers KO1 and KO4 each contained tails homologous to the 5′ and 3′ ends of the target chromosomal integration site, respectively. Primer tail lengths were from 80- to 125-bp long depending on the individual gene being deleted (precise tail lengths are noted in Table S1
). Additionally, the 3′ end of the ‘left’ amplicon and the 5′ end of the ‘right’ amplicon had overlapping complimentary regions. Amplification was performed in 1 × Phusion HF buffer containing 2% dimethyl sulfoxide, 1 mM MgCl2
, 200 μM dNTPs, 0.5 μM of each primer, 125 ng pCT468 and 0.04 U Phusion™ DNA polymerase (New England Biolabs) in a total reaction volume of 100 μL. Thermocycling consisted of incubation at 98 °C for 40 s followed by 35 cycles of successive incubations at 98 °C for 10 s and 72 °C for 2 min. After thermocycling, a final extension was performed at 72 °C for 8 min.
Fig. 1 Gene deletion and amdS marker recycling strategy. (a) A linear gene disruption DNA fragment was assembled by three rounds of PCR. The fragment contained the Aspergillus nidulans amdS gene flanked by directly repeating 300-bp segments of its native 3′ (more ...)
The two halves of the disruption fragment were ‘knitted’ together by an additional round of PCR. In this strategy, complimentary regions in the left and right amplicons annealed to each other and were extended by the polymerase to form a full-length disruption fragment template that was subsequently amplified by extension of primers KO5 and KO6. Primers KO5 and KO6 also contained tails of additional chromosomal targeting sequence (~80–125 bp) that elongate the targeting sequence first introduced by primers KO1 and KO4. Thus, final amplified disruption fragments contained from 160- to 250-bp chromosomal targeting sequence on each end, depending upon the specific primer lengths used for each gene. The reaction conditions for knitting PCR were identical to those above, with the exception that 500 ng each of the left and right amplicons was used as template, and an extension of 3 min at 72 °C was used during thermocyling.
Whole-cell PCR was used to assess the integrity of a target chromosomal locus either after integrative transformation of cells with a disruption fragment or after out-recombination of the amdS marker, using primer pairs that direct amplification of specific diagnostic DNA fragments (). Candidate K. lactis strains were patched onto YCB agar plates containing 5 mM acetamide and incubated overnight at 30 °C. A sterile pipette tip was used to scrape approximately 1 mm2 of cells into 25 μL of a 1 mg zymolyase mL−1 solution in 30 mM sodium phosphate (the Associates of Cape Cod, East Falmouth, MA). The cells were incubated at 25 °C for 1 h to allow for cell wall digestion, after which the cells were lysed and DNA was denatured by incubation at 98 °C for 10 min. The temperature was lowered to 80 °C and 75 μL 1 × ThermoPol buffer containing 200 μM dNTPs, 0.8 μM of each ID primer and 5 U Taq DNA polymerase (New England Biolabs). Thermocycling consisted of 30 cycles of successive incubations at 95 °C for 30 s, 50 °C for 30 s and 72 °C for 2 min. After cycling, a final extension was performed at 72 °C for 10 min.
Construction of protease-deficient strains
After its assembly by PCR, 2 μg of a gene disruption DNA fragment was introduced into K. lactis GG799 competent cells as described by the manufacturer (New England Biolabs) followed by selection of transformants by growth on YCB agar medium supplemented with 5 mM acetamide for no more than 3 days at 30 °C. Successful disruption of a target chromosomal locus was assessed by whole-cell PCR (see previous section).
To recycle the amdS marker, a strain harboring an amdS+ disrupted target allele was grown in the absence of selection on YPD agar to permit recombination between the directly repeating UTR regions that flank the amdS gene (). Null mutants lacking the amdS gene were then isolated by three rounds of restreaking on YCB agar supplemented with 10 mM fluoroacetamide (Sigma-Aldrich) and 0.1% (w/v) ammonium sulfate, and incubation for 2–3 days at 30 °C.
Analysis of cell health and growth
Growth curves and measurement of cellular biomass produced during culturing of null mutant strains was performed by growing strains in 250-mL shake-flasks containing 100 mL YPGal medium at 30 °C for 72 h. Cell growth was measured by OD600 nm in triplicate in an Ultraspec 2100 Pro spectrophotometer (GE Healthcare, Piscataway, NJ). After 12, 24, 48 and 72 h of growth, 10 mL of each culture were removed and cells were pelleted by centrifugation at 4000 g for 10 min. Cell pellets were washed once in water to remove medium components and dried in disposable preweighed aluminum pans (ThermoFisher Scientific, Waltham, MA). The dry cell mass (g L−1) was calculated.
Null mutant strains were examined for defects in cell wall integrity by assessing their growth compared with wild-type GG799 cells on YPD agar medium supplemented with the cell wall-perturbing compounds Congo Red (10 μg mL−1
; Sigma-Aldrich) or Calcofluor white (100 μg mL−1
; Sigma-Aldrich) at 30 °C for 2–3 days (Kopecká & Gabriel, 1992
; Ram et al., 1994
Protease activity in SCM of each mutant strain was measured by two methods. In each assay, protease activity was determined directly from SCM prepared from cells grown to saturation. Measurements were normalized to each culture's final OD600 nm to account for any slight differences in cell growth. All reactions were performed in triplicate.
Total protease activity in SCM was measured using IRDye 800RS casein (LI-COR Biosciences, Lincoln, NE) as a fluorogenic substrate. Reactions were carried out by incubating 50 μL of SCM with 19 pmol of the IRDye 800RS casein substrate (0.126 μM final concentration) in 150 μL of 0.05 M Tris-HCl buffer (pH 7.2) containing 0.05% Tween 20 (v/v) and 0.01% sodium azide (w/v) at 30 °C for 20 h in the dark. The fluorescence intensity of reactions was measured using the 800-nm channel of an Odyssey® Infrared Imaging System (LI-COR).
Activity of subtilisin-type and yapsin-like proteases in SCM was assayed using the chromogenic substrate Z-Tyr-Lys-Arg-pNA (Bachem, Switzerland). The reaction was carried out by incubating 50 μL of SCM with 0.15 mM substrate in 50 mM Tris-HCl (pH 7.2) containing 1 mM CaCl2 in a total volume of 100 μL at 30 °C for 24 h. The reaction was terminated by the addition of EDTA to a final concentration of 10 mM. Liberation of p-nitroanilide (pNA) was measured at 405 nm in a SpectraMax M5 spectrophotometer (Molecular Devices, Sunnyvale, CA).
Gaussia princeps luciferase expression and assay
Secreted expression of G. princeps
luciferase (Gluc) in K. lactis
GG799 cells has been reported previously (Read et al., 2007
). Briefly, DNA encoding the G. princeps
luciferase ORF was cloned downstream of DNA encoding the K. lactis
α-mating factor secretion leader in the integrative K. lactis
expression vector pGBN19 to yield pGBN19-Gluc. In the present study, 2 μg of pGBN19-Gluc was linearized by SacII digestion and introduced into each of the protease null mutant strains using a lithium acetate transformation procedure (Read et al., 2007
), after which transformants were selected by growth on YPGal medium containing 200 μg G418 mL−1
. Strains harboring a single-copy insertion of pGBN19-Gluc into the LAC4
locus of the K. lactis
chromosome were identified by whole-cell PCR as described previously (Read et al., 2007
To assay Gluc enzyme activity present in SCM, strains secreting the Gluc protein were grown in triplicate 20 mL YPGal cultures for 40 h at 30 °C with shaking. Gluc activity was measured by mixing 25 μL of SCM and 50 μL of 1 ×Gaussia luciferase assay buffer (New England Biolabs) in a Microfluor black flat-bottom microtiter plate (Thermo Labsystems, Franklin, MA). Luminescence was immediately measured in an LMax luminometer (Molecular Devices) in relative light units (RLU). To limit any effect that variation in culture density had on luciferase abundance, RLU were normalized to the cell density of each culture (OD600 nm units).
Human interferon Hy3 expression
For secreted expression of interferon Hy3 in K. lactis
, the Gateway destination vector pDest-920 was created by Gateway conversion of the integrative K. lactis
expression vector pKLAC1 (Colussi & Taron, 2005
) as follows. Vector pKLAC1 was digested with XhoI (New England Biolabs) and the cohesive ends were filled in with Klenow DNA polymerase (New England Biolabs) to produce a blunt-ended DNA fragment that was ligated to the Gateway reading frame B cassette (Invitrogen, Carlsbad, CA). The ligation reaction was used to transform Escherichia coli
DB3.1 cells. Colonies were selected on Luria–Bertani medium containing 100 μg ampicillin mL−1
and 15 μg chloramphenicol mL−1
, and cloned vectors were screened by restriction digest for insert orientation. Correct clones were sequenced through the Gateway cassette junctions to ensure that the reading frame was maintained.
Interferon Hy3 Gateway entry clones were generated as described previously (Esposito et al., 2005
). These clones contained sequences for reconstitution of the Kex protease site (KR↓) and two Ste13p cleavage sites (EA↓EA↓) immediately upstream of the start codon of the interferon Hy3 gene (GenBank AF085805). Entry clones were recombined into the expression vector pDest-920 using Gateway LR recombination (Invitrogen) as per the manufacturer's protocols to generate pDest-920-IFN. Assembled pDest-920-IFN clones were verified by restriction digestion, and high-quality DNA was prepared using the GenElute XP Midiprep Kit (Sigma-Aldrich). Ten micrograms of pDest-920-IFN were digested with SacII (New England Biolabs) and the linear expression cassette was isolated, concentrated using a QiaQuick spin column (Qiagen, Valencia, CA) and used to transform each of the K. lactis
protease null mutant strains.
Fed-batch K. lactis fermentation
Yeast defined fermentation medium (YDFM) was composed of (per liter) 11.83 g KH2PO4, 2.29 g K2HPO4, 30 g glucose, 1 g MgSO4·7H2O, 10 g NH4SO4, 0.33 mg CaCl2·2H2O, 1 g NaCl, 1 g KCl, 5 mg CuSO4·5H2O, 30 mg MnSO4·H2O, 8 mg Na2MoO4·2H2O, 10 mg ZnCl2, 1 mg KI, 2 mg CoCl2·6H2O, 0.4 mg H3BO3, 30 mg FeCl3·6H2O, 0.8 mg biotin, 20 mg Ca-pantothenate, 15 mg thiamine, 16 mg myo-inositol, 10 mg nicotinic acid and 4 mg pyridoxine. The phosphate buffer and glucose were sterilized in the bioreactor after which the remaining components were added from sterile stock solutions after cooling.
A 1 L seed culture of the K. lactis
background carrying an integrated vector for expression of interferon Hy3 was grown to mid-log phase (OD600 nm
) in YPD medium, after which 100 mL was used to inoculate 1 L of YDFM in a Bioflow 110 fermentor (New Brunswick Scientific, Edison, NJ). The culture was grown for 18.75 h (OD600 nm
) until a decline in the growth rate was detected using a BugEye 100C noninvasive biomass monitor (BugLab, Danville, CA) indicating a depletion of nutrients from the batch phase. At this point, a glucose feed was started. The glucose feed medium consisted of (per liter) 500 g glucose, 10 g MgSO4
O, 16.5 g CaCl2
O, 1 g NaCl and 1 g KCl. Trace metals and vitamins were added to 1.5 and four times the concentration present in YDFM, respectively. Glucose feeding was performed for 4 h at 0.38 mL min−1
, after which the feed was stopped and a galactose feed was initiated to induce production of interferon Hy3. The galactose feed medium had the same composition as glucose feed medium, except 500 g galactose was substituted for glucose. Galactose feeding was performed for 24 h at 0.38 mL min−1
after which the culture was chilled to 10 °C for collection of SCM containing interferon Hy3. After chilling, the culture was centrifuged at 4000 g
for 15 min to remove cells. The SCM was filtered by passage through a Sartopore 2 capsule (0.2 μM; Sartorius Stedim, Aubagne, France) and was immediately stored at 4 °C.
Western blotting was used to assess the quality of the Gluc and interferon Hy3 proteins secreted from various K. lactis strains. SCM (10 μL) was analyzed by denaturing polyacrylamide gel electrophoresis (SDS-PAGE) on a 10–20% Tris-glycine polyacrylamide gel (Cosmo Bio Company, Tokyo, Japan). Separated proteins were transferred to nitrocellulose membrane (Whatman GmbH, Dassel, Germany). For detection of Gluc, the membrane was probed with an anti-GLuc antibody (1 : 3000 dilution; New England Biolabs), followed by a horseradish peroxidase (HRP)-conjugated anti-rabbit secondary antibody (1 : 2000 dilution; Cell Signaling Technology, Danvers, MA). For detection of interferon Hy3, the membrane was probed with a rabbit polyclonal antibody (1 : 1000 dilution) generated against the C-terminal tail of Hy3 (Covance, Princeton, NJ) followed by an HRP-conjugated anti-rabbit secondary antibody. Protein-antibody complexes were visualized using either LumiGlo™ (Cell Signaling Technology) or West Pico (ThermoFisher Scientific) detection reagents.