Sarah Keane, assistant professor, Chemistry and Biophysics

In typical lab courses, students run experiments and write up lab reports, which are only read by the professor or teaching assistants. Students may feel isolated from the broader science community where sharing your results beyond your research group and publishing are essential. When labs went remote in the “age of COVID,” students may have felt even further removed from the scientific process without the ability to interact with an instrument or pour a solution into a test tube.

The virtual approach to teaching labs in spring 2021 gave Sarah Keane an opportunity to deeply connect her students with the scientific community. An assistant professor in the Departments of Biophysics and Chemistry, she taught Biophysics 450, an upper level class for undergraduates and graduate students that covers essential laboratory skills, including modern research methods such as Nuclear Magnetic Resonance (NMR) spectroscopy, a tool for identifying molecular structure.

Twenty or so of Keane’s students analyzed never-before-recorded NMR data on RNA molecules, added this new information to a database for NMR scientists, and published their findings in a paper in Biomolecular NMR Assignments. The course module immersed students in the middle of a real research project. “Instead of teaching about NMR spectroscopy in an abstract way, I find the best way of teaching is to have students work with data,” says Keane. 

In this course module, Keane and students added chemical shifts for RNA molecules to the Biological Magnetic Resonance Data Bank (BMRD). Chemical shifts are like a fingerprint of every hydrogen and carbon atom in a molecule. These shifts can be measured using 2D NMR spectroscopy. The shift for each carbon and hydrogen atom is read out from a peak in the spectra.

In her research, Keane and her graduate students students identified 12 specific RNA sequences whose shifts were missing from the BMRB database. They synthesized these RNA, measured their 2D NMR spectra, and assigned the peaks to each atom in the molecules.  For the course,  Keane deleted some of the assignments and gave the spectra to the undergraduate and graduate students to independently assign the peaks. 

 Focus on RNA

The importance of RNA molecules in medicine and biology has come into stark focus with the advent of mRNA vaccines against the SARS-CoV-2 virus. Recent studies have also shown that RNA also plays important roles in protecting DNA and researchers continue to uncover more functions of RNA molecules.  RNA researchers rely on accurate predictions of the 3-D structures to develop therapeutics or drugs to target RNA or to uncover ways the RNA might be used in the cell. 

 RNA molecules are “strings” of nucleotide bases—adenine, cytosine, uracil, and guanine— in various sequences. Each hydrogen and carbon in a unique chemical environment, such as in one of these bases, has a unique chemical shift.  Software exists to make accurate predictions of RNA molecule’s 3D structure, but the programs base their predictions on known chemical shifts in the  BMRB. RNA scientists need this database to be robust in order to keep creating amazing new RNA technologies, but the database’s library of RNA molecules is fairly sparse compared to more popularly studied proteins. Keane and her students have now filled in some gaps in the database of chemical shifts for RNA molecules.

Unambiguously identifying these chemical shifts in RNA is a laborious process because many of the hydrogen and carbons are in such similar chemical environments that the shifts can show up near or on top of each other.  For example, the structures of adenine and guanine consist of very similar carbon rings with three group changes between them (highlights in pink, tan, and teal show differences). Because the hydrogen and carbons in these structures are in such similar environments, their chemical shifts are very similar.

In addition to the environment within a base, the base pairs directly next to a certain base in the RNA sequence can change the chemical shift. So, in a triplet sequence like Uracil-Adenine-Adenine, the flanking Uracil and Adenine affect the central Adenine’s chemical shifts. The shifts can even be subtly affected by which bases flank the sides of a core three-base-pair sequence. These subtle changes to the chemical shifts affected by base pairs flanking triplet sequences are precisely what Keane and her students determined in their experiments.

Contributions to science

This dearth of chemical shift data in the BMRB provided an opportunity for Keane and her students.  The class uploaded their findings to the BMRB, expanding this database and connecting this project to the larger RNA NMR community.  Now other researchers may be able to predict the chemical shifts of new RNA molecules implicated in disease or unknown cellular functions which could lead to new treatments or new insights in cell biology, increasing  the power of the predictive software and making  these tools even better for future RNA scientists.

Aside from contributing to the database, this work also provided another benefit to the students—they learned about publishing a scientific paper. Instead of a lab report, the students helped prepare a manuscript to report on the structure of these 12 previously unidentified chemical shifts. While participating in the complete manuscript preparation was voluntary, the students that did help are co-authors on the paper published in Aguust 2021 in Biomolecular NMR Assignments.

Next Step:  2D to 3D

Despite the difficulty of using  2D NMR with RNA molecules, this module took many of the students all the way from a limited knowledge of 2D NMR to a having a scientific publication based on novel 2D NMR data. Keane plans to include this module in future versions of Biophysics 450 and update it as lab courses go back to being taught in person. Keane says she would love to eventually include a second module of the course where students perform some of the basics of 3D structure calculations using the data they upload to the BMRB. “Ultimately something I'd like to see worked in is using the computational resources that Michigan has and building in a structure determination module,” she says.

Keane also hopes to have the students collect some of the data in the future once lab courses are in person. “To be able to put a sample on the spectrometer and for them to see the facility and just learn a little bit more about using the NMR, I think that would be ideal,” she says.

Learn More

The work was supported by Keane's CAREER grant from the National Science Foundation.

Keane Lab

The paper published by the students: Liu Y#, Kotar A#, Hodges TL, Abdallah K, Taleb MH, Bitterman BA, Jaime S, Schaubroeck KJ, Mathew E, Morgenstern NW, Lohmeier A, Page JL, Ratanapanichkich M, Arhin G, Johnson BL, Cherepanov S, Moss SC, Zuniga G, Tilson NJ, Yeoh ZC, Johnson BA, Keane SCNMR chemical shift assignments of RNA oligonucleotides to expand the RNA chemical shift databaseBiomol NMR Assign 15(2):479-490 (2021). (# indicates co-first author)

--Writer Zechariah Pfaffenberger is a PhD Candidate in the Biteen Lab