How do you study for a preliminary exam? I don’t know, I was hoping you could tell me.

Table of Contents:

I might write something insightful up here at some point. For now, just trying to make a daily log of what I’ve been doing.

Step 1: Setting goals to stay focused and motivated

It seems like common knowledge that it’s important to have goals, but I often overlook them when starting a project.

My goal with my preliminary exam isn’t just to pass the exam, which

Study Notes


If I’m going to make progress, I need deadlines. All my deadlines are going to be for 4PM EST.

Tuesday, August 9 (5 weeks): Rough draft of slides and presentation Tuesday, August 16 (4 weeks): Schedule practice prelims for next week Tuesday, August 23 (3 weeks): Monday, September 12 (1 day): No more preparation


Techniques Log

Flow cytometry

RNA sequencing

Northern blot


For example, we want to see whether mazF inhibits ribosome formation. So, pulse 3H-Uridine and chase with cold uridine, run a polysome gradient and you can see that 3H-uridine is incorporated into ribosomes in absense of mazF, but not after expression of mazF.

Papers to-read

This section will eventually go-away, as I read these papers and put summaries in the daily logs.

Hockenberry 2018 for some hypotheses about why translation initiation varies.

Giess 2017 for another method that may help identifying translation start sites.

Why is initiation important?

Hersch 2014 can affect stalling.

Does CDS Sequence matter?

Verma 2019 early elongation events matter for efficiency.

How does rRNA maturation work?

Daily Log

Wednesday, 3 August 2022

Reading(s) for today:

1. Wade 2019, mBio

Summmary: This is a letter in response to Nigam 2019, about the generation of proteins after mazF toxin induction by nalidixic acid (NA).

In the 42 genes identified by Nigam 2019, none have more ACA upstream of the start codon than in the set of all E. coli genes. For some of these genes, ACAs identified by Nigam are not in the mRNAs at all.

Nigam sees GFP changes upon NA treatment, and Wade & Laub argue that this is probably a non-mazF-dependent effect, especially since many of the identified genes are known to be stress-responsive genes.

2. Nigam 2019, mBio

Summary: This is Nigam et al.’s reply to Wade & Laub’s critique above. They say that they don’t claim that ACA is enriched before the start codon in these genes. They claim that the stress transcription factor $\sigma^{32}$ is a result of modified ribosomes.

I’m not convinced by this response, so I’m not going to go further on the original paper–I think they’re detecting upregulation of a general stress response.

Tuesday, 2 August 2022

Progress for today

Not much progress outside of reading today.

Reading(s) for today:

1. Oron-Gottesman 2016, mBio

Summary: Used a fluorescent reporter +/- ACA sequences with induction of mazF with nalidixic acid to show some sequence dependencies of mazF.

Relevance to project: They have a leaderless construct, but not very relevant. Maybe worth looking at the EDF-like sequence in S1 for lmRNA expression?

A leaderless GFP mRNA, has 17 out-of-frame ACA sites that don’t interfere with expression after mazF induction. Adding an ACA site in the ORF prevented expression.

Really not impressed by the quality control in this paper. E.g., it sites Fig S1 about the effect of nalidixic acid on their reporter, but S1 has nothing to do with nalidixic acid–that’s S2. & Their “OD600” axis is in the hundreds–is it OD * 100?

Basically we’re seeing that induction of mazF reduces the GFP level in constructs with in-frame ACA sites. There’s no difference in a mazEF knockout, and it increases GFP level in WT and when you add an AC before the start codon (they call this leaderless due to cleavage, but don’t show that).

They make an argument that only in-frame ACA sites are cleaved, substantiating their claim by comparing PCR band intensities.

Then they find that their GFP-up/down phenotype +/- mazF induction can be reserved by making a mutation to the EDF-like sequence in bS1.

They conclude that there may be a bias away from ACAs being in frame, to protect from mazF cleavage.

I’m not convinced by this paper–the figures are not very clean, there are not multiple veins of evidence for each conclusion. It highlights that I really need to make sure I can think critically about the techniques raised in the controversy over the formation of stress-induced translation machinery, should also read Wade 2019 and the response Nigam 2019 and Vesper 2011, and Culviner 2018.

2. Culviner 2018

Summary: Used high-throughput RNA sequencing and ribosome profiling to carefully quantify mazF products. Leaderless RNAs aren’t upregulated in mazF stress.

Relevance to project: The Moll lab has argued that mazF creates lots of lmRNAs and specalized ribosomes. This paper shows that that’s not the case. How do we explain results from Moll and colleagues?

Take a $\Delta$mazF strain, put mazF on an ara-inducible promoter, low-copy plasmid. Induce mazF for just 5 minutes, take paired-end sequencing reads, and look for regions where density decreases. Confirm patterns with single-gene RT-qPCR.

82% of genes were highly-cleaved (2-fold or more downregulated, compared to empty vector) after mazF induction.

They also compare this to 5’-OH sequencing, but that’s not well-correlated–likely because of different 5’-end stabilities.

Not all ACAs are cleaved at the same rate, so they took some sequence logos and think that there’s a ~7-nt region.

They confirmed that these flanking sequences are important by selecting some of them across different RNA-seq scores, taking them out of context into a reporter, and then qPCR quantifying them, and that correlated as-expected.

They find a few genes that, by this method, may be leaderless after mazF induction (just 41). They measure ribosome footprints for all genes, and find that none of those leaderless genes have any substantial increase in footprints.

To put the nail in the coffin, they added an ACA site between the RBS and start codon of YFP, and did not see any increase in flourescence after mazF induction.

They also saw traffic jams of ribosomes upstream of identified cleavage sites.

As far as specalized ribosomes: they see clear cleavage products at the ACA sequences in nascent 16S rRNA sequences, which inhibits rRNA maturation. They pulse-chased hot uridine

Monday, 1 August 2022

I want to focus on a few things over the course of this study period:

  • Write my project proposal
  • Better understand and revise my project proposal
  • Map between my proposal and common techniques / ways of thinking from 1st year course material
  • Record my progress each day

Progress for today:

  1. Start to write a messy version of the preliminary exam proposal
    1. Not much progress here, just got the header/abstract done

Reading(s) for today:

  1. Nikolic 2022, BMC Research Notes

Summary: Used a fluorescent reporter +/- ACA sequences and +/- a leader under expression of mazF.

Relevance to project: They use a “leaderless” construct, but they don’t specify the promoter (native GFP promoter?) or show a 5’ start, but that may have been shown in reference paper (Oron-Gottesman 2016, will read next). In both plasmids, they are not able to see expression of leaderless GFP over background, except in the high-copy-number-plasmid after 6h mazF induction. So, their lmRNA GFP is not highly expressed.

Relevant data points:

  1. Their “leaderless” GFP is expressed at the same level as no-reporter, whereas canonical GFP is ~40-fold in exponential phase.

Figure summary:

Figure 1 shows GFP (under leaderless or canonical mRNA, both without ACA sequences) after induction of mazF. In a high copy plasmid, there’s clear increase in GFP concentration compared to no-plasmid control.

Figure 2 shows that the amount of increase of flourescence after mazF induction is lower for mCherry (with ACA) than GFP (without ACA)