Paul Keim

Laboratory production of botulinum neurotoxin (BoNT) began more than 90 y ago for medical, military, and later pharmaceutical applications, creating one of the longest-running examples of microbial cultivation under sustained human control. A single Clostridium botulinum Group I lineage, Army Hall A (AHA), was established by the US military in 1942 and later gave rise to Hall A-hyper (HAH), the strain widely used for pharmaceutical BoNT production. We analyzed more than 1,000 C. botulinum genomes to identify AHA’s closest relatives and to infer the most recent common ancestor (MRCA) shared with its nearest lineage. Relative to this MRCA, AHA accumulated nearly 100 genetic changes, including 46 single-nucleotide substitutions, 44 small insertions or deletions, and several large deletions and structural variants that led to the loss of more than 80 genes. A nonsense mutation in mutS generated a hypermutator phenotype that accelerated mutation rates and increased genetic diversity. This event is inferred to have occurred early in the lineage’s domestication and appears to have facilitated laboratory-adaptive traits, including the lack of sporulation and increased BoNT yield. Competition experiments under standard growth conditions confirmed substantial laboratory adaptation, with AHA exhibiting a strong fitness advantage over a close relative. Together, these results reconstruct the genomic trajectory of a bacterium evolving under prolonged human-mediated selection and provide a genome-resolved example of microbial domestication. The findings show how laboratory conditions, industrial selection, and changing mutation rates can shape bacterial evolution, and they offer a general framework for understanding domestication and laboratory adaptation across microbial systems.

Laboratory production of botulinum neurotoxin (BoNT) began more than 90 years ago for medical, military, and later pharmaceutical applications, creating one of the longest-running examples of microbial cultivation under sustained human control. A single Clostridium botulinum Group I lineage, Army Hall A (AHA), was established by the U.S. military in 1942 and later gave rise to Hall A-hyper (HAH), the strain widely used for pharmaceutical BoNT production. We analyzed more than 1,000 C. botulinum genomes to identify AHA’s closest relatives and to infer the most recent common ancestor (MRCA) shared with its nearest wild lineage. Relative to this MRCA, AHA accumulated nearly 100 genetic changes, including 46 single-nucleotide substitutions, 44 small insertions or deletions, and several large deletions and structural variants that led to the loss of more than 80 genes. A nonsense mutation in mutS generated a hypermutator phenotype that accelerated mutation rates and increased genetic diversity. This event occurred early in the lineage’s domestication and appears to have facilitated laboratory-adaptive traits, including the lack of sporulation and increased BoNT yield. Competition experiments under standard growth conditions confirmed substantial laboratory adaptation, with AHA exhibiting a strong fitness advantage over a close wild relative. Together, these results reconstruct the genomic trajectory of a bacterium evolving under prolonged human-mediated selection and provide a genome-resolved example of microbial domestication. The findings show how laboratory conditions, industrial selection, and changing mutation rates can jointly shape bacterial evolution, and they offer a general framework for understanding domestication and laboratory adaptation across microbial systems. Significance: Human-mediated species domestication has been central to the development of civilization, with well-known examples in plants and animals. Microbes have also been domesticated, often unintentionally and before the advent of modern microbiology. The military and pharmaceutical applications of botulinum neurotoxin led to the sustained cultivation and eventual domestication of a high toxin-producing Clostridium botulinum strain that remains widely used today. We reconstruct the genomic changes along this trajectory and identify a key early mutation that generated a hypermutator phenotype, increasing genetic variation available for human-directed selection. These findings reveal fundamental evolutionary processes shaping bacterial genomes under long-term human control. This genome-resolved example of microbial domestication offers a general framework for understanding laboratory adaptation in both evolutionary and applied microbiology.

Valley fever, a disease caused by Coccidioides spp., is a fungal respiratory disease with an expanding range. Methods to culture the pathogen from soil, especially Coccidioides posadasii , are very challenging, limiting the genomics knowledge of environmental strains. In this study, we designed and tested a targeted DNA capture and enrichment system for the characterization of Coccidioides genomes without the need to culture. In this system, RNA probes are hybridized to Coccidioides DNA in a complex sample, followed by DNA amplification, sequencing, and analysis. Our enrichment system was targeted toward coding region sequences in C. posadasii str. Silveira and tested on control DNA spiked into soil; DNA hybridized to probes was then sequenced and correctly placed into a reference phylogeny, based on the known placement of the whole-genome sequence. We then applied the enrichment system to a range of sample types (soil, air filters, rodent tissue) from a site in Mesa, Arizona, USA. The enriched samples were sequenced and placed into the C. posadasii phylogeny to understand the phylogenetic diversity within the Mesa site over time. The results demonstrate that low DNA signal in most sample types was boosted after enrichment. Enriched sequences from air filters collected at multiple time points from the Mesa site linked two different isolates collected from fatal cases of Coccidioidomycosis in a pig-tailed macaque colony housed at the Mesa site. This represents the first time that environmental C. posadasii DNA was directly linked to Coccidioidomycosis and demonstrates the power of this approach for genomic epidemiology. IMPORTANCE: All human cases of Valley fever are acquired through environmental exposure, so surveillance and characterization of the pathogen in soil are critical for risk mitigation efforts. Current databases are biased toward human clinical isolates, and little is known about the genomics of environmental strains of Coccidioides posadasii . In this study, we designed, tested, and validated a probe enrichment system that amplifies trace DNA in a complex sample. Sequenced DNA can be used to link environmental exposure with human cases, directing public health agencies to interventions that limit human exposure. This use case was demonstrated in this study, as trace DNA trapped on air filters was linked to a fatal case of primate Coccidioidomycosis at a site in Arizona. The probe enrichment system described in this study represents a powerful tool to better understand the genomic composition of environmental C. posadasii strains, which can aid in public health investigations.

Melioidosis is an emerging infectious disease caused by the intracellular bacterial pathogen, Burkholderia pseudomallei. Infection of humans and animals can occur through multiple routes of bacterial entry resulting in life-threatening pneumonia and/or bacteremia. B. pseudomallei is inherently resistant to multiple antibiotics and mortality rates are high despite therapeutic intervention. Development of an effective vaccine could protect at-risk individuals, such as those who reside in highly endemic areas, military personnel, diabetics, and travelers. Despite decades of pursuit, no candidate vaccine for melioidosis has advanced beyond pre-clinical testing in rodent models. Here, we demonstrate that an outer membrane vesicle (OMV) vaccine prevents pulmonary disease in non human primates (NHP) using survivability, biotelemetry, and clinical pathology assessments. Using two independent serologic assays, we demonstrate that vaccination in rhesus macaques is associated with systemic and mucosal IgG and IgA to OMV surface proteins and polysaccharides. We also show that NHP immune sera promotes opsonophagocytosis by macrophages and that human sera responses to several key OMV antigens are associated with survival in melioidosis. Collectively, these results attest to the potential efficacy of the OMV vaccine and lay the groundwork for its advancement to human phase 1 clinical trials.

Genomic diversity in pathogen populations is foundational for evolution and adaptation. Understanding population-level diversity is also essential for tracking sources and revealing detailed pathways of transmission and spread. For bacteria, culturing, isolating, and sequencing the large number of individual colonies required to adequately sample diversity can be prohibitively time-consuming and expensive. While sequencing directly from a mixed population will show variants among reads, they cannot be linked to reveal allele combinations associated with phylogenetic inheritance patterns. Here, we describe the theory and method for using population sequencing directly from a mixed sample, along with a minimal number of individually sequenced colonies, to describe the phylogenetic diversity of a population without haplotype reconstruction. To demonstrate the utility of population sequencing in capturing phylogenetic diversity, we compared isogenic clones to population sequences of Burkholderia pseudomallei from sputum of a single patient. Our results point to the pathogen population being highly structured, suggesting that for some pathogens, sputum sampling may preserve structuring in the lungs and thus present a noninvasive alternative to understanding colonization, movement, and pathogen/host interactions. We also analyzed population sequences of Staphylococcus aureus derived from different people and different body sites to reveal directionality of transmission between hosts and across body sites, demonstrating the power and utility for characterizing the spread of disease and identification of reservoirs at the finest levels. We anticipate that population sequencing and analysis can be broadly applied to accelerate research in a wide range of fields reliant on a foundational understanding of population phylogenetic diversity.

Background: Valley fever (VF) is a common cause of community-acquired pneumonia in the American Southwest. Diagnosis is challenging due to VF's similarities to other pulmonary diseases. The VF T-cell response is important to control the disease, and measuring this response holds promise as an improved diagnostic. We aim to identify of VF T-cell epitopes, and to develop disease and efficacy diagnostics for the clinic and vaccine trials. Methods: We identified VF-specific T-cell epitopes by using their ability to bind the major histocompatibility complex (MHC) class II molecules and T-cell stimulation. We performed MHC binding prediction on known and unknown Coccidioides antigens and then empirically tested the predictions on unknown antigens with an in vitro multiplex MHC binding analysis. Peripheral blood mononuclear cells from patients with clinical Coccidioides infections and from endemic or nonendemic healthy controls were stimulated with the identified peptides and evaluated for immunologic memory responses. Results: A total of 108 Coccidioides peptides were identified by MHC class II binding. The 108 peptides (NAU108) were synthesized, pooled, and jointly evaluated for immunogenicity in VF-positive individuals. The peptides reactivated memory CD4+ T cells in VF-confirmed and endemic VF specimens when compared to nonendemic control samples as determined by activation marker and cytokine secretion. Conclusions: This study identified peptides that, when pooled, had immunogenic properties in humans infected with Coccidioides that can be used to distinguish infected individuals from endemic area healthy controls or nonexposed individuals outside the endemic area. This approach holds relevance for the development of diagnostic assays for VF.

Identifying improved treatments for severe and refractory coccidioidomycosis (Valley fever) is needed. This endemic fungal disease is common in North and South America, and cases have increased substantially over the last 30 years. The current standard of care, oral daily fluconazole, often fails to completely eradicate Coccidioides infection; however, the high cost of identifying new compounds effective in treating Valley fever is a barrier to improving treatment. Therefore, repurposing existing pharmaceutical agents in combination with fluconazole therapy is an attractive option. We screened the Library of Pharmacologically Active Compounds (LOPAC) small molecule library for compounds that inhibited fungal growth in vitro and determined IC ₅₀ values for a subset of compounds. Based on these findings, we tested a small subset of these agents to validate the screen, as well as to test the performance of fluconazole in a combination therapy approach, as compared with fluconazole alone, in a murine model. We observed that combination therapy of tamoxifen:fluconazole and sertraline:fluconazole significantly reduced the burden of live fungus in the lung compared with fluconazole alone, and we observed reduced or nonexistent dissemination. These results suggest that tamoxifen and sertraline may be repurposed as adjunctive agents in the treatment of this important fungal disease. IMPORTANCE: Developing new drugs, especially for regional orphan diseases, such as Valley Fever, is a slow and costly endeavor. However, there is a wealth of FDA-approved drugs available for repurposing, offering a more economical and expedited approach to improve treatment. Those existing compounds with antifungal properties can become novel therapies with relative ease: a considerable advantage for patients in need of alternative treatment. Despite the scope of remaining tasks, our comprehensive screening of potential candidates has revealed promising combinations for further exploration. This effort outlines a practical pipeline for Valley fever drug screening and identifies viable drug combinations that could impact patients more rapidly than single drug development pathways.

Background Early reports showed that patients with COVID-19 had recrudescence of previously resolved coccidioidomycosis (Valley fever, VF), and there were indications that coinfection had more severe outcomes. We therefore investigated serial infection of Coccidioides posadasii and SARS-CoV-2 in a K18-hACE2 mouse model to assess disease outcomes. Methods In our model, we challenged K18-hACE2 mice sequentially with a sub-lethal dose of SARS-CoV-2 and 24 hours later with low virulence strain of Coccidioides posadasii, and vice versa, compared to mice that only received a single infection challenge. We performed survival and pathogenesis mouse studies as well as looked at the systemic immune response differences between treatment groups. Results Here we show that co-infected groups have a more severe disease progression as well as a decrease in survival. Importantly, results differ depending on the SARS-CoV-2 variant (WA-1, Delta, or Omicron) and infection timing (SARS-CoV-2 first, C. posadasii second or vice versa). We find that groups that are infected with the virus first had a decrease in survival, increased morbidity and weight loss, increased fungal and viral burdens, differences in immune responses, and the amount and size of fungal spherules. We also find that groups coinfected with C. posadasii first have a decrease fungal burden and inflammatory responses. Conclusions This is the first in vivo model investigation of a coinfection of SARS-CoV-2 and Coccidioides. Because of the potential for increased severity of disease in a coinfection, we suggest populations that live in areas of high coccidioidomycosis endemicity may experience higher incidence of complicated disease progression with COVID-19.

Recombination of short DNA fragments via horizontal gene transfer (HGT) can introduce beneficial alleles, create genomic disharmony through negative epistasis, and create adaptive gene combinations through positive epistasis. For non-core (accessory) genes, the negative epistatic cost is likely to be minimal because the incoming genes have not co-evolved with the recipient genome and are frequently observed as tightly linked cassettes with major effects. By contrast, interspecific recombination in the core genome is expected to be rare because disruptive allelic replacement is likely to introduce negative epistasis. Why then is homologous recombination common in the core of bacterial genomes? To understand this enigma, we take advantage of an exceptional model system, the common enteric pathogens Campylobacter jejuni and C. coli that are known for very high magnitude interspecies gene flow in the core genome. As expected, HGT does indeed disrupt co-adapted allele pairings, indirect evidence of negative epistasis. However, multiple HGT events enable recovery of the genome’s co-adaption between introgressing alleles, even in core metabolism genes (e.g., formate dehydrogenase). These findings demonstrate that, even for complex traits, genetic coalitions can be decoupled, transferred, and independently reinstated in a new genetic background—facilitating transition between fitness peaks. In this example, the two-step recombinational process is associated with C. coli that are adapted to the agricultural niche. IMPORTANCE: Genetic exchange among bacteria shapes the microbial world. From the acquisition of antimicrobial resistance genes to fundamental questions about the nature of bacterial species, this powerful evolutionary force has preoccupied scientists for decades. However, the mixing of genes between species rests on a paradox: 0n one hand, promoting adaptation by conferring novel functionality; on the other, potentially introducing disharmonious gene combinations (negative epistasis) that will be selected against. Taking an interdisciplinary approach to analyze natural populations of the enteric bacteria Campylobacter , an ideal example of long-range admixture, we demonstrate that genes can independently transfer across species boundaries and rejoin in functional networks in a recipient genome. The positive impact of two-gene interactions appears to be adaptive by expanding metabolic capacity and facilitating niche shifts through interspecific hybridization. This challenges conventional ideas and highlights the possibility of multiple-step evolution of multi-gene traits by interspecific introgression.

Escherichia coli is a diverse pathogen, causing a range of disease in humans, from self-limiting diarrhea to urinary tract infections (UTIs). Uropathogenic E. coli (UPEC) is the most frequently observed uropathogen in UTIs, a common disease in high-income countries, incurring billions of dollars yearly in treatment costs. Although E. coli is easily grown and identified in the clinical laboratory, genotyping the pathogen is more complicated, yet critical for reducing the incidence of disease. These goals can be achieved through whole-genome sequencing of E. coli isolates, but this approach is relatively slow and typically requires culturing the pathogen in the laboratory. To genotype E. coli rapidly and inexpensively directly from clinical samples, including but not limited to urine, we developed and validated a multiplex amplicon sequencing assay, called ColiSeq. The assay consists of targets designed for E. coli species confirmation, high resolution genotyping, and mixture deconvolution. To demonstrate its utility, we screened the ColiSeq assay against 230 clinical urine samples collected from a hospital system in Flagstaff, Arizona, USA. A limit of detection analysis demonstrated the ability of ColiSeq to identify E. coli at a concentration of ~2 genomic equivalent (GEs)/mL and to generate high-resolution genotyping at a concentration of 1 × 10 ⁵ GEs/mL. The results of this study suggest that ColiSeq could be a valuable method to understand the source of UPEC strains and guide infection mitigation efforts. As sequence-based diagnostics become accepted in the clinical laboratory, workflows such as ColiSeq will provide actionable information to improve patient outcomes. IMPORTANCE: Urinary tract infections (UTIs), caused primarily by Escherichia coli , create an enormous health care burden in the United States and other high-income countries. The early detection of E. coli from clinical samples, including urine, is important to target therapy and prevent further patient complications. Additionally, understanding the source of E. coli exposure will help with future mitigation efforts. In this study, we developed, tested, and validated an amplicon sequencing assay focused on direct detection of E. coli from urine. The resulting sequence data were demonstrated to provide strain level resolution of the pathogen, not only confirming the presence of E. coli , which can focus treatment efforts, but also providing data needed for source attribution and contact tracing. This assay will generate inexpensive, rapid, and reproducible data that can be deployed by public health agencies to track, diagnose, and potentially mitigate future UTIs caused by E. coli .