DEVELOPMENT OF A COST-EFFECTIVE METHOD TO CONTROL BACTERIAL CONTAMINATION OF BIOFUEL ETHANOL FERMENTATION USING LYTIC ENZYMES

• Sponsoring Institution: National Institute of Food and Agriculture
• Project Status: COMPLETE
• Funding Source: SMALL BUSINESS GRANT
• Reporting Frequency: Annual
• Accession No.: 1019566
• Grant No.: 2019-33610-29780
• Cumulative Award Amt.: $100,000.00
• Proposal No.: 2019-00748
• Project Start Date: Jul 15, 2019
• Project End Date: Oct 14, 2020
• Grant Year: 2019
• Program Code: [8.8]- Biofuels and Biobased Products

Recipient Organization
IBIOCAT, INCORPORATED
2408 BEDFORD DR
CHAMPAIGN,IL 618207704

Performing Department
(N/A)

Non Technical Summary
One of the major challenges to the profitability of bioethanol plants is yield loss from microbial fermentation process upsets. Bacterial contamination is a prevalent problem in the U.S. commercial fuel ethanol fermentations as these are typically performed under non-sterile conditions (Connolly, 1997; Bischoff et al, 2009; Rich et al, 2017). Due to the non-sterile conditions that exist at fuel ethanol production plants, it is estimated that a production facility can experience up to 4% in ethanol production loss. Other independent surveys and trade magazines estimate an average industry wide loss of 2% annually. A recent published article in the Ethanol Producer Magazine reports a $4.5 million loss in revenue for a 100-million-gallon annual production ethanol plant (Lewis, Sept. 2016). Using that figure and the U.S. Energy Information Agency (EIA) projected annual production of 15.8 billion gallons of ethanol for 2017, one can calculate a total industry wide loss of $711 million. This figure includes in addition to loss in ethanol production output, other costs for cleaning and downtime and represents a significant loss to an industry that operates with thin margins (shown in Figure in red color) and sometimes without breaking even as illustrated in Figure 1 for the period of October 2014-August 2018 (Iowa State University; accessed 9/2/18). The industry continues to look for ways to reduce contamination as this would significantly improve industry profitability, reduce environmental footprint by reducing energy use to dry the co-product that is sold as animal feed, reduce energy required for distillation in beer still and increase efficiency of carbohydrate conversion to ethanol. The animal feed co-product will have a small increase in protein level with reduced Maillard reaction product formation that contribute to browning and less efficient uptake of protein by animals. These benefits are measurable and consistent with USDA NIFA goals for the SBIR program for improved agricultural feedstock utilization. The proposal aims to demonstrate the use of a USDA patented bacteriophage lysins to control lactic acid bacterial (LAB) contamination of fuel ethanol fermentation. Once an effective blend of lysins is demonstrated in Phase I the goal is to move forward with cost effective fermentation production of these lytic enzymes by an enzyme company or other commercial companies that currently supply antimicrobials to the fuel ethanol industry.

Animal Health Component

100%

Research Effort Categories

Basic

(N/A)

Applied

100%

Developmental

(N/A)

Classification

Knowledge Area (KA)	Subject of Investigation (SOI)	Field of Science (FOS)	Percent
511	1510	1100	100%

Knowledge Area
511 - New and Improved Non-Food Products and Processes;

Subject Of Investigation
1510 - Corn;

Field Of Science
1100 - Bacteriology;

Keywords

bacteriophage lysins

biofuels

fuel ethanol fermentation

lactic acid bacteria

lytic enzymes

Goals / Objectives
This project aims to improve production of bioethanol at the existing ethanol biorefineries by developing an optimal blend of lytic enzymes that significantly reduces losses from bacterial contamination.The objectives include 1) selection of lytic enzymes; 2) optimization of blend for use in the early stages of the ethanol fermentation process, and 3) testing of lytic enymes under industrial conditions with ethanol plant feedstocks.

Project Methods
Objective 1- Growing of Bacterial and Yeast Cultures Lactic acid cultures will be revived from lyophilized cultures in MRS Broth (De Man et. al, 1960) or in pre-prepared Liver Infusion Broth (Difco No. 226920) at 35 g/L, as described (https://nrrl.ncaur.usda.gov/cgi-bin/usda/medium/12/nrrl_medium_12.pdf). Stock cultures will be grown and maintained on MRS agar slants. Broth cultures will be typically incubated in stationary mode at the appropriate temperatures (28-37 °C, depending on strain optimum growth temperature) to an optical density of 1.0 when measured at A620nm. E. colistrains engineered to produce endolysins will be grown in Luria-Bertani broth (LB; Difco Laboratories) at 37° C with shaking, while lactobacilli will be grown in de Man, Rogosa and Sharp broth (MRS; Difco Laboratories) at 37° C without shaking. S. cerevisiaetest strains will be grown overnight in a yeast extract peptone (YP; Becton Dickinson and Co., Sparks, MD) medium that is supplemented with 4% (w/v) glucose at 30° C with shaking at 200 rpm using a New Brunswick gyratory shaker.Objective 2 - Purification of lysins from E. coli Lysin preparations will be prepared by growing E. coli strains engineered to produce several lysins as described (Roach et al., 2013). The lysins will be recovered from cell cultures as detailed below. All lysin proteins that are over expressed inE. coli may be purified using nickel column chromatography as described (Becker et al., 2009). The purified pET21a that harbor the lysin genes of interest, will be transformed intoE. coliBL21 (DE3) (Invitrogen, Carlsbad, CA) using heat stock at 42° C for 30 s. The BL21 (DE3) transformants will be cultured at 37° C in 1 L of modified LB (mLB) medium (15 g/l tryptone, 8 g/l yeast extract, 5 g/l NaCl) that is supplemented with 150 μg/ml ampicillin. Mid log-phase (OD600 nmof 0.4-0.6) cultures will be induced with 1 mM IPTG (isopropyl-beta-D-thiogalactopyranoside), followed by 10° C overnight incubation. The induced cells will be pelleted, resuspended in 10 ml of lysis buffer (50 mM NaH2PO4, 300 mM NaCl, 10 mM imidazole, 30% glycerol, pH 8.0), and sonicated on ice for 15 min (1 s pulses separated by 1 s rests). After centrifugation at 9000 x g for 30 min, the proteins will be purified from the cleared supernatant using immobilized metal ion affinity chromatography with a nickel-NTA Superflow resin (QIAGEN, Valencia, CA). The resin will be washed with 40 column volumes (CV) of lysis buffer, and additionally with 15 CV of wash buffer (50 mM NaH2PO4, 300 mM NaCl, 20 mM imidazole, 30% glycerol, pH 8.0). The His6-tagged proteins will be eluted with the buffer (50 mM NaH2PO4, 300 mM NaCl, 250 mM imidazole, 30% glycerol, pH 8.0) to achieve >90% purity. The protein column eluents will then be filter sterilized (0.22 μm), and concentration measured spectrophotometrically using a NanoDrop ND-1000 (NanoDrop Technologies, Wilmington, DE). The lysins' purities will be determined via sodium dodecyl sulfate-polyacrylamide gel electrophoresis (SDS-PAGE). Testing of the crude E. coli cell lysates will also be carried out to determine if the crude preparations can be used as is without the need for purification in order to reduce cost and reduce material losses during chromatographic separation.The purified proteins will be analyzed using 15% SDS-PAGE, with or without 300 ml equivalent of mid-log phase (OD600 nm= 0.4-0.6) lactobacilli cells that are pelleted, and washed twice in buffer (50 mM NaH2PO4, 150 mM NaCl, pH 8.0) prior to addition to the pre-polymerized gel. Gels will be subjected to electrophoresis at 150 volts (~1 h). SDS-PAGE gels will be stained using Coomassie blue dye and zymograms will be washed with excess H2O for 1 h followed by incubation at 24° C in 50 mM Tris-HCl, 1% Triton X114, pH 5.5, until translucent bands are visible for images to be taken (~18 h).Objective 3 -Testing of Lysin Preparation The turbidity assays will be performed in a Molecular Devices Spectra Max 340 plate reader. Strains of Lactobacillus will be grown to mid-log phase (OD600 nm= 0.4 - 0.6) at 37°C, washed in buffer (20mM phosphate, 150mM NaCl, 30% glycerol: pH 8.0), pelleted, and frozen at −80°C. After that the cells will be thawed on ice and resuspended to OD600 nm= 2.0 in buffer (20mM phosphate, 150mM NaCl: pH 5.5, unless otherwise stated). Lysins concentration will be standardized to 1 μM per well and the assay started by the addition of 100 μl aliquots of cell suspensions having an initial OD600nm=1. Following that, absorbance (OD600nm) readings will be taken every 30 s for 30 min and specific activities determined on a sliding scale as described by Becker, et al. (2009) and reported as OD600nm/min/μM. Control samples with cells alone (no enzyme) will also be included, and 'cells alone' specific activities subtracted from experimental sample specific activities used as control to determine the degree of cell autolysis.Objective 4 -Testing of Lysins with Commercial Ethanol Mashes Fermentation mashes will be secured from an ethanol corn wet mill and from an ethanol dry mill plant for testing of endolysin preparations. The fermentation mashes will be obtained after the liquefaction step to insure sterility prior to testing in the laboratory under defined conditions. The fermentation mashes will also be sterilized for 20 minutes by autoclaving at 121° C. Fermentations will be carried out in triplicate in 250 ml non-baffled shake flasks. Fifty ml aliquots of mash will be dispensed prior to sterilization into the flasks and the flasks will be cooled to 32° C. Commercial-grade glucoamylase enzyme will be added at the recommended standard dose and yeast inoculated to a final yeast cell count of 107 cells/ml. The flasks will be placed on a gyratory shaker (New Brunswick Shaker) that is set at 150 rpm. Lactic acid bacteria (L. fermentumstrain BR0315-1 andL. reuteristrain B-14171) will be added at different inoculation rates 102-107 to reproduce low to high level contamination conditions and endolysin added at the determined effective dose. The flasks will be incubated for 72 hrs. and sampled at 24, 48 and 72 hrs. to measure ethanol, organic acid, sugars and glycerol production by LC analysis. 'No-lysin' controls will also be included.Testing in continuous mode will require a bench two to three-stage cascade fermentation set up with a total residence time of 72 hrs. The fermentation will be carried out using a commercial strain of yeast and L. fermentumstrain BR0315-1 inoculation at the outset with intermittent dosing of lysin preparations as needed. Samples will be collected throughout the run on a daily basis from the final stage with sampling of all stages after the first 72 hours and at the end of three weeks. In this fermentation, 250-1,000 ml flasks will be used with fermentation volume set at 100-400 mls with a daily feed rate of 100-400 mls. The medium for this fermentation will consist of a wet mill fermentation mash and the enzyme glucoamylase and yeast will be set at recommended industrial enzyme doses and yeast inoculation rates in use. A 'no-lysin' control will be run for one week prior to inoculation with Lactobacillus. In the second week, the Lactobacillus strain will be introduced thereby giving it time to establish. In the third week, the lysin preparation will be added to the fermentation. Aliquots will be checked for bacterial counts by plating on 1.5% MRS agar containing 100 μg/ml cycloheximide (to inhibit growth ofS. cerevisiae) and incubated anaerobically using the AnaeroPack System (Mitsubishi, Tokyo, Japan) at 37° C for 24 hrs.

Progress 07/15/19 to 10/14/20

Outputs
Target Audience:The project targets the fuel ethanol industry and the companies that provide technical services and products that support it. The U.S. Energy Information Administration (EIA) projects that in 2021, the ethanol industry will average 970,000 barrels per day which is slightly lower than the 2020 output. The economic importance of the fuel ethanol to the US economy is highlighted by a studyby the Renewable Fuels Association (RFA) that indicated that ethanol production in 2019 accounted for more than 68,600 direct jobs accross the U.S. and added $43 billion to the gross domestic product and $23 billion in household incme. The goal of this project is to determine the feasibility of using bacteriophage endolytic bacterial cell wall enzymes that are produced by recombinant strains of E. coli to control lactic acid infection of the fuel ethanol fermentation. USDA patented work and prior published articles indicatethat bacteriophage derived endolysins expressed in E. coli have a broad activity against a range of gram positive bacteria including strains of Lactobacillus (LAB) that are recoveed from fuel ethanol fermentation. It is envisioned that phage derived endolysin may provide a cost effective alternatives to chemicals, reduce reliance on the use of antibiotics and improve profit margin of ethanol production by reducing feedstock yield losses from bacterial contamination. Changes/Problems:Several challenges were encountered during this work that needed to be addressed during this project. -Strains of LAB used in this work were not found to be as well adapted to growth under low pH conditions used in fuel ethanol fermentations. Under these conditions selective pressures, poor yeast nutrition, presence of high levels for both acetic and lactic acid in recycled water used in the make, other inhibitors and microaerophilic condition aid in the selection of biofilm of LAB that is hard to reproduce in the lab in batch, in semi and in continous fermentation. During the work we had to continously select and adapt LAB strains to low pH conditions. Several of these adapted LAB strains weretested with endolysin preparations. In some cases wealso employed high LAB inoculation rate to insure survival and growth of LAB in the test media at low pH. We also resorted to pre-inoculate lab and industrial media with LAB to insure the bacteria is well established prior to yeast addition as a healthy yeast grows much better at low pH and over time dominates the fermentation. -The low pH used in fuel ethanol fermentation and the high ethanol titer that is achievedare not optimalfor endolysin activity andcan contribute to a significant loss of activity. For the best endolysin tested (LysKB317), a high doseof protein is necessary as activity at pH 4.5 is less than 50% of that at the optimal pH of 6.0 following 30 minutes of incubation as reported byLu et al. (Biotechnol Biofuels, 2020, 13:157) using an invitro assay. Lu et al. also reported that ethanol concentration of over 5% v/vreduces LysKB317 activity by 45-54%. In a typical fuel ethanol plant fermentation over 10% ethanol (w/v) is routinely achieved with over 5% v/v achieved within the first 16 hours of the onset of the fermentation. It would be necessary to test the use of LysKB317 at an ethanol plant location to establish minimum loading necesary. At a production facility, the addition of endolysin can be staged differently as it can be added to yeast during the yeast propagation stage where the starting pH is higher and ethanol is below 5%. While we did not test that in the lab, based on prior industrial production experience this would be feasible, and more cost effective as LAB contamination is most likely to occur in the fermentation during the propagation stage. Testing at a production facility would require significant scale up of endolysin production and that need to be carried out as a Phase II project activity. -High protein loading is necessary to show activity against low LAB cell count (<10^7 cells/ml). Even at thehigh endolysin protein loading may not be sufficient when bacterial contamination is >10^7. Our approach with testing of endolysins used a worst case scenairo in which a high level of contamination is present. At a production facility a better targeting for the use of endolysins would require lower endolysin protein loading. It is also envisioned that endolysins will be produced in a different production hostfor commercial viability. This will need to be validated and explored further in a Phase II as it was not withinthe scope of Phase I. -During the project while we strived to carry out the work in a timely fashion, but delays in the availability of USDA NIFA funds and lock down at and work facility access restrictions at the U of IL and at the USDA NCAUR lab as of March 2020, have added signifiant time and costto the project. In spite of that, the PI workedthroughout the lockdown and carried out experiments at the ibiocat lab at the U of IL reseach park in Champaign, IL. The limited amount ofendolysin preparations and delay in getting analytical results did not help but we managed to keep things under control with the help from our USDA partners at NCAUR Peoria. The PI also sought and was granted two no-cost extensions. For a small startup company this project has been costly as rent was paid to the U of IL Research ParkEnterprise Works throughout the lock down. ThePI provided additional $25,000 in persnal funds to keep the project going. What opportunities for training and professional development has the project provided?This Phase I SBIR was a collaborative project with Dr. Chris Skory of USDA NCAUR lab in Peoria, IL. A postdoctoral student, Shao Lu, was hired to work on this project on a part time basis by the Peoria lab to assist with endolysin preparations. How have the results been disseminated to communities of interest? Nothing Reported What do you plan to do during the next reporting period to accomplish the goals? Nothing Reported

Impacts
What was accomplished under these goals? In preparation for accomplishing the three stated goals of the project, ibiocat inc. carried out 7 major experiments during the period of 7/15-12/31/19 to establish a working experimental model for use to test endolysin treatment of LAB cells in yeast ethanol fermentation. A significant effort and resources were dedicated to reviving, handling, preservation and characterization of Lactobacillus strains.The USDA NCAUR culture collection was contacted to source strains of Lactobacillus (LAB) that can be used to test endolysin preparations. The strains selected were originally isolated from ethanol fermentations. The included the following species of LAB: L. acidophilus (B-4495); L. agilis (B-14856); L. animalis (B-14176); L. brevis (B-4527); L. casei (B-1922); L. coryniformis(B-4391); L. delbrueckii (B-763); L. fermentum (B-1840); L. gasseri (B-4240); and L. hilgardii (B-1843). Two other strains were also obtained from our collaborator at USDA NAUR, Dr. Chris Skory, were L. reuteri (B-14171) and L. fermentum (BR 0315-1). These two strains have been extensively tested by the USDA Peoria lab that havebeen the subject of several publications including most recent work (Lu et al., Biotechnol Biofuels, 2020, 13:157). The strains were revived, transferred, and stocks prepared. Several of these strains are heterofermentative and produce under microaerophilic conditions in addition to lactic acid (LA), ethanol (ETOH) and acetic acid (AA). The AA is known to be a major contributor to stress in yeast during alcoholic fermentation and in addition to reducing yields from sugar can also lead to stuck fermentations. An additional 15 experiments were completed 1/1-12/15/20 with the goal of validating the use of endolysins in ethanol fermentation in shake flasks and in bench scale fermenters using industrial media. The results ofthese experiments are incorporated in the following responses. 1)Selection of lytic enzymes In preliminary work at ibiocat inc, we tested several endolytic enzymes which includeda commercial preparationof egg white lysozyme in liquid form (Lallemand, Lallzyme Lyso Easy)that have activity on the NAM-NAG linkage in bacterial cell wall peptidoglycan that is currently sold to the wine industry.We also testedpurified and crude preparations of LysA which cleaveat the same site as lysozyme (Lyz)and another preparation of LysA2that cleaves one of the peptide side chains in peptidoglycan. Both LysA and LysA2 were reported on by Roach et al. (Biotechnol Biofuels, 2013, 6(1). In addition to Lallemand lysozyme, a Sigma sourced lyophilized lysozyme provided by the USDANCAUR lab in Peoria, as well as another recent endolysin LysKB317 that is produced in a recombinant E.coli host (Lu et al., 2020,Biotechnol Biofuels,13:157). Other samples of endolysins that were provided on 1/13/2020 included preparations ofLysgaYandλSA2. These preparations were provided in purified and as whole cell free lysate in 150-450ul quantities containing 200-450ug/ml of protein as determined by the Qubit Protein Assay, In early experiments, LysKB317 gave consistently the best results when compared to the other endolysins and in an experiment using 48 hr cultures of L. brevis B-4752 and L. fermentans BR 0315-1 and LysKB317 showed greater activity of this endolysin on L. fermentans cells with >95% reduction in A620 as compared with L. brevis B-4527 cells which had a 34.5 % reduction A620. An additioal shipment of LysKB317 and LysA was provided by USDA on 6/22/20 for fermentation scale up testing. 2) Optimization of blend for use in the early stages of the ethanol fermentation process This research did not demonstrate synergistic effect that supports the working hypothesis that we set out to test for the development of an optimal blend using more than one endolytic activity against gram positive cell wall peptidoglycan polysaccharide backbone and the peptide side chains. The most active endolysin was a LysKB317 (isolated from the bacteriophage EcoSau)preparation produced in a recombinant E. coli host that is thought to cleave at theβ-1,4-glycosidic backbone of the peptidoglycan N-acetyl-glucosamine-N-acetylmuramic acid (NAG-NAM). This preparation showed consistently better activity on LAB cell walls than other endolysin preparations (LysA, LysA2,LysgaY, λSA2) that weredescribed by Roach et al. (Biotechnol Biofuels, 2013,6(1)20). In preliminary work at ibiocat inc, wetested two commercial preparations of egg white lysozyme (Sigma, Lallemand) that have activity on the NAM-NAG linkage in bacterial cell walls as well as crude preparations of LysA which cleaveat the same site as lysozyme. AnotherLysA2 preparation that cleaves one of the peptide side chains in peptidoglycan was also tested. We demonstratedthat the activity of LysKB317 could not be complemented with either Sigma lysozyme,LysA or LysA2. We also demonstrated that LysA2, LysA and lysozyme are also active against the LAB strains tested but in most cases LysKB317 had greater activity and broader lytic activity against several LAB ethanol contaminants tested. 3) Testing of lytic enzymes under industrial conditions with ethanol plant feedstocks. Ethanol fermentations were carried in sterile 50 ml tube, 500 ml Erlenmeyer flasks, and in two Bellco 1.0 liter fermentation vessels using ingredients from ethanol plants (ADM, ICM). ADM CR15 dextrin (Clinton, IA), ADM Corn Steep Liquor (CSL, 8.6 % DS, Decatur corn wet mill), ADM Corn Flour (ADM Cedar Rapids, IA), wet stillage and dry mill corn mash (ICM, KY Commonwealth Agri Energy Plant (Hopkinsville, KY) and wet stillage (Jack Daniels Distillery, Lynchburg, TN). Due to the variability of plant sourced wet material and microbial instability, frozen stocks of CSL were stored at -80 C till used. The dextrin CR15 was used primarily in make up as the carbon sourceto insure greater consistency in experimental results as there is considerable variation in plant sourced cooked mashes. Due to the presence of liquified starch, plant corn mashes cannot be stored frozen as starch retrogrades and becomes inaccessible to AMG enzymes leading to yield losses.All fermentations were carried out at a pH of around 4.5 and a temperature of30 C with low agitation (150 rpm) to insure good mixing in a New Brunswickshaker incubator. The alpha amylase and glucoamylase starch enzymes used for liquefaction and saccharification were obtained from several companies (AB Enzymes, ADM plant grade enzymes, Danisco DuPont, Amano, Several Chinese suppliers) and used at recommended doses. Several commercial ethanol yeast strains were screened on CR15-CSL medium (Lallemand, Lesaffre, Pakmaya Yeast Turkey, others) prior to strain selection. The results of the fermentationexperiments established that LysKB317 is the mosteffective endolysin preparation against LAB infection. At high protein dosing, it can reduceethanol losses over untreated fermentation spiked with LAB (losses can be as high as 10% in heavily LAB spiked flasks that are untreated with endolysins).Lysozyme was also found to be effective under ethanol fermentation conditions(low pH, ethanol >5% v/v) but a precipitate was formed indicating possible poor solubility and stability of the enzyme. Lyz was therefore used at a high dose using a stock solution of 2g/20ml of buffer that is diluted 10X prior to use.Similarily LysA2 was also found to be effective but did not have a wide range of activity on LAB strains tested. Based on experimental results greater consistency and broader activity of LysKB317 this supports greater potential for use of this endolysin provided a cost effective manufacturing process can be developed for this protein. This may require a different host than E. coli as many commercial starch enzymes are produced in over 100 g/l protein concentration by mold fermentation. This can be envisioned as a part ofthe focus of a Phase II project.

Publications

Progress 07/15/19 to 07/14/20

Outputs
Target Audience:The project targets the fuel ethanol industry and the companies that provide technical services and products that support it. The U.S. Energy Information Administration (EIA) projects that in 2021, the ethanol industry will average 970,000 barrels per day which is slightly lower than the 2020 output. The economic importance of the fuel ethanol to the US economy is highlighted by a studyby the Renewable Fuels Association (RFA) that indicated that ethanol production in 2019 accounted for more than 68,600 direct jobs accross the U.S. and added $43 billion to the gross domestic product and $23 billion in household incme. The goal of this project is to determine the feasibility of using bacteriophage endolytic bacterial cell wall enzymes that are produced by recombinant strains of E. coli to control lactic acid infection of the fuel ethanol fermentation. USDA patented work and prior published articles indicatethat bacteriophage derived endolysins expressed in E. coli have a broad activity against a range of gram positive bacteria including strains of Lactobacillus (LAB) that are recoveed from fuel ethanol fermentation. It is envisioned that phage derived endolysin may provide a cost effective alternatives to chemicals, reduce reliance on the use of antibiotics and improve profit margin of ethanol production by reducing feedstock yield losses from bacterial contamination. Changes/Problems:Several challenges were encountered during this work that needed to be addressed during this project. -Strains of LAB used in this work were not found to be as well adapted to growth under low pH conditions used in fuel ethanol fermentations. Under these conditions selective pressures, poor yeast nutrition, presence of high levels for both acetic and lactic acid in recycled water used in the make, other inhibitors and microaerophilic condition aid in the selection of biofilm of LAB that is hard to reproduce in the lab in batch, in semi and in continous fermentation. During the work we had to continously select and adapt LAB strains to low pH conditions. Several of these adapted LAB strains weretested with endolysin preparations. In some cases wealso employed high LAB inoculation rate to insure survival and growth of LAB in the test media at low pH. We also resorted to pre-inoculate lab and industrial media with LAB to insure the bacteria is well established prior to yeast addition as a healthy yeast grows much better at low pH and over time dominates the fermentation. -The low pH used in fuel ethanol fermentation and the high ethanol titer that is achievedare not optimalfor endolysin activity andcan contribute to a significant loss of activity. For the best endolysin tested (LysKB317), a high doseof protein is necessary as activity at pH 4.5 is less than 50% of that at the optimal pH of 6.0 following 30 minutes of incubation as reported byLu et al. (Biotechnol Biofuels, 2020, 13:157) using an invitro assay. Lu et al. also reported that ethanol concentration of over 5% v/vreduces LysKB317 activity by 45-54%. In a typical fuel ethanol plant fermentation over 10% ethanol (w/v) is routinely achieved with over 5% v/v achieved within the first 16 hours of the onset of the fermentation. It would be necessary to test the use of LysKB317 at an ethanol plant location to establish minimum loading necesary. At a production facility, the addition of endolysin can be staged differently as it can be added to yeast during the yeast propagation stage where the starting pH is higher and ethanol is below 5%. While we did not test that in the lab, based on prior industrial production experience this would be feasible, and more cost effective as LAB contamination is most likely to occur in the fermentation during the propagation stage. Testing at a production facility would require significant scale up of endolysin production and that need to be carried out as a Phase II project activity. -High protein loading is necessary to show activity against low LAB cell count (<10^7 cells/ml). Even at thehigh endolysin protein loading may not be sufficient when bacterial contamination is >10^7. Our approach with testing of endolysins used a worst case scenairo in which a high level of contamination is present. At a production facility a better targeting for the use of endolysins would require lower endolysin protein loading. It is also envisioned that endolysins will be produced in a different production hostfor commercial viability. This will need to be validated and explored further in a Phase II as it was not withinthe scope of Phase I. -During the project while we strived to carry out the work in a timely fashion, but delays in the availability of USDA NIFA funds and lock down at and work facility access restrictions at the U of IL and at the USDA NCAUR lab as of March 2020, have added signifiant time and costto the project. In spite of that, the PI workedthroughout the lockdown and carried out experiments at the ibiocat lab at the U of IL reseach park in Champaign, IL. The limited amount ofendolysin preparations and delay in getting analytical results did not help but we managed to keep things under control with the help from our USDA partners at NCAUR Peoria. The PI also sought and was granted two no-cost extensions. For a small startup company this project has been costly as rent was paid to the U of IL Research ParkEnterprise Works throughout the lock down. ThePI provided additional $25,000 in persnal funds to keep the project going. What opportunities for training and professional development has the project provided?This Phase I SBIR was a collaborative project with Dr. Chris Skory of USDA NCAUR lab in Peoria, IL. A postdoctoral student, Shao Lu, was hired to work on this project on a part time basis by the Peoria lab to assist with endolysin preparations. How have the results been disseminated to communities of interest? Nothing Reported What do you plan to do during the next reporting period to accomplish the goals? Nothing Reported

Impacts
What was accomplished under these goals? In preparation for accomplishing the three stated goals of the project, ibiocat inc. carried out 7 major experiments during the period of 7/15-12/31/19 to establish a working experimental model for use to test endolysin treatment of LAB cells in yeast ethanol fermentation. A significant effort and resources were dedicated to reviving, handling, preservation and characterization of Lactobacillus strains.The USDA NCAUR culture collection was contacted to source strains of Lactobacillus (LAB) that can be used to test endolysin preparations. The strains selected were originally isolated from ethanol fermentations. The included the following species of LAB: L. acidophilus (B-4495); L. agilis (B-14856); L. animalis (B-14176); L. brevis (B-4527); L. casei (B-1922); L. coryniformis(B-4391); L. delbrueckii (B-763); L. fermentum (B-1840); L. gasseri (B-4240); and L. hilgardii (B-1843). Two other strains were also obtained from our collaborator at USDA NAUR, Dr. Chris Skory, were L. reuteri (B-14171) and L. fermentum (BR 0315-1). These two strains have been extensively tested by the USDA Peoria lab that havebeen the subject of several publications including most recent work (Lu et al., Biotechnol Biofuels, 2020, 13:157). The strains were revived, transferred, and stocks prepared. Several of these strains are heterofermentative and produce under microaerophilic conditions in addition to lactic acid (LA), ethanol (ETOH) and acetic acid (AA). The AA is known to be a major contributor to stress in yeast during alcoholic fermentation and in addition to reducing yields from sugar can also lead to stuck fermentations. An additional 15 experiments were completed 1/1-12/15/20 with the goal of validating the use of endolysins in ethanol fermentation in shake flasks and in bench scale fermenters using industrial media. The results ofthese experiments are incorporated in the following responses. 1)Selection of lytic enzymes In preliminary work at ibiocat inc, we tested several endolytic enzymes which includeda commercial preparationof egg white lysozyme in liquid form (Lallemand, Lallzyme Lyso Easy)that have activity on the NAM-NAG linkage in bacterial cell wall peptidoglycan that is currently sold to the wine industry.We also testedpurified and crude preparations of LysA which cleaveat the same site as lysozyme (Lyz)and another preparation of LysA2that cleaves one of the peptide side chains in peptidoglycan. Both LysA and LysA2 were reported on by Roach et al. (Biotechnol Biofuels, 2013, 6(1). In addition to Lallemand lysozyme, a Sigma sourced lyophilized lysozyme provided by the USDANCAUR lab in Peoria, as well as another recent endolysin LysKB317 that is produced in a recombinant E.coli host (Lu et al., 2020,Biotechnol Biofuels,13:157). Other samples of endolysins that were provided on 1/13/2020 included preparations ofLysgaYandλSA2. These preparations were provided in purified and as whole cell free lysate in 150-450ul quantities containing 200-450ug/ml of protein as determined by the Qubit Protein Assay, In early experiments, LysKB317 gave consistently the best results when compared to the other endolysins and in an experiment using 48 hr cultures of L. brevis B-4752 and L. fermentans BR 0315-1 and LysKB317 showed greater activity of this endolysin on L. fermentans cells with >95% reduction in A620 as compared with L. brevis B-4527 cells which had a 34.5 % reduction A620. An additioal shipment of LysKB317 and LysA was provided by USDA on 6/22/20 for fermentation scale up testing. 2) Optimization of blend for use in the early stages of the ethanol fermentation process This research did not demonstrate synergistic effect that supports the working hypothesis that we set out to test for the development of an optimal blend using more than one endolytic activity against gram positive cell wall peptidoglycan polysaccharide backbone and the peptide side chains. The most active endolysin was a LysKB317 (isolated from the bacteriophage EcoSau)preparation produced in a recombinant E. coli host that is thought to cleave at theβ-1,4-glycosidic backbone of the peptidoglycan N-acetyl-glucosamine-N-acetylmuramic acid (NAG-NAM). This preparation showed consistently better activity on LAB cell walls than other endolysin preparations (LysA, LysA2,LysgaY, λSA2) that weredescribed by Roach et al. (Biotechnol Biofuels, 2013,6(1)20). In preliminary work at ibiocat inc, wetested two commercial preparations of egg white lysozyme (Sigma, Lallemand) that have activity on the NAM-NAG linkage in bacterial cell walls as well as crude preparations of LysA which cleaveat the same site as lysozyme. AnotherLysA2 preparation that cleaves one of the peptide side chains in peptidoglycan was also tested. We demonstratedthat the activity of LysKB317 could not be complemented with either Sigma lysozyme,LysA or LysA2. We also demonstrated that LysA2, LysA and lysozyme are also active against the LAB strains tested but in most cases LysKB317 had greater activity and broader lytic activity against several LAB ethanol contaminants tested. 3) Testing of lytic enzymes under industrial conditions with ethanol plant feedstocks. Ethanol fermentations were carried in sterile 50 ml tube, 500 ml Erlenmeyer flasks, and in two Bellco 1.0 liter fermentation vessels using ingredients from ethanol plants (ADM, ICM). ADM CR15 dextrin (Clinton, IA), ADM Corn Steep Liquor (CSL, 8.6 % DS, Decatur corn wet mill), ADM Corn Flour (ADM Cedar Rapids, IA), wet stillage and dry mill corn mash (ICM, KY Commonwealth Agri Energy Plant (Hopkinsville, KY) and wet stillage (Jack Daniels Distillery, Lynchburg, TN). Due to the variability of plant sourced wet material and microbial instability, frozen stocks of CSL were stored at -80 C till used. The dextrin CR15 was used primarily in make up as the carbon sourceto insure greater consistency in experimental results as there is considerable variation in plant sourced cooked mashes. Due to the presence of liquified starch, plant corn mashes cannot be stored frozen as starch retrogrades and becomes inaccessible to AMG enzymes leading to yield losses.All fermentations were carried out at a pH of around 4.5 and a temperature of30 C with low agitation (150 rpm) to insure good mixing in a New Brunswickshaker incubator. The alpha amylase and glucoamylase starch enzymes used for liquefaction and saccharification were obtained from several companies (AB Enzymes, ADM plant grade enzymes, Danisco DuPont, Amano, Several Chinese suppliers) and used at recommended doses. Several commercial ethanol yeast strains were screened on CR15-CSL medium (Lallemand, Lesaffre, Pakmaya Yeast Turkey, others) prior to strain selection. The results of the fermentationexperiments established that LysKB317 is the mosteffective endolysin preparation against LAB infection. At high protein dosing, it can reduceethanol losses over untreated fermentation spiked with LAB (losses can be as high as 10% in heavily LAB spiked flasks that are untreated with endolysins).Lysozyme was also found to be effective under ethanol fermentation conditions(low pH, ethanol >5% v/v) but a precipitate was formed indicating possible poor solubility and stability of the enzyme. Lyz was therefore used at a high dose using a stock solution of 2g/20ml of buffer that is diluted 10X prior to use.Similarily LysA2 was also found to be effective but did not have a wide range of activity on LAB strains tested. Based on experimental results greater consistency and broader activity of LysKB317 this supports greater potential for use of this endolysin provided a cost effective manufacturing process can be developed for this protein. This may require a different host than E. coli as many commercial starch enzymes are produced in over 100 g/l protein concentration by mold fermentation. This can be envisioned as a part ofthe focus of a Phase II project.

Publications

Progress 07/15/19 to 07/14/20

Outputs
Target Audience: Nothing Reported Changes/Problems: Nothing Reported What opportunities for training and professional development has the project provided? Nothing Reported How have the results been disseminated to communities of interest? Nothing Reported What do you plan to do during the next reporting period to accomplish the goals?1) Develop an effective endolytic blend and validate against LAB strains that are common ethanol fermentation contaminants. 2) Establish effectivenss of endolytic enzymes on LAB in yeast ethanol fermentation in shake flasks 3) Test effectiveness of endolysin blend against LAB in continous ethanol fermentation

Impacts
What was accomplished under these goals? Objective 1: carried out preliminary work to establish effectiveness of several endolysin preparations against 4 selected LAB strains. Early results established that a new preparation active againist the CHO backone of cell walls of LABlysed 80-90 % of bacterical cells at high cell loading in test tube assay under conditions selected at 35 C , pH 7.2 buffer. These results while encouraging need to be reproduced under ethanol fermentation conditions (pH 4.5;T 30-32 C). Objective 2: Testing ofcombination of two best endolysin preparationwith/without lysozyme that have activity towards the CHO backbne and side peptides, have not shown synergiestic effects . Further work is necessary on optimization ofendolytic blend. Objective 3: Several shake flask experiments have tested the impact of LAB one yeast ethanol fermentation with industrial media have been completed. Work on continous fermentation is in progress.

Publications

Progress 07/15/19 to 03/14/20

Outputs
Target Audience: Nothing Reported Changes/Problems: Nothing Reported What opportunities for training and professional development has the project provided? Nothing Reported How have the results been disseminated to communities of interest? Nothing Reported What do you plan to do during the next reporting period to accomplish the goals?1) Develop an effective endolytic blend and validate against LAB strains that are common ethanol fermentation contaminants. 2) Establish effectivenss of endolytic enzymes on LAB in yeast ethanol fermentation in shake flasks 3) Test effectiveness of endolysin blend against LAB in continous ethanol fermentation

Impacts
What was accomplished under these goals? Objective 1: carried out preliminary work to establish effectiveness of several endolysin preparations against 4 selected LAB strains. Early results established that a new preparation active againist the CHO backone of cell walls of LABlysed 80-90 % of bacterical cells at high cell loading in test tube assay under conditions selected at 35 C , pH 7.2 buffer. These results while encouraging need to be reproduced under ethanol fermentation conditions (pH 4.5;T 30-32 C). Objective 2: Testing ofcombination of two best endolysin preparationwith/without lysozyme that have activity towards the CHO backbne and side peptides, have not shown synergiestic effects . Further work is necessary on optimization ofendolytic blend. Objective 3: Several shake flask experiments have tested the impact of LAB one yeast ethanol fermentation with industrial media have been completed. Work on continous fermentation is in progress.

Publications