Antibiotic Interactions in Ethanol Production: Sharing a Decade of Data

Evidence shows that effective drug combinations decrease the risk of bacterial contamination and antimicrobial resistance. Using antibiotics with two modes of action can provide wider microbe coverage and longer lasting fermentation activity.
By Jenny Forbes and Cam Fowler | November 09, 2021

Recently, producers have asked Phibro Ethanol questions about antibiotic interactions. Ethanol producers want to know: Do drugs such as virginiamycin and penicillin work in synergy or do they have an antagonistic relationship? Is there a technical risk to using these drugs in combination day in and day out? In other words, are our historic practices damaging? The available information overwhelmingly demonstrates that the antimicrobial activities of virginiamycin and penicillin work well together. Specifically, evidence shows that effective drug combinations decrease the risk of bacterial contamination and antimicrobial resistance.

Understanding Drug Interactions
Although the fields of human and animal medicine spend significant energy researching drug interactions (including antibiotic interactions), little research has been conducted specific to industrial applications. That said, medical research finds that drugs commonly used to treat infections and chronic disease in humans and animals often have negative (antagonistic) or positive (synergistic) effects when combined. Indeed, it is commonly accepted that antibiotics frequently interact, but few studies have been done comparing treatment interactions in commercial ethanol plants.

Despite the unknowns, ethanol producers often treat fermentations with virginiamycin, penicillin, or blends of these two antibiotics. Virginiamycin is a streptogramin type of antibiotic that inhibits the 50S ribosomal subunit of bacteria. Inhibiting the 50S ribosomal subunit interferes with protein synthesis and stops bacterial growth (referred to as (R) throughout the article). In contrast, penicillin is a beta-lactam antibiotic. Penicillin and antibiotics in the beta-lactam class work by interfering with bacterial cell wall production, often resulting in bacterial lysis and death (referred to as (W) throughout the article).

To address whether it’s safe or helpful to combine antibiotics in fermentation, this article will summarize data from scientific technical journals and present data generated in Phibro’s laboratory over the last decade. A number of independent, peer-reviewed technical studies indicate that combining drugs similar to virginiamycin and penicillin have a neutral to synergistic effect. Phibro’s data demonstrate that in more than 20,000 unique samples, there were no cases of antagonistic response with these two drugs. To conclude, the article will share Phibro’s guidance on how best to conduct similar research within a laboratory setting.

Journal Research
Antibiotic interaction studies with penicillin or penicillin derivatives such as ampicillin are relatively common due to its importance in human medicine. However, studies using virginiamycin are not as available. The next best substitute is to look at studies that examine antibiotics with the same activity profile as virginiamycin. The study, “Functional classification of drugs by properties of their pairwise interactions1” (Yeh P. et al., 2006) mapped the interactions of different classes of antibiotics. The drug interaction results can be seen in Figure 1, which was adapted from the study. In this graphic, a red line indicates synergism, a green line indicates antagonism, and no line indicates an additive effect.

The study tested the following R antibiotics (which can be considered similar in action to virginiamycin): Chloramphenicol (CHL), Clindamycin (CLI), Erythromycin (ERY), Spiramycin (SPR), and Fusidic acid (FUS). The study examined the following W antibiotics (which can be considered similar to penicillin):  Piperacillin (PIP), Ampicillin (AMP), and Cefoxitin (FOX). 

Figure 2, which again was adapted from this study, demonstrates drug interactions and pairwise interactions.  The bar graph demonstrates bacterial presence with no drug (represented with "0"), each drug used individually, and the combination of drugs. It shows that a combination of erythromycin and ampicillin (which represent virginiamycin and penicillin respectively) have a very high level of synergy.  In fact, the interactions between W and R antibiotics are predominantly synergistic, with a few conditions additive or inconclusive. None of the W and R combinations in this study created an antagonistic response. This should give ethanol producers confidence that products such as virginiamycin and penicillin can be added together with very low risk of a negative drug interaction.

Phibro Ethanol Laboratory Research
Phibro Ethanol has been evaluating drug interactions for approximately a decade. We instigated formal testing in 2012 and we began populating our database mid-2013. Phibro’s studies evaluated 4,090 unique samples with a minimum of five different drug combinations studied per sample. The result is more than 20,000 samples evaluated from mid-2013 to mid-2021. In all these tests, there was not one instance of antagonistic drug interaction. 

Phibro Ethanol’s laboratory studied virginiamycin and penicillin interactions using bacterial consortia (a mix of many bacteria strains) taken from thousands of customer samples by high-throughput antimicrobial sensitivity analysis using modified OmniLog5 technology. Phibro’s testing uses bacteria taken from industrial samples and monitors the bacterial growth in the presence of various concentrations of antibiotics and different combinations of antibiotics. It is a tool utilized frequently to determine the effectiveness of various antibiotic blends and helps achieve the best microbial control at the best antibiotic dose. These growth curves with blends and single antibiotic components rarely show antagonism or synergism. Instead, Phibro’s OmniLog results show that most blends look like an average of the two components used (additive behavior). Figure 3A shows a 24-hour Omnilog where the blends (represented by PhibroXact, a combinatory product containing virginiamycin and penicillin) have an additive effect. If synergism or antagonism were present, you would expect PhibroXact results to be above (antagonism) or below (synergism) the lines for the individual components. Additive effects are quite common while synergistic impact (Figure 3B) is noticed occasionally.

Clinical studies often focus on single species of bacteria and not a consortium, making it challenging to utilize research studies to generalize performance in industrial fermentations. Typical fuel ethanol fermentations contain a consortium of bacteria, which helps to explain the predominance of additive results in OmniLog testing. In fermentation, using antibiotics with two modes of action can provide better coverage of a variety of microbes, and have longer lasting activity throughout fermentation. Indeed, utilizing drug interactions has been implemented for decades in the fuel ethanol industry.

How to Conduct Laboratory Evaluations
In addition to external research and internal Phibro research supporting the use of combinatory antimicrobial products, Phibro can provide guidance on conducting independent laboratory research in this area. There are two predominant considerations: 1) bacterial selection criteria and 2) virginiamycin application. As previously referenced, there is a natural discussion surrounding the use of bacterial isolates vs. bacterial consortia. When possible, Phibro believes strongly in using a bacterial consortium for analyses. This more closely represents ethanol plant conditions. It is highly critical to control all existing bacteria, not simply one bacterial isolate. That said, there are occasionally times when using bacterial isolates is ideal. In those instances, it is best to choose a strain that commonly represents the bacterial populations found in ethanol fermentations and is thus process relevant. Phibro’s data presented in the pie chart (Figure 4) show the predominant bacterial genera found in ethanol plants. Because Phibro usually only identifies specific bacterial isolates when ethanol producers experience bacterial infections, these data, representing samples from nearly 100  unique plants, were usually collected during times of bacterial contamination.

When selecting bacteria to study in the laboratory, it’s critical to choose wisely. For example, the genus Pediococcus (in bold in Figure 4) can be described as a “bad actor” and should be avoided in laboratory settings because it represents only a tiny fraction of relevant bacterial contaminants. Some Pediococcus species (and a number of other species) can also produce biofilms. When a bacterial organism produces a biofilm, treatment of that organism is challenging because the biofilm layer protects the organism. The ethanol industry is well aware that if biofilms become present within the process, a combination of thorough cleaning and antimicrobial treatment are necessary. All things considered, use of these types of organisms in laboratory settings can result in misleading conclusions.

Along with selecting the most useful bacteria, studies hoping to test antibiotic efficacy need to consider how to best utilize virginiamycin. Phibro Animal Health Corporation is the world’s only manufacturer of virginiamycin. Post-virginiamycin production, the antibiotic active is designated either as an ethanol process aid (Lactrol®) or for use in animal feeds. Flow agents are incorporated that are specific to the end-use designation. Lactrol is formulated with hydrophilic agents while virginiamycin for animal feed is encapsulated for moisture protection until ingested by the animal. Lactrol is only sold by Phibro or licensed distributors and is only available under the trade-name Lactrol. Virginiamycin products used in the ethanol industry, other than Lactrol, will have apparent solubility concerns. This makes lab testing of non-Lactrol virginiamycin troublesome and can lead to conflicting results. Bottom line, if a laboratory wants to test virginiamycin’s efficacy related to ethanol production, Lactrol should be the product of choice.

In conclusion, both external data and Phibro data overwhelmingly support the use of virginiamycin and penicillin in combination in a commercial ethanol production process. To adequately test this principle in a laboratory setting, proper selection of bacterial consortia or isolates is critical, as is the selection of the antimicrobial agent. Ethanol producers should feel comfortable using the tools available to them.

Co-authors: Jenny Forbes
Vice President, Products & Services
Phibro Ethanol
[email protected]

Cam Fowler
Senior Director, Technical Affairs
Phibro Ethanol
[email protected]


1 YEH, JUN- JUN & I TSCHUMI, ARIANE & KISHONY, ROY. (2006). YEH P, TSCHUMI AI, KISHONY R. Functional Classification of Drugs by Properties of Their Pairwise Interactions. Nature Genetics. 38. 489-94.

2 YEH, PAMELA J ET AL. “Drug Interactions and the Evolution of Antibiotic Resistance.” Nature Reviews Microbiology VOL. 7,6 (2009): 460-6.

3 Tyers M, Wright GD. Drug combinations: a strategy to extend the life of antibiotics in the 21st century. Nature Reviews Microbiology. 2019. 17(3):141-155.

4 J. B. GUNNISON,” Studies on Antibiotic Synergism and Antagonism: the Effect in vitro of Combinations of Antibiotics on Bacteria of Varying Resistance to Single Antibiotics.” 1953. Journal of Bacteriology. Aug;66(2):150-8


6 Cokol, M., Chua, H.N., Tasan, M., Mutlu, B., Weinstein, Z.B., Suzuki, Y., Nergiz, M.E., Costanzo, M., Baryshnikova, A., Giaever, G., et al. Systematic exploration of synergistic drug pairs. Mol Syst Biol. 2011. 7: 544-544.

7 Farha MA, Brown ED. Chemical probes of Escherichia coli uncovered through chemical-chemical interaction profiling with compounds of known biological activity. Chem Biol. 2010. 17(8):852-62.

8 Borisy AA, Elliott PJ, Hurst NW, Lee MS, Lehar J, Price ER, Serbedzija G, Zimmermann GR, Foley MA, Stockwell BR, Keith CT. Systematic discovery of multicomponent therapeutics. Proc Natl Acad Sci . 2003. 100: 7977–7982.