Lab News

Written by Edward Mabry (upcoming M.S. student)

Week 14 (4/23/21) – Edward led our discussion on long-range structure mapping of Dengue and Zika viruses published in 2019 by the Wan lab.

Goals

The goals of the paper were to develop a better understanding of how the genome of Dengue and zika viruses are structurally organized, including long range interactions and structure conservation, and how these interactions from in cells versus in virions.

Method

The authors used NAI-MaP structure probing within virions and within Hela cells and computationally analyzed mutation rates in resultant cDNA to model RNA secondary structures.

They also performed pairwise interactome mapping by utilizing the SPLASH protocol. They performed SPLASH both in virions, in solution and in infected Hela cells to determine long range interaction mapping via crosslinking of the pairwise interactions and proximity ligation.

The authors analyzed their data with R-scape for covariation across the serotypes and viral species. They also analyzed structure and sequence conservation across viral genomes to create an identification method to find regions of functional interest.

Mutations that disrupted long range interactions were tested in interferon-deficient mice for their impact on viral fitness. Disruption of long range interactions diminished viral fitness.

Results

Utilizing NAI-Map and SPLASH, the study was able to identify a number of highly structured and conserved functional regions within the genomic RNA and large number of pairwise interactions. Many of these, in addition to being long range, showed high heterogeneity, suggesting that many regions can have many different structures and bind to different partners in the viral genome.

Comparative analysis showed that long-interactions were disrupted within cells significantly compared to those within virions. However, when the interactions identified within virions were disrupted via mutagenesis, viral replication was severely reduced and the long-ranges pairings were deemed important for viral fitness.

Discussion

One of the important topics focused on by our lab was the use of SPLASH to identify long distance pairwise interactions, which is something that is difficult to study in SHAPE MaP. This could be utilized to determine long-range pre-mRNA structures to understand transcript and/or gene regulation, even within long introns.

The study shows strong data for the regions of functional interest having a strong effect on viral fitness and does produce a better understanding of how the viruses in the Flaviviridae family are organized.

There was some confusion on figure labeling (such as the structure model in Figure 2b), but overall this was an informative paper utilizing a new procedure for RNA structure analysis.

Week 13 (4/14-4/16/21) – We attended the 2021 RNA Biology Symposium organized by the National Cancer Institute. This Symposium spanned 3 days with a range of topics including small and noncoding RNAs, translation and RNA modification. A major theme of the Symposium was phase transition, which we discussed in detail in February for our Journal Club on phase transition mediated by the SAR-CoV-2 nucleocapsid protein. Of the many informative presentations at the NCI Symposium we had several favorites. One favorite presentation was “Genome Regulation by long-noncoding RNAs” by Dr. Howard Chang. Dr. Chang described the in-depth work his group has performed on the XIST lncRNA and embryonic and somatic X-inactivation, some of which has been published. Another amazing presentation was “RNA in Genomic Medicine, diagnosing rare disease and COVID-19 implications” by Dr. Diana Baralle. Dr. Baralle’s recent publication on differential expression of ACE2 isoforms during viral infection provides insight into her presentation for those unable to attend.

Upcoming (4/23/21) – Edward Mabry (M.S. student) will lead our discussion on long-range structure mapping of Dengue and Zika viruses published in 2019 and led by the Wan lab.

Written by Luke Hatfield (Research Technician)

Week 12 (4/9/21) – We discussed PANDORA-Seq, a technique to detect modified small RNAs developed by multiple labs and led by Dr. Qi Chen.

Goals

• Develop an experimental method to overcome the current shortcomings of analyzing small and micro-RNA analysis using current RNA-seq protocols.
• Refine a bioinformatics pipeline for detection of microRNA expression in different experimental conditions.
• Exhibit PANDORA-seq’s ability to detect and identify small non-coding RNA (sncRNA) elements such as micro-RNA (miRNA), ribosomal RNA-derived small RNAs (rsRNAs), and RNA-derived small RNA’s (tsRNAs) at greater resolution than seen before.

Method

• Combine two existing enzymatic methods used to process small non-coding RNAs (sncRNAs): treatment with AlkB and T4PNK.
• Compare alternative treatment states (AlkB treatment first, T4PNK second, vice versa) No noticeable difference was found in administering the treatments in a particular order.
• Every sample set was tested in a compounding manner: first was traditional treatment, then the AlkB treatment, then the T4PNK treatments, finally the PANDORA-seq treatment in order to demonstrate the increased effectiveness of combining the treatments across sample types.
• These conditions were tested across six mouse cells (brain, liver, spleen, mESCs, sperm, and sperm heads) and three human cells (HeLa cells, Primed hESCs, and naïve hESCs).

Results

• PANDORA-seq showed that there are tissue specific responses to the treatments, revealing a landscape of miRNAs, tsRNAs, and rsRNA’s previously unquantified in traditional RNA-seq analyses.
• PANDORA-seq also helped illuminate the change in miRNA dynamics during iPSC induction in mESCs and hESCs, showing fluctuating levels as the cells progressed from Day 0 to Day 6.

Discussion

• PANDORA-seq demonstrated a very real ability to effectively capture small and micro-RNA in ways that traditional RNA-seq has previously been incapable.
• An extra chapter on the specific packages and bioinformatic analysis performed would be insightful to the data.
• The biological role of tsRNA and rsRNA are relatively unclear.
• Future iterations of this method and pipeline could potentially focus on identifying a wider variety of small RNA, such as alu-derived miRNA.

Upcoming (4/16-4/18/21) – We are attending the NCI RNA Symposium. There are lots of great speakers and registration is free!

Written by Austin Herbert (upcoming graduate student)

Week 11 (4/2/21) – Mutation of splicing factor genes has shown to be a key factor promoting cancer cell proliferation in colorectal carcinoma. Determining the effects of splicing factor mutations across different cancer types is essential for deconvoluting any shared or unique mis-splicing patterns in cancer. We discussed work by Seiler et. al surveying the mutational profiles and the subsequent effects of variation on splicing factor genes in 33 different types of cancers. 119 splicing factor genes with putative driver mutations were identified in one or more cancer types. Splicing genes were most frequently mutated as mutually exclusive events. Of the 33 cancer types, bladder carcinoma and uveal melanoma were found carrying more driver splicing mutations than expected by chance in like SF3B1, SRSF2, and RBM10. Finally, using and in-silico gene set enrichment analysis, Seiler et al were able to predict that splicing factor mutations may be associated with the deregulation of cell-autonomous pathways and immune infiltration. By identifying splicing factor mutation profiles across so many cancer types, Seiler et al has set the groundwork for experiments characterizing the functional consequences of such mutations and their overall contributions to cancer biology.

Upcoming (4/9/21) – We are deciding between Pandora-Seq and DRACO!

Week 10 (3/26-3/27/21) – We attended talks at the 5th Annual RNA Symposium on “Processing RNA” from the University of Michigan RNA Biomedicine Institute. Although conferences remain virtual for now, the RNA Biomedicine Institute put together a really good symposium. One of the highlights for us included a keynote talk by RNA structure guru Dr. Kevin Weeks. Dr. Weeks described his ongoing collaboration with Dr. Tony Mustoe on the aptly named DANCE-Seq chemical probing analysis protocol/algorithm. I’m looking forward to reading about DANCE-Seq and the insights it can supply when it is published! Another highlight for us was the combination of splicing talks. These included a keynote address by Dr. Tracy Johnston and a blitz talk by graduate student Cathy Smith. Dr. Johnston gave an amazing molecular biology seminar digging into how chromatin modifications influence splicing. I was particularly impressed with Cathy Smith’s system for understanding how variation influences splicing. In addition, we all agreed that the overall format of the Symposium was well thought out. A good example of how “less is more”. The organization committee scheduled intense sessions, but only from 11-2:30. For us this was just the right amount of time to have a diverse set of speakers, but not an overwhelming amount of information.

Upcoming (4/2/21) – We will discuss a 2018 paper by Seiler, et al. on splicing factor variants in 33 different cancer types.

Week 9 (3/19/21) – We discussed AGO3’s biological activity in the context of other human argonaute proteins with a focus on the 2012 publication by Hu, et al. on the special role of AGO3 in binding Alu-derived small RNAs and mediating gene repression. This work was done by the Rosenfeld lab.

We are planning some summer projects on AGO3 activity. We discussed the redundant roles for AGO1-4 in microRNA guided gene repression and then focused on the little that is known about specific functions of AGO3. Interestingly, a class of Alu elements are up-regulated by retinoic acid (RA) treatment in pluripotent cells. RA causes cells to differentiate along neural lineages. Dicer seems to target RA-induced Alu RNAs to create small RNAs between 28-65 nucleotides long. These Alu-derived RNAs are protected by AGO3 specifically. Several mRNAs contain potential Alu-derived RNA binding sites, including transcripts involved in cellular differentiation like nanog. Hu, et al. show that nanog mRNA is depleted after treatment with RA, in part due to AGO3-mediated activity. This was a comprehensive analysis of Alu-derived small RNAs in differentiation and a clear specific function for AGO3. In the same year, other researchers showed that AGO1 and AGO3 double-knock out mice are viable. I would be very interested to know if loss of AGO3 also resulted in loss of Alu-derived small RNAs and altered the ability of pluripotent cells to differentiate in mice. Little follow-up has been done in this field with the exception of some studies on the G isoform of polymerase III, which is involved in transcription of up-regulated Alu elements during differentiation.

Upcoming (3/25 and 3/26) – We are attending the 5th Annual RNA Symposium on “Processing RNA” from the RNA Biomedicine Institute at the University of Michigan.

Written by Luke Hatfield (research technician)

Week 8 (3/12/21) – We discussed Sharma, et al.’s paper on the kinetics of DAZL RNA binding from the Jankowsky laboratory at Case Western.

Goals

• Develop a method for assessing cellular binding and dissociation kinetics of RNA-protein interaction at binding sites on target RNA’s on a transcriptome wide scale.
• Determine how and how long RNA-Binding Proteins (RBP’s) interact with binding sites.
• Assess DAZL binding kinetics on many different RNA classes, before specifically assessing its regulatory effects on target mRNA’s.

Method

• KIN-CLIP (Kinetic cross-linking and immunoprecipitation)
  • Involved the use of a pulsed femtosecond ultraviolet laser to time-resolve RNA-protein cross-link
  • Followed up by immunoprecipitation and high-throughput sequencing

Results

• RBP binding is infrequent – according to their results RBP binding to RNA-binding sites occurs only around six times every minute
• RBP’s interact with mRNA-binding sites in clusters. The causes of these proximal clusters and their saturation points were both identified as important variables.

DAZL Binding

• It was observed that DAZL RBP’s bind to the 3’-UTR in mRNAs in clusters. This cluster formation suggests that RNA structure or the proximal binding site of other proteins plays a significant role in determing where the clusters are.
• It was observed that DAZL’s likelihood of binding was correlated with the number of binding sites in a cluster, as well as the cluster’s proximal location to polyadenylation sites.

DAZL Function

• Revealed that the degree of regulatory function DAZL has on a specific class of mRNA is dependent on the binding patterns determined by the previous experiment.
• mRNA DAZL cluster characteristics strongly correlated with 3’-UTR length, the number of cluster sites on a 3’-UTR, the proximity of a cluster to a polyadenylation site, and the cumulative difference for a RBP to bind at a given site in low and high concentration environments.
• Identified 21 mRNA groups that showed DAZL regulatory effects based on the previously defined conditions.

Final Thoughts

• More experimental consideration should be given to identifying the regulatory effects of binding RBP’s and how they can influence downstream genotypic and even phenotypic expression. Diving into more detail about the specific processes the DAZL regulation was accomplishing would have been an excellent third point to make in this article but may have been restrained due to time or impact.
• Uncovering the secrets of DAZL-RNA binding interactions will help develop methods and solutions for a widely applicable system of characterizing regulatory programs for other RBP’s.

Upcoming (3/19/21) – We will discuss a 2012 paper (Hu, et al. from the Rosenfeld lab) on AGO3’s role in cellular differentiation via small RNAs derived from Alu elements.

Written by Austin Herbert (upcoming PhD student)

Week 7 (3/5/21) – We discussed Park, et al.’s paper on leveraging in silico approaches for predicting effects of non-coding variants on the dysregulation of RNA-binding proteins (RBPs) in respect to psychiatric diseases. RBPs have a diverse set of functions and play a key role in post-transcriptional regulation. They function in RNA splicing, transcript stability, localization and translation, all critical roles in wild-type metabolism of RNA. By coupling their already published program “Seqweaver” with GWAS data, the authors were able to accurately predict the effect of non-coding variants on RBP dysregulation. The validity of their in-silico method was proved through discovering a novel link between an RBP-disrupting variant in DDHD2 and a multi-ancestry-associated locus that increases the risk of schizophrenia. Finally, their published data set capturing common and ultra-rare variants will further advance the biochemical characterization of RBP-dysregulation and its contribution to complex diseases.

Upcoming (3/12/21) – Luke Hatfield will pick an article for discussion.

Week 6 (2/26/2021) – We discussed Zeng, et al.‘s paper on a new method for ribosomal RNA (rRNA) depletion headed by the scientists in the Molecular Oncology lab at the University of Chicago. We recently were preparing transcriptomes for RNA sequencing and used a commercial depletion kit to decrease the amount of rRNA in our sample. While we were happy with the results of our commercial depletion, it was prohibitively expensive to perform biological replicates for multiple samples. We’d also like to troubleshoot new transcriptome protocols without rRNA being a cost bottleneck. So, we decided to discuss Zeng, et al., where the authors perform standard rRNA depletion using both hybridization probes to pull down rRNA (Ribominus Euk v2 platform) and hybridization probes designed for targeting by RNase H based digestion of RNA:DNA hybrids (NEBNext rRNA Depletion platform). As RNase H methods are already less expensive and perform better overall, most comparisons of the new protocol developed by Zeng, et al. were to the NEBNext rRNA depletion protocol. In this method – reverse transcriptase-mediated ribosomal RNA depletion (RTR2D), the authors very cleverly used reverse transcription to fill-in DNA hybrids across rRNA species and form full RNA:DNA hybrids that can be targeted by RNase H. This drastically cuts down on the number of probes required to perform depletion. They show that their method does not impact non-targeted RNA, like mRNA and lncRNAs. Using Nutlin3A (MDM2 inhibitor, p53 pathway) Zeng, et al. show that RTR2D consistently identifies differentially expressed transcripts with NEBNext rRNA samples, possibly even more consistently due to strong RTR2D rRNA depletion. We will definitely be trying this protocol in lab!

Upcoming (3/5/21) – Looking forward to discussing Park, et al.’s paper on RNA binding proteins and complex disease, specifically psychiatric disorders like schizophrenia.

Week 6 (2/19/2021) – We discussed Gawroński, et al.’s paper on light sensitive RNA structures for our journal club this week. Translational regulation is a common post-transcriptional regulatory mechanism where RNAs produce different amounts of protein. One of the great advantages of translational regulation is how easily it can be changed. The cell does not need to make more RNA, it just instantly changes the quantity of protein produced from an existing pool of transcripts. In plants, light is an important environmental signal. Several photosynthesis-related RNAs are translational regulated based on light conditions. It is not clear how RNA structure is influenced by light or by light-induced protein binding. Gawroński, et al. show that RNAs with weak Shine-Delgarno (SD) sequences have RNA structural changes linked with translational increases. They use the gene psbA as an example of a weak SD RNA, where RNA structure can block translation. Under high light conditions, the upstream SD region in psbA changes structure, opening up the SD and start codon to recognition by the ribosome. The authors hypothesize that protein binding causes these structural changes. They also demonstrate that rbcL, an RNA with a strong SD sequence, does not have significant structural changes upstream of the start codon, despite being translationally up-regulated in high light conditions. I would love to learn more about why and how protein binding to the RNA is controlled by light conditions. This meticulously planned study is applicable across species, not just in plants. I could see similar experiments in any organism for understanding the interplay between RNA structure and translational regulation during environmental response.

Upcoming (2/26/21) – Deciding between Sharma, et al. and kinetics of RBPs, Taliun, et al. and massive genetic variation or Zeng, et al., on mastering inexpensive library prep.