EBRC Seminar Series – May 13, 2021 (2:00 PM ET)
May 13, 2021 | Virtual
Please join us for an exciting seminar on May 13, 2021, from 2-3:30 PM ET. This is the second of three seminars in the 2021 EBRC Seminar Series.
Speaker abstracts are below. The seminar is open to all, so please feel free to share this information with your colleagues.
The seminar will be held on Zoom using the following link for all sessions:
Zoom link: https://berkeley.zoom.us/j/97626552307?pwd=alVlS3dXM0lZYklYeE9zVXljWUI0UT09
Meeting ID: 976 2655 2307
Aindrila Mukhopadhyay (Lawrence Berkeley National Lab)
“Non-canonical crRNAs derived from host transcripts enable multiplexable RNA detection by Cas9”
Chunlei Jiao (Helmholtz Institute for RNA-based Infection Research)
CRISPR nucleases are guided by CRISPR RNAs (crRNAs) that are naturally derived from CRISPR arrays. Here, we discovered that crRNAs could derive from cellular transcripts outside of the CRISPR-Cas locus in Campylobacter jejuni, and we exploit this discovery to achieve a novel means of multiplexed RNA detection. The discovery came from sequencing RNAs that preferentially bound to the Cas9 in C. jejuni (CjeCas9), revealing an unexpected set of RNAs that shared a motif complementary to the anti-repeat of the system’s tracrRNA. The size and composition of these bound RNAs resembled that of crRNAs, indicating that the full-length version of these RNAs base pair with the anti-repeat portion of the tracrRNA and undergo processing to form crRNA-like RNAs. We call these processed RNAs non-canonical crRNAs (ncrRNAs). Using a cell-free transcription-translation (TXTL) system, we found that some ncrRNAs could drive efficient and sequence-specific DNA cleavage by CjeCas9 and its natural tracrRNA. Given the known sequence flexibility within the repeat:anti-repeat stem for single- guide RNAs, we hypothesized that the anti-repeat portion of the tracRNA could be reprogrammed to convert any RNA-of-interest into a functional ncrRNA that guides Cas9 to its DNA target. Using DNA cleavage assays in TXTL and in E. coli, we found that reprogrammed
tracrRNAs(Rptrs) designed to pair with different regions of an mRNA yielded efficient DNA cleavage not only for the CjeCas9 but also for the S. pyogenes Cas9 and the Streptococcus thermophilus CRISPR1 Cas9. Finally, based on the capability of Rptrs to link any RNA-of-interest to sequence-specific DNA cleavage, we established a multiplexed RNA diagnostic platform called LEOPARD (Leveraging Engineered tracrRNAs and On-target DNAs for PArallel RNA Detection). With LEOPARD, we achieved multiplexed detection of RNAs from different viruses including SARS-CoV-2 and other respiratory viruses in one reaction. We further distinguished SARS-CoV-2 and its D614G variant with single-nucleotide specificity in patient samples. These findings establish a previously unknown source of crRNAs and demonstrate the practical utility of LEOPARD for detecting numerous biomarkers in one test.
“Developing a mathematical framework for controlling complex biological systems”
Marcella Gomez (UC Santa Cruz)
In this talk, we refer to the achievement of an intended and predicted response in a biological system as controlling biology. Such efforts are often guided by classical mechanistic models from first principles. The level of complexity of these systems makes it extremely difficult to develop mechanistic models that can account for all possible interactions and predict biological response. To overcome these challenges, we propose to move away from first principal methods and instead identify key leverage points in the targeted biological pathways that can be directly up or down regulated by external signaling molecules. These molecules can be controlled by a feedback algorithm and delivered in situ by a bioelectronic device. In recent work, we successfully implemented feedback control on human‐induced pluripotent stem cells (hiPSCs) to regulate the cell’s resting potential known to affect cell physiology and functions such as proliferation, differentiation, migration, and apoptosis, as well as cell–cell communication and large‐scale morphogenesis. This was achieved without use of a model nor training data. We further outline an approach to extend the method to more complex biological systems. In particular, we consider the task of accelerating wound healing.
“The Promoter Calculator – A Sequence-to-Function Biophysical Model of Transcriptional Initiation for Sigma70 Promoters with Any Sequence”
Travis La Fleur (Penn State)
Engineering synthetic promoters with precision control has remained a challenge due to our inability to predict how a promoter’s sequence and DNA context controls its function and mRNA output. Here, we developed an accurate sequence-to-function model of transcriptional initiation that enables the automated design of synthetic promoters and the a priori prediction of cryptic promoters within natural systems – both of which are needed to advance Synthetic Biology towards genome-scale functional design. To do this, we combined oligopool synthesis, library-
based cloning, and next-generation sequencing to construct and characterize 14,206 rationally designed sigma70 promoters in vitro. Measurements include transcriptional start site frequencies and overall mRNA levels. This approach enables highly-parallel characterization of constitutive promoter activity without confounding factors such as unintentional transcriptional regulation, non-sigma70 activity, and mRNA decay. These measurements, in combination with machine learning, were used to parameterize a thermodynamic model quantifying the
interactions controlling transcription rate. For demonstration, the “Promoter Calculator” was used to accomplish three major tasks – accurate prediction of thousands of sigma70 promoters across various conditions, the de novo design of novel sigma70 promoters, and the identification of cryptic promoters internal to a genetic circuit. 4,350 highly non-repetitive promoters and 6,165 genome-integrated promoters characterized in vivo were accurately predicted by the model with Spearman Rank-Order Coefficients of .68 and .78, respectively. Promoters designed de novo using the Promoter Calculator were characterized in vivo exhibiting a 683-fold range in expression resulting in a Pearson Correlation Coefficient of .85. The Promoter Calculator was used to analyze a genetic circuit containing 11 circuit promoters and 29 cryptic promoters. The model identified 29 sigma70 promoters out of 40 observed promoters (72.5% accurate), including 10/11 circuit promoters and 19/29 cryptic promoters. The Promoter Calculator facilitates context-aware, rational promoter design without relying on a fixed table of pre-characterized sequences while serving as a powerful tool in promoter identification.