Tuesday, July 3, 2012

Argonaute I and Argonaute II


Structure of Yeast Argonaute with Guide RNA
Kotaro Nakanishi, David E. Weinberg, David P. Bartel, Dinshaw J. Patel
Nature (2012), 486, 368-373

This week brings exciting news for the RNA-world.
The group of Dinshaw Patel has published a new Argonaute structure which includes its guide RNA.
The structure can be found at the Protein Data Bank under the 4f1n identifier and it was solved at 3.2 Angstrom resolution.
Some weeks ago I posted here the news of Schirle and Macrae's new 2.3 Angstrom structure of Argonaute II.
So, now both Argonaute I (in the left side of the movie) and Argonaute II (in the right side of the movie) with their corresponding guide RNA's can be compared side by side, and they are quite similar in shape.
If one does a partial alignment of the structures using the default alignment parameters of pymol, one gets that the two structures align with an RMSD of 3.8 Angstroms.
Again, a movie is quite useful and we have made a very short one as you can see below:

video

Monday, June 11, 2012

Altritol-RNA Duplex at 1.96Å and P5abc RNA Folding Pathway


How Does Hydroxyl Introduction Influence the Double Helical Structure: The Stabilization of an Altritol Nucleic Acid:Ribonucleic Acid Duplex
Margriet Ovaere, Jiri Sponer, Judit E. Sponer, Piet Herdewijn, Luc Van Meervelt
NAR (2012), XX, 1-11

Altritol Nucleic Acids (ANA) are those whose ribose sugar has been modified to have and additional methylene group between the O4' and C1' atoms in the sugar of an RNA:


A duplex can then be made between RNA strands and ANA strands and the results for the nice duplex elucidated with X-Ray crystallography at 1.96 Ångstrom resolution are reported by Ovaere and coworkers. The structure can be found at the Protein Data Bank (PDB) under PDB_ID:3OK2.

One can compare the RNA-ANA, to plain old RNA, an RNA-LNA duplex, and a LNA-LNA duplex as illustrated in the following picture.




Folding Path  of P5abc RNA  Involves Direct Coupling of  Secondary and Tertiary Structures
Eda Koculi, Samuel S. Cho, Ravi Desai, D. Thirumalai, and Sarah A. Woodson
NAR (2012), XX, 1-11

Once again the collaboration of Sarah Woodson with one of her former graduate students brings out beautiful results to further our understanding of RNA folding.

In this case the energy diagram talks by itself.





Thursday, June 7, 2012

Nano-DNA Tech., Femtosecond X-ray of the Future, and Yet Another Small-Molecule Detector Made of DNA.

In this week's Chemical and Engineering News three reviews concerned nucleic acids. A first review claims that a group at Harvard Medical School has a more effective technique (faster) than Ned Seeman's to produce DNA shapes. A second review is about computer simulations which predict that data can be extracted from an spectroscopy, previously thought of as destructive, and a third review concerns yet another technique to use DNA to detect small ligands.


Complex Shapes Self-Assembled from Single-Stranded DNA Tiles
Bryan Wei, Mingjie Dai, and Peng Yin
Nature (2012), 485, 623-626

The authors are all part of an institute called, "Wyss Institute for Biologically Inspired Engineering", which seems to be a newish institute at Harvard, probably started in 2008, and which got a $125 million dollar donation from Hansjörg Wyss, a billionaire who owns a company called Synthes which makes implants and various medical devices. They claim that up to now the process of engineering DNA shapes is done using "a long scaffold strand which is folded by hundreds of short auxiliary strands into a complex shape", and their method uses the simplest tile form, they call it an SST (single-stranded tile). Surprisingly they reference Luc Jaeger's work (who has developed the single-stranded assembly technique for RNA's since at least 1996, called RNA Tectonics)  just in passing. As part of their results they show an AFM image of the assembled 2D DNA-shapes:



Potential for Biomolecular Imaging with Femtosecond X-ray Pulses
Richard Neutze, Remco Wouts, David Van der Spoel, Edgar Weckert, Janos Hajdu
Nature (2000), 406, 752-757

This oldish paper was highlighted because what they proposed in 2000 has now become a reality.
If one is interested ever in showing examples of the power of theory, and computational modeling has on future achievement in science, this is an excellent example. In the experimental paper which proves the predictions (Science, (2012), 337, 362-364), they are using a so-called Serial Femtosecond Crystallography (SFX) which allows them to obtain a 1.9 Ångstrom resolution structure of an enzyme called lysozyme, from microcrystals which would be useless for conventional X-ray crystallography, an using a coherent X-ray imaging instrument (LCLS) form the Stanford Linear Accelerator Center(SLAC):



The PDB  structures have  the following accession codes:
4ET8, 4ET9, 4ETA, 4ETB, 4ETC, 4ETD, 4ETE.

Tuesday, May 15, 2012

Two Fresh Reviews on Nucleic Acid Simulations.

Two new reviews have just come out headed by common names in the field of modeling nucleic acids like Michael Levitt, Jiri Sponer, and Tom Cheatham.


Modeling Nucleic Acids
Adelene YL Sim, Peter Minary, Michael Levitt
Current Opinion in Structural Biology (2012), 22, 1-6

This is a very short review, so, if you're really looking for a REVIEW on modeling nucleic acids this is not a recommended read.
Levitt's group highlights what they call, degrees of freedom (DOF), as if it was a whole field, weird..., "degrees of freedom are degrees of freedom", not a subject field.
Anyhow, that DOF part is where it gets interesting, as they point out to their recent work which goes in the direction of rigid-body analysis, although, again, they wanna call it different.
They cite two of their recent articles, one called "Modeling and Design by Hierarchical Natural Moves", and the other called "Conformational Optimization with Natural Degrees of Freedom: A Novel Stochastic Chain Closure Algorithm".

The most interesting part of the review is that they claim to have solved, what they call the "lever-arm" problem, which I suspect to be closely related to the crank-shaft effect (neatly illustrated by Mr. "I solved tRNA", i.e., Sung‐Hou Kim). They go on to say that the trick is to use only degrees of freedom which affect local conformation (?) but this selection breaks the continuity of the helix, and so that forces them to use a chain-closure algorithm. All in all it's quite hard to follow what they mean, so it leaves one intrigued to look for answers in their other papers which seem to be more meaty.



Molecular Dynamics Simulations of G-DNA and Perspectives on the Simulation of Nucleic Acid Structures
Jiří Šponer, Xiaohui Cang, Thomas E. Cheatham III
Methods (2012), XX, XX-XX

This one is a proper review with specific details on molecular dynamics using explicit solvent (water).
It seems to tell in detail the very valuable information on the miss-happenings of the Orozco corrections (ff9 + parmbsc0) to the Amber force-field for nucleic acids which result in ladder-like conformations.
It also gives a good outline of the up-to-date developments of the AMBER and CHARMM force-fields, and mentions that CHARMM seems not to be stable on G-DNA simulations, of course, just using one methodology for the solvent cage, so, remains to be seen what happens by using something different to particle mesh ewald.

Monday, May 7, 2012

More Insights into RNA Silencing. The Structure of Argonaute 2 at 2.3 Angstrom Resolution.

The Crystal Structure of Human Argonaute2
Nicole T. Schirle and Ian J. Macrae
Science (2012), 336, 1037

Schirle and Macrae have solved the structure of the protein-RNA complex called Argonaute2 using X-ray spectroscopy. This protein-RNA complex is part of the RNA Induced Silencing Complex, or RISC for short. As its name reveals RISC's task is that of silencing. What RISC silences is RNA, meaning that it blocks a process that otherwise would have been accomplished had RISC not been there. The process blocked is translation of messenger RNA (mRNA) into proteins, and this is why the process is also known as post-transcriptional regulation, since mRNA is the target, and this one molecule has to be produced in transcribing DNA to RNA.

It seems that the trick of RISC is that of carrying a short RNA (20-25 nucleotides) which will bind in a complementary fashion to an specific mRNA sequence and then make it a target for degradation. The structure solved  by the researchers at Scripps Research Institute in la Jolla, California, is that of the molecular complex present in humans.

The following movie shows the molecular complex using the protein data bank (PDB) structure 4ei1:

video


The script to produce the images (which are put together into a movie) using the molecular viewing program pymol follows:

#########################################
# Mauricio Esguerra                     #
# May 4, 2012                           #
#                                       #
# General pymol script to beautify      #
# molecules.                            #
#                                       #
# This script can be invoked without    #
# X11 display with:                     #
# pymol -qc bla.pml >& bla.log &        #
#########################################


#########################################
# Molecule Specific Commands
#########################################
load 4ei1.pdb
load blocks.r3d
hide everything
create protein, chain A
create nucleic, chain B and (not resn U)
remove 4ei1 in protein
remove 4ei1 in nucleic
show cartoon
show cgo
show surface, protein
set transparency, 0.5, protein
set cartoon_ladder_mode, 0
set cartoon_rect_width, 1.2
set cartoon_rect_length, 0.9
set cartoon_cylindrical_helices, 1
set stick_radius, 0.14
show sticks, nucleic


#########################################
# General settings
#########################################
set orthoscopic, 1


#########################################
# Protein Secondary Structure Coloring 
#########################################
color blue, ss h
color red,  ss s
color grey, ss l+''
color black, nucleic


#########################################
# Movie Making
#########################################
unset movie_auto_interpolate
unset movie_loop

mset 1 x480

# Zoom to full and call it frame 1
zoom complete=1
frame 1
mview store

# zoom and stay
zoom nucleic, complete=1
frame 70
mview store
frame 90
mview store

mview interpolate




# from frame 101 to 315 turn in y
frame 200
turn y, 140
mview store

# from frame 316 to 480 turn in y
frame 315
turn z, 90
mview store
mview interpolate, power=1

zoom state=-1
frame 480
mview store
#frame 480
#mview store


set movie_loop
mview interpolate

mview smooth

#mplay
#mstop

#Make images for movie
viewport 1024,768
set ray_trace_frames=1
set cache_frame=0
mpng mov

A nice perspective called:
"Guided Tour to the Heart of RISC"
written by Emine Kaya and Jennifer A. Doudna accompanies the discovery reports in the same issue of science.

Monday, April 16, 2012

Binding of Mg2+ and Na+ to RNA, and a new group II intron structure.

Non-specific binding of Na+ and Mg2+ to RNA determined by force spectroscopy methods.
C. V. Bizarro and A. Alemany and F. Ritort
NAR (2012), XX, 1-14

Pulling experiments were done in an RNA hairpin with a stem of 20 base-pairs and a GNRA (GAAA) tetraloop.

Following is the sequence that was annealed and cloned into the pBR322 DNA plasmid:
gcgagccataatctcatct GTTT ccagatgagattatggctcgc

In Tables one and two the authors show results of a persistent length ( P ) determination using the inextensible worm-like chain model where the range of single stranded RNA P, is between 7.5 and 12.7 Angstroms, for Na+ concentrations between 1050mM and 50 mM, and between 7.5 and 12.7 Angstroms, for Mg2+ concentrations between 10mM and 0mM (with fixed [Na+]=50mM).
The results are illustrated in the following graph which is rescaled to illustrate one of the main conclusions of the authors:

"Our findings demonstrate the validity of the approximate rule by which the non-specific binding affinity of divalent cations is equal to that of monovalent cations taken around 100-fold concentration for small molecular constructs"


As one of the final conclusions the authors illustrate in the supplementary material how counterion condensation theory and tightly bound ion (TBI) theory predict the free energies of formation of the RNA hairpin depending on ion concentration for the monovalent and divalent cases, concluding that for the monovalent case counterion condensation theory and TBI are equivalent, but for the divalent case counterion condensation theory fails.



Crystal structure of a group II intron in the pre-catalytic state.
Russell T Chan, Aaron R Robart, Kanagalaghatta R Rajashankar, Anna Marie Pyle and Navtej Toor
NSMB(2012), XX, 1-3

From the same author who got you the very first crystal structure of a group II intron (PDB_ID:3bwp), say, yes! Navtej Toor!, we are getting a second structure from the same intron in a pre-catalytic state, the new structure has a PDB_ID:4ds6, which as of this post hasn't been released yet to the wild. The group II intron is supposed to be an ancestor to the spliceosome and that is one of the reason for it being so interesting. With the previous structure, which was a post-catalytic state structure and this one they propose a full three step structural mechanism for the splicing reaction summarized in the following figure:

Friday, April 13, 2012

April 2012 Review

The structure of triplex DNA in the gas phase.
Arcella A, Portella G, Ruiz ML, Eritja R, Vilaseca M, Gabelica V, Orozco M.
JACS (2012), XX, 1-28


Discrete RNA libraries from pseudo-torsional space.
Humphris-Narayanan E, Pyle AM.
JMB (2012), Mar13

This seems to be an update on using virtual bonds as ramachandran criteria for structure simplification. Again, no mention whatsoever of Wilma's work.


Thursday, April 5, 2012

Catching up on RNA News.

The last post in this blog happened in January and it shied away from the meaty content of RNA.
This is my attempt to catch up, and hopefully to produce at least a brief reminder of RNA related news, mainly for personal use, sorry if this comes out too cryptic for the general public.


Very nicely sold article from the abstract description. It still is to be seen how this new parametrization of fixed charges to dielectric constants works as compared to the traditional TIP3 and SPC, time will tell.
Fennell CJ, Li L, Dill KA.
J Phys Chem B. 2012 Mar 7. [Epub ahead of print]
PMID: 22397577 [PubMed - as supplied by publisher]


Four low-resolution structure determination techniques are evaluated in combination with the MC-SYM MC-FOLD pipeline. The techniques are hydroxyl radical footprinting (OH), methidiumpropyl-EDTA (MPE), multiplexed hydroxyl radical cleavage (MOHCA), and small-angle X-ray scattering (SAXS). There are two sets of interactions proposed to compare the predicted vs. the experimental structures named low and high. The high includes tertiary interactions. As I understand it the experimental data is fit to the MC-SYM MC-FOLD pipeline results to find the best fit between them.
Parisien M, Major F.
J Struct Biol. 2012 Feb 23. [Epub ahead of print]
PMID: 22387042 [PubMed - as supplied by publisher]


These guys sound like physicists using jargon. In this case they are using the Rosetta software to bridge protein structure prediction and RNA loop models, nothing new, but they seem like a powerful group.
Sripakdeevong P, Kladwang W, Das R.
Proc Natl Acad Sci U S A. 2011 Dec 20;108(51):20573-8. Epub 2011 Dec 5.
PMID: 22143768 [PubMed - indexed for MEDLINE]


For me it was interesting to see someone combining OPLS and AMBER (old, not the newish Orozco params for nucleic acids), using a molecular dynamics program I didn't know about called MOIL (from a group in U. Texas@Austin headed by a Ron Elber).
It was also surprising to see a non-referenced value for the persistence length of dsRNA as being about 500 base-pairs, that is, > 1500 Angstroms, not even close to the old Hagerman experimental values (around 700 Angstrom), being used as justification for keeping their ARNA's frozen in the simulation, makes you wonder what will happen if they're not kept "frozen". Later I noticed that this was an error and they meant 500 Angstrom, which still is not right, but at least that's the DNA persistence length.
It is interesting also how they sort of compare their results for the radial distribution functions against counterion condensation theory.
The Ionic Atmosphere around A-RNA: Poisson-Boltzmann and Molecular Dynamics Simulations
Serdal Kirmizialtin, Alexander R.J. Silalahi, Ron Elber†, Marcia O. Fenley
Biophysical Journal
Volume 102, Issue 4, 22 February 2012, Pages 829–838
http://www.sciencedirect.com/science/article/pii/S0006349512000628

RNA and Its Ionic Cloud: Solution Scattering Experiments and Atomically Detailed Simulations
Serdal Kirmizialtin, Suzette A. Pabit, Steve P. Meisburger, Lois Pollack, Ron Elber
Biophysical Journal
Volume 102, Issue 4, 22 February 2012, Pages 819–828
http://www.sciencedirect.com/science/article/pii/S0006349512000860

Tuesday, January 24, 2012

Posting latex equations in blogger.

When using the mathjax javascript code found at:

http://www.mathjax.org/docs/1.1/start.html#mathjax-cdn

One can post mathematical equations using latex as in the following example:

LaTeX Code:
When a \ne 0, there are two solutions to ax^2 + bx + c = 0 and they are
x = {-b \pm \sqrt{b^2-4ac} \over 2a}.


RENDERED Equations:
When \(a \ne 0\), there are two solutions to \(ax^2 + bx + c = 0\) and they are
$$x = {-b \pm \sqrt{b^2-4ac} \over 2a}.$$

Great, so now I can make posts with simple matrix equations if necessary.

Tuesday, January 10, 2012

Latest News from Chemical and Engineering News

To start the new year we look at the headlines related to nucleic acids in Chemical and Engineering News.
In the first week of the year two nucleic acid notes have been written in C&EN. One concerns genetic engineering modifying DNA sequences by using Zinc-finger nuclease proteins, and the other also concerns genetic engineering but this time at the expression level by suggesting a protocol to program gene expression using so-called "RNA-devices".


Targeted chromosomal duplications and inversions in the human genome using zinc finger nucleases
Hyung Joo Lee et al.
Genome Research advanced publication (2012)

Jin-Soo Kim and coworkers at Seoul National University have designed Zn-finger nucleases that target specific genomic sequences in human embryonic kidney cells and are able to induce sequence duplication and inversions.


Model-Driven Engineering of RNA Devices to Quantitatively Program Gene Expression
James M. Carothers, et al.
Science334 1716-1719 (2011)

The following is just a copy and paste of the abstract from the article due to how hermetic it seems to my understanding.
"The models and simulation tools available to design functionally complex synthetic biological devices are very limited. We formulated a design-driven approach that used mechanistic modeling and kinetic RNA folding simulations to engineer RNA-regulated genetic devices that control gene expression. Ribozyme and metabolite-controlled, aptazyme-regulated expression devices with quantitatively predictable functions were assembled from components characterized in vitro, in vivo, and in silico. The models and design strategy were verified by constructing 28 Escherichia coli expression devices that gave excellent quantitative agreement between the predicted and measured gene expression levels (r = 0.94). These technologies were applied to engineer RNA-regulated controls in metabolic pathways. More broadly, we provide a framework for studying RNA functions and illustrate the potential for the use of biochemical and biophysical modeling to develop biological design methods."