Discussing X Chromosome Dosage Compensation in Drosophila & Mammals with Professor Asifa Akhtar
July 30, 2019
Dr. Asifa Akhtar is a Senior Group Leader and Director at the Max Planck Institute of Immunobiology and Epigenetics in Freiburg, Germany. Her lab focuses on dosage compensation in Drosophila melanogaster, and they also explore the mechanisms used by other organisms to achieve balanced levels of sex chromosome transcription in males and females.
In her recent interview on Active Motif’s Epigenetics Podcast, Dr. Akhtar talked about how the male-specific lethal (MSL) complex, and the histone acetyltransferase MOF in particular, contributes to the regulation of the dosage compensation process. Furthermore, she also talked about some potential functions of the conserved MSL proteins in humans and how they are similar to and different from their fruit fly counterparts.
Listen to the full interview with Professor Asifa Akhtar on Active Motif’s Epigenetics Podcast.
Professor Asifa Akhtar on How She Got Started in Science
Dr. Stefan Dillinger: It’s always interesting to learn how top researchers got their start in science. How did you become interested in biology in the first place? Were you interested in biology ever since you could remember?
Professor Asifa Akhtar: Science was always one of these things that was generally interesting to me as a student, but what I distinctly remember is at the time when I was approaching A levels, I was learning about very basic descriptions of DNA and genetic information, and transcription, and I thought that was pretty cool and interesting. But then when I really started doing biology at the university and was doing an internship during this time to do these small, short-term experiments, that's when I realized that actually doing experiments was extremely exciting and also discovering something new was extremely exciting.
Up until then, I didn't really know what I wanted to do, but these little, short-term research projects really enlightened me and fueled my interest in research. And I thought, “this is exactly what I want to do, and I want to do a Ph.D.” And that was really the concrete moment.
Stefan: When you started your Ph.D., how did you select your research topic? Was it more like you found your topic, or was it more the other way around, that your topic found you?
AA: During my time in university, I was interested in studying transcriptional regulation and learning how things are regulated. So I was looking for labs that were studying transcription regulation and that's how I ended up looking at Imperial Cancer Research Fund (now known as Cancer Research UK), because there were some leading scientists in that field that were working there.
And, interestingly, my first stop at ICRF was in the lab of David Bentley, but after a few months he moved to Toronto, and at the time, I didn't want to move countries. And, luckily, on the same floor we had Richard Treisman, who we used to have joined group meetings with, and he was kind enough to adopt me. And so, the rest of my Ph.D. was done in his lab. It was a fantastic experience to be in his lab, and I'm very grateful for that.
Similarities and Differences in X Chromosome Dosage Compensation in Flies and Humans
Stefan: And now you're working on the MSL complex and also on dosage compensation. In an earlier episode of our podcast I spoke to Edith Heard about X inactivation in mammals and the mechanism. However, the mechanism is different in Drosophila, the organism that you work on. How do flies differ from humans in this process, and why?
AA: “Why?” is an interesting question, but it's definitely different, and very excitingly different. In mammals, females will activate one of their X chromosomes randomly, and then this decision, once it's made, is remembered. For flies, the males have a single X chromosome, compared to the females with two X chromosome, similar to the case for humans, but male flies basically do double duty and overexpress their single X chromosome by two-fold to up-regulate.
So, you basically have opposite outcomes of dosage compensation in the mammalian system versus the Drosophila system. In mammals, there is inactivating in females, and in Drosophila, there is up-regulating in males.
I think flies are pretty clever, actually, to do it this way because they take care of two things. They don't only take care of extra autosome ratio, because all the autosomes come in pairs and in the male then the sex chromosome is singular, but also take care of male to female differences between X and Y chromosomes.
But despite this difference in the overall outcome, one is activating and one is repressing, there are actually also similarities. If you look at it from the chromosomal perspective, the majority of the genes on the X chromosome are getting cis-regulated in one way or the other, so there is some similarity there. Also, the two large noncoding RNAs in the Drosophila system are very involved in regulating this transcriptional mechanism. This is very reminiscent of Xist, where it actually coats the inactive X chromosome and is involved in repression.
If you take it one step back, there are similarities in how large non-coding RNAs are cis-acting on the chromosome that is going to be regulated. If you look at it from a bigger picture perspective, X chromosomes in both flies and mammals provide a very interesting system to look at gene regulation, how genes are dosage compensated, up-regulated, or down-regulated.
But also, interestingly, in both systems, not every gene is dosage compensated, so despite this regulation how do certain genes escape? What are the boundaries of what to regulate? The logic behind this is very interesting and complicated. If you just look at the linear piece of DNA it's not very easy to understand what genes escape this regulation. It's not that every second gene is compensated, for example.
And how that logic is worked out by the cell is basically the biggest challenge that we would like to understand.
Stefan: Would you say to sum this all up that it's basically similar overall outcomes, balanced X chromosome expression between males and females, but mediated by different mechanisms?
AA: Yeah, I think the strategies that are used are very distinct, but at the global level, what is similar is that you're operating on a chromosomal level. The proteins are different, but maybe the underlying principle to regulate a chromosome is that level of similarity that I'm talking about.
Regulation of Dosage Compensation by the MSL Complex
Stefan: The complex you're focusing on is the MSL complex, which is one of the main components of this dosage compensation pathway. In a paper in 2011 in NSMB described the structure of this MSL complex and how it interacts together with MOF. How does this work?
AA: Maybe just before I go into the structure, let me introduce to you what the MSL complex is and how it was identified. The MSL complex stands for male-specific lethal complex. The heroes of genetics, father figures in the field who have been working on this, revealed factors through genetic screens that are important for dosage compensation because their absence causes male-specific lethality. So if you don't have these factors, males are dead. The MSL complex contains both proteins and also large non-coding RNAs.
What is also very interesting is that not only do these factors give a very interesting phenotype when mutates, the MSL complex proteins and RNAs localize very beautifully to the X chromosome when you stain for them using antibodies or you do RNA FISH for the RNA. This type of staining allows you to immediately recognize what is X and what is autosome.
The visualization of this entire process makes you think, how is this achieved, how is this complex able to recognize, in this milieu of all the chromosomes, that the X chromosome is where I want to go? And the Drosophila system is even more interesting than other organisms from that perspective because we have been able to take pieces of the X chromosome and put them on an autosomal location to ask whether the X chromosome be recognized only at its normal location, or whether the MSL complex can find it out of context? It's like finding your mommy if you were shopping on a busy street and suddenly lost her.
But this complex does that. You will be able to recognize very beautifully this one sharp band where the X chromosomal sequences were residing, so it's able to recognize it out of context.
This is just to give you a brief overview of the history of how these proteins were originally identified, which is namely by genetics, but in the meantime, there was plenty of biochemical evidence, and we have also contributed to this, that this really is a core complex of four proteins that interact, MSL1, MSL2, MSL3, and MOF, but in addition, there is an RNA helicase and these two roX RNAs that are part of this complex, and altogether they form this very robust entity that is regulating the X chromosome.
Now, how the structure is assembled, which was the original question you asked, is actually also intricate. We don't understand everything about this, but by using substructures of the complex, we know that MSL1 is basically the scaffold protein of the complex. MSL1 dimerizes and this dimerization is used by MSL2, which is the ubiquitin ligase the complex, to dock on. You can imagine MSL2 as being like an inverted Y chromosome. Basically, the bottom part is where the dimerization happens, but then it has its two legs and this is where MOF and MSL3 reside, so they don't need dimerization for interaction. So, it's a heterodimeric complex that works together and, of course, these structures are based on sub-complexes and it will be really fantastic to actually do other cryo-EM microscopy or crystallography to see how the whole complex would work together.
Stefan: How does the MSL complex recognize and bind the X chromosome? Is DNA sequence-specific, or more chromatin or histone modification based?
AA: This is one of the biggest challenges that is still unsolved in the field. There are certain DNA sequences present on the X chromosome that have the potential to recruit the MSL complex, but such sequences are also present on autosomes, so it's not clear how X chromosome binding specificity is achieved. Is it purely DNA-based? It's very difficult to imagine that it's only DNA-based. I think it's the composition of many different factors that make it happen, and definitely these non-coding RNAs that are expressed from the X chromosome play a very important role in that specificity because they are male-specifically expressed and there are certain male-specific factors that are also expressed that will help localize the rest of the proteins that are not male-specifically expressed to target.
What is MOF and How Does it Regulate Gene Expression?
Stefan: You published a paper in 2012 in the journal Cell which investigated the function of MOF. What is the function of MOF and what makes it so special for Drosophila?
AA: MOF always comes to us and brings us surprises. MOF stands for “males absent on the first” and it was given this name because the early genetic experiments demonstrated that males with mutations in this gene were not viable.
MOF is an acetyltransferase, it’s the major histone H4 lysine 16 (H4K16) acetyltransferase in both flies and mammals, so if you remove it, bulk H4K16ac is decreased in both of these systems. MOF is an activator of gene expression in both human and fly systems.
From the main phenotype and from we had known before, we thought that MOF only had an X chromosome-specific function. But by really performing detailed binding analyses in males and females, we realized that actually it's a protein that not only binds to the X chromosome, but also binds to promoters on autosomes, so it seems to actually have two functions. In subsequent experiments, we were also able to show that MOF resides in two distinct complexes. In addition to the MSL complex that everybody knew about, MOF is also part of a new complex that we called the nonspecific lethal complex (NSL) because all of the proteins associated with this complex are essential. This is the complex that targets all of these autosomes with MOF. So MOF basically has two lives in the cell.
Of course, the challenge for us is to understand how the tug-of-war between these two complexes happens. Is MOF sometimes in one complex and sometimes in the other complex? Is there an exchange of MOF between the two complexes? Do they ever see each other in the cellular milieu? These are questions that we are trying to answer at the moment.
We are also trying to understand in the lab is what the contributions are of the MSL versus NSL complexes in the mammalian system. It looks like the majority of the function at the promoters in the mammalian system is dictated by the NSL complex.
Prefer listening over reading? Check out the full interview with Professor Asifa Akhtar on Active Motif’s Epigenetics Podcast.
Stefan: What are the functions of MOF in the MSL and NSL complexes? Does it do different things in the different complexes?
AA: I think the thing that is really very interesting is that these proteins are not just something unique that just happens in flies. These two complexes also exist in the mammalian system. Evolution conserved two complexes that are use MOF as their catalytic subunit within them, and it’s very interesting to ask what’s happening in each complex. What are the differences and similarities?
This is one of the things that is really interesting in the lab, we used both the Drosophila system and the mammalian system to look at what has been evolutionarily conserved and what has been diversified for function. And now we are using, more and more, not only fly models, but also mouse models, and have made several knockout mice for the MSL complex members, and interestingly, most of these proteins are also essential in the mammalian system, which is very exciting.
Another aspect that we realized, again MOF bringing us surprises, when we moved into the mammalian system was its chromosomal localization was different than in flies. In the Drosophila system, as I told you already, the X chromosome provides a very visual system to look at biology because the X chromosomal territory is so beautiful to see. Moving into the mammalian system, we were trying to stain the cells to see where MOF localized, especially because X inactivation happens very differently than dosage compensation in flies, and since this in an activator complex, we wanted to know where it localized.
In addition to its canonical function in the nucleus, we were surprised to realize that MOF also localizes to the mitochondria, which is the powerhouse of the cell. So again, another function that we had not anticipated and, together with MOF, there were also some other NSL complex members localizing to the mitochondria.
That observation started another new chapter in this story. How is the communication between the nucleus and mitochondria being achieved? I find this really interesting because there are only two organelles in our cell that have a genome, the nuclear genome and the mitochondrial genome, and we found this NSL complex is located in both of these compartments.
Stefan: Is this a connection between nuclear gene expression and mitochondrial gene expression?
AA: Exactly. And that's what we showed, that MOF actually resides in the mitochondria, not only alone, but with these proteins as well, and affects mitochondrial transcription in the human cells and is very involved in metabolic control.
The type of question we are trying to understand now is, what is the communication between metabolic control and gene expression? Could it be that MOF complexes or sub-complexes that exist in the mitochondria somehow crosstalk with the amount of metabolic activity that is needed in the cell and help regulate gene expression in the nucleus? MOF binds most of the housekeeping genes, and most of the housekeeping genes are metabolic genes, so how is that regulation achieved? These are really interesting questions that now surface that we had never thought about before.
I think the move from the Drosophila to the mammalian system was very interesting for us from many aspects. Looking at this communication between organelles is very interesting and we are trying to understand how this is happening because the nucleus is not working alone.
There are all these things that are happening in the cytoplasm, which of course help regulate gene expression, or vice versa, and investigating how gene expression is dictating some of the organellar functions will be very interesting in the future.
Stefan: Your research has moved from the basic biological mechanisms of dosage compensation in flies to nutrition and environmental effects on gene expression in humans. That’s quite a big transition. How did this progression change the direction of research in your lab?
AA: Yeah. I'm very interested in this crosstalk between epigenetic and metabolic regulation. Of course, when you invoke metabolic demand into the system, it's interesting to see that every organ in our body needs metabolic control differently and responds differently to nutrients and energy levels. It could be that when we knock out MOF and its components in different cell types, we have different context-specific outcomes because the metabolic demands of these cells are different.
There could also be different results at different times of the day, so circadian rhythms could be very interesting to look at. We haven't really looked at that yet. But again, it makes us think about this entire set of complexes in a very different manner than we used to think about them before.
These research questions added new dimensions to our lab. The core is still trying to understand this big question of how the X chromosome is regulated, but this fundamental question, although it only affects two-fold regulation, has a lot of intricate fine-tuning, which is beautiful to understand.
But to be able to understand this completely, I think we had to diversify into different areas to be able to then maybe come back from the lessons we learned, maybe in the mammalian system, because we were not even able to explore this in the Drosophila system. I think that the challenge, but also maybe a good strategy to go forward, is to actually learn how this very important histone acetyltransferase is working in different systems and then synthesize the bigger picture.
The Involvement of Non-Coding RNAs in Dosage Compensation
Stefan: You mentioned previously that it's not only proteins in this complex, but also non-coding RNAs. How do the non-coding RNAs contribute?
AA: The involvement of non-coding RNAs is one of the very exciting aspects of dosage compensation, in general, both X inactivation and dosage compensation in Drosophila. In the flies, as I said before, there are two large non-coding RNAs that contribute to assist regulation. There is no sequence similarity between the two, but structurally, they have stem-loops that are very important for docking the proteins, the MSL proteins, and what we think is that they are very important for the spreading of the complex along the X chromosome.
I would say maybe there is a combination of maybe the structural information in the non-coding RNAs, the way these proteins and RNA complexes interact. What I would love to know is whether the interaction of these RNAs with the underlying DNA maybe causes some kind of special structure that helps keep the genes open to help cis-regulation.
These are all questions that we are trying to understand, and also how the RNAs are helping assemble and spread the complex, are questions that we don't really fully understand. But their absence causes male-specific lethality, so they are definitely important and redundant, so you only need one of the RNAs to execute the function. The question is how this redundancy is built, knowing that they are very different in sizes. One RNA is about 600 base pairs and the other RNA is about three and a half kilobases.
So how are they functioning? We are looking at early development in flies to see whether this makes a difference, because the RNAs are expressed at different times, to reveal their differences. This phenomenon has been known for years, and these are very interesting questions that have yet to be answered.
Hi-C and Chromatin Conformation Get into the Dosage Compensation Game
Stefan: In 2015 in a publication in Molecular Cell, you expanded this study and then this area into a Hi-C and the TAD area, and then the spreading along the X chromosome. What were the findings there?
AA: The trigger for doing those kinds of experiments was this frustrating observation that if you just look at the linear piece of DNA, you don't find the logic of where the complex is binding and what genes are going to be regulated. There must be something more than just looking at the genes to be compensated and that's why we tried to do Hi-C experiments to look at whether chromatin conformation plays a role in regulating dosage compensation. The naive hypothesis was that maybe the X chromosome would look different in males versus females, and maybe this will make a difference. But yet, the very interesting and surprising, but at first is a disappointing finding was that the X chromosome looks exactly the same between males and females.
But what was very interesting is that there are sites on the X chromosome that would be called high-affinity sites, which from other chromatin-binding experiments are known to show a very high signal for the MSL complex. And we found that those regions cluster in space.
And from that came the idea to propose this conformation-based affinity model. If you just look at just the pure DNA sequence, these sequences are present also somewhere else on autosomes, but because of their clustering on the X chromosome, they may be able to help basically trap the complex, and this generates specificity. By actually doing this clustering, you will be able to bring pieces of DNA together that, in space, are in proximity, but if you stretch them out, they will not be next to each other. That's what you see actually when you do further experiments and look for the relationship between whether these sites that are present on the X chromosome and whether the genes are going to be compensated. What you see is that the proximity of the genes to these sites correlates with the extent of dosage compensation.
We think chromosomal topology provides a very important way to organize the X chromosome and contributes to the specificity that we see. But again, as I said to you before, I don't think it's just one answer that we will get. It's not purely the DNA sequence. It's not purely the genome topology. It's everything coming together in the right dose leads to dosage compensation.
How Do Male X Chromosomes in Drosophila Know How Much to Increase Gene Expression?
AA: One aspect of this process that maybe people don't really think about, especially for the fly system, is that it's not about lack of activation. The female genes are also transcriptionally active. It's just that the males are two times more active. So how do you fine-tune this activity? You're going from activity, to a little bit more activity.
What is the sensing mechanism? How does the cell know that it has made two times more?
We know that at least we have some clues because the genes that are on the X chromosome are important and essential, and some of them are already haploinsufficient genes, so the dosage of these genes is actually important.
In a recent study, we have also shown that the components of this complex are not exclusively binding to X chromosomal sites, but also autosomal sites, and when these are autosomal sites, you look at which genes are being bound. They are, again, evolutionarily conserved haploinsufficient genes, interestingly bound by the MSL orthologs in the mammalian system.
It looks like there is a hidden logic to kind of sense haploinsufficient genes in the fly system and the mammalian system. We have more clues now than we had a few years ago, so it's a very exciting time to be investigating the regulatory principle behind that logic. I think that the clue will come from looking at orthologs in the mammalian and Drosophila systems because with orthology we will be able to figure out what is really essential to keep the functional aspect conserved.
Currently what I think is that the core MSL complex, which is composed of proteins that are expressed in males and females, is the evolutionary complex that is involved in general transcriptional regulation. This system was hijacked for a specialized function to regulate the X chromosome by involving the non-coding RNAs and evolving male-specific factors that will now help to do one specialized function. But this general transcription regulation, to maybe look at a dosage-sensitive gene, is what is also conserved in the mammalian system and maybe was the ancient function of the complex.
I find having the non-coding RNAs is a very elegant way to go about this entire thing because you can rapidly evolve these RNAs and by using these RNA-binding proteins and non-coding RNAs, you're able to diversify a function and come up with new strategies during evolution, so I think that's a clever way to do it. We don't have all the clues, but this is how I think, from the data that we have right now, very likely this is how the logic evolved.
What Does the Future of Dosage Compensation and the MSL Complex Look Like?
Stefan: This conversation has taken us on an exciting journey through your scientific career, but can you give a summary of the scientific finding you consider to be your most important or most exciting one? And also give a little outlook into the future of what you are going or want to do.
AA: I think trying to understand X chromosomal regulation and how these two-fold effects are central themes that I think that my lab has contributed to from a biochemical perspective, genetic perspective, and genomics perspective, and this will continue to excite us for years to come.
Diving into the mammalian system and looking at the interface of the fly and the Drosophila system was a very exciting new area for my lab because it is also bringing us new knowledge and new aspects that were totally underappreciated about the complex.
One aspect that I'm very excited about and we didn't really discuss is that we are not stopping at mice. We also are moving into the human system I have been very fortunate to collaborate with a cohort of clinicians recently where we identified mutations in MSL3, which is a core component of the MSL complex, and were able to define a new syndrome where the loss of this protein that is part of the core complex leads to extremely reduced histone H3 lysine 16 acetylation in human patients, which is associated with developmental disorder, a neurological disorder, and the challenges to try to understand how we can understand from basic molecular fundamental biology, information that we can then help these patients.
For me, the biggest challenge will be to look at this interface, use fundamental basic research to be able to then address how, in these syndromes, where more and more MSL and NSL complex members are affected, what is going wrong. It looks like even de novo mutations where you are heterozygous for the wild type protein is already sufficient to cause a disease state. Of course, these proteins are essential, so the question is, what can we learn from these disease situations to come a bit closer to understanding biology.
Since MOF is a histone acetyltransferase, maybe in the future we will be able to find compounds that are specifically able to reduce this activity or enhance this activity, because both MSL and NSL complex members are involved in not only syndromes, but also in diseases like cancer. In various disease states, the cells up- or down-regulate components of these complexes, so it will be really interesting to find ways to moderate their activity to be able to help and also go into the avenue of epigenetic therapy.
I think these are big challenges in front of us, but I think we can move forward in a very interesting manner by combining these aspects and learning from what the new things this complex is telling us as we move into different systems and not shy away from unexpected results.
Check out the full interview with Professor Asifa Akhtar on Active Motif’s Epigenetics Podcast to learn more about her views on dosage compensation in Drosophila and other organisms.