Bacterial strains and culture conditions. All bacterial and yeast strains are described in Table 1 and unless otherwise stated were grown in the following media: Escherichia coli all were grown in LB Miller (LB broth) culture medium (Difco Laboratories, Inc.); Limosilactobacillus fermentum was grown in Lactobacilli MRS (MRS broth; Difco Laboratories, Inc.) media; and S. cerevisiae strains were grown in synthetic media with glucose (SDCAA; 2% D-glucose, 0.67% yeast nitrogen base with ammino acids supplement minus tryptophan (-Trp), 0.5% peptone, 0.54% sodium phosphate dibasic anhydrous, and 0.86% sodium phosphate monobasic monohydrate; Sigma-Aldrich Inc.) without induction. To induce yeast constructs, synthetic media SGCAA (-Trp) containing galactose (2% D-glucose was replaced with 2% galactose) was used to grow cells overnight. Control S. cerevisiae EBY100 strain was grown in SDCAA with complete amino acids supplementation. Antibiotics were added to LB media where appropriate at the following concentrations: carbenicillin (Car; Sigma-Aldrich, Inc.) at 100 µg/mL or kanamycin (Kan; Sigma-Aldrich, Inc.) at 50 µg/mL. E. coli cultures were grown at 37°C with agitation at 200 rpm and L. fermentum at 37℃ stationary. S. cerevisiae strains were grown at 30℃ with agitation at 200 rpm.
Construct, strains, and plasmids.
Endolysin LysKB317 (GenBank: KP027015.1) gene template from pRham N-His Kan::LysKB317 (Table 1; [32]) was PCR amplified using primers described in Table 2. Amplified LysKB317 was inserted into reversed PCR linearized pYD5 vector (generously gifted by Dr. Mark Dumont [51]) using Gibson cloning kit (New England BioLabs Inc.). To construct pYD5::LysKB317, endolysin gene was inserted between Aga2 signal cleavage site and upstream of V5-eptipote and the glycine serine (G4S)3 linker (Fig. 1A and Fig. 1C) exposing the N-terminus enzymatic domain of LysKB317 upon expression. To construct pYD1::LysKB317, synthetic genes containing V5-epitope tag, (G4S)3 linker and endolysin LysKB317 in the backbone of pET21a + listed in Table 1 (GenScript) were used as DNA template for PCR amplification using primers listed in Table 2. The vector pYD1 (Addgene plasmid # 73447) was linearized with primers (Table 2) and PCR amplified endolysin fragment was inserted between the end of Aga2p and stop codon after the 6xHis-tag deleting original (G4S)3 linker, enterokinase site, V5 tag and His6 tag using Gibson cloning and exposing the C-terminus cell wall binding domain of LysKB317 upon expression. Constructs were transformed into E. coli Top10 chemically competent cells (Invitrogen) and verified in-house by Sanger sequencing prior to S. cerevisiae EBY100 transformation.
Bacterial and yeast culture growth curve. Relative backscatter at gain = 10 setting was read in a FlowerPlate 48-well microtiter plate with clear flat bottom and aluminum sealing cover (Millipore Sigma) using a BioLector plate reader (Beckman Coulter). Triplicate of each strain at OD600 nm of 0.6, was seeded at 1:100 ratio (10 µL bacteria or yeast to 990 µL appropriate media) with a final volume of 1 mL measured for 72 h at 30℃.
Immunostaining and microscopy. A confocal microscope (Nikon Eclipse C1Plus; Nikon) was used to image surface-displayed endolysin LysKB317 in S. cerevisiae. Induced or uninduced yeast cells were grown in appropriate SGCAA (induced) or SDCAA (uninduced) media at 30℃ with 200 rpm overnight. One hundred microliter of yeast cells were washed twice with 100 µL 1x phosphate buffer saline (PBS) + 3% bovine serum albumin (BSA) buffer, pH 7.4 at 800 x g for 2 min. Primary rabbit anti-LysKB317 (1:500; ThermoFisher) and secondary Goat anti-rabbit Alexa 488 (1:5,000; Invitrogen) were incubated with the samples at room temperature for 30 min with gentle mixing. Cells were washed with 1x PBS + 3% BSA before taking 5 µL resuspended yeast sample for visualization using 488 nm argon laser 513/30 band pass emission filter on the photomultiplier tubes (PMT) detector.
Western blot analysis. Galactose induced yeast cells were harvested and boiled in a water bath with 2x Laemmli sample buffer at 1: 1 ratio for 10 min (Bio-Rad). Protein samples were separated on a sodium dodecyl sulfate (SDS) – polyacrylamide gel (Bio-Rad Any kD tris-glycine Mini-Protean TGX precast protein gel) per manufacturer’s instructions. Following SDS-PAGE, proteins were transferred onto a 0.2 µm pore size low fluorescence polyvinylidene difluoride (PVDF) membrane (Bio-Rad) using a Trans-Blot turbo transfer system (Bio-Rad). Transfer of protein was verified using Ponceau S Staining (Cell Signaling Technology, Inc.). Membrane non-specific blocking was performed with Blocker BSA in TBS (ThermoFisher) at room temperature for 1 h. Rabbit anti-LysKB317 antibody (1:500; ThermoFisher) was applied and incubated overnight with gentle swirling at 4℃ in tris-buffered saline (TBS) containing 0.1% Tween-20 (Sigma-Aldrich). Secondary goat anti-rabbit antibody conjugated to Alexa 488 at 1:5,000 ratio (ThermoFisher) was applied and incubated with gentle swirling at room temperature. Membrane protein bands were detected using ChemiDoc XRS + imaging system (Bio-Rad).
Live/dead viability assay. Induced/uninduced yeast cells of 500 µL at OD600 nm = 2 and target bacteria of 500 µL at OD600 nm = 2 were washed separately with nano-pure water twice at 800 x g for 2 min and dilute bacteria and yeast at 1:1,000 with nano-pure water. Live/dead stain (Invitrogen) were added immediately to the washed bacteria and incubated at room temperature for 10 min before mixing and the bacteria to the yeast in a sterile fresh 1.5 mL Eppendorf tube. In triplicate biological repeats (n = 3), 5-µL of yeast-bacteria sample were transferred into a SD025 chamber slide (Nexcelom) and the live and total bacteria were numerate every 5 min with Cellometer X2 (Nexcelom) using a 0.5–5 µm cell size gating parameter.
Small-Scale corn fermentation. Fermentation method was adopted from Khatibi et al. with some modification [19, 45]. Briefly, corn mash (approximately 33% solids) was obtained from a commercial dry grind ethanol fermentation facility and stored at -20℃ until use. In each 25 mL Erlenmeyer flask, 12.0 mL of autoclaved corn mash was pretreated with 10 µL glucoamylase (Allcoholase II Liquid 300; Alltech) and 200 µL 12% (NH4)2SO4 overnight at 37℃ with agitation at 200 rpm. An addition of 6.8 mL of 25% galactose for induction or nano pure sterile water (for non-induced) was added to the flasks to final concentration of 20% solids corn mash and 10% galactose as described previously [19, 52]. Inoculation of yeast strains (induced/uninduced or untransformed), 1 mL of OD600 nm = 160 and bacteria, 200 µL of OD600 nm = 8 (or nano pure water for control flask without infection), were added for final volume of 20 mL per fermentation flask. Rubber stoppers with G20 1½ PrecisionGlide Needle (Becton Dickinson) to vent gas (CO2) were used as closures. Fermentation flasks were incubated at 30℃ with 50 rpm for 72 h. At time 0 h, 24 h, 48 h, and 72 h, 500 µL of sample was removed for analysis and the flasks were immediately returned to incubator. Samples at each time point were diluted at 1:100,000 ratio in fresh sterile 1.5 mL Eppendorf tubes with nano-pure water, plated using Eddy Jet 2 spiral plater (E mode 50; IUL Instruments) on MRS-agar containing 10 µg/mL cycloheximide and incubated at 37℃ overnight. Colonies were enumerated using a Flash & Go plate reader (IUL Instruments) to determine CFU/mL where ten colonies or more were counted. Remainder of 500 µL samples were spun down using a bench top centrifuge at max speed for 1 min and 200 µL supernatant removed for HPLC analysis.
HPLC metabolite analysis. A high-performance liquid chromatography (HPLC) system (Shimadzu) with 300 mm Aminex HPX-87H column (0.5 mL/min 5 mM H2SO4 at 65℃; Bio-Rad) with refractive index detector (RID) was used to detect fermentation metabolites (acetic acid, ethanol, glucose, and lactic acid; [45]).
Expression and purification of His-tagged LysKB317. A single colony of E. coli harboring pRham N-His Kan::LysKB317 (Table 1) was isolated and inoculated into 5 mL LB media with Kan at 37°C and 200 rpm shaking overnight. This was used to inoculate fresh 25 mL LB with Kan at 1:100 ratio, which was then incubated at 37°C and 200 rpm until OD600 nm reached 0.6 [32]. Culture was induced with 0.2% (w/v) L-rhamnose (Sigma-Aldrich) for at least 16–18 h. Cells were centrifugated at 8,000 x g for 15 min at 4°C and the spent media was decanted. Cell pellets were then lysed using B-Per (Invitrogen) with DNase I, RNase I and lysozyme (10 U/mL; ThermoFisher) and incubated at room temperature for 20 min. Whole cell lysate were centrifuged at 15,000 x g for 5 min. and soluble supernatant pipetted into His-Spin column (ZymoResearch) following the manufacture’s protocol for Ni-NTA protein purification. Purified protein concentration was quantified using Qubit Protein Assay Kit (ThermoFisher) and subsequently used as a positive control for SDS-PAGE and western blot analysis.
Measurement of spent synthetic SDCAA media pH. Spent media approximately 500 µL was pipetted into a 1.5 mL Eppendorf tube in triplicates. A micro-pH electrode (Hl1093B, Hanna Instruments) was inserted into the media to measure the pH.
Statistical analysis. Where appropriate, experimental results were analyzed using two-tailed student t-test with equal variance to determine statistically significant at *p < 0.05 (Microsoft 365 Excel version 2202).