Bacterial contamination of fuel ethanol fermentations by lactic acid bacteria (LAB) can have crippling effects on bioethanol production. Producers have had success controlling bacterial growth through prophylactic addition of antibiotics to fermentors, yet concerns have arisen about antibiotic resistance among the LAB. Here, we report on mechanisms used by 32 LAB isolates from eight different US bioethanol facilities to persist under conditions of antibiotic stress. Minimum inhibitory concentration assays with penicillin, erythromycin, and virginiamycin revealed broad resistance to each of the antibiotics as well as high levels of resistance to individual antibiotics. Phenotypic assays revealed that antibiotic inactivation mechanisms contributed to the high levels of individual resistances among the isolates, especially to erythromycin and virginiamycin, yet none of the isolates appeared to use a β-lactamase. Biofilm formation was noted among the majority of the isolates and may contribute to persistence under low levels of antibiotics. Nearly all of the isolates carried at least one canonical antibiotic resistance gene and many carried more than one. The erythromycin ribosomal methyltransferase (erm) gene class was found in 19 of 32 isolates, yet a number of these isolates exhibit little to no resistance to erythromycin. The erm genes were present in 15 isolates that encoded more than one antibiotic resistance mechanism, suggestive of potential genetic linkages.

Corn-based fuel ethanol facilities mix enzymatically treated, gelatinized corn starch with water to generate a "mash" that is used as the substrate in large-scale (∼500,000 gallon) yeast-based fermentations. In contrast to other food and beverage fermentations (e.g., cheese, wine), bioethanol production is presumed to be optimal when bacteria are absent from the fermentation-thus maximizing conversion of glucose to ethanol-yet the facilities are not sterilized. Culture-based analysis has suggested that lactic acid bacteria occupy this niche and, under certain circumstances, can outcompete the dedicated fermentation yeast for nutrients. Here, we use 16S rRNA gene amplicon sequencing to probe bacterial community structure during bioethanol fermentation. Nineteen total batches from five corn-based fuel ethanol fermentation facilities were analyzed. From each batch, five samples were taken. This includes the contents of the yeast propagation tank at inoculation, three samples taken at intervals during the fermentation, and a sample taken at the end of fermentation. Bacterial community structure was compared with time, between facility, between fermentor, between batches from the same fermentor, and against environmental variables within each fermentation. Communities were dominated by members of the Firmicutes and Proteobacteria phyla, with lactic acid bacteria dominating the communities in two of the five facilities. In the other facilities, Proteobacteria (largely members of the Pseudomonas and Escherichia-Shigella genera) outcompete the lactic acid bacteria. In most cases, the yeast propagation tank inoculum imparted a rich bacterial community, but the batches vary regarding whether this inoculum was the primary driver of the fermentation community structure.

An Enterobacter cloacae strain (E. cloacae F3S3) that was collected as part of a project to assess antibiotic resistance among bacteria isolated from bioethanol fermentation facilities demonstrated high levels of resistance to antibiotics added prophylactically to bioethanol fermentors. PCR assays revealed the presence of canonical genes encoding resistance to penicillin (ampC) and erythromycin (ermG). Assays measuring biofilm formation under antibiotic stress indicated that erythromycin induced biofilm formation in E. cloacae F3S3. Planktonic growth and biofilm formation were observed at a high ethanol content, indicating E. cloacae F3S3 can persist in a bioethanol fermentor under the highly variable environmental conditions found in fermentors.