Wright et al. on the Process of Mutation

In this thread I would like to explore a specific paper written by Wright et al.

quote:

---

Proc Natl Acad Sci U S A. 1999 Apr 27;96(9):5089-94.
Hypermutation in derepressed operons of Escherichia coli K12.
Wright BE, Longacre A, Reimers JM.
Source

​

Division of Biological Sciences, University of Montana, Missoula, MT 59812, USA.

​

Abstract

​

This article presents evidence that starvation for leucine in an Escherichia coli auxotroph triggers metabolic activities that specifically target the leu operon for derepression, increased rates of transcription, and mutation. Derepression of the leu operon was a prerequisite for its activation by the signal nucleotide, guanosine tetraphosphate, which accumulates in response to nutritional stress (the stringent response). A quantitative correlation was established between leuB mRNA abundance and leuB- reversion rates. To further demonstrate that derepression increased mutation rates, the chromosomal leu operon was placed under the control of the inducible tac promoter. When the leu operon was induced by isopropyl-D-thiogalactoside, both leuB mRNA abundance and leuB- reversion rates increased. These investigations suggest that guanosine tetraphosphate may contribute as much as attenuation in regulating leu operon expression and that higher rates of mutation are specifically associated with the derepressed leu operon.

---

The full text (in .html format) can be found here.Once the topic is promoted I will describe the methods used and the results in a way that a layperson can relate to. After this, we will determine if these findings demonstrate a process of random or guided mutation.Just to get it out of the way, I define random mutations as changes in the DNA sequence that are blind to the needs of the organism. IOW, mutations are random with respect to fitness. I am not saying that mutation rates are constant through time, nor am I saying that each base has an equal chance of being substituted, inserted, or deleted. With respect to the paper, I will attempt to demonstrate that the same mechanisms that produce reversions in leuB- organisms will also cause deleterious mutations in very important and vital genes as well as mutations which do not change the fitness of offspring.

I am assuming all of you understand the relationships between replication (DNA to DNA), trascription (DNA to RNA), and translation (RNA to protein). This is often called the Central Dogma. If you don't understand these concepts then use google for a quick refresher course. This is basic biology stuff, so it shouldn't be hard to find or to understand.What I would like to do first is describe the different E. coli strains, genes, and the biochemistry that affects gene expression. This is vital for understanding the data in some of the figures and tables.Strains and gene knockouts:CP78: this strain lacks functional genes to produce the amino acids arginine (argH-), histidine (his-), threonine (thr-), and most importantly leucine (leuB-).CP79: Same as CP78 but with one difference. It has a deleterious mutation in the relA gene. This relA gene product (i.e. relA protein) is responsible for producing guanosine tetraphosphate (ppGpp).78AL: Same as CP78 but with a new promoter region for the leuB gene. This will be very important in my future posts, so pay careful attention to this. The replaced the portion of the leuB gene that controls its expression with a promoter that responds to another chemical, IPTG. This allows them to control the expression of the gene independently of leucine concentrations in the surrounding media. If they want more leuB gene products all they need to do is add IPTG. For those who are interested, they are replacing the native promoter with a modified lac promoter. A google for "lac promoter" should find hundreds of thousands of hits describing how this promoter works. For those who do not know much about promoters now would be a good time to do a google for that lac promoter. I would suggest this site, especially the animation tab which has narration describing how the lac system works.So how do all of these genes interact with one another and the environment? Since the paper focuses on the leuB gene and derepression of the leuB gene we will focus on that.First, what is derepression? This is where a gene is normally kept in the off position until a derepressor removes whatever is stopping the gene from being transcribed or creates a situation which turns the gene on. In the case of leuB, the gene is repressed when there is leucine present in the growth media. This prevents the bacteria from using energy to produce its own leucine. When the cell senses that its amino acid stores are low (be it arg, his, thr, or leu) it increases the expression of relA which then increases the intracellular levels of ppGpp. This is called the stringent response. In addition to ppGpp increases, there is also another system that senses leucine stores specifically (the paper doesn't mention this second level of control, but I can try and find it if someone is interested). In order to remove the repression on the leuB gene (i.e. derepression) you have to have both an increase in ppGpp and the lack of leucine. The result is an increase in leucine through "de novo" pathways instead of scavenging from the environment.To sum up this section, when you have low leucine levels it kicks in two systems to derepress the leuB gene: the stringent response and the leucine specific system.This is why they are comparing CP78 and CP79. They want to see how changes in ppGpp affect the overall mutation rate in leuB. By using a non-functional mutant of relA they are able to remove ppGpp from the picture and see what affects derepression has. 78AL was also used to move to a completely different repressor system to further verify that single stranded DNA is what is being hypermutated.There are other mutants and rescued strains, but they are not central to the main argument. These are more of a check to verify certain findings. If need be, I will include those other strains once we discuss the data.If I have any of this information wrong PLEASE correct me.Edited by Taq, : confused stringent response with leucine specific systemEdited by Taq, : No reason given.Edited by Taq, : edited strain names

Good catch. It is now fixed. That is a very fair criticism. For those who are interested, this paper outlines the features of the tac promoter.I decided to leave the details out because I am trying to make this accessible to the scientific layperson. From my reading of the paper, the important bit is that the gene can now be controlled with IPTG instead of ppGpp. That is what I want to stress in further posts.Edited by Taq, : No reason given.

It wouldn't distinguish between the two, but they do control for this by measuring the mutation rate in non-starvation conditions. They did not see any mutants in these controls, at least not at the concentrations they were plating at.

I don't think the authors ever stated that it was. What they are trying to show is a major increase in mutations under specific conditions. I think their methodology and results can show and do show an increase in mutations for actively transcribed genes. I think the authors would agree that if you plated enough bacteria that you would find a leuB- reversion in the non-starved controls.

To help clarify the 3rd post in this thread, let's look at figure 2 from the paper: Caption:

quote:

---

Figure 2
leuB mRNA abundance in E. coli CP78 (relA+, ppGpp+) and CP79 (relA−, ppGpp−). Cells were grown to log phase (log) in minimal medium or were grown to log phase and were washed and transferred to minimal medium without arginine (−arg), threonine (−thr), or leucine (−leu) for 15 min. Total RNA was recovered, and the level of leuB transcript was determined by nuclease protection assay and densitometry (see Materials and Methods).

---

This figure is straightforward. It demonstrates that in CP78 (relA+) the amount of leuB mRNA is tied to leucine, and only leucine, starvation. In media that lacks threonine or arginine there is no significant increase in leuB over control (which is the clear bar labelled "log"). However, there is no signficant increase in leuB in CP79 (relA-). This indicates that relA, and by extension ppGpp, is important for the leucine starvation specific derepression of leuB.Any questions so far (I am talking to you zi ko and shadow71)?

That is correct. It took me 3 or 4 tries to figure out what you were talking about, hehe.If you want to work ahead I will be focusing on Figure 3 as well as Tables 1 and 2 for my next post. As a brief preview, pyrD and glpK have the opposite expression profile as leuB, and spoT is the gene responsible for ppGpp degradation. spoT knockouts will have higher ppGpp levels due to lack of degradation.

If you can keep from ruining the punch line it would be most appreciated. Of course, this is a public thread so anyone can say whatever they want (as long as it is on topic). I am trying to go through the paper slowly so others have time to ask for clarification and ask questions. If it was presented all at once it would be too overwhelming, IMHO.

It's been a couple of days, so this first figure will be a bit of a refresher. caption:

quote:

---

Figure 3
Ribonuclease protection assay of E. coli CP78 (relA+, ppGpp+) and CP79 (relA−, ppGpp−). Total RNA was recovered, and the levels of specific transcripts were determined by nuclease protection assay and densitometry (see Materials and Methods). Total RNA [5 μg (lanes 1—4) and 10 μg (lanes 5—8)] was hybridized simultaneously with probes for leuB, pyrD, and glpK mRNA. E. coli CP78 and CP79 cells were grown to log phase (L) and then were washed and starved for leucine for 60 min (S). RNA markers 200—500 nt in length were run in the last lane. The leuB hybridized fragment is 436 nt, pyrD is 313 nt, and glpK is 244 nt.

---

It would take a few posts to explain how this blot was produced. Thankfully, the actual method is not that important, but the results are. The authors picked 3 different genes that should respond differently to ppGpp levels. As has been discussed, leuB is derepressed when ppGpp is elevated. pyrD is just the opposite. It is repressed by ppGpp. glpK is only controlled by the growth rate of the organism and is not sensitive to ppGpp. L stands for log phase (unstarved) and S stands for starvation. They tested both the relA- and relA+ strains. The results are just as they should be, so nothing too surprising here.Now we get to the meat of the paper.Table 1
Correlation between leuB mRNA levels determined as described for Fig. ​Fig.22 and leuB− reversion rates(too lazy to transcribe the data here so click on this link)They measured two things. The amount of leuB mRNA compared to the total amount of mRNA for all genes. If this ratio increases then leuB has been derepressed. They also measured the number of leuB- clones that regained their ability to make leucine which are called leuB- reversions. The leuB- gene had a very specific mutation, "a C-to-T transition resulting in a serine-to-leucine substitution at amino acid residue 286 of the LeuB protein" according to the paper. The rate at which that T was reverted back to a C is the reversion rate. Also keep in mind the units that the reversion rates are given in, which is 10^-9. So for the first number given in that column, 0.15, that is 0.15 reversions per billion bacteria. Yes, billion with a B.The table demonstrates that the reversion rate does correlate with the amount of leuB mRNA. It also correlates with conditions where ppGpp are elevated (in the relA+ and spoT strains with leucine starvation conditions). The highest reversion rate in this table is 2.5 reversions per billion bacteria. This is the almighty guided mutation, 2 reversions per billion bacteria. Powerful, isn't it?

I have done a million Western's, so I tend to use the word "blot" where it isn't appropriate. I am much more of a protein guy than a gene jockey. I thought it would be confusing to describe how the RNase protection assay worked when the results are pretty straightforward. But thanks for the clarification anyway. Most appreciated. I don't know if it was discussed in this paper or a related paper, but it is known that the relA2 (i.e. relA-) genotype is leaky. relA2 is not an allelic replacement knockout, but relA with a specific mutation. I think this shows up more in the spoT knockouts. For the purposes of this discussion, I am more than willing to conceed that lueB derepression is tied to leucine starvation and is partly dependent on ppGpp, leaky alleles or not. I would agree with your cynicism. Many of the papers I read do not show entire blots, only a slice of the blot that includes the band of interest. I have always hated that. Sometimes there is important data elsewhere on the blot. What one author calls "non-specific" binding/bands I call important information.

Now would be the time to ask questions. There are many knowledgable people participating in this thread that would be happy to answer them.

How does this specifically relate to the paper we are discussing? Is the process of transcribing and translating the leuB gene any different than how the rest of the genes in the genome are treated?Also, did you have any questions about the information presented thus far? Do you understand how leuB is derepressed and how reversion rates are related to mRNA abundance?

Just so that we are on the same page, I will fully agree that the process of evolution is non-random. If evolution were random then deleterious mutations would have just as much a chance of being passed on as beneficial mutations. This is not the case.However, one of the mechanisms within the process of evolution is random, and that process is mutation. I will show that the processes that produced the leuB reversions in the Wright et al. paper are incapable of determining if the mutations are helpful or harmful to the overall fitness of the organism.At this point, I have discussed a table that listed the number of leuB reversions due to a single nucleotide substitution. The rate at which this occurs is 2 per billion cell divisions. Just 2 bacteria out of every 1,000,000,000 are able to gain an advantageous mutation in their leuB gene. If this process is a guided process, then why is the reversion rate so low?I think I will hold off on discussing new material in this paper until Monday to give people time to discuss why the reversion rate is 2 per billion cells.Edited by Taq, : No reason given.

Perhaps you could explain how it is lacking? Are you saying that fitness is not a factor in the propagation of a genome?

The two aren't the same, but they are both measures of fitness. Obviously, an organism that has one offspring is more fit, on average, than an organism that has none.It is worth mentioning that the authors are using different environments to test the fitness of the organisms. For example, the bacteria in these experiments are descendants of a leuB- strain that is not able to produce it's own leucine. Some of those descendants do spontaneously acquire the ability to make their own leucine, and therefore reproduce at a higher rate than the other descendants. Is this a valid definition of increased fitness in your eyes?