Lecture 11: Bacterial genome organization
Today:�
Bringing growth (mostly) to a close
Bringing growth (mostly) to a close
What have we learned?
Doubling time is a function of the nutrient quality.
To double faster with better nutrients, bacteria dedicated more of their mass to ribosomes.
When cell density is high or nutrients run out, growth slows to nothing
We can understand these dynamics with a simple model
Stable point: any initial bacterial density will evolve to this stationary phase!
When we extend to 2 species, we see three different stable states depending on how they compete!
1
2
coexistence
bistability
single-species dominance
We’ll now shift from the scale of huge populations of cells to molecules
vs.
DNA
RNA
protein
RNA polymerase
Ribosome
Our mathematical model predicted that there were characteristic states of the many cell system, e.g. stationary phase or stable coexistence points.
What if instead of species inhibiting each other it is genes inhibiting each other through gene regulatory network?
We’ll see that exactly the same math can be used to understand states of gene expression!
Now we’ll go from cells to molecules
This scale still very mysterious!
The molecule to start with is DNA!
Two major categories of bacterial DNA organization
1. Longitudinal or “ori-ter”
Caulobacter crescentus
Origin of DNA replication
Direction of replication
Replicated origin moved to other pole
ori-ter recapitulated in new cell
Two major categories of bacterial DNA organization
2. Transverse or “left-ori-right”
Slow-growing E. coli
tightly packed
Less condensed crossover region
Some species have variations on these two patterns or switch between them under different conditions.
How is the bacterial DNA arranged in the cell?
A bacterial genome has ~1-10 million bases in it.
In equilibrium, a double-stranded DNA with 1,000,000 base pairs will be a sphere (think ball of yarn) with a diameter of ~300 µm
~300 µm
This has to fit into a ~1 µm cell!!!!
How is the DNA organized?
Is it just randomly crammed in?
If not, what drives the organization?
How do we measure this?
How is the genome so condensed?
“Structural Maintenance of Chromosomes” protein complexes (SMC)
It’s not just physical confinement by the cell wall. DNA-binding proteins condense the genome. Like histones in eukaryotic nuclei.
Etc.
Do these complexes tend to condense the genome into an organized structure?!
How do you measure DNA organization?
1. Fluorescence imaging
DNA extremely dense, structures too small even for superresolution techniques
But you can do some simple, very clever experiments.
Measuring DNA organization with fluorescence
E. coli genome
ori – origin of replication
lac – lac operon
moveable label: different strain for each locus
Random
Regulated Pattern
What do they see?
Locations along the genome have a consistent position within cells!
Variance of intra-locus distance increases with distance between loci
Width of distribution increases
Cell position scales linearly with genomic position!
Average packing density of 1.6 Mbp / µm
ori
crossing region
Loci are most precisely located relative to each other than within the cell
“Single Locus” is the variance in a locus’ absolute location in the cell
“Inter-Locus” is the variance in a locus’ position relative to ori
constant
lower, i.e. more precise
Becomes less precise as you move away from ori
Genomic loci are more precisely positioned relative to their neighbors than relative to the cell.
Suggests major role of intra-nucleoid interactions in shaping structure.
This simple experiment revealed much about E. coli DNA organization!
How can we probe chromosome structure across the whole genome instead of discrete loci?
2. Chromosome capture
Find these kind of spatial interactions
Associate them with particular sequences
“Chromosome Capture”
DNA
Chromosome capture data
Resolution of ~5000-10000 bases
How do you read Hi-C plots?
Position along genome
Position along genome
Imagine the genome were a circle with no non-adjacent points contacting each other
You would only ever see frequent association with adjacent sequences in Hi-C
Hi-C Map
DNA
How do you read Hi-C plots?
Position along genome
Position along genome
What if instead there were a large loop on the right?
Hi-C Map
You’ll now see a large Hi-C signal corresponding to the areas that physically interact.
DNA
Off-diagonal features on the Hi-C map are interpreted as parts of the genome that are physically close, but far apart in genomic distance.
What does a DNA loop look like on these plots?
DNA
Genome position
Genome position
DNA
Genome position
Genome position
Hairpin loop
Simple loop
dot
Square region on diagonal
What does Hi-C tell us about bacterial genome organization?
Three broad domains
What does Hi-C tell us about bacterial genome organization?
Also a series of square domains of various sizes along the diagonal.
What are these?
Most of these domains are repeated in different strains of Bacillus, suggesting this is a structure of the genome.
Spatially interacting sequences, probably looping
With statistical analysis, Marbouty, et al. find “barriers” in the genome, regions with few contacts that define borders between domains of enriched contact.
Barriers are associated with highly transcribed genes or regions acquired via horizontal gene transfer
Transcription may lead to supercoiling, leading to close associations of regions in between highly transcribed genes
A model with twisted hairpins can recapitulate the Hi-C map
“plectonemes”
Are the barriers more highly-transcribed regions?
(This is C. crescentus, not B. subtilis, but it also has ori-ter structure)
Does transcription drive chromosome structure?
Little structure observed
(transcription inhibitor)
Does transcription drive chromosome structure?
Location of poorly expressed gene van
Insertion of highly expressed rsaA into van locus
What have we learned?
Bacterial genomes are not random, but highly organized.
Hairpin loops are interspersed by highly transcribed regions
Highly transcribed gene
Next: a bit of evolution and then how genes are expressed from the genome