New & Noteworthy

Squeezing Out a Tiny Bubble of DNA

November 20, 2017

---

Have you ever tried to walk back to the lab with too many plates between your thumb and middle finger? And had them be so compressed that they shoot out of your hands, spilling all over the floor? Yeah, me too.

While this situation is bad for researchers, it turns out that the same sort of force may be helpful in cells by forcing the DNA open just a bit to get the ball rolling on transcribing a gene into messenger RNA (mRNA). The DNA is squeezed between two points so that a bubble of DNA pops open, leaving some single stranded DNA for RNA polymerase II (RNAP II) to get a hold of.

In a new study in Nature Structural & Molecular Biology, Tomko and coworkers show that Ssl2p, the double-stranded DNA translocase subunit of the basal transcription factor TFIIH, hydrolyzes dATP or ATP to force open around 5 or 6 base pairs of DNA. Once pried open, this small DNA bubble is expanded to 13 base pairs in the presence of NTP hydrolysis, presumably RNAP II beginning to transcribe the DNA. This stable open complex is now ready for promoter clearance and elongation.

These authors teased apart this mechanism using single-molecule magnetic-tweezers (henceforth referred to as tweezers). The idea is to stretch DNA out between two anchor points, add various components of the RNAP II machinery and various reagents, and to measure the changes in the stretched out DNA. Under the right conditions, these length changes directly reflect DNA unwinding.

Getting a gene read in a eukaryote like Saccharomyces cerevisiae is no easy task. A complicated mass of proteins called the preinitiation complex needs to form on the promoter first.

Tomko and coworkers loaded a subset of this complex, which included the TATA binding protein (TBP), TFIIB, TFIIF, TFIIH, and RNAP II, onto a piece of DNA with a single TATA box-containing followed by a strong initiation site. The authors arranged it so the 2.1 kilobases of DNA they used in their experiments was negatively supercoiled as is found in vivo.

In the first set of experiments, they added NTPs and dATP to their reactions. Most of the DNA did not show any changes as measured by the tweezers, but around 5% did.

There were two distinct populations of DNA in the subset that were active. Around 1/3 of the DNA showed  clear open and closed DNA transitions with the open DNA stretching over about 13 base pairs. The other 2/3 of the DNA showed longer-lived, smaller bubbles of around 5 or 6 base pairs of DNA. The authors interpreted the first as stable open complexes and/or elongation complexes and the second set as bubbles of DNA forced open by the preinitiation complex.

They repeated the experiment in the absence of NTPs and saw only the smaller bits of open DNA. And when they left out both NTPs and dATP, they saw no opened DNA at all.

So the small regions of opened DNA are dependent only on dATP hydrolysis making the Ssl2p subunit of TFIIH, a polypeptide known to hydrolyze dATP, the prime culprit for causing them. Which also makes sense, as Ssl2p is a DNA translocase that has been implicated in previous experiments in forcing the DNA to shorten between the preinitiation complex and downstream DNA. These experiments also showed that NTPs needed to be hydrolyzed to proceed to the next step, a stable open complex.

Tomko and coworkers next wanted to determine if these smaller bits of open DNA play a meaningful role in transcription. They tackled this question with a couple of different experiments that both rely on the idea that opening the 5-6 base pairs of DNA is a necessary first step on the way to transcription.

In the first experiment, the authors saw that these smaller regions of open DNA had longer lifetimes at lower NTP concentrations. Remember, NTP hydrolysis is required to proceed to the next step, the 13 bp, more stable open complex. By decreasing the concentration of NTPs, the authors slowed this step down, causing the preceding step of the smaller 5-6 base pairs of opened DNA to hang around longer.

In the next set of experiments, they engineered DNA bubbles into the DNA that varied between 3 and 12 base pairs and tested what size bubble would allow transcription to proceed in the absence of TFIIH, the basal or general transcription factor they implicated in forcing these DNA bubbles open. Previous work had shown that TFIIH is unnecessary if the DNA has a 12 base pair region that is already opened.

They found that while 3 base pairs of opened DNA was insufficient to allow transcription in the absence of TFIIH (and dATP), an open region of 4 base pairs was. This is consistent with the idea that the 5-6 base pairs of opened DNA they saw in their experiments are relevant and that they are caused by the action of TFIIH.

Reading a gene is nontrivial in eukaryotes like us, and our friend yeast. The DNA is squeezed by TFIIH within the preinitiation complex to open just enough of the DNA for RNAP II to get a toe hold and get transcription rolling. A much better result than when a similar force gets your stack of plates rolling on the lab floor.

Don Ho probably isn’t singing about tiny, Ssl2p-mediated DNA bubbles but they too are fine.

by Barry Starr, Ph.D., Director of Outreach Activities, Stanford Genetics

Categories: Research Spotlight

Tags: open complex, Preinitiation, RNA polymerase II, SSL2, TFIIH

Yeast Reveals a Vicious Circle Between Sugar and Cancer

November 10, 2017

---

The housing crisis that started in 2006 is a classic example of a vicious circle. As prices went down, some people couldn’t afford their mortgages and so were foreclosed upon. These foreclosed houses entered the market and caused prices to drop further which caused more foreclosures and so on. Only a Herculean effort by the government and the Federal Reserve was able to break this cycle.

Vicious circles can happen in biology too.

In a new study in Nature Communications, Peeters and coworkers have used our friend Saccharomyces cerevisiae to uncover one of these in cancer cells. And it turns out that this yeast is the perfect model organism for this study. (Funny how often that is true…)

Both S. cerevisiae and cancer tend to utilize their sugar through fermentation instead of respiration. This process in yeast is obviously great news for beer and wine drinkers. However, it is not such great news for people with cancer.

In the Warburg effect (named after the scientist who discovered it), the more aggressive a cancer is, the more sugar it ferments. This new study suggests that the consumption of sugar via fermentation and the aggressiveness of the cancer are related in a vicious circle.

Fermentation creates the byproduct fructose-1,6-bisphosphate (Fru1,6bisP) which, according to this study, activates the oncogene Ras which causes the cell to grow faster. As it grows faster, it ferments sugar faster creating more Fru1,6bisP which activates more Ras and so on. The cancer cells grow out of control until doctors apply some Herculean treatment to put a stop to it.

While yeast and cancer cells have a lot in common, wild type yeast isn’t quite up to cancer’s standards when it comes to cancer’s unbridled intake of glucose into the glycolysis pathway. These authors turned yeast a bit more cancer-like in this regard by deleting the TPS1 gene. Tps1p is a yeast hexokinase that tightly regulates yeast’s glucose intake into glycolysis.

Yeast cells without TPS1 deal poorly with glucose and need to be grown in galactose. When Peeters and coworkers added glucose to tps1Δ cells that had been previously grown in galactose, the cells activated Ras, a potent oncogene. This activation caused the cell to undergo apoptosis as assayed by cytochrome c release from the mitochondria, exposure of phosphadityl serine on the plasma membrane and generation of reactive oxygen species.

Deletion of hexokinase 2 eliminated this activation of Ras and suppressed the apoptosis. This suggests that perhaps some buildup of an intermediate metabolite in the glycolytic pathway might be to blame for the Ras activation.

The authors tested this hypothesis by removing the cell wall of wild type yeast cells and adding glycolysis metabolites to the resulting spheroplasts. They found that one metabolite, Fru1,6bisP, activated Ras. Supporting this finding, they found that deletion of both PFK1 and PFK2, two genes that encode phosphofructokinase 1, the enzyme responsible for making Fru1,6bisP, eliminated Ras activation in tps1Δ cells. Together, these data support the idea that Fru1,6bisP is the culprit behind Ras activation in yeast.

All well and good, but what about human cancer cells? Is something similar going on there? You have probably guessed that the answer is yes.

The authors first depleted the level of Fru1,6bisP in two different cell lines by starving them of glucose for 48 hours. They then added back glucose, which has been shown to increase the levels of Fru1,6bisP in cells, and measured the activation level of Ras and two of its downstream targets, MEK and ERK. All three were transiently activated in both HEK293T and Hela Kyoto cell lines. Looks like Fru1,6bisP activates Ras in cancer cells as well.

So this might explain the Warburg effect that I mentioned earlier. The more glucose a cancer cell uses during fermentation, the more Fru1,6bisP it generates. And the more Fru1,6bisP that is made, the more Ras gets activated prompting the cell to grow faster. Which of course results in more glucose getting fermented and so on.

I don’t have time to go into the work these authors did to try to uncover the mechanism by which Fru1,6bisP activates Ras, but suffice it to say that it does not appear to be a direct interaction with Ras. Instead, it appears to work at least partially by disrupting the interaction between Ras and one of its guanine nucleotide exchange factors, Cdc25p (or Sos1 in humans).

And the story is not yet complete. There is still a lot of work to do in pinning down the specifics of this vicious circle. Yeast will undoubtedly be instrumental in helping us work through the rest of the details.

by Barry Starr, Ph.D., Director of Outreach Activities, Stanford Genetics

Categories: Research Spotlight

Tags: cancer, fermentation, PFK1, PFK2, TPS1, Warburg Effect

Alliance of Genome Resources Opens Door with Version 1.0 Release

November 03, 2017

---

The Alliance of Genome Resources (the Alliance) announces the release of the Alliance of Genome Resources website 1.0 – providing unified access to comparative genetics and genomics data from the Alliance data resources (www.alliancegenome.org). The focus of the Alliance is to facilitate the use of these data towards better understanding of human biology and disease.

The Alliance brings together the efforts of the major National Institutes of Health (NIH) National Human Genome Research Institute (NHGRI)-funded Model Organism Database (MOD) groups, and the Gene Ontology (GO) Consortium, in a synergistic integration of expertly-curated information about the functioning of cellular systems.

The MODs were created in the early days of the Human Genome Project in support of the major experimental models for human biology. The MODs currently included in the Alliance are the Saccharomyces Genome Database (SGD), FlyBase, WormBase, Mouse Genome Database (MGD), Rat Genome Database (RGD), and Zebrafish Information Network (ZFIN).  In addition, the Alliance includes the Gene Ontology (GO) Consortium. Now these groups will merge key activities and data representations, coordinating data retrieval and analysis, within a comparative perspective. Other MODs and related resources will be added to the Alliance going forward.

As part of this initial release, Alliance working groups have focused on the ability to easily access pages that summarize details of genes and diseases, with extensive representation of orthology data, and with access to multi-track JBrowse capabilities primarily for visualization of sequence data. Users recover gene details, functional information, and disease associations within a comparative perspective. As the integration of the MOD and GO teams progress with inclusion of additional data, the vision going forward includes the incorporation of other model organism information resources and other bioinformatic nodes within a common data platform, facilitating data recovery, analysis, and integration.

Categories: News and Views

Tags: Alliance of Genome Resources

Something from Nothing: A New, Structured Protein from Noncoding DNA

November 01, 2017

---

In the Rick and Morty episode Mortynight Run, a gaseous life form pulls off what alchemists had been trying to do for centuries. It literally creates gold out of thin air!

Our ever faithful friend, Saccharomyces cerevisiae can’t do that (at least not yet). But what it has done is create a protein, Bsc4p, out of noncoding intergenic DNA. And while not gold, it is a fully functioning protein.

In a new study, Bungard and coworkers show that this recently evolved gene, BSC4, encodes a protein that can fold into at least a partially defined structure. This matters as there was some debate about whether newly generated proteins could attain a defined structure or if they would remain as intrinsically disordered proteins (IDPs). A reasonable debate, given how rare structure is among amino acid sequences and how plentiful IDPs are in a cell.

Bsc4p is a great protein to study in this regard as there is very strong evidence that it has evolved relatively recently in S. cerevisiae, but not in other closely related species. And it definitely does work in a S. cerevisiae cell. While not essential, some genetic studies (including this one) indicate that it plays an important role in DNA damage repair pathways.

Here is an example from S. paradoxus of the noncoding sequence that the S .cerevisiae BSC4 gene almost certainly sprang from:

gtg TCT GTA ATT CTA CGG AAA AGT AAA CAA AAA AAC TGT AAT TGC ATA ACG AGC AAT TTA TAT ACA ATA CAC ATA GAA AGA CTT TCG CTC tga TGT CCG AAC TGC CAT TGT CAT TGG AGA AAA TCC TTA TGT GGA GTG GAG TTC CCT GCA GGT TAT TTT CAG AGA AAA CGT GGT TAC AAA AAG GGA CCA GAT TCG CCC tag CTT ACA ACT CGC TTG AAT CAT CTT TAT GCC AGA CCT TTC AAC GCC GCG ACC CCA AAA ACA taa ATG CTG AGT CAC CAT GGT GCT GGG CGC TGT CGC TGT CGC GCT GTT CCT TTC CGA GAA AAG CAC GGC AAC AAC AAC AAC AGT CCA TAT GAC CAA AAA AAA AAT AAC CGC AAA TGG CAG tga AAT GCA ATT ATC ATT GTA TAC GA?

In order for this sequence to become a gene coding a protein, at a minimum the first lowercase, dark orange codon needs to be mutated to ATG, a start codon, and the rest of the lowercase, dark orange codons need to be mutated away from being stop codons.

This seems to be part of what happened in S. cerevisiae:

ATG TCT ATT GTG CTA CGG AAG AGT AAC AAA AAA AAC AAA AAC TGC ATA ACA AGC AAG TTT TAT ACA ATA CAC ATT ATA AAA ATT TCT ACT CCG GTG TTC CGA GCT CCC ATT GCC ATT GGA GAA AGC CCT TAT GTG GAG TGG AGC TGC CTA CAG GTT GTT TTC AGG AAA GAC ATG GTT ACA AAA AAG ACG ACA TTC GCC CAA CTT ATC ACT CGC TTG AAC CAC TTT TTA TGC CAA GCC CTT AAA CGC CGC GAC TCA AAA ACA TAC ATA CTG TGC CGC ACG GCA GTT TTT GGC GCT ATG ACA CCC TTT TCC CCA AGA AAA TCG CAT ATT AAC AAC AAA TTA CCC ATG CAA CCC AGG AAA AAA AAA ATA GTC ATT ATA TAC GTA GTG CGC TTT CAT TGA

Through a few small changes, we now have a 131 amino acid polypeptide where before we had some noncoding DNA between LYP1 and ALP1.

Bungard and coworkers use a variety of techniques to show that this newly evolved protein has structure. Not as much structure as many proteins that have been around longer, but more than many of those IDPs.

Consistent with protein structure, Bsc4p forms compact oligomers under native conditions that are partially resistant to proteolysis, has a far UV circular dichroism (CD) spectra consistent with beta sheets, has a buried tryptophan, Trp47, that becomes solvent accessible under denaturing conditions, as measured by tryptophan fluorescence, and has a near UV spectra consistent with a hydrophobic core. However, they found no evidence of any significant interactions between the secondary structures to form a single three dimensional shape, in the protein.  In other words, no evidence of a tertiary structure.

And that wasn’t the only sign that Bsc4p wasn’t a mature, fully structured protein. For example, that near UV CD that showed a hydrophobic core, was weak in intensity, which is consistent with at least “partially molten character.” And Bsc4p bound certain dyes: Congo red, Thioflavin T, and ANS, in a way consistent with some molten globule and/or amyloid character.

So we have a bit of a mixed bag with Bsc4p. One way to think about it is as a young protein still developing its ultimate three dimensional structure. Or, it could be that for the job it does, this is all the structure it needs.

In any event, it is definitely a newly evolved protein with at least some structure which shows that this can indeed happen. Sometimes Mother Nature can make structured proteins from noncoding DNA. Like that gaseous being on Rick and Morty, producing gold out of thin air.

by Barry Starr, Ph.D., Director of Outreach Activities, Stanford Department of Genetics

Categories: Research Spotlight

Tags: BSC4, newly evolved gene, noncoding DNA, protein structure

Transcription Factors: Kings of the Genome Jungle

October 20, 2017

---

Imagine that Jane is in trouble on the edge of the jungle. She needs to be saved soon or she will be sent back to Europe (which she does not want).

Tarzan knows he can’t get there in time by running along the jungle floor. But of course he has a trick up his sleeve—swinging from vine to vine!

He gets there in time, fends off her would-be kidnappers, and saves the day. All because he transferred from one strand of vine to another instead of “sliding” along the jungle floor.

A new study by Wollman and coworkers shows that at least two transcription factors in Saccharomyces cerevisiae, Mig1p and Msn2p, seem to share a lot in common with Tarzan. In a process termed intersegment transfer, they get to where they need to in the genome by “swinging” from DNA segment to DNA segment instead of just sliding along the DNA.

And transcription factors like these need to get to the right place in time to save the cell from outside threats. Just like Tarzan had to swing from vine to vine to save Jane.

This sort of approach would not necessarily work well with just a single transcription factor with a single DNA binding domain. It would sort of be like a one-armed Tarzan—it is hard to take advantage of swinging from vine to vine without at least two arms!

The authors argue that Mig1p gains its “extra arms” through joining together into a cluster. Now each Mig1p can bind DNA and drag the other transcription factors with them. A neat solution to the one arm problem.

Their basic approach is to use fluorescence microscopy to follow a GFP-Mig1p fusion protein in a single cell, something that has only become possible recently. What they found was that there were two populations of transcription factors—a diffuse set of smaller molecules and distinct, larger clusters made up of multiple Mig1p’s.

The clusters appeared to be the ones doing the work in the nucleus. Kinetic studies showed that they stayed in one place in the nucleus for over 100 seconds which is consistent with the clusters and not the monomers being bound to the DNA.

OK so this transcription factor tends to clump up into clusters and it looks like these clusters are the ones regulating gene expression. They also showed that a second transcription factor, Msn2p, did the same thing.

The authors next set out to see if this approach made sense for genetic regulation by running simulations of Mig1p finding its sites in the nucleus as either monomers or as clusters. It made sense to form clusters.

A lot of previous work has been done in S. cerevisiae in terms of the three dimensional map of the genome and where Mig1p DNA binding sites were located in this mesh of DNA. And in the course of their studies, Wollman and coworkers were able to estimate how many Mig1p molecules were in yeast cells and how many were in clusters.

They now had all the information they needed to run their simulations. When they crunched the numbers, they found that clusters fit their data much better than monomers (R² = 0.75 vs. R² < 0).

The final step was to work out what part of Mig1p was involved in forming the clusters. To do this, the authors compared Mig1p and Msn2p, the second transcription factor they studied that also formed into clusters, and looked for structural regions they might have in common.

What they found was both proteins had a highly disordered region. For Mig1p, it was at the C-terminus and for Msn2p it was at the N-terminus.

The hypothesis is that these disordered regions, which are both at the opposite end of the protein from the DNA binding domain, interact and form ordered structures that enable clusters to form.  Wollman and coworkers used circular dichroism to show that when Mig1p was put in conditions that favor cluster formation, there was a transition consistent with unstructured protein becoming structured.

What we seem to have is a cluster of transcription factors connected in the middle with their DNA binding domains pointing out. This rolling cluster can more easily hop from DNA strand to DNA strand to find the right spots to bind.

Without this mechanism, Mig1p couldn’t get to where it needs to in time. It is as if Tarzan had 6-9 arms circling his body so he could get to Jane even more quickly.

With the wide range of tools available and our deep understanding of how yeast works and how a yeast cell is organized, our trusted ally S. cerevisiae again teaches us something fundamental about how our biology works.

Mig1p may be able to swing like Tarzan but it can’t yell like him. Or like Carol Burnett!

by Barry Starr, Ph.D., Director of Outreach Activities, Stanford Department of Genetics

Categories: Research Spotlight

Tags: free diffusion, intersegment transfer, MIG1, Saccharomyces cerevisiae, transcription factor

Divide and Prosper

October 09, 2017

---

A Swiss Army knife is a marvelous thing. It is a bunch of useful tools all wrapped up in something small enough to put in your pocket.

Need to cut a string? It has the scissors to do that. How about cutting that little piece of wood? Yup, the tiny saw can do that. It has lots of tools that can do lots of things.

Now of course, it isn’t perfect. You would not want to use those scissors for cutting fabric or to use that saw to cut a 2 by 4. And there really isn’t any easy way to make a better pair of scissors or a better saw in the Swiss Army knife without wrecking the other tools in it.

If you make a better pair of scissors, you’ll have a worse saw. And vice versa.

That is why if you want to improve the saw, you need to isolate it away from the other tools. This holds true for the scissors and yes, even the knife.

Multifunctional proteins can have similar constraints. Tweak one function and it can mess up the other functions. A better saw makes a worse pair of scissors.

Nature doesn’t usually improve on a protein’s function by pulling out the part of the protein that does a specific job and making it better. No, instead, she duplicates the gene with the instructions for the protein. Now Nature can optimize each function of each gene copy—both improved functions can still happen albeit with separate proteins.

It is like having two Swiss Army knives and improving the saw on one and scissors on the other. You still have scissors even when your improvements on your saw have wrecked the attached scissors. They are just on the other knife.

In a new study in GENETICS, Hanner and Rusche explored this idea of “subfunctionalization” using the Saccharomyces cerevisiae proteins Sir3p and Orc1p. The genes that encode these proteins are thought to have arisen from a single ancestral gene after a whole-genome duplication around 100 million years ago.

The ORC1 gene is highly conserved in eukaryotes and serves an important role in DNA replication. In addition to this role, in S. cerevisiae, it also works to silence the mating type loci by recruiting the SIR complex to either HMRa or HMRα. Sir3p is also involved in silencing genes including HMRa or HMRα, but it does its job by recognizing deacetylated histones.

The basic strategy of this study is to compare Sir3p from S. cerevisiae with KlOrc1p, an Orc1p homolog from the yeast Kluyveromyces lactis that is involved in both DNA replication and gene silencing.

Since K. lactis did not undergo the whole genome duplication that S. cerevisiae did, the idea is that KlOrc1p represents the multifunctional ancestral protein (or at least what it has evolved into over the past 100 million years or so). Sir3p is the saw while KlOrc1p is the saw as part of the Swiss Army knife.

A MATa strain of S. cerevisiae lacking Sir3p cannot mate because its HMRα locus is not repressed. And a MATα strain of S. cerevisiae lacking Sir3p cannot mate because its HMRa locus is not repressed.

In the first experiment, Hanner and Rusche wanted to see if Orc1p from yeasts that did not undergo the whole genome duplication could rescue strains of S. cerevisiae lacking Sir3p. In other words, do these non-duplicated proteins have Sir3p functions or did Sir3p evolve new functions once freed from its DNA replication constraints?

These authors showed that KlOrc1p could not rescue mating in a strain of S. cerevisiae lacking Sir3p (but still containing its own Orc1p). They took this even further and showed that Orc1p from two other non-duplicated yeasts, Lachancea kluyverii and Zygosaccharomyces rouxii, could not rescue the S. cerevisiae strain either. Sir3p in S. cerevisiae has functions that none of these “ancestral” Orc1p’s have.

The next step was to create chimeras between Sir3p and KlOrc1p to identify the parts of Sir3p that developed new functions.  Both proteins consist of four broad subdomains: AAA+, winged helix, N-terminal bromo-adjacent homology (BAH), and a rapidly evolving linker.

The researchers swapped domains between the two proteins and found that a Sir3p with an AAA+ subdomain from the KlOrc1p did not allow a S. cerevisiae strain lacking Sir3p to successfully mate. So Sir3p has gained some new function in AAA+ that the ancestral protein lacked. The authors were able to narrow down this region a bit more—the AAA+ subdomain lacking its first two alpha helices worked too.

Hanner and Rusche used chromatin immunoprecipitation assays to show that this chimeric protein did not make it to the HMRa or HMRα loci or to other repressed genes. They also looked at mRNA levels and showed that genes that should have been repressed weren’t. This Sir3p with the AAA+ subdomain of KlOrc1p clearly cannot do the job of Sir3p.

And it isn’t because of the loss ATPase function of the AAA+ domain in Sir3p. While Sir3p has an ATPase domain, it does not have any discernible ATPase activity. When Hanner and Rusche replaced the KlOrc1p ATPase domain with the ATPase domain of Sir3p, this protein still couldn’t rescue the S. cerevisiae strain deleted for SIR3. The ATPase activity of KlOrc1p does not explain its inability to rescue the strain.

It is known that the AAA+ domain of Sir3p interacts with Sir4p. Another possible explanation for the failure of KlOrc1p to complement the loss of Sir3p may be that KlOrc1p has over the last 100 million or so years lost the ability to interact with the S. cerevisiae version of Sir4p. This does not seem to be the reason as adding in the K. lactis version of Sir4p, KlSir4p, to the S. cerevisiae strain deleted for Sir3p, did not allow the strain to mate. Inability to interact with Sir4p does not seem to explain the failure of KlOrc1p to rescue the strain.

The authors hypothesize that once freed of the constraints of the multifunctional Orc1p, the AAA+ region of Sir3p gained the ability to interact with deacetylated histones. This will need to be tested in future studies.

There you have it. Once freed of having to focus on two things, Sir3p was able to evolve new functions like recognizing deacetylated histones. Like a saw being freed from a Swiss Army knife so that it can sing.

Don’t try this with the saw from a Swiss Army knife.

by Barry Starr, Ph.D., Director of Outreach Activities, Stanford Department of Genetics

Categories: Research Spotlight

Tags: gene duplication, ORC1, SIR3, subfunctionalization

Sensing Disturbances in the Membrane

September 25, 2017

---

In the original Star Wars movie, Darth Vader knows Obi-Wan is aboard the Death Star because he feels a “disturbance in the force.” The highly conserved protein Ire1p does something similar with disturbances in the membrane of the endoplasmic reticulum (ER) in Saccharomyces cerevisiae.

Ire1p isn’t sensing the presence of a Jedi master of course. Instead, it is sensing when the membrane is under stress through either the buildup of unfolded proteins in the ER lumen or membrane and/or through changes in the membrane itself.

And unlike Darth Vader, who upon sensing the presence of Obi-Wan seeks him out and destroys his physical form with a light saber, Ire1p instead forms dimers and larger oligomers when it senses a disturbance. This activates the unfolded protein response (UPR), which results in the activation of a bunch of genes so that the cell can start to deal with its unfolded proteins and perturbed membrane mess. (And no, unfortunately, the cell does not do this with tiny light sabers!)

Previous work had established the part of Ire1p that responds to unfolded proteins. But scientists had yet to nail down what part of the protein senses membrane problems. Until now that is.

In a new study in Molecular Cell, Halbleib and coworkers have not only worked out what part of Ire1p senses disturbances in the ER membrane in our old friend S. cerevisiae, but they may also have figured out how the protein does it.

At first blush, it would seem to make sense that the single, highly conserved transmembrane helix (TMH) of Ire1p would be important for sensing membrane issues. After all, it is in intimate contact with the membrane!

Previous mutation studies showed that the TMH alone probably isn’t the key player. Halbleib and coworkers confirmed this result by replacing all thirteen amino acids of the TMH in Ire1p with leucines and showing that this mutated protein was still able to respond to membrane problems caused by either too much dithiothreitol (DTT) or too little inositol.

A closer look at the protein revealed a conserved region of Ire1p that overlaps with the TMH and is predicted to form an amphipathic helix (AH). CD spectroscopy showed that an oligopeptide of identical sequence to this region formed an alpha helix in micelles but not in aqueous solution—it needs a membrane environment to become an alpha helix.

These authors used mutational analysis to show that the hydrophobic side of this AH is the part of the Ire1p involved in sensing problems in the ER membrane. For example, two specific mutations on the hydrophobic side, F531R and V535R, negatively affected a yeast cell’s ability to respond to disturbances in the ER membrane. Yeast containing Ire1p with a mutation on the hydrophilic side, R537E, however, had no such problem and responded fine. Ire1p function was also compromised when the researchers extended their leucine replacement of the transmembrane domain by three more residues into the AH.

So Halbleib and coworkers have seemingly identified the key part of Ire1p that can sense changes in the membrane. Now, they set out to examine how this part of the protein actually does the sensing.

Right away the authors could see that the mutants that disrupted the ability of Ire1p to sense membrane disturbances also did not oligomerize under membrane-perturbing conditions in a living yeast cell. This makes sense as oligomerization is required to marshal the cellular forces needed to deal with these ER problems.

They confirmed that oligomerization could be affected by membrane composition with some in vitro experiments. They first made what they called a minimal sensor construct—the amphipathic helix and the transmembrane helix attached to the maltose binding protein. They reconstituted this sensor into eight different liposomes with varying lipid compositions.

Using continuous-wave paramagnetic resonance spectroscopy they were able to show that the ability of their minimal sensor to oligomerize depended upon the membrane composition. The more similar the composition was to a stressed membrane, the more likely their sensor was to oligomerize. This is exactly what we would expect if the AH was the key player in sensing membrane problems.

So certain mutations and membrane situations can affect the oligomerization state. But this still doesn’t really get at why Ire1 proteins oligomerize in perturbed membranes.

To try to figure this out, they turned to molecular dynamic simulations.

They found that the AH tended to compress the lipid bilayer and that the more stressed the membrane, the more energy this compression costs. They also found that if two Ire1p molecules happened to bump into each other and dimerize, that the energy costs went down.

In other words, stressed membranes tend to force Ire1p molecules to oligomerize to reduce the energy cost of being in the membrane. And of course oligomerization triggers the unfolded protein response (UPR)!

Ire1p senses disturbances in the membrane because the energy cost of remaining a monomer in stressed membranes is too high. Much more plausible than the midi-chlorians that supposedly let Darth Vader sense disturbances in the force!

The unfolded protein response (UPR) is involved in many diseases including cancer and diabetes. 

by Barry Starr, Ph.D., Director of Outreach Activities, Stanford Genetics

Categories: Research Spotlight

Tags: IRE1, membrane sensing, UPR

Cheerio Containment

September 11, 2017

---

In the store the other day I saw a kid in a stroller covered in Cheerios, an opened, near empty Ziploc-type plastic bag next to him. It looked like one of those cheap, knock off brands that are much harder to close properly.

Other than keeping the Cheerios out of reach, his parents had two options here. The first was to use two bags, putting the cheaper re-sealable bag into another one. The odds are much better that at least one of them would be closed properly.

The second option was to get a re-sealable bag with a better zipper, like an actual Ziploc. With a better baggie, the parents had a much better chance of making sure that the bag was sealed.

In a new study out in GENETICS, Lawrence and coworkers show that yeast cells go for the second approach when it comes to the histone H1 and gene expression regulation. Rather than regulating gene expression by controlling the number of molecules of H1 in a cell, the yeast instead regulates transcription by controlling how well H1 can bind DNA. And it looks like the strength of binding is influenced by the acetylation of H1’s more famous histone cousins—the histones found in the nucleosome.

Almost everyone has heard of the histones that make up the nucleosome, that structure of DNA wrapped around a core of eight histone proteins. H1 is a bit less famous. It binds the linker DNA, or the DNA between some of the nucleosomes.

When bound, H1 tends to repress nearby genes. It acts like a clamp that keeps the nucleosome in place.

This gene repression is probably why our friend Saccharomyces cerevisiae has only one H1 per 4-37 nucleosomes as compared to the one H1 per nucleosome in vertebrates. After all, S. cerevisiae’s genome has a much higher gene density. If it had one H1 for every nucleosome it might not have any genes turned on!

You may have noticed that scientists don’t have a good handle on an exact number of H1 molecules in a yeast cell. One of the first things these authors did was a careful study to try to nail down just what that number is. They came up with 18.9 +/- 1 nucleosomes/H1.

The next step was to figure out how a yeast cell controls H1 to regulate gene expression. There are two general ideas about how a cell might go about this.

In the first, the cell controls the amount of H1 it makes. The more H1, the more genes get repressed.

In the second, H1 is controlled by the acetylation of the core histone proteins. The acetylation loosens the core histone’s grip on the DNA making it harder for H1 to bind.

It would be very hard to do experiments in most beasts to try to tell which model is correct. For example, mammals have 11 genes for H1 proteins. This makes changing the overall amount of H1 nontrivial to say the least.

This is where the ultimate model organism, S. cerevisiae, can come in and save the day yet again (#APOYG). Since this yeast has only one H1 gene, HHO1, it is much simpler to tweak the amount of H1 protein in a cell.

Lawrence and coworkers put HHO1 under the control of the GAL promoter and found that in the presence of galactose, the extra H1 had a severe effect on growth. These yeast were unhappy with around three times as much H1 protein, even though there was not much of an effect on chromosomal structure as measured by micrococcal nuclease assays.

The researchers next looked for genes which when deleted, rescued the growth defect of yeast making too much H1. They did this by transforming their GAL-HHO1 plasmid into the collection of ~4700 nonessential gene knockout strains.

Histone deacetylases (HDACs) were the most interesting class of genes they found that could rescue the growth defect. By deleting an HDAC gene, histone acetylation increases (because the deacetylation decreases), implying that H1 can be regulated by the acetylation status of histones. The loss of the HDAC increased the amount of acetylation, which made it harder for H1 to bind, and so repress transcription.

They confirmed this finding in a couple of ways with the most interesting one using a couple of different mutants of histone H3. One mutant had its acetylation sites in the H3 tail changed to glutamines, which mimics a histone in a perpetual state of acetylation. This mimic of an acetylated H3 tail partially rescued the growth defect of too much H1.

So it looks like the effect of H1 can be regulated by the acetylation status of its more famous histone kin. More acetylation results in a looser grip, which makes for more transcription. Like a cheap plastic baggie with a zipper with a looser grip making Cheerios more likely to spill, H1’s looser grip due to more acetylation results in more transcription and more transcripts being let loose into the cell.

by Barry Starr, Ph.D., Director of Outreach Activities, Stanford Genetics

Categories: Research Spotlight

Tags: gene regulation, H1, HHO1, histones, nucleosome

That Sweet Spot of Mistranslation

August 03, 2017

---

Anyone at a party knows that a little alcohol can make you charming, but a lot can doom any relationship from blossoming (just listen to Drunk Uncle!). In fact, too much can destroy a party! You need to have enough alcohol to curb social inhibitions, but not so much you overwhelm them.

According to a new study out in GENETICS by Berg and coworkers, something similar sometimes seems to be true when the genetic code evolves. A variety of beasts, including the yeast Candida albicans, have slightly different genetic codes—one or a few codons code for a different amino acid than usual. This is often the result of a mutated tRNA, the molecule that carries the right amino acid to the right codon.

These authors found that mutating only the anticodon, the part of the tRNA that recognizes the codon, is not a great way to head down the path to a new genetic code. That mutated tRNA leads to too many of one amino acid being replaced with another. Like the boorish drunk who kills the party because he has had too much to drink, the cell is overwhelmed with too many of the wrong amino acids scattered across its proteins and dies.

What Berg and coworkers found was that having fewer of these tRNAs with a mutated anticodon allowed for a tolerable level of amino acid substitutions. This means that this tRNA can hang around until it is helpful, like when it can suppress a new mutation of a key amino acid in a key protein. 

The authors dubbed these low level mutated tRNAs as “phenotypically ambivalent intermediate tRNAs”—tRNAs that are in the process of changing the genetic code for at least one codon. It may be that the variants of the genetic code found in nature arose this way.

They started out with a strain of Saccharomyces cerevisiae with a mutation that inserted a proline in the wrong place of the TTI2 protein.  This strain does extremely poorly in the presence of 5% ethanol.

The authors then tried to create and/or isolate suppressor mutations in a serine tRNA that could allow the strain to grow in the presence of 5% alcohol. The idea is that this tRNA would now carry a serine to that troublesome proline codon.

They started off by changing the anticodon of a serine tRNA to UGG. Now, the cell would put a serine in at CCA proline codons.

When they transformed a plasmid carrying this tRNA into their yeast strain, they got very few colonies. This obnoxious tRNA overwhelmed the cell by changing too many prolines to serines. It ruined the party!

 They next set out to find a way to bring this bad boy under control. They mutated the serine tRNA with the proline anticodon using UV mutagenesis and found four mutants that allowed this yeast strain to grow in 5% ethanol.

Each mutant tRNA had a single mutation: G9A, A20bG, C40T, and G26A. Berg and coworkers set out to figure out why the cells now tolerated the mistranslation they couldn’t handle before.

What they found was that at least for two of them, G9A and G26A, the cells dealt better with the mutated serine tRNA because there was less of them around. The toxic drunk had become the tipsy charmer!

Well, maybe not quite charming, but at least something that could be dealt with. Both mutated tRNAs affected cell growth in the absence of ethanol, with the G26A version having the more severe effect—a reduction in growth by 70%.

Most likely there was less of the G26A variant because it was a victim of the rapid tRNA decay (RTD) pathway. The G26A variant affected the growth rate much less in the absence of alcohol in a strain deleted for MET22, a key gene in the RTD pathway.

By looking at the crystal structure of a serine tRNA in complex with its aminoacyl tRNA synthetase from Thermus thermophilus, Berg and coworkers predicted that the G9A mutation should result in a poorly folded tRNA. They found this was indeed the case when they compared the melting curves of the G9A mutant and the tRNA lacking the G9A mutation.

So what we have are some tolerable, but by no means benign mutations. For example, the G26A is quickly selected against in the absence of ethanol.

This makes it hard to imagine how these sorts of mutations might arise and one day permanently alter a genetic code. The key to understanding how this might happen is a set of experiments Berg and coworkers did that showed that both G26A and G9A have little or no effect on cell growth in the absence of a mutated anticodon. In other words, tRNAs can exist in a poised state, ready to easily adapt with a single change to the anticodon if need be.

And it turns out that poised tRNAs may exist in the real world. For example, human tRNAs have a lot of variation. Perhaps these are around to one day save a cell with a mutation that would normally be deadly.

As this work (and real life) shows, too much of a good thing can be bad. This is true of alcohol (remember high school or college?) and true of some mutant tRNAs. Yeast can teach us about the tRNAs, the rest we need to learn on our own. #APOYG

by Barry Starr, Ph.D., Director of Outreach Activities, Stanford Genetics

Categories: Research Spotlight

Tags: MET22, mistranslation, tRNA, TTI2

Now Yeast Even Finds Fungal Pathogens!

July 27, 2017

---

Some products have lots and lots of uses. Sometimes they are designed that way like the hypothetical Shimmer, a floor wax and dessert topping from Saturday Night Live. And sometimes they can be used in more situations than they were originally designed for like Windex in the movie My Big Fat Greek Wedding, or the most versatile thing out there in real life, duct tape. (Click here and here for some fun duct tape uses).

Wait, did I say duct tape was the most versatile thing out there? Maybe I spoke too soon.

Our friend Saccharomyces cerevisiae is in contention with duct tape for that title. Of course it is used to make bread, beer and wine, but it has lots of other uses too. Just a very few include using it to: make antimalarial drugs, understand how our cells and a surprising variety of human diseases work, understand how evolution can happen, understand how cancer forms and progresses, and make ambergris (whale puke). In the near future, it may even help deal with climate change by making biofuels out of agricultural waste, and pain management by making opioids.

And now a new study by Ostrov and coworkers adds another use—the simple and inexpensive detection of fungal pathogens. This is a big deal because if it works the way the authors think it will, this new use could one day save the lives of a significant fraction of the two million or so people who die from fungal pathogens each year.

In a nut shell, these authors have engineered S. cerevisiae to be attracted to different kinds of fungus by monkeying with its pheromone detection system.

S. cerevisiae comes in two mating types, a and α. The a-type cells make the secreted mating pheromone “a-factor” and STE2, a receptor that responds to “α-factor.” And α-type cells make the secreted mating pheromone “α-factor” and a receptor, STE3, that responds to the “a-factor.” So, thea-type cell makes something that makes it attractive to α-type cells and vice versa.

The end result of the factor/receptor interaction is that a number of genes are regulated so that the two haploid cells, a-type and α-type, can merge to become an a/α diploid. Fertilization yeast-style.

What these authors have done is to switch out the STE2 receptor from S. cerevisiae with the receptors from other fungal species in a-type cells so that the newly engineered cell responds to the presence of the new fungal species’ pheromone. For example, they replaced S. cerevisiae STE2 with the STE2 from Candida albicans so that the newly engineered cell now responds to pheromones from C. albicans.

Now of course that doesn’t make much of a sensor! No, they also need a readout to know that the engineered yeast has detected the presence of C. albicans.

They started out using a fluorescent reporter to see whether their engineered system responded to C. albicans mating pheromone. The cells fluoresced when the C. albicans mating pheromone was around at low levels but not with any of nine other fungal mating pheromones. So the system is definitely working—it can specifically identify C. albicans.

But to make it more useful in the developing countries where fungal pathogens are the biggest problem, they wanted to create a system with a more visible readout. For this they turned a bright red pigment called lycopene.    

They needed to add three genes from Erwinia herbicola to get yeast to make lycopene: crtE, crtB, and crtI. They placed the first two under constitutive promoters (pADH1 for crtE and pTEF1 for crtB) and the third gene, crtI, under the control of the pheromone-inducible promoter from FUS1. So, lycopene is not made unless the appropriate pheromone is around to turn on the crtI gene.

This new system worked as well as the fluorescent one.

For the next step, they replaced the S. cerevisiae STE2 with STE2 genes from nine other fungal species involved in human disease and/or food or agricultural spoilage and showed that their system worked for all of them. These new pathogens included: Candida glabrata, Histoplasma capsulatum, Lodderomyces elongisporus, Botrytis cinerea, Fusarium graminearum, Magnaporthe oryzae, Zygosaccharomyces bailii, and Zygosaccharomyces rouxii.

And they didn’t stick to just the well-characterized fungal species either. They were also able to create a system that worked for Paracoccidioides brasiliensis, a less-well studied fungus that causes paracoccidioidomycosis (PCM), a disease endemic to Latin America.

It may be that just understanding a fungal species’ mating system is enough to create this sort of biosensor. More species will have to be tested to see just how universal the simple swapping out of a single gene will be.

In their final step, Ostrov and coworkers tested whether they could create a simple, inexpensive dipstick test by spotting their engineered strains onto filter paper. For the initial test, they focused on two of their engineered strains: the one that detects the presence of C. albicans and the one that detects the presence of P. brasiliensis. They found that the system worked by simply putting the yeast-spotted filter paper into soil, urine, serum, and blood that had been supplemented with synthetic pheromone.

The yeast with C. albicans STE2 turned red with C. albicans pheromone, but not with P. brasiliensis pheromone and vice versa. And to increase its utility further, the authors found that the filter paper stayed functional even after being stored for 38 weeks at room temperature.

The impact of fungal pathogens on human health is understudied and underappreciated despite the toll they take on people’s lives, especially in the developing world. For example, as one recent review points out, “…at least as many, if not more, people die from the top 10 invasive fungal diseases than from tuberculosis or malaria.”

The basic research done on S. cerevisiae has allowed these scientists to devise an ingenious way to identify fungal pathogens. That magical combination of basic research and the most versatile eukaryote out there, S. cerevisiae, is poised to help humanity yet again.  #APOYG

by Barry Starr, Ph.D., Director of Outreach Activities, Stanford Genetics

Categories: Research Spotlight

Tags: crtB, crtE, crtI, fungal pathogens, MATA, MATALPHA, mating, STE2, STE3

Next