Paper describing the structure of LFA-1

From: Narayanan Ramasubbu

Dear All:
I would like to know the literature on the crystal structure of Leukocyte Function-Associated Antigen One (LFA-1, CD11a/CD18).
I have the structure of I domain but not for the entire molecule. I would greatly appreciate if people can point me to the right article/reference.

Thanks
Subbu

----------
From: Ed Pozharski

One way to get that kind of information is to run your sequence against
pdb using NCBI BLAST, and then follow up the papers associated with the
PDB entries that come up.

PubMed search is also not a bad starting point.  If all else fails,
there is always google.  I am sure that with right keywords (e.g.
"structure of Leukocyte Function-Associated Antigen One") a review
article will pop up (if it exists) among the first five or so hits.

--
Oh, suddenly throwing a giraffe into a volcano to make water is crazy?
                                               Julian, King of Lemurs

Program announcement: Nautilus

From: Kevin Cowtan

Here's an alpha release of my nucleic acid model building program for the early adopters among you to try out. There's no GUI, there's only a Linux binary for now, the features are rather limited, and it's only been tested on synthetic data. On the other hand, it seems to work for me.

Feedback would be very welcome. Please try it out if you can and send me the logfile (with filenames and cell parameters redacted if you prefer). Anything it builds, you get to keep.

It is available from here:
http://www.ysbl.york.ac.uk/~cowtan/nautilus/nautilus.html

..

'Nautilus' is a program for automatic model building of nucleotide
structures in electron density maps. It will trace a map with no model,
extend an existing model, or add nucleotide chains to an existing
non-nucleotide model.

'nautilus' does not currently perform refinement - you will need to
refine and recycle for further model building yourself. Neither does
it assign sequence - the model is built as ploy-U.

This is an alpha release. It may not work at all. It has only
been tested on synthetic data with simulated errors.
The API will change significantly in subsequent releases.

DNA cif files

From: Gregory Bowman
Date: 16 September 2011 16:26

I have a question about the bond angle restraints in the DNA cif files. I recently submitted a protein-DNA complex to the PDB, and found out that many of the glycosidic bond angles (atoms O4'-C1'-N9/N1) were outside the accepted range. I switched from CCP4 6.1.13 to 6.2.0 (installed/updated with fink on Mac OSX 10.6.8), and this "fixed" the problem in that now there is a narrower distribution of bond angles, but the target/ideal values are different in these two CCP4 monomer libraries (108.4 versus 107.8):

=====

/sw64/lib/ccp4-6.1.13/data/monomers

file=a/AD.cif

Ad       A   'Adenosine                           ' DNA                32  21 .

Ad       N9     C1*    O4*     108.400    3.000

file=c/CD.cif

Cd       C   'Cytidine                            ' DNA                30  19 .

Cd       N1     C1*    O4*     108.400    3.000

file=g/GD.cif

Gd       G   'Guanosine                           ' DNA                33  22 .

Gd       N9     C1*    O4*     108.400    3.000

file=t/TD.cif

Td       T   'Thymidine                           ' DNA                32  20 .

Td       N1     C1*    O4*     108.400    3.000

=====

/sw64/lib/ccp4-6.2.0/data/monomers

file=d/DA.cif

DA       DA  '2'-DEOXYADENOSINE-5'-MONOPHOSPHATE  ' DNA                34  22 .

DA       "O4'"  "C1'"  N9      107.800    0.800

file=d/DC.cif

DC       DC  '2'-DEOXYCYTIDINE-5'-MONOPHOSPHATE   ' DNA                32  20 .

DC       "O4'"  "C1'"  N1      107.800    0.800

file=d/DG.cif

DG       DG  '2'-DEOXYGUANOSINE-5'-MONOPHOSPHATE  ' DNA                35  23 .

DG       "O4'"  "C1'"  N9      107.800    0.800

file=d/DT.cif

DT       DT  'THYMIDINE-5'-MONOPHOSPHATE          ' DNA                34  21 .

DT       "O4'"  "C1'"  N1      107.800    0.800

Why are these values different in the two libraries? Am I looking at the wrong files or the wrong bonds?

Thanks!

Greg

----------
From: Ian Tickle


Hi, it depends what you mean by 'different'.  The SUs of the new set are much lower than the old values (0.8 deg vs. 3.0) so one would assume that these are improved estimates based on new data.  The difference between the old and the new is (108.4 - 107.8) = 0.6 deg with a SU of the difference of sqrt(3^2 + 0.8^2) = 3.1 deg, so difference/SU = 0.6/3.1 = 0.2 sigma - hardly what one could call 'significantly different'.

Cheers

-- Ian

Silver staining Coomassie stained gels

From: K Singh

Dear All,
Can anyone suggest me a protocol for silver-staining the PAGE that is
already stained with Coomassie.
Kris

----------
From: Ed Pozharski

Just destain it and then use the standard silver-staining protocol.  If
for some reason you want to have both stainings superimposed, take two
pictures and then combine them - it should be easy to do by
superimposing the ladders, just make sure that pictures are of the same
size (maybe tricky if you don't have an imager, but nothing GIMP
couldn't overcome).

--
"Hurry up before we all come back to our senses!"
                          Julian, King of Lemurs

----------
From: Artem Evdokimov

Destain it really well. One easy peasy option is to put the gel in water and add 1 or 2 ml of cheap Q sepharose to it then boil. Gel will destain in 2-3 minutes of boiling.

----------
From: Soisson, Stephen M

Now that's a nifty trick I hadn't heard of before!

----------
From: Artem Evdokimov

It is nice indeed, that's how we stainand destain our gels in less than 8 minutes :-) there was at one time a proprietary product like tis called magic beads or somesuch. But regular resin works ;-)

----------
From: Dima Klenchin


There is absolutely nothing wrong with silver staining of a Coomassie-stained gel without destaining. It only prevents "blowout" of most intense bands already well-visible with Coomassie. The only thing to do is to equlibrate the gel in water before going with silver. At least that's my experinece with the silver staining protocol that is based on tungstosilicic acid (aka Bio-Rad Silver Stain Plus described here for DNA in agarose but works equally well for proteins in PAAG: http://www.ncbi.nlm.nih.gov/pubmed/2446526 ).

- Dima

about alternate conformations

From: Ming
Date: 16 September 2011 15:08

Hi,

I was using Refmac from CCP4 to refine a protein's crystal structure. The methionine has half selenium and half sulfur. I was trying to make alternate conformations and let refmac do the refinement. But it keeps giving me error message as follows:

There is an error in the input coordinate file
At least one the chains has 2 residues with the same number
Check above to see error
===> Error: Problem with coordinate file
<B><FONT COLOR="#FF0000"><!--SUMMARY_BEGIN-->
 Refmac_5.5.0109:  Problem with coordinate file

And here is my modified pdb file.

HETATM  403  N  AMSE A 129      ***           N 
HETATM  404  CA AMSE A 129      ***           C 
HETATM  405  CB AMSE A 129      ***           C 
HETATM  406  CG AMSE A 129      ***           C 
HETATM  407 SE  AMSE A 129      ***          SE 
HETATM  408  CE AMSE A 129      ***           C 
HETATM  409  C  AMSE A 129      ***           C 
HETATM  410  O  AMSE A 129      ***           O 
ATOM    411  N  BMET A 129      ***           N 
ATOM    412  CA BMET A 129      ***           C 
ATOM    413  CB BMET A 129      ***           C 
ATOM    414  CG BMET A 129      ***           C 
ATOM    415  SD BMET A 129      ***           S 
ATOM    416  CE BMET A 129      ***           C 
ATOM    417  C  BMET A 129      ***           C 
ATOM    418  O  BMET A 129      ***           O 

I already confirmed with pdb that this is the right format for this case. But refmac doesn't work with it. I wonder if there is any other changes I should make for refmac?

Thank you,


Ming

----------
From: Ed Pozharski



Do you have the evidence that your incorporation ration was 50%?

On a practical side, try giving the SeMET a different chain ID.  You can
change it back manually after refinement.  Assuming that the side chain
conformation is the same, you can probably just have the selenium atom
added and hope that due to partial occupancy refmac will not apply any
vdw terms.

--
"Hurry up before we all come back to our senses!"
                          Julian, King of Lemurs

----------
From: Kevin Jin

I did not make it clear. The protocol we used for protein expression always limit the occ of MSE as 0.75. You may estimate your occ of MSE by ED. You can check PDB for JCSG structures.

 

Cheers,

 

Kevin

Fwd: [ccp4bb] Why Does Cross-linking Mean Anything?

From: Jacob Keller
Date: 15 September 2011 21:10

Dear Crystallographers and Biochemists,

cross-linking, say with gluteraldehyde, is an oft-used method of
demonstrating a protein's oligomeric state in solution. I have a
difficulty with this, however: theoretically (and in practice!), one
can tune the amount of cross-linker to get what ever result is
desired, such that any protein with some exposed lysines can be
cross-linked in any oligomeric state. How, then, does one evaluate the
power of this evidence? Maybe one should do a gradient of
gluteraldehyde concentrations, then plot the deviation of the observed
cross-linked oligomerization from a theoretical null hypothesis? Seems
like this could be done, but I have never seen this in the
literature...

Best,

Jacob Keller

--
*******************************************
Jacob Pearson Keller
Northwestern University
Medical Scientist Training Program


----------
From: Ed Pozharski
Date: 15 September 2011 21:18


Right - just do it side-by-side with a protein known to be monomeric of
roughly the same size/lysine content...  And what is the "critical
concentration" of gutaraldehyde at which the false positives appear in
your experience?

--
"I'd jump in myself, if I weren't so good at whistling."
                              Julian, King of Lemurs

----------
From: Jacob Keller


The critical concentration depends on protein concentration, time of
reaction, brand of gluteraldehyde, day of week, color of my shirt....

No, I don't know--I have seen cross-linking gradients in Nature and
such in which several oligomeric states can be seen up to the one the
author asserts is the physiological one. This is a nice experiment for
proving one's point on paper, but maybe not for establishing the
truth? Maybe a control with some SDS would be appropriate (although
this would probably perturb the lysines). Or maybe the experiment
should be done in a lysate, and then western-blotted?

Jacob

----------
From: Herwig Schuler

Dear Jacob,

agree, it's a mess. From what I read, the glutataldehyde concentration should be low (<0.01%) and the x-linked complex that you get should not occur in high salt conditions (reasoning that 1.2M KCl would break the average complex apart). Have seen papers where more selective zero length crosslinkers have been used - the Pierce catalog used to have lots of them - it seems it boils down to the same problem, eventually you will find one that "works" but you will need independent evidence to convince yourself.

I typically make really nice MW size ladders with my monomeric negative control proteins, though.

Best, Herwig

* * * * * * * * * * * * * *
Herwig Schüler, PhD
PI, Structural Biochemistry
Structural Genomics Consortium, MBB
Karolinska Institutet
Scheeles väg 2
S-17177 Stockholm

----------
From: R. M. Garavito

Jacob,

One of the problems with glutaraldehyde is the its chemistry is so bizarre.  It actually forms quite long transient polymers in solution.  You also have to ask yourself why formaldehyde also "fixes" tissues.  This is why glutaraldehyde works so well for tissue fixation for EM as opposed to our usual bivalent crosslinkers we use in biochemistry experiments.  Check out the old EM literature about discussions of glutaraldehyde chemistry.

Moreover, the Schiff's base linkage glutaraldehyde is slowly reversible.  You need to reduce it to make it permanent.  I think that glutaraldehyde is a very poor choice for a precise bivalent crosslinker, but as a broad spectrum crosslinker (hitting lysines and a free amino terminii that are different distances apart), glutaraldehyde is great.  As it is highly volatile (its unique smell tells you you've had the bottle open too long), you can crosslink crystals by vapor diffusion in an hour.

So I would be cautious about interpreting any crosslinking results using glutaraldehyde, except the obvious (i.e., oligomers may indicate the native tertiary state of a protein or complex).

Cheers,

Michael

****************************************************************

R. Michael Garavito, Ph.D.

Professor of Biochemistry & Molecular Biology

513 Biochemistry Bldg.   

Michigan State University      

East Lansing, MI 48824-1319

****************************************************************

Fortran runtime error in Procheck on Mac OS X 10.6.8

From: Alexander Batyuk
Date: 15 September 2011 07:48


Dear colleagues,

I have a problem running procheck on Mac OS X 10.6.8. It stops with the following error:

..................................................................

Stereochemical quality plots and residue-by-residue listing

At line 2639 of file /sw64/src/fink.build/ccp4-6.2.0-101/ccp4-6.2.0/src/procheck/pplot.f (unit = 14, file = 'rama.sum')
Fortran runtime error: Sequential READ or WRITE not allowed after EOF marker, possibly use REWIND or BACKSPACE


I would appreciate any help.

Thank you and best wishes,

Alex


--
Alex Batyuk
The Plueckthun Lab
www.bioc.uzh.ch/plueckthun

----------
From: Saul Hazledine


Hello Alex,
 I know nothing about Procheck but the following may be of interest to you. There is a similar error message produced during ARP/wARP model building as described in the message below:

https://www.jiscmail.ac.uk/cgi-bin/webadmin?A2=CCP4BB;30c74e5a.1109

This error was found to be caused by the behaviour of gfortran 4.6.x which is currently used by Fink.

If your problem is related to the ARP/wARP issue then you should be able work around it by installing CCP4 from the DMG file provided on the CCP4 site. This DMG file contains executables compiled by the Intel Fortran compiler.

Saul Hazledine

Structure problem

From: #HEW KAI LI KELLY#
Date: 13 September 2011 09:55

Hi,

I am facing some problems in solving my structure now, so I am wondering if anyone is able to give me any tips and tricks on this matter.

My protein-DNA complex structure diffracted to 1.5A. There are 4 missing residues, 2 on each terminal. There is no twinning in the data. The angles, the bonds, the rotamers and the Ramachandran plot are okay too. I am using molecular replacement for the phasing and the sequence homology between my protein and my homologous model is 33%. The electron density map for the protein looks very nice and there is also nice density for the DNA. Rfree converged from the initial 39%. However, Rfree refused to go down any further and it's still around 30-31%. Does anyone have any suggestions for me? Thank you in advance!

Warmest Regards,

Kelly Hew

----------
From: Eleanor Dodson


Are you absolutely sure of the spacegroup?

Eleanor

----------
From: Vellieux Frederic


Hi there,

In crystallography there are so many places where you can have problems (and need to solve these problems) that I cannot list them all.

"There is no twinning in the data" - you probably mean "the data does not seem to indicate the presence of twinning but there might be twinning"; what about the space group ? What about your crystal appearing to have one space group for one component (example the protein) but the space group for the other component (e.g. DNA, which could be partially disordered, that happens) being different - and you have processed the data in the apparent space group for the protein ? The crystal could contain the protein:DNA complex plus one of the components needed for the whole thing to pack (and form the crystals); the model might be sufficiently different from "your" structure that all the loops plus a good part of the core is improperly positioned - or you have domain and subdomain "motions" etc etc.

There is one symptom: R-free seems to be stuck. What the reason is for this is unknown. There are cases in the literature where molecular replacement leads to this behaviour and where the crystallographers have to use experimental phasing (and understand what the problem was with molecular replacement afterwards, when the structure is solved).

Fred.

*
*
*
*
*
*

----------
From: Ed Pozharski


And you have built the DNA already, right?


--
Oh, suddenly throwing a giraffe into a volcano to make water is crazy?
                                               Julian, King of Lemurs

----------
From: Kay Diederichs

density for the DNA. Rfree converged from the initial 39%. However,

Hi,

pls check out <http://strucbio.biologie.uni-konstanz.de/ccp4wiki/index.php/Refinement>

To get more specific help, you'll have to tell us much more - number of residues and bases, spacegroup, ... (at least) everything that would end up in "Table 1" of your paper describing the structure, and in the header of the PDB file.

HTH,

Kay
--
Kay Diederichs              .



----------
From: Yuri Pompeu


Just echoing what has been said.
I would make sure you have the right space group.
It may be worthwhile tyring to find a MR solution in different space groups with different compositions.
Another imporatant thing is how complete is your model?
Do you have all the protein and DNA modeled in? How many waters (you should see plenty at 1.5 A)
How good is the difference map?
These are all things that should be checked before panic sets in....

Cheers

refmac and DNA bond angles

From: Gregory Bowman
Date: 15 September 2011 00:10

I'm running into some geometry problems with my DNA model after refinement with refmac (version 5.5.0109), and would appreciate any feedback. 

The problem is that the angles for many of the glycosidic bonds are 2 to 4 degrees off of the ideal values, and so are several standard deviations outside what's expected. Our data (from MR) is ~2 Å, and the density is well defined for the DNA. I was thinking that perhaps refmac was not recognizing my DNA and so just taking the poor geometry from one model and enforcing it for the next, but in the log file it does not indicate that a new molecule has been found, and I do not feed in a cif file from previous runs. Another indication that refmac recognizes the DNA bases is that it says it is renaming them. The DNA bases are named DA, DC, DT, DG, and the header of the output pdb file says:

MODRES       DC B    1  Cd                                              RENAME
MODRES       DC B    2  Cd                                              RENAME
MODRES       DA B    3  Ad                                              RENAME
MODRES       DT B    4  Td                                              RENAME
(etc...)

In the output, strangely (to me), it is actually not renaming them. That is, it is keeping the "DC" etc names.

In coot, I'm measuring the glycosidic bond (O4'-C1'-N9) to be 104.6, whereas in AD.cif, it is listed as 108.4.

The bond lengths are fine, and there are no distortions of the protein model.

Thanks,

Greg

----------
From: Gregory Bowman

OK, I updated ccp4 to 6.2.0 (and refmac to 5.6.0117) and now the angles coming out close to the ideal values. Also, I see now the the "MODRES" in the pdb header from before was tell future refmac runs to rename those residues - which confused the newer refmac version as it thought the DT (renamed Td) were somehow "DY". By just deleting these MODRES lines it was fine.

--
Department of Biophysics
Johns Hopkins University
302 Jenkins Hall
3400 N. Charles St.
Baltimore, MD 21218

glycerol on three-fold axis

From: Jacqueline Vitali
Date: 14 September 2011 20:03

Dear colleagues,

Has anyone seen glycerol on a three-fold axis?

THis is possible as the glycerol can be disordered but I want to know if there is an actual case.

Any information would be greatly appreciated.

Jackie Vitali

Ramachandran plot difference between Coot and Morprobity (or Phenix)

From: Xiaopeng Hu
Date: 2011/9/30
Dear all,

I just notified that there is a big difference between the Ramachandran plot analysis results produced by Coot and Morprobity (or Phenix). For the structure I am working now, Phenix(Morprobity) gives out Ramachandran outliers 0.2%, favored 95.2%, whileas Coot gives out Outliers 1.24%, Allowed 5,14% and Prefered 93.62%.I am wondering if there is a simply explain for the difference which I don't know? Or I just made some silly mistakes?

Best wishes,

xiaopeng

----------
From: Edward A. Berry
What does ProCheck say?

----------
From: Frank von Delft

Oh god no don't ask Procheck , its Rama plot is a complete disaster zone - for one thing, it's * ancient*. Doesn't discriminate between amino avoid types. Grrr. To be avoided at all costs.

Sent from tiny silly touch screen

----------
From: Mario Sanches
Hi Xiaopeng,

If you are feeding both programs exactly the same file then you are not doing anything wrong. Notice that, if you calculate the Ramachandran plot on Molprobity and then you go on coot, do a round of manual refinement, and only then calculate the Ramachandran, then that can be the source of the difference. But I would bet that it is because different programs have their own definition of what is allowed and what is an outlier. 

I personally use the Ramachandran plot on coot just to guide my refinement, but use another program (Molprobity is pretty good) to do a thorough validation.

Good luck,

Mario Sanches

--
Mario Sanches
Postdoctoral Researcher
Samuel Lunenfeld Research Institute
Mount Sinai Hospital
600 University Ave
Toronto - Ontario
Canada
M5G 1X5
http://ca.linkedin.com/in/mariosanches

----------
From: Nat Echols

It's not you - I believe that Coot and Molprobity/Phenix both use the Richardson lab's data, but this isn't the first complaint I've heard about discrepancies in the statistics, so I suspect that the cutoffs are slightly stricter in Coot.  Will check this later today.

-Nat

Workshop in Athens

INTERNATIONAL

WORKSHOP

on

Macromolecular crystal growth and optimisation methods

 

National Centre for Scientific Research "DEMOKRITOS"

Athens, GREECE

 

31^st October  - 3^rd November  2011

 

 

Tentative List of Speakers

 

Professor Naomi Chayen, Professor of Biomedical Sciences, Imperial College London, United Kingdom. In charge for the E.U. "TOPCRYST" project for Imperial College

The joys and challenges of crystallising proteins

 

Professor John R. Helliwell, Professor of Structural Chemistry and Chair, School of Chemistry, University of Manchester, United Kingdom

The lessons of diffraction resolution and the study of crustacyanin

 

Professor E. Eliopoulos, Professor in Biochemistry, Dept of Agricultural Biotechnology, Agricultural University of Athens, Athens, Greece

Bioinformatics as a tool for crystallisation of membrane proteins

 

Mr. Fabrice Gorrec, Automation Crystallization Scientist, Structural Studies, MRC Laboratory of Molecular Biology, Cambridge, United Kingdom

Robotics, procedures and innovations for macromolecular crystallisation

 

Dr. Demetres D. Leonidas, Associate Professor of Biochemistry, Dept. of Biochemistry and Biotechnology, University of Thessaly, Greece.

High-throughput structure based drug design: crystallisation

 

Dr. Irene Margiolaki, Lecturer, Department of Biology, Section of Genetics, Cell Biology and Development, University of Patras, Greece and Visiting Scientist, European Synchrotron Radiation Facility (ESRF), Grenoble, France

Macromolecular powder diffraction: current status and future prospects

 

Dr. Kyriacos Petratos, Principal Researcher, Protein Structure &  Function division, IMBB-FoRTH, Heraklion, Greece

Experimental phasing of diffraction data

 

Dr. Roberto Steiner, Principal Investigator, Randall Division of Cell and Molecular Biophysics Group Leader, King's College London, United Kingdom

There's more than one way to skin a cat (and to crystallize proteins): simple alternative strategies that worked for us (hopefully for you as well)

 

Dr. Marcus Swann, Cchem, MRSC, Business Manager, Farfield Group Ltd, Manchester, United Kingdom. In charge for the E.U. "TOPCRYST" project for Farfield Group Ltd

Characterising biomolecular assemblies using Dual Polarisation Interferometry

 

Dr. Emmanuel Saridakis, Research Fellow, N.C.S.R. "Demokritos", Institute of Physical Chemistry, Athens, Greece. Co-ordinator of the E.U. "TOPCRYST" Project

Dual Polarisation Interferometry as a diagnostic tool for protein crystallisation

 

Professor Socrates Tzartos, Professor of Immunobiology, Dept. of Pharmacy, University of Patras, Greece & Head of the Laboratory of Molecular Neurobiology and Immunology, Department of Biochemistry, Hellenic Pasteur Institute. Athens

Membrane proteins: the paradigm of nicotinic acetylcholine receptors in muscle and nerve

 

Dr. Spyros E. Zographos, Researcher, National Hellenic Research Foundation , Institute of Organic & Pharmaceutical Chemistry, 11635 Athens, Greece

The purification of proteins: Objectives and strategy, choice of source, purification methods (an overview of chromatographic methods), and protein purification examples

 

Dr. Lata Govada, Research Associate, Crystallisation Group, Biomolecular Medicine, Imperial College London, United Kingdom

When we twist... so we twist ... for better diffracting crystals

 

Dr. Sahir Khurshid, Research Associate, Crystallisation Group, Biomolecular Medicine, Imperial College London, United Kingdom

Theory and practice of protein crystallisation

 

 

The Workshop will also comprise demonstrations of Dual Polarisation Interferometry, crystallisation setup methods and crystallisation robotics.

 

Lunch and some dinners will be provided.

Limited financial support will be available to the attendants.

 

Limited to 20 participants. Applications to be sent to:

Dr. Emmanuel Saridakis

Institute of Physical Chemistry, N.C.S.R. "Demokritos"

Athens 15310, Greece

 

esaridak@chem.demokritos.gr

fax: +30-210-6511766