12: PARPs, Sigma-factors, protein splicing, RNAi in flies, protein highlighting

It's been a while since I've updated, and I've probably forgotten most of the wealth of information I had learned in the past few weeks. Granted, I gave up science for almost a week while I was in Ireland.

So, for one class, I read quite a few papers/reviews about PARPs, or poly-ADP-ribose polymerases. I'm supposed to write a proposal on the topic, which was very difficult, because there's so much speculation:

These enzymes basically tag their targets (which are sometimes even themselves!) with strings of ADP-ribose, in irregular linear and branched ways. The tagging is thought to disrupt protein-protein interactions, but is often transient, as there's an accompanying PARG, or glycohydrase that gets rid of the modificaiton. The best studied PARP is PARP1, which is somehow involved in DNA-damage response. It's also part of a new kind of programed cell death, I think they called it parthanatos (don't quite remember). I got really interested in PARP4 or vPARP, because it's part of this giant organelle-like structure called a vault.

So vaults are ribonucleoprotein complexes about 3x larger than the ribosome, and present in many types of eukaryotic cells at >10,000 copies. They were only discovered in the late 80's, and are composed of major vault protein (MVP), vPARP, TEP-1 (telomerase-like protein), and vRNA. They form this large canister-like structure, but they don't know what it does - many implicate it in drug resistance, transport of proteins between nucleus and cytoplasm, response to cancer-causing compounds, and formation of the nuclear pore complex. Anyway, vPARP isn't essential for forming the structure, and knockout mice are just a bit more sensitive to oncogenic agents, but I want to get an idea about whether the poly-ADP-ribosylation affects formation of these structures. One group even observed that vPARP can form filament-like structures that recruit MVP, so I want to do some in vitro structural work to figure out what the heck is going on!

Recent topics in Chemical Biology class:
A genome-wide transgenic RNAi library for conditional gene inactivation in Drosophila
Dietzl, et al. Nature 2007

This collaboration created a GENOME-WIDE RNAi library for flies...seemed pretty useful if you're into that sort of research. They had a pretty easy system for targetting tissue-specific expression, and although they threw around a large number of percentages and controls, it seemed like it was a pretty good tool.

Development of a Tandem Protein Trans-Splicing System Based on Native and Engineered Split Inteins
Shi, and someone else JACS 2005

So, protein-splicing is pretty cool, and I think Tom Muir (whom I interviewed with at Rockefeller) was one of the big proponents of it. Basically, it works like DNA splicing, except with amino acids - the intein is the removeable element, and it catalyzes it's own removal. This paper showed how to put two different inteins together, and provided a way to do FRET within a single protein.

Regulated Fast Nucleocytoplasmic Shuttling Observed by Reversible Protein Highlighting
Ando et al. Science 2004

I thought this was really NEATO! They engineered this fluorescent protein from some sea creature to be able to be reversibly turned on and off, in a FRAP-type way. They then used this to observe the kinetics of protein transport back and forth across the nuclear membrane. They called the protein "Dronpa - after dron, a ninja term for vanishing, and pa, which stands for photoactivation." Gotta love scientists - I don't think I'll ever forget the name of this protein.

We also talked about evolving RNAs to catalyze enzymatic reactions, we read this pointless paper:

Self-Sustained Replication of an RNA Enzyme
Lincoln, et al. Science 2009

where they engineered an RNA ligase to make copies of itself in solution indefinitely - only one problem - they needed to throw in oligonucleotide substrates, so I thought this was worse than the example we talked about in class, where they engineered an RNA to ligate itself and replicate in vitro at least using reverse transcriptase. These experiments are trying to get closer to supporting the RNA world hypothesis, but I would really dislike working on this type of research.

In Micro200, we talked about toxin-antitoxin systems in bacteria, which are a really cool way to force bacteria to hold on to specific genetic sequences. The way most of them work: there are two genes near each other, one encodes a stable toxin, while the other encodes an unstable antitoxin that typically binds to the toxin and sequesters it. Some of the toxins are pretty nasty, for example, MazF is an endonuclease that cleaves any mRNA with an ACA in it (pretty common!) My homework assignment involves this protein in particular - under certain kinds of stress, MazF will selectively kill certain members of an E. coli population while allowing others to live, through an ROS (reactive oxygen species)-dependent and independent pathway. A recent paper suggests that under stress, MazF alters protein expression to favor expression of smaller proteins that may or may not be involved in ROS-defense. Connected to all of this is a quorum-sensing type molecule called extracellular death factor (EDF), a small molecule that they identified as NNWNN. Where does this come from? Apparently, the middle of a random gene (zwf, G6P-1-DH) that gets cleaved (probably via MazF), gets translated as NNWDN, gets acted upon by asparagine synthetase A (AsnA), which converts D to N. Might've been confusing, but REALLY cool. : )

We've also been talking about sigma factors, which are components that assist RNA polymerase in regulating the expression of specific genes in bacteria. These are a way to cluster genes required for adaptation to a particular environmental condition. There are also anti-sigma factors that bind to sigma factors and stop them (there are also anti-anti-sigma factors, and I think even anti-anti-anti-sigma factors, haha, yay life-MAPKKK!)

I read:
A Third Recognition Element in Bacterial Promoters: DNA Binding by the Alpha-subunit of RNA Polymerase
Ross, et al. Science 1993

and

Identification of the activating region of catabolite gene activator protein (CAP): Isolation and characterization of mutants of CAP specifically defective in transcription activation
Zhou, et al. PNAS 1993

which talk about the binding of RNA polymerase alpha-subunit in binding to UP elements, in the region around -40 to -60 before a gene, and then identify specific residues in catabolite gene activator protein (CAP) that bind to DNA and bend it to influence gene expression at the promoter.

Right now? I'm reading:
Structure and function of a transcriptional network activated by the MAPK Hog1
Capaldi, et al. Nature Genetics 2008

probably about transcriptional activator networks in yeast
and
“Proteotyping”: Population Proteomics of Human Leukocytes Using Top Down Mass Spectrometry
Roth, et al. Analytical Chemistry 2008

which sounds self-explanatory.

Dan Kahne gave a good talk on Friday, however, I've probably forgotten most of the details already - he talked about research in his lab on the folding and incorporation into the outer membrane of beta-barrel proteins in gram-negative bacteria. If you think about how these proteins have to first cross the inner membrane and the peptidoglycan layer before getting folded and into the outer membrane, it seems like a pretty difficult process. I've got a few papers lying around that talk about the response bacteria have for detecting misfolding of these proteins, involving DegP or DegS or something, and some more anti-sigma factors that I'll probably learn about in the next day or three. Anyway, there's apparently this complex of 5 proteins, two of which are essential, that are involved in this process. YaeT is one of them, a beta-barrel itself. People still debate about how beta-barrel proteins get into the membrane, because unlike alpha-helical membrane proteins which can probably screw in, beta-barrels will end up having unsatisfied hydrogen bonds at the ending beta sheets before the barrel is closed. Beta-strand augmentation is one strategy Dan thinks the system employs, where the protein attaches a beta strand to YaeT or something. He went through a bunch of data invovling deletion mutants of various periplasmic domains for YaeT, with interesting results. Definitely a problem I want to stay informed about.

I'm learning so much, and it's a shame, because I'm forgetting most of it. I hope all this summarizing on here will help me. : (

11: CRISPRs, Enviropig, Brainbow, Lysozyme

CRISPR — a widespread system that provides acquired resistance against phages in bacteria and archaea
Sorek, et al. Nature Rev Microbiol 2008

CRISPRs are clustered, regularly interspaced short palindromic repeats found in bacterial genomes. Together, with a leader sequence and CRISPR-associated (CAS) genes, they believe that this system helps prevent phage infection....P.S. I didn't know phage were 5-10x more numerous than bacteria!

So, yea, they think this is a bacterial homolog (paralog?) of eukaryotic RNAi. They're still working out the details, but there are already patents on this for three applications, namely strain spoligotyping (spacer oligotyping), engineering phage resistance in industrial bacterial strains used to make dairy products, and engineering gene knockdown in bacteria. The cool thing is that because the spacers are numerous and repetitive, you could knockdown several genes with this machinery....but apparently, it's still not proven that it has RNAi function.

CRISPR Interference Limits Horizontal Gene Transfer in Staphylococci by Targeting DNA
Marraffini, et al. Science 2008

So, I think the title says it all, but to go on, they found a clinical strain of Staph. epidermis that has CRISPR loci with part of the nickase gene inside, a gene found in all staphylococal conjugative plasmids (recall conjugation is a form of either intra- or interspecies horizontal gene transfer).
Blah, blah, blah, complicated genetics that I don't really understand right now, and ...oh, what's this?

Apparently, they proved that CRISPR doesn't target RNA - it targets DNA! Also interesting is their concluding few lines:

"CRISPR function is not limited to phage defense, but instead encompasses a more general
role in the prevention of HGTand the maintenance of genetic identity, as with restriction-modification systems.A primary difference between restriction modification and CRISPR interference is that the latter can be programmed by a suitable effector crRNA. If CRISPR interference could be manipulated in a clinical setting, it would provide a means to impede the ever-worsening spread of antibiotic resistance genes and virulence factors in staphylococci and other bacterial pathogens."

Perhaps I'll follow up on this after a careful re-read and class discussion tomorrow morning.

Man, I just can't keep up on here with the amount of science that I take in everyday. I still haven't gotten to write entries about any of my lab work. And just yesterday, I was skimming an article (probably in Nature) about genetically modified animals and the legal implications of their potential approval. They talked about how there weren't too many companies working on it, but that they had developed the Enviropig, whose salivary glands secrete an enzyme that helps reduce the pig's phosphorus waste.(Picture from the UCSD news website)

And in today's Chemical Biology lecture, we discussed how GFP was developed (for which the 2008 Nobel Prize was awarded in Chemistry) as well as its colored variants. While I could go into detail on how it works, I'll just mention that the coolest application I thought was the "Brainbow", where this lab at Harvard used a Cre-recombinase system to label 1000s of murine neuronal cells with varying combinations of CFP, YFP, GFP, and RFP. They claim this will help them dissect the complex network of connections between cells in the brain. Certainly looks pretty : )


Disulfide Isomerization After Membrane Release of Its SAR Domain Activates P1 Lysozyme
Xu, et al. Science 2005

These authors argue about the activate-ability of lysozymes using cysteine-accessibility experiments.

10: Checkpoints, Flagella

Checkpoints: Controls That Ensure the Order of Cell Cycle Events
Hartwell & Weinert Science 1989

This is an older article, but it provides a good framework to think about control of the cell cycle. Here are two initial questions the authors posed:
"What principles does the cell use to establish an ordered pathway of events? Does the existence of such order imply the existence of control mechanisms that enforce order?"

They basically have two models they propose:
One is substrate-product assembly - basically, objects have inherent properties that cause them to come together in the proper way. They cite phage assembly as an example, where the proteins are created simultaneously but require the baseplate to come together first before the outer coating can assemble around it. In this example, the order of the process is controlled by the materials themselves

The other model is subtly different - phage DNA is processed in a matter that depends on a control mechanism. The DNA will not be processed by terminase unless proheads are in place and a special inhibitor is put in place to make sure the system abides by this principle.

So, in the early days of thinking about the cell cycle, I guess scientists were unsure which was true (although it seems so obvious now). But, they were able to use chemical treatment and special mutants to deduce things. They argue that, although you can't prove it for certain, if these types of treatments relieve the dependence relationship, then there's most likely a control mechanism in place. I wonder if they were the first people to coin "checkpoint". Anyway, they go through a few examples of checkpoints, namely how mitosis depends on DNA replication, how replicons depend on each other, how mitosis is dependent on cell growth, how spindle pole replication is dependent on DNA replication, etc. They also mention that they observe embryonic stem cells to be different from somatic cells in these areas of control, and speculate on why that would be - are they sacrificing fidelity for quicker division? They also mention the SOS response, which I will get to probably in the next entry.

So, after class discussion, I no longer felt that I understood the difference between the two models - I thought the presence of an extrinsic sensor/regulator was the key part - yet there are some systems in which the lines start to blur - for example, SlmA binds to nucleoid DNA and prevents FtsZ activity - it's both the sensor and the executioner.

Sensing Structural Intermediates in Bacterial Flagellar Assembly by Export of a Negative Regulator
Hughes, et al. Science 1993

Background: In this paper, they talk about how flagellar assembly is grouped into 3 classes of genes: the early and middle genes are responsible for assembling the basal body and hook, which are attached together and anchored in the membrane. These genes induce the expression of the late genes that build off of the hook by exporting flagellin, hook-associated proteins (HAPs) and the Cap protein. These middle genes affect expression of late genes through encoding an alternate sigma factor (fliA, sigma 28), but if there's a defect in assembly, flgM will negatively modulate this activity.

Results: They come up with a valid model, where if the base of the flagellum is correctly set up, it can export flgM, thus lifting the repressiong of fliA and corresponding late genes that can finish assembling the flagellum. What's interesting is that normally fliA and flgM exist in equilibrium when flagella are not being assembled, but if the flagellum gets sheared, it allows for flgM export and thus rebuilding. They came to this conclusion through a few experiments, namely analyzing spent media for exported flgM, analyzing motility, and determining whether fliC (flagellin) can get exported too. They use some mutants in these experiments, either defective in some of the flagellar components or with a lacZ fusion to the flgM that they hypothesized clogged up the export (though they don't really prove this is what's happening, this paper has yet to be disproven).

A Molecular Clutch Disables Flagella in the Bacillus subtilis Biofilm
Blair, et al. Science 2008

So, bacteria switch between being motile and burying themselves in biofilms, but how is this coordinated? I mean, you wouldn't want to just start lopping off your flagella when you're getting ready to settle down - they're expensive and take a lot of time/effort to make again.

So, these two processes are alternately controlled by the transcription factor SinR in B. subtilis, which represses the eps (Extracellular polysaccharide) operon. Mutants in sinR are nonmotile - but are the flagella just getting stuck in the sticky goo secreted?

No, sinR epsH mutants don't make EPS and are still stuck. But, after searching for suppressors of the nonmotile phenotype, they found mutants in EpsE, a family II glycosyltransferase. And when they continued to look for suppressors of suppressors, they found mutations in FliG, the protein that makes the C-ring at the base of the flagellum. This protein is responsible for transducing the rotational movement generated by MovA/MovB (yes, that's right, flagella rotate because of the proton motive force conducted through these channels, anchored to the peptidoglycan skeleton) to the flagellum. So, they wondered whether EpsE was acting either as a brake, to stop the flagellum from rotating, or if it was acting like a clutch and detaching the flagellum from the motion-generating apparatus. After immobilizing bacteria and inducing expression of EpsE, they measured that flagella were not rotating as normal, but they weren't locked in position - they still underwent Brownian motion. Thus, EpsE serves as a clutch!

Regulated Pole-to-Pole Oscillations of a Bacterial Gliding Motility Protein
Mignot, et al. Science 2005
So, there are all kinds of different ways that bacteria can move, some of which are still not understood. Some bacteria use pili-directed movements, where they shoot out a pilus kind of like a grapling hook, and then try to reel themselves in. One poorly understood way is through gliding.

In this paper, they used FRAP to show that they found another oscillatory protein that goes back and forth, from pole to pole (recall MinD in E. coli), called frzS (they were in Myxococcus xanthus).

9: Kinetic Isotope Effects, Enzymatic Chlorination, Bacterial Communication

Recently, in Chemical Biology lecture, I learned about Kinetic Isotope Effects, where replacing hydrogen atoms with deuterium in a substrate can influence the rate of reaction.

The principle is that heavier atoms have a lower potential energy, so the energy required to access the transition state is greater. Thus the rate of reaction is quicker for the lighter atom. You can compare the rates of the deuterated reaction to the normal reaction and get a ratio that determines the nature of transitions at that position.
For example, if the ratio is greater than 2, then you know that the C-H bond breaks.
And if the ratio is between 1 and 1.5 or so, then you know that that carbon transitioned from sp3 to sp2 hyrbidization.
Finally, if the ratio is less than 1, then you know the opposite - that carbon went from sp2 to sp3.

Measurement of the r-Secondary Kinetic Isotope Effect for the Reaction Catalyzed by Mammalian Protein Farnesyltransferase
J. Pais, et al. JACS 2006

Anyway, this paper used the kinetic isotope effect to analyze the kinetics of farnesyltransferase, which is a pretty important enzyme required for helping lots of proteins associate with the cell membrane (including Ras!). Because product release is the limiting step for the reaction, they had to use low substrate concentration to mimic single turnover kinetics. So, they radiolabelled FPP and determined the extent of reaction at various time points. Because the observed KIE decreased with % of reaction, they had to extrapolate backward to approximate 0% of reaction.

They concluded from their data that C1 undergoes a dissociative transition state, which really doesn't mean anything to me....*shrug*. And the KIE was different for another peptide that they used, so I don't know what the deal is. Interestingly, Mg2+ speeds up the reaction, but doens't have an effect on the KIE, suggesting that the ion is not involved in the conformational rearrangement of FPP that they propose precedes the reaction.

Chlorination by a Long-Lived Intermediate in the Mechanism of Flavin-Dependent Halogenases
Yeh, Walsh, et al. Biochemistry 2007

This was an article by that guy here I really like. Basically, they characterize this enzyme RebH that creates 7-chlorotryptophan in the biosynthesis of rebeccamycin, an antitumor compound.

They initially thought that HOCl is generated at the active site via FAD, O2, and NaCl, and that the OCl- ion is what performs the chlorination. However, they demonstrate that the enzyme retains the chlorine on a lysine nitrogen, and uses that instead. They did this a number of ways, including making arguments that lysine blocks OCl- from channeling through the enzyme to the tryptophan substrate via a crystal structure they got, some radiolabelled HPLC reactions, and showing that OCl- could just float away if that was responsible, yet enzymatic activity remained for hours. I liked it - very clear and made sense.

And then, today on the front page of the New York Times.com, I read a summary of research that fed boys and men a diet of foods with high glycemic index and found a correlation with acne, presumably due to the triggered release of insulin, androgen, and other hormones that stimulate oil production.

Bacterially Speaking
Bassler/Losick Cell 2006

This was a cool review covering bacterial signaling. Some of this was review for me, after writing my review on Quorum Sensing in Staph. aureus last semester, but there are some pretty cool examples.

I knew about Vibrio fischerii, which produce bioluminescence in deep sea squids once they sense a high enough population through secreted acyl-homoserine lactones. Most of the gram-positive signals are peptides (including the Staph. cyclic AIPs or autoinducing peptides). Interestingly, there are some molecules, such as AI-2, that serve in cross-species communication.

Some molecules (quinolones from Pseudomonas) are hydrophobic, and so the bacteria will actually pinch off a portion of its membrane along with the molecule to encapsulate it and send it along. Amazingly, they can also package other types of molecules with it that serve to kill other types of bacteria that intercept the signal. Other bacteria secrete anti-quorum sensing molecules, such as proteases or lactonases that degrade signal, and we humans might even have a few circulating through the bloodstream (called paraoxonases, whose endogenous substrates still haven't been identified)! Other bacteria will just eat up the signal of neighbors (e.g. E. coli) to prevent them from communicating.

Many of these signals activate two-component systems, usually a histidine kinase that phosphorylates the aspartate of a response-regulator protein, that often comes in the form of a transcription factor, instituting development programs or other plans by influencing the expression of over >100 genes in many cases.

There are other examples - B. subtilis that are nutritionally deprived will start sporulation. About half of the population will not sporulate initially - these members will get targetted by a sporulation-induced killing system that basically kills these neighbors and uses them for food, until it's absolutely necessary to form spores.

There are more friendly examples of interspecies communication, including the Nod system we learned about in AP Bio where Rhizobium will grow on plant mycorrhizae that secrete the signal.

Nutritional Control of Elongation of DNA Replication by (p)ppGpp
Wang, et al. Cell 2007

I haven't read this paper yet, though I'm not sure I will get to it. In fact, I probably won't. But, according to the abstract, (p)ppGpp serves as a signal of nutrient deprivation that quickly and potently stops bacterial DNA replication by binding to primase. Awesome....I guess.

8: Bacterial DNA Replication

First, something to add to last entry: I never knew that some plasmids encode stable toxin/unstable antitoxin gene pairs that force the host to retain them or die....very intriguing...

Anyway, it looks like I will finially be learning about bacterial transcription and replication in greater detail, something that needed to happen sooner or later.

DNA replication initiation: mechanisms and regulation in bacteria
Mott, et al. Nature Reviews Microbio 2007
So, bacterial DNA replication initiates at oriC. This sequence is preceded by a lot of special sequences that bind various factors. Many of these are called DnaA boxes, because they help bind DnaA, the initiator protein. This protein oligomerizes in its ATP-bound form, and modifies the local topology of the DNA (ahh, DNA topology is complicated: major groove, minor groove, writhe, twist, supercoiling, toroidal wrap), so that the helicase DnaB can bind. There's variation between species, both in the DnaA binding elements and in DnaA structure, but it remains clear that cooperativity assists in recruting enough protein to oriC to help unwind the DNA.

DnaA has four domains - domain 1 assists in interactions between DnaA and itself, as well as with DnaB. Domain II is a poorly conserved linker, domain III possesses ATPase activity as well as the primary oligomerization determinants, and domain IV binds the DnaA boxes.

There are primarily two ways that E. coli regulates replication firing at oriC. The first is sequestration, which the next paper is about - briefly, GATC sites normally exist in a fully methylated state, except when replication first occurs, after which these sites become hemimethylated due to semi-conservative replication. SeqA binds to these hemimethylated site and prevents DnaA binding. Interestingly, SeqA also forms a filamentous oligomeric structure.
The second is regulation of DnaA availability/activity. The datA locus, found somewhat close to oriC, binds DnaA with much more affinity, and deletion of this locus promotes overinitiation, suggesting that this region helps as a sink for excess DnaA. Many genes also contain DnaA binding regions that are involved in influencing transcription. Notably, this occurs at the dnaA gene itself - the sites are fully occupied and repressive only in the presence of ATP-DnaA.
There's also RIDA, or regulatory inactivation of DnaA. Hda (a homologue of DnaA) associates directly with the sliding clamp of DNA polymerase III and promotes DnaA ATPase activity.

Interestingly, bacteria have histone-like proteins that seve similar roles in DNA compaction/organization into the nucleoid, and these factors (HU, Fis, IHF, H-NS) are required for initiation synchrony by introducing bends into DNA.


SeqA: A Negative Modulator of Replication Initiation in E. coli
Lu, et al. Cell 1994

One of the ways that bacteria can regulate replication (and make sure it doesn't recur to give too many copies) is through methylation of GATC sites by Dam methylase.

This group first created/identified mutants in seqA, and developed an assay to determine how long GATC sites remained hemimethylated. They found that seqA mutants methylate GATC sites upstream of oriC quicker than wild-type, but other sites remained the same.

Then, they developed an assay to count the number of replications initiated by blocking cell division. They found that seqA mutants had all kinds of odd numbers of initiations. Also, the mutants grew abnormally, which the authors deduced was a result of excess replication forks getting stalled and inducing the SOS response.

The MatP/matS Site-Specific System Organizes the Terminus Region of the E. coli Chromosome into a Macrodomain
Mercier, et al. 2008 Cell

Background: Just like in eukaryotes, bacterial DNA needs to be highly condensed in order to fit inside the cell and efficiently segregate during cell division.

Some previous studies seem to suggest that the E. coli chromosome is ordered into 4 macrodomains (Ori, Left, Right, and Ter) with interspersed nonstructured regions. This group identified a signature sequence matS that appears numerous times throughout Ter that may assist with packaging, via a protein binding partner named MatP.

To identify a sequence that appeared often enough to serve as a strong candidate, they used a word-search type program, where they searched the genome for 11-word long repeats and used statistics to narrow down on a few that were common in Ter but rare in other macrodomains. They used this to come up with a 13-mer consensus sequence, and found that this same sequence also appeared in other bacteria.

Then, they used electrophoretic mobility shift assays (EMSAs) to observe that something binds to matS. So, they set up a screen to identify a gene responsible for a protein to bind to matS (basically, if the gene is present, it will bind to matS in a promoter region for a construct and confer streptomycin resistance), and they called it matP. Once they purified and characterized MatP, they showed its binding to matS using EMSA and calculated the affinity of the interaction and compared to MatP affinity for other sites.

For in vivo studies, they used ChIP with anti-MatP to identify sites. They extended this by doing ChIP-on-chip. Deletion of MatP lead to severe defects in chromosome segregation. Using flow-cytometry, they saw that matP deletion mutants had a few more replication initiations (rifampicin runout experiments - rifampicin binds RNA polymerase) and took longer to replicate than wild-type.

Here's a cool assay: They used attL-attR sites inserted at various parts of the genome to determine the recombination frequency between different macrodomains. They saw that matP mutants had a higher rate of recombination between Right and Ter. Next, they used a GFP-MatP fusion to visualize MatP localization in vivo. Goes from the new pole, to midcell, and before division, splits in two, and then the two foci move away. They also inserted parS/parM-mCherry sites in various macrodomains, and attempted to visualize near/co-localization with MatP.

Ok, again, I have to stop for now. These Cell papers are just WAY too long X_X.

7: Bacterial Division and Bacterial Actin, Bacterial Plasmid Segregation, and Heart Cycles

So, it's been a while since my last update, and I've learned alot. I'm going to try to briefly summarize a bunch of things.

A Division Inhibitor and a Topological Specificity Factor Coded for by the Minicell Locus Determine Proper Placement of the Division Septum in E. coli
de Boer, et al. Cell 1989

They identified this MinCDE locus that's connected to cell division in E. coli.
Highlights:
Without MinCDE, E. coli forms mini-cells (divides in wrong places and pinches off parts lacking DNA and other material)
MinC+D without E inhibits cell division, leading to long, long cells.
Overexpressing MinC+D with E around inhibits cell division.
And finally, overexpressing E leads to minicells.

Something's going on, where E directs cell division, but C and D prevent E from gathering at places other than the middle of the cell.

Well, that's weird, but maybe the next paper can help clarify things:
Rapid pole-to-pole oscillation of a protein required for directing division to the middle of Escherichia coli
Raskin, PNAS 1999

This paper is pretty cool. They visualized GFP-tagged MinD.....and saw that it very rapidly moves in clumps back and forth from one pole of the cell to the other! Imagine seeing that in your microscope...

MinE is necessary for this oscillation, but not MinC. This cool technique basically verified the previous model.


Spatial Regulators for Bacterial Cell Division Self-Organize into Surface Waves in Vitro
Loose, et al. Science 2008

So, even with this model, the specific interactions between these proteins still isn't well understood. But this recent paper showed something cool:
If you just isolate fluorescently-labelled pure MinD+E and put them on a pseudomembrane, they will spontaneously organize themselves into waves (and they made movies!), where MinE basically sweeps MinD along, plucking it out of the membrane. This helps to explain MinD oscillations back and forth.

Amazingly, after class, this type of system helps to explain why successive cell divisions always occur orthogonal to each other - I don't know the eukaryotic division machinery that well, but : )



Prokaryotic DNA segregation by an actin-like filament
Mùller-Jensen, et al. EMBO J 2002

So, in one of my previous entries, I talked about a bacterial protein MreB that acts a lot like actin.
Just to review, that's what actin polymerization looks like. Actin monomers with ATP bound (Actin G for globular) bind to the barbed end of an actin filament (Actin F for filament), and the ATP is hydrolyzed to ADP. Sometimes there are nucleation factors that promote filamentation at a given site. These actin proteins have conserved domains for where the ATP binds.

Ok, so first, they identified this actin homolog, ParM, that exists on sequenced plasmids. When they express the plasmid in E. coli with fluorescently-labelled polyclonal antibodies for ParM, they see filaments from pole-to-pole (40% of the time), 20% have them at foci in the center, and the remaining 40% have fluorescence nonspecifically everywhere. They quantified the filament at roughly 15,000-18,000 molecules/cell. They found that parC and ParR are required for the filaments. This is strange, because parC is a promoter. This suggests that ParR bound to parC is a nucleation site for ParM filamentation. Interestingly, a previous group that used a GFP-ParM fusion didn't see filaments, and this was probably an artifact of the construct!

Using ParM mutants, they showed that ATPase activity wasn't required for forming filaments. D170 mutants formed inflexible, rigid filaments that extended the entire length of the cell (this residue is in the phosphate-chelating domain). And then, using ultracentrifugation, they demonstrated that Mg2+ and ATP were required for filament formation, which they also analyzed under an EM. They subsequently used light scattering to analyze filamentation kinetics in vitro, and showed using a non-hydrolyzable ATP analog that depolymerization was dependent on ATP hydrolysis. Similarly, they threw in the complex, and observed differential light scattering with signficant added ParR (but still <1000 style="font-weight: bold;">Reconstitution of DNA Segregation Driven by Assembly of a Prokaryotic Actin Homolog
Garner, et al. Science 2007

ParM is similar to actin based on filament structure, but they differ in that ParM nucleation is rapid, polymerization is symmetrical and bidirectional (as opposed to polarized actin growth), and finally, that it's dynamically unstable.

So, this group purified the machinery and watched it go in vitro using fluorescently-labelled ParM, and parC DNA attached to microbeads. They saw that filaments formed aster-looking arrays to a maximum length of 3um, unless they added nonhydrolyzable ATP, in which case they grew much longer. They also observed some beads that became connected to each other by ParM bundles that interact with parC at both ends, and these beads were pushed apart from each other over time. The end that attaches to parR/parC becomes resistant to depolymerization, and when they cut a filament with a laser, the frayed ends start to depolymerize.

Oh wow, this is pretty cool. Then they put the machinery in microtube channels they created, and watched as the filaments oriented themselves along the long axis of the channels, and pushed beads apart, until they hit the ends of the channel. They then used "speckle microscopy" and photobleaching to show that nascent polymerization occurs at the parR/parC interface (I think speckle microscopy is where they rapidly mix in the protein with a newly colored conjugated fluorophore)

Cool stuff!

Today, I led a paper discussion on:
An Expanded Eukaryotic Genetic Code
Chin, et al. Science 2003

In this paper, they evolved E. coli Tyr-tRNA synthetase to accept unnatural amino acids and incorporate them into stop codons in yeast via E. coli tRNAcua. They had this complicated series of screens where they used positive and negative selection to screen over 10^8 library members, and miraculously isolated ones for 5 different unnatural amino acids. They then demonstrated that the system works by expressing SOD with the unnatural incorporations, though they could only get 20% of the yield of native protein.

The second paper we discussed was:
Improving Nature's Enzyme Active Site with Genetically Encoded Unnatural Amino Acids
Jackson, et al. JACS 2006

In this paper, they mutated various active site residues to unnatural amino acids in the enzyme nitroreductase from E. coli (which converts the prodrugs CB1954 and LH7 into active DNA alkylating agents poised to kill cancer cells) in an attempt to improve enzymatic activity. While they found ones that improved activity, apparently they didn't do a good job justifying why (they came up with ideas, but didn't show evidence). It was interesting though to hear the reaction from my TF - basically, he didn't take them seriously because they came from a small place. He recognized that their resources limited the experiments they could do, but didn't consider it real science, because it didn't come from some place like Harvard. I spoke up and said I know exactly where they come from!

Finally, I was reading about pharmacogenetics/pharmacogenomics (perhaps I'll save that for another time), and they mentioned Torsades de pointes, which is basically a cardiac arrhythmia (specifically a tachycardia), in this class' case, potentially induced by a large number of drugs depending on patient genotype.

So, of course, I discovered that on wikipedia, and I found something about long QT syndrome, which I remember hearing about during some discussion of genetics. So, I looked up what a QT interval is, and through wikipedia, learned about both the polarizaion/depolarization of heart cells, as well as the path of electricity that induces cardiac contraction (and produces bloodflow!). The heart is SO amazing. Obviously, I won't remember everything, but here's a bit (I hope this gif works):

The P wave illustrates the generation of the action potential at the sinoatrial (SA) node and its conduction to the atrioventricular (AV) node. This enables spatial resolution of atrial and ventricular contractions, otherwise blood could have backflow. The signal then propagates through the bundle of His (PR interval) and through the Purkinje fibers (QRS complex). Finally the ventricles repolarize during the T wave. Thus, the QT interval is the distance between the initiation of the QRS complex and the end of the T wave (though there is some disagreement about how to calculate this accurately from ECGs).

Wow, that was long. I have more to talk about, but I'll save it for next time!

6: More Bacterial Cell Wall Stuff

I've just had a drink or two, and I'm rushing, but we'll see how this goes:

Control of Cell Shape in Bacteria: Helical, Actin-like Filaments in Bacillus subtilis
Jones,et al. Cell 2001

Overview
So, in eukaryotes, cell shape is most often dictated by the cytoskeleton (mostly actin filaments). But what about in bacteria? This paper analyzed B. subtilis mutants in MreB and Mrel, which happen to have a spherical cell phenotype.

Results
First they demonstrated that MreB was essential for B subtilis (even though it's not for E. coli), and phenotypic analysis of cells depleted in MreB showed they had an abnormal, rounded and wide morphology compared to wild-type rods. Mrel mutants, however, had abnormal curvature, but were viable, and the authors hypothesized that this protein is involved in coordinating the FtsZ ring involved in cytokinesis after mitosis.

So, they tried looking at GFP-MreB under the microscope, but it was dispersed throughout the cell. Then they tried using a monoclonal antibody for c-Myc-MreB, and found that it formed helical structures (they have a hard time showing this with 2D microscope images). Using quantitative immunoblotting, they estimated ~8,000 molecules/cell. They used the same technique to visualize Mrel double helices in vivo. Finally, they made some mutations in MreB, and showed that these altered cell morphology. After analyzing some sequences, they saw that MreB/MreI were absent from coccoid bacteria, and proposed a model where these proteins fulfill the role of actin in bacteria (sequence alignment shows few gaps between the proteins) and control cell axes through assembly in an ATP-dependent manner.

Control of Cell Morphogenesis in Bacteria: Two Distinct Ways to Make a Rod-Shaped Cell
Daniel, et al. Cell 2003

Summary
In this paper, they used a fluorescently-labelled vancomycin probe to explore peptidoglycan structure, assembly, and its effect on cell shape.

They found nascent peptidoglycan assembly was directed in a helical manner dependent on Mrel, which is a homolog of MreB. Without Mrel around, they hypothesized that B. subtilis switches to a polar mode of growth (basically, directed by cell division).

Distinct penicillin binding proteins involved in the division, elongation, and shape of Escherichia coli K12
Spratt 1975 PNAS

Summary
This paper is pretty old, and couldn't get to much specific detail, but was able to use fluorography of displaced radiolabelled antibiotics to identify penicillin binding proteins with roles in cell shape and division. The following is pretty ridiculous:
"The binding proteins were detected by fluorography (14) on Kodak RPRoyal x-ray film for 48 days at -70C."
Wow. If it took me 48 days straight to get data, I think I'd give up on science.