An integrated, modular approach to data science education in microbiology.

We live in an increasingly data-driven world, where high-throughput sequencing and mass spectrometry platforms are transforming biology into an information science. This has shifted major challenges in biological research from data generation and processing to interpretation and knowledge translation. However, postsecondary training in bioinformatics, or more generally data science for life scientists, lags behind current demand. In particular, development of accessible, undergraduate data science curricula has the potential to improve research and learning outcomes as well as better prepare students in the life sciences to thrive in public and private sector careers. Here, we describe the Experiential Data science for Undergraduate Cross-Disciplinary Education (EDUCE) initiative, which aims to progressively build data science competency across several years of integrated practice. Through EDUCE, students complete data science modules integrated into required and elective courses augmented with coordinated cocurricular activities. The EDUCE initiative draws on a community of practice consisting of teaching assistants (TAs), postdocs, instructors, and research faculty from multiple disciplines to overcome several reported barriers to data science for life scientists, including instructor capacity, student prior knowledge, and relevance to discipline-specific problems. Preliminary survey results indicate that even a single module improves student self-reported interest and/or experience in bioinformatics and computer science. Thus, EDUCE provides a flexible and extensible active learning framework for integration of data science curriculum into undergraduate courses and programs across the life sciences.

Prognostic Value of the Hematopoietic Cell Transplantation Comorbidity Index for Patients Undergoing Reduced-Intensity Conditioning Cord Blood Transplantation.

The Hematopoietic Cell Transplantation Comorbidity Index (HCT-CI) has been validated as a tool for evaluating the risk of treatment-related mortality (TRM) in HLA-matched sibling and matched unrelated donor bone marrow and peripheral blood stem cell transplantation patients. However, the role of the HCT-CI after cord blood transplantation (CBT) has not been fully investigated. In this analysis, we sought to evaluate the predictive value of the HCT-CI in patients undergoing reduced-intensity conditioning (RIC) CBT. Between 2006 and 2013, HCT-CI scores were prospectively tabulated for patients with hematologic malignancies sequentially enrolled on multicenter RIC CBT studies coordinated by the Fred Hutchinson Cancer Research Center: 151 patients with acute myeloid leukemia/myelodysplastic syndrome (n = 101), chronic myeloid leukemia (n = 3), acute lymphocytic leukemia (n = 24), non-Hodgkin lymphoma (n = 8), Hodgkin lymphoma (n = 3), and other hematologic malignancies (n = 12) underwent RIC CBT and were included. Two patients received a single CBT and the remaining 149 received a double CBT. All patients received cyclosporine and mycophenolate mofetil for graft-versus-host disease prophylaxis. Median HCT-CI for the whole group was 3 (range, 0 to 8). Using the HCT-CI categories of low (0), intermediate (1 or 2), and high risk (>3), there was no significant difference in TRM between the 3 groups. However, when the patients were divided into 2 groups, HCT-CI ≤ 3 or > 3, the incidence of TRM at 3 years after transplantation was 26% (95% confidence interval [CI], 17 to 36) in the HCT-CI ≤ 3 group versus 50% (95% CI, 30 to 67) in the HCT-CI > 3 group (P = .01). Overall survival for patients with HCT-CI ≤ 3 was 40% (95% CI, 27 to 51) versus 29% in patients with HCT-CI >3 (95% CI, 12 to 48) (P = .08). Our study demonstrates that HCT-CI score > 3 is associated with an increased risk of TRM at 3 years after transplantation in patients undergoing RIC CBT. Because of the significant risk of TRM in patients with HCT-CI > 3 compared with risk for those with HCT-CI ≤ 3, patients with an HCT-CI score >3 should be counseled before undergoing RIC CBT.

Development of a data science CURE in microbiology using publicly available microbiome datasets.

Scientific and technological advances within the life sciences have enabled the generation of very large datasets that must be processed, stored, and managed computationally. Researchers increasingly require data science skills to work with these datasets at scale in order to convert information into actionable insights, and undergraduate educators have started to adapt pedagogies to fulfill this need. Course-based undergraduate research experiences (CUREs) have emerged as a leading model for providing large numbers of students with authentic research experiences including data science. Originally designed around wet-lab research experiences, CURE models have proliferated and diversified globally to accommodate a broad range of academic disciplines. Within microbiology, diversity metrics derived from microbiome sequence information have become standard data products in research. In some cases, researchers have deposited data in publicly accessible repositories, providing opportunities for reproducibility and comparative analysis. In 2020, with the onset of the COVID-19 pandemic and concomitant shift to remote learning, the University of British Columbia set out to develop an online data science CURE in microbiology. A team of faculty with collective domain expertise in microbiome research and CUREs developed and implemented a data science CURE in which teams of students learn to work with large publicly available datasets, develop and execute a novel scientific research project, and disseminate their findings in the online Undergraduate Journal of Experimental Microbiology and Immunology. Analysis of the resulting student-authored research articles, including comments from peer reviews conducted by subject matter experts, demonstrate high levels of learning effectiveness. Here, we describe core insights from course development and implementation based on a reverse course design model. Our approach to course design may be applicable to the development of other data science CUREs.

Propelling a Course-Based Undergraduate Research Experience Using an Open-Access Online Undergraduate Research Journal.

The University of British Columbia has developed a course-based undergraduate research experience (CURE) that engages students in authentic molecular microbiology research. This capstone course is uniquely built around an open-access online undergraduate research journal entitled Undergraduate Journal of Experimental Microbiology and Immunology (UJEMI). Students work in teams to derive an original research question, formulate a testable hypothesis, draft a research proposal, carry out experiments in the laboratory, and publish their results in UJEMI. The CURE operates in a feed forward manner whereby student-authored UJEMI publications drive research questions in subsequent terms of the course. Progress toward submission of an original manuscript is scaffolded using a series of communication assignments which facilitate formative development. We present a periodic model of our CURE that guides students through a research cycle. We review two ongoing course-based projects to highlight how UJEMI publications prime new research questions in the course. A journal-driven CURE represents a broadly applicable pedagogical tool that immerses students in the process of doing science.

Development of a Peer-Reviewed Open-Access Undergraduate Research Journal.

Dissemination of results is a fundamental aspect of the scientific process and requires an avenue for publication that is specifically designed to suit the nature of the research being communicated. Undergraduate research journals provide a unique forum for students to report scientific findings and ideas while learning about the complete scientific process. We have developed a peer-reviewed, open-access, international undergraduate research journal that is linked to a course-based undergraduate research experience. We reflect on lessons learned and recommend effective approaches for the implementation and operation of a successful undergraduate research journal.

Crystal structure of the major periplasmic domain of the bacterial membrane protein assembly facilitator YidC.

The essential bacterial membrane protein YidC facilitates insertion and assembly of proteins destined for integration into the inner membrane. It has homologues in both mitochondria and chloroplasts. Here we report the crystal structure of the Escherichia coli YidC major periplasmic domain (YidCECP1) at 2.5A resolution. This domain is present in YidC from Gram-negative bacteria and is more than half the size of the full-length protein. The structure reveals that YidCECP1 is made up of a large twisted beta-sandwich protein fold with a C-terminal alpha-helix that packs against one face of the beta-sandwich. Our structure and sequence analysis reveals that the C-terminal alpha-helix and the beta-sheet that it lays against are the most conserved regions of the domain. The region corresponding to the C-terminal alpha-helix was previously shown to be important for the protein insertase function of YidC and is conserved in other YidC-like proteins. The structure reveals that a region of YidC that was previously shown to be involved in binding to SecF maps to one edge of the beta-sandwich. Electrostatic analysis of the molecular surface for this region of YidC reveals a predominantly charged surface and suggests that the SecF-YidC interaction may be electrostatic in nature. Interestingly, YidCECP1 has significant structural similarity to galactose mutarotase from Lactococcus lactis, suggesting that this domain may have another function besides its role in membrane protein assembly.

Crystal structure of a bacterial signal Peptide peptidase.

Signal peptide peptidase (Spp) is the enzyme responsible for cleaving the remnant signal peptides left behind in the membrane following Sec-dependent protein secretion. Spp activity appears to be present in all cell types, eukaryotic, prokaryotic and archaeal. Here we report the first structure of a signal peptide peptidase, that of the Escherichia coli SppA (SppA(EC)). SppA(EC) forms a tetrameric assembly with a novel bowl-shaped architecture. The bowl has a dramatically hydrophobic interior and contains four separate active sites that utilize a Ser/Lys catalytic dyad mechanism. Our structural analysis of SppA reveals that while in many Gram-negative bacteria as well as characterized plant variants, a tandem duplication in the protein fold creates an intact active site at the interface between the repeated domains, other species, particularly Gram-positive and archaeal organisms, encode half-size, unduplicated SppA variants that could form similar oligomers to their duplicated counterparts, but using an octamer arrangement and with the catalytic residues provided by neighboring monomers. The structure reveals a similarity in the protein fold between the domains in the periplasmic Ser/Lys protease SppA and the monomers seen in the cytoplasmic Ser/His/Asp protease ClpP. We propose that SppA may, in addition to its role in signal peptide hydrolysis, have a role in the quality assurance of periplasmic and membrane-bound proteins, similar to the role that ClpP plays for cytoplasmic proteins.

Protective activity of the Bordetella pertussis BrkA autotransporter in the murine lung colonization model.

This study examined the vaccine potential of the autotransporter protein BrkA of Bordetella pertussis in the sublethal intranasal murine respiratory challenge model of infection. Five different acellular pertussis (Pa) vaccines, containing different pertussis-component antigens but all comprizing diphtheria (D) and tetanus (T) toxoids, were tested. A two-pertussis-component DTPa vaccine containing pertussis toxoid (PT) and filamentous hemagglutinin (FHA) induced only limited bacterial clearance. However, a three-pertussis-component DTPa vaccine containing PT, FHA and a recombinant BrkA protein (rBrkA) was found to be as efficacious in protecting mice against colonization by B. pertussis strains Tohama I and 18-323 as the commercial Infanrixtrade mark vaccine that also includes PT and FHA but pertactin (PRN) instead of rBrkA. Vaccination of mice with rBrkA as the only B. pertussis antigen did not protect against colonization by B. pertussis. We also demonstrated that BrkA is ubiquitously expressed by highly prevalent clinical isolates of B. pertussis and suggest that new acellular pertussis vaccine formulations that include BrkA have equivalent efficacy as currently available DTPa vaccines against B. pertussis infections.

Phage display and crystallographic analysis reveals potential substrate/binding site interactions in the protein secretion chaperone CsaA from Agrobacterium tumefaciens.

The protein CsaA has been proposed to function as a protein secretion chaperone in bacteria that lack the Sec-dependent protein-targeting chaperone SecB. CsaA is a homodimer with two putative substrate-binding pockets, one in each monomer. To test the hypothesis that these cavities are indeed substrate-binding sites able to interact with other polypeptide chains, we selected a peptide that bound to CsaA from a random peptide library displayed on phage. Presented here is the structure of CsaA from Agrobacterium tumefaciens (AtCsaA) solved in the presence and absence of the selected peptide. To promote co-crystallization, the sequence for this peptide was genetically fused to the amino-terminus of AtCsaA. The resulting 1.65 A resolution crystal structure reveals that the tethered peptide from one AtCsaA molecule binds to the proposed substrate-binding pocket of a symmetry-related molecule possibly mimicking the interaction between a pre-protein substrate and CsaA. The structure shows that the peptide lies in an extended conformation with alanine, proline and glutamine side chains pointing into the binding pocket. The peptide interacts with the atoms of the AtCsaA-binding pocket via seven direct hydrogen bonds. The side chain of a conserved pocket residue, Arg76, has an "up" conformation when the CsaA-binding site is empty and a "down" conformation when the CsaA-binding site is occupied, suggesting that this residue may function to stabilize the peptide in the binding cavity. The presented aggregation assays, phage-display analysis and structural analysis are consistent with AtCsaA being a general chaperone. The properties of the proposed CsaA-binding pocket/peptide interactions are compared to those from other structurally characterized molecular chaperones.

Identification of secretion determinants of the Bordetella pertussis BrkA autotransporter.

The autotransporters comprise a functionally diverse family of gram-negative proteins that mediate their own export across the bacterial outer membrane. They consist of an amino-terminal passenger region called the "alpha-domain" and the structural hallmark of the autotransporter family, a carboxy-terminal transporter region usually referred to as the "beta-domain." The passenger region can be quite diverse and constitutes the effector functions of these proteins, whereas the C-terminal region is conserved and is responsible for translocating the passenger moiety across the outer membrane. BrkA is the 103-kDa autotransporter protein in Bordetella pertussis that is cleaved to yield a 73-kDa N-terminal alpha-domain and a 30-kDa C-terminal beta-domain. We have previously shown that a recombinant form of the beta-domain of BrkA is capable of forming channels in artificial membranes. Here, we define two additional secretion determinants of BrkA. N-terminal sequencing of the 73-kDa BrkA passenger from B. pertussis and Escherichia coli revealed that BrkA has a 42-amino-acid signal peptide. In addition, deletion analysis of BrkA identified a 31- to 39-amino-acid region found immediately upstream of the beta-domain that was essential for surface expression. This 31- to 39-amino-acid linker region, together with the beta-domain, defines the minimal BrkA translocation unit. The linker region may also serve to anchor the BrkA passenger to the bacterial surface.

A conserved region within the Bordetella pertussis autotransporter BrkA is necessary for folding of its passenger domain.

Autotransporter secretion represents a unique mechanism that Gram-negative bacteria employ to deliver proteins to their cell surface. BrkA is a Bordetella pertussis autotransporter protein that mediates serum resistance and contributes to adherence of the bacterium to host cells. BrkA is a 103 kDa protein that is cleaved to form a 73 kDa alpha-domain and a 30 kDa beta domain. The alpha domain, also referred to as the passenger domain, is responsible for the effector functions of the protein, whereas the beta domain serves as a transporter. In an effort to characterize BrkA secretion, we have shown that BrkA has a 42 amino acid signal peptide for transit across the cytoplasmic membrane, and a translocation unit made up of a short linker region fused to the beta-domain to ferry the passenger domain to the bacterial surface through a channel formed by the beta-domain. In this report, we provide genetic, biochemical and structural evidence demonstrating that a region within the BrkA passenger (Glu601-Ala692) is necessary for folding the passenger. This region is not required for surface display in the outer membrane protease OmpT-deficient Escherichia coli strain UT5600. However, a BrkA mutant protein bearing a deletion in this region is susceptible to digestion when expressed in E. coli strains expressing OmpT suggesting that the region is required to maintain a stable structure. The instability of the deletion mutant can be rescued by surface expressing Glu601-Ala692in trans suggesting that this region is acting as an intramolecular chaperone to effect folding of the passenger concurrent with or following translocation across the outer membrane.