D.S.Soriano's (photo ca. 1991)
|Peptidomimetic/ Molecular Informatics Site|
-I am a chemistry professor with U. Pittsburgh-Bradford in PA., U.S.A. -
U. Pittsburgh- Bradford
-I serve as Managing Editor of the "J.Drug Education and Awareness"-
-I serve as a consulting editor for Bentham Science Publishers' "Letters in Drug Design and Discovery"-
(My technical consulting services are available in all aspects of computer-aided
drug design and analysis as well as hardware /clustering/ networking)
For a look at one of our recent publications:
"Beta- Catenin Complexed With TCF-4 and Some of the Processes
Involved with Computer-Aided Drug Design and Protein Analysis"
(T. Unger, R. Lawrence and D.Soriano , "J. Undergraduate Chemistry Research"
2003, 2, 39. )
"A Molecular Informatics Analysis of a κ Opioid Receptor Model and Two Ligands"
(J.Skinner, A.Smith and D.Soriano submitted "J. Undergraduate Chemistry Research")
Rendering of Human Aldose Reductase Active-Site with
Inhibitor / Key Amino- Acid Interaction Shown:
Trp20, His110, Trp111 and Leu300 converging on inhibitor.
(Current Modeling Project in my Molecular Informatics Class
see: J. Med. Chem. ,2000, 43, 1062-1070. )
Image of human Aldose Reductase with "Alpha Site Finder' option of 'MOE'
Image of the hydrophilic/ hydrophobic 'Alpha- Site Finder' Prediction
for Human Aldose Reductase Active- Site (coenzyme and site occupied by zopolrestat inhibitor in pink; white=hydrophobic , red = hydrophilic predicted binding sites).
-Modeling Study of the active-site is being conducted with C.C.G.'s "Molecular
Operating Environment" (M.O.E.) -
(we thank the"Chemical Computing Group" for the use of their software on our campus!)
alpha-helix sketch from Linus Pauling's 1948 notebook
Purpose/ Rationale of Site: To assist students in bridging the gap from organic chemistry to computational aspects of polypeptide secondary structure in biochemistry and to introduce MOLECULAR INFORMATICS. The backbone conformations of peptides (e.g., alpha-helix, beta-sheet, beta- turn, etc.) provide critical templates for the 3D structures when interacting with the given receptor-site, but the overall shape and intrinsic stereoelectronic properties needed for molecular recognition, signal transduction, etc. depends on the arrangements of the residue side-chains in 3D space ( see : Tet. Letters ,43, 2137, 2002). Care must be exercised in setting up, modeling and interpreting even the simplest structures. Problems are selected and developed for my Computational Chemistry (Chem. 1404) class. I hope to demonstrate the use of modeling techniques with selected peptides/ peptidomimetics in order to illustrate the above points. For a nice introduction to the peptide/peptidomimetic secondary structural problem visit:
Students can see how I set up peptide modeling problems with the Wavefunction modeling program. I place selected output molecular conformers in a zip-file which they can download (here) and study with the downloadable "Deep View" program.
It is important for students to develop sound computational skills at the secondary level of polypeptide investigation before heading on to tertiary analysis of proteins. Molecular modeling is now an integral technique in molecular pharmacology, medicinal chemistry research and an essential component of the rapidly emerging disciplines of chem/bio- informatics and computational biology. We also hope, in the near future, to present a database of peptide/peptidomimetics conformational energetics based on Pittsburgh Supercomputer Center calculations .
For an introduction to molecular modeling (for organic students using Wavefunction products) , I suggest:
"Introduction to Molecular Modeling" ( cd-rom) , D.S.Soriano , J.Blair and J.Slimick
Available from: Nova Science Publishing (Huntington, N.Y.) ISBN: 1590330935 (2002).
Students are introduced to Glaxo's "Deep View" freeware protein modeling software (download below) using selected polypeptide tutorials. A good way to learn the art and science of protein modeling! This is done with Glaxo permission.
"Deep View " Website:
I. DOWNLOAD THE "DEEP VIEW" SOFTWARE, .PDB DATABASE
(PREPARED BY ME) AND SIMPLE TUTORIAL WHICH WILL GET YOU
STARTED WITH "DEEP VIEW" :
( BE CERTAIN TO START WITH 'README!' FIRST -UPON DOWNLOAD)
II. For a good source of red/cyan 3D glasses (for 3D view):
This is Ala-Gly-Ala Tripeptide / Zwitterionic Form
(above anaglyph image was constructed with "Biodesigner" software- download below)
Sometimes, you will want to take a break between calculations. Below is the
tripeptide gly-ala-gly which was built and saved as a space-filling model.
I then exported the image to Microsoft's "Photo Editor" and created a stained- glass
III.Visit Professor Gail Rhodes' site for an excellent introductory tutorial for "Deep View":
IV. (optional) : Download and install the 3D plug-in below:
After successful upload of the plug-in, you will obtain a user-interactive
3D model by clicking on the link below:
(also look at 3D image at the bottom of this page!)
|Do you want a
FREE downloadable chemical drawing/ 3D package?
"ACD ChemSketch 5.0"
"ACD Chem Sketch 5.0" is the property of "ACD Labs "(Toronto, Ca.)
(software can be downloaded from this site with their permission)
V. You can also Download Piotr Rotkiewicz's "Biodesigner"
It can be used, for free,by academic people, provided the software is not
altered in any way. The rendering effort by Piotr is superb!
View .pdb files with his red/cyan anaglyph 3D. It is the best that I have seen!
Weekly Peptide Modeling & Visualization Studies!!
(utilizing Wavefunction's PC Spartan '02 )
Computational evaluations of selected peptide conformational preferences in polar solvent/gas phase
utilizing PM3 semi-empirical calculations.
( Platform Used for Calculations: Gateway PC/ Windows NT/392K ram/647 mHz. Processor)
Visit Wavefunction's web-site:
Modeling Problem #1 gly-ala-gly tripeptide
- image of the conformer generated by energy minimization (view with 3D glasses)
(starting with input of the classic alpha-helix angles)
- zwitterionic form-
-alpha helix secondary angle values:-
psi angle: -57 deg., phi angle= -48 deg. (classic angles- input)
( beta sheet angles are: psi =135 deg.; phi= -140 deg.)
calculated angles for the above conformer: (after minimization)
phi= -88 deg., psi= 39 deg. (generated values for the N terminus side of the peptide).
Be certain you can clearly identify and measure the psi and phi dihedral angles!
Use the dihedral angle measurement option of "Deep View" to verify!
Next Part of the Evaluation: a "conformational search" for the global
minimum (lowest energy conformation) was conducted with
the "Spartan Pro '02 " semi-empirical "PM3" program (gas-phase, i.e. non-polar environment)
-Approx. 2 hrs (wall clock) time of computation was required-
-97 conformers were outputed-
-the conformers were ranked in order of energies-
-a 9.9. Kcal difference from #1 to #97 was reported-
Conformers 1(most stable), 5 and 97 (least stable) are available
for you to Download below : (view them with the "Deep View" program
and measure and verify the reported phi and psi angles!!
shown below is the global minimum (#1) -use 3D glasses to view:
(angles are measured from N terminus!)
psi angle: 120.82 deg.
phi angle: -99.82 deg.
Heat of Form.: -147.60 Kcal/ mol
-the global minimum does not resemble initially- minimized conformer-
Modeling Problem #2 ala-pro-ala
- location of global minimum( lowest energy shape), however, not trivial-
-built structure, minimized, conducted PM3 (gas-phase)
-1.5 hrs. (wall clock) calculation-
-87 conformers reported-
-ranked in energies order-
-download the zip.file (below) which contains
conformers #1, 3 & 82-
-study the conformers and measure dihedral angles-
-would you have predicted #1 as the global minimum?-
-simple peptide still illustrates problem of rotational
degrees of freedom analysis-
-question: why did you select #3 to include?
-answer: no particular reason; just a plausible conformer for the molecule. Good for comparisons!-
Modeling Problem # 3 Thyroliberin TSH/PRL Releasing Hormone
see: J. Med. Chem., 15, 479-482,1972 Nature 256, 750, 1975 Biopolymers ,13, 2615, 1974
-tripeptide containing L-pyroglutamyl-L-histidyl-L-prolinamide structure-
-conformational analysis investigated , in water and polar aprotic solvent, with NMR-
-therefore, some experimental evidence for conformational preference in polar solvents-
here is a plausible conformer suggested by PM3 (2.5 minute calculation):
-we will model the free base and protonated forms (imidazole ring as free- base or protonated) since conformer which is "bio-active" could be either or both-
-we will also model current thyroliberin conformationally -restricted analogs in this tutorial (see: J.Med. Chem., 39,1571-74,1996)-
-a search was then initiated for a stable of conformer including the global minimum-
-PM3 ( gas-phase) calculations were employed and calculations took approx. 30 hrs. for the protonated form-
Results ( protonated form):
-90 conformers reported. Heat of formations ranged from: 31. 38 (global minimum) to 41.37 (least stable conformation) Kcal/mol.-
-Of the first 13 conformers, 11 have a trans relationship between proline and pyroglutamyl rings. -
-An extended conformation is exhibited for the histidyl residue with folding back of the residue toward the prolyl ring-
-This prediction is in full agreement with the nmr results reported in "Biopolymers", 13, 2628, 1974.-
Shown here is an alignment of the first 5 lowest energy conformers with the histidyl ring selected for molecular overlay:
Download the file below which contains the first thirteen (13) conformers( most stable) for you to view with "Deep View":
(Content of the .zip file Containing the 13 Lowest Energy Conformers )
|001.pdb||31.38 Kcal/mol||(Global minimum)|
|005.pdb||33.66||(13th most stable conformer of 90 reported)|
you may wish to align two at a time with the "BioDesigner " software as well!
-results for the free-base conformational search will be reported the first week of April '02-
-I will also post some evidence that the bio-active conformation has a trans relationship between the terminal residue side-groups-
Modeling Problem #4 gly-pro-gly-arg peptide
intramolecular H-bonding? Loop?
1. Build gly-pro-gly-arg. We will look for evidence of an intramolecular "loop". Loops can initiate changes from ,for example, alpha to beta motifs.
2. Model peptide at PH=7, so arg guanidinium group
is protonated (use Spartan's advanced build to create
the protonated side-group). The peptide has a +1 net charge @
PH=7 in water.
The crude image( above) should still show you interaction
between the guanidinium proton, Gly N-term. proton
and C term. carboxylate (look in the center of the peptide).
3. I built and minimized the structure and subjected the result to PM3 "conformational search". When you study the conformers I have left for you in the zip file (below) be sure to look for and measure the distance between any groups that seem to have hydrogen bonds- Spartan takes the importance of H- bonds into account! A typical intramolecular bond is generally 1.7-2.1 Angstroms in distance. These intramolecular bonds and electrostatic interactions between (+) and (-) charges can form recognizable loops in a secondary structural pattern.
4. The Calculations took 101 hrs (approx. 4.25 days) to run on my PC. Twenty-eight (28) conformers were reported. The heats of formation ranged from -75.82 to -66.14 Kcal/mol. The conformers had area ranging from 416-441 square Angstroms. The global minimum ( molecule #1 saved in zip -file as "a.pdb") had a double interaction between the C terminal carboxylate and one of the arginine gunaidinium NH hydrogens; the carboxylate also strongly interacted with one of the N terminal NH+ protons as seen in the initially- built and minimized structure in step #1. I arbitrarily selected 6 conformers to illustrate the range of properties for the tetrapeptide. The calculations carried out were PM3 (gas-phase).
|conformer number||calculated Heat of Form. (Kcal/mol)||Calculated surface area (sq. Ang.)||Zip-file I.D.|
|1(most stable conformer)||-75.82||440.5||a.pdb|
|28( least stable conformer)||-66.13||438.7||e.pdb|
DOWNLOAD the zip-file and examine the six conformers with "Deep View":
Modeling Problem #5: Conformationally Restricted Peptidomimetic
consideration of a conformationally restricted Pro-Gly type:
-researchers at the above site reported only MM2 calculations regarding the psi and phi predicted angles.-
-I decided to carry out the analysis (conformational search) using AM1 gas-phase and aqueous environments-
approx. 1.0 of wall- clock computation was required.
-note the pertinent psi and phi angles (above)-
-15 conformers were reported-
-gas phase heat of formation spread was only 2.1 Kcal/mol with no conformational preference-
-aqueous phase , however, identified conformer #13 as the global minimum with two intramolecular H bonds-
the prolyl carbonyl oxygen, oxazoline nitrogen and carboxamide H are circled in the above image. These atoms are involved in H bonding.
-carboxamide H / prolyl carbonyl O @ 2.21 Angstroms; oxazoline nitrogen/ carboxamide H @ 2.6 Angstroms-
-there is a trade-off between proline ring planarity (as an energy cost) vs. intramolecular H -bonding in water-
Download the following zip-file with the 15 conformers. Examine the molecules for intra-H bonding and also measure the proline ring planarity vs. H-bonding when it occurs for a given conformer. Use red/cyan glasses and Deep-View+/or Biodesigner.
Download the following zip-file which contains the results with X=S. Twenty (20) conformers reported with an 11Kcal/mole preference for the global minimum with intramolecular H- bonding (aq.) !
Modeling Problem # 6 Is Five-Ring Proline Essential for Beta-Turns??
The influence of cyclic secondary amino acid ring size on beta- turn stability will now be modeled. Why was proline , a five ring, selected vs. ,say, a four or six- membered analog for inclusion in proteins ? In an important finding, Hayashi and co-workers reported NOESY nmr data which strongly suggests that the five ring is needed for creation of essential beta-hairpin turns. Without the latter, there would be no facile tool for alpha/ beta motif pattern change in proteins.
See: Tetrahedron Letters 38, 3039, 1997.
I will model the three tetrapeptides with MM/ AM1 and evaluate the possible correlation between the experimental and theoretical models.
Modeling Problem #7: The use of a peptidomimetic template to Initiate alpha-Helix Chain Growth
Modeling Problem #8: Selective Hexapeptide CCK-A Agonist Anorectic Agents
See: J. Med. Chem., 40, 4302, 1997 (coming soon!)
VI. Secondary Sequence Predictions for Peptides/ Proteins Available on the Web:
Visit the following web-site and employ the "NNPredict"
program which will predict the % helix and extended sheet for an entered
primary sequence of amino acids.
Start with this great "user- friendly"
program and then proceed to the additional sites list below.
We will later build and model selected polypeptides and compare results
with these programs.
Modeling Problem # 9: Type I/ II Turns as a Function of linker Stereochemistry
See: J.Am.Chem.Soc., 117,5169, 1995
We will model these three model systems and determine how the Beta Turn type changes as a function of the linker chain/amino acid R group stereochemistry
VII. For another nice "user- friendly" polypeptide secondary
structure predictive site ("AgaDir") :
I would advise undergrads to start with these two programs and then
progress to the Columbia U. site (below).
VIII. Columbia University maintains a web-site which contains
"PredictProtein"- a program which will predict secondary
structure for a polypeptide with 17, or more, residues- great
for many neuropeptides! (I have included instructions on how to rapidly download proteins from the Protein Data Bank ,for tertiary viewing, using "Deep View" in "readme" section).
If the 3D plug-in ( see optional step above) was installed correctly
you will be able to interrogate the user-interactive peptide below:
(also try hitting the 'reload' button- this can help!). Use the R-click on the structure to interact!)
Visit my other page for a free, downloadable nmr
tutorial. It was developed here at Pitt- Bradford with my colleagues:
We also have a Discussion Board section for you there!
David Soriano Ph.D.
University of Pittsburgh-Bradford
Contact me with questions or comments!
"The University of Pittsburgh recognizes the value and potential of personal publishing on the Internet and so allows and encourages students, staff, and faculty to produce their own personal Web pages; however, the University accepts no responsibility for the content of personal pages".
08/28/03 11:37:20 AM