Biocomputation and Bioinformatics PDF Print

Computational Design of Thermostable Proteins (Makhatadze)

Industrial enzymes are used in a variety of commercially significant applications such as detergents, starch processing, textile manufacturing, oil and gas production, pulp and paper processing, and production of baked goods, beer, wine, and dairy products. In addition to being potentially more efficient catalyst, the use of enzymes is much friendlier on the environment. However, the use of enzymes in the industrial applications has been limited because most enzymes lose biocatalytic properties at extreme temperatures that are found in many industrial processes.


proteins-figure-1                        proteins-figure-2


The technology developed by us allows rational design of the proteins with increased thermostability. Using computer algorithms we are able to predict modifications in the amino acid sequence of proteins that will lead to an increase in the thermostability. Experimental validation of the technology on several different proteins has been performed.




I. Per residue energies of interactions as predicted by computational modeling. II. The 3D-structures of the five proteins. III. Experimental characterization of the stabilities of the designed and wild type proteins. 

Biochemistry 2006 45, 2761




I. Per residue energies of interactions as predicted by computational modeling. II. The 3D-structures of the two enzymes (AcPh and Cdc42). III. Experimental characterization of the stabilities of the designed and wild type AcPh and Cdcd42 (panel A-C) and the effect of temperature on activity (panel D). VI. Activity assays of AcPh at different temperatures and stimulation of the GTPase activity of the Cdc42 proteins by the increasing concentrations of Cdc42GAP.

Black - wild types, red and/or blue – designed proteins. 

PNAS 106, 2601 (2009)

Characterization of the interaction interface between rhodopsin and transducin (Garcia)  

G-protein coupled receptors (GPCRs) comprise the largest family of eukaryotic membrane proteins that act as sensory molecules in various cellular signaling pathways. The function of these 7-transmembrane domain proteins is mediated by their intracellular protein partners, the heterotrimeric (αβγ) G-proteins. Notably, the receptor-bound G-protein complex is yet to be characterized in full atomic detail. Substantial evidence supports the notion that rhodopsin and other GPCRs may form an activation-competent complex with its G-protein partner, transducin, in the dark state. Here, we use Molecular Dynamics simulations at the μsec timescale as a powerful method to characterize the interaction interface of the complex and to evaluate the presence and structural details of dynamics in the dark-adapted state. 




Figure 1. Open-book view of the interaction interface between rhodopsin (left) and transducin (right). The location of regions of interaction between rhodopsin and the heterotrimeric G-protein are shown on an open-book representation of the complex that is split along the plane of the membrane. Blue color indicates regions on rhodopsin that interact predominantly with Gα, yellow with Gβ while green with both protein subunits. Interactions are shown only if they are observed with a probability greater or equal to 1/e in our simulation ensemble. 


garcia_figure3                 garcia_figures1 


Figure 2. Modes of interdomain interactions. Patterns of interacting residues between rhodopsin (right) and Gα (left) are further classified into four modes of interdomain interactions (a-d). Averages are calculated within each cluster and are displayed on the open-book representation. The sizes of the clusters are 41%, 16%, 20% and 23% respectively. In each cluster, the protein subunits engage using distinct structural features that are highlighted on the structure. Regions of interaction in the two subunits are colored according to the probability of contact formation in our MD ensemble. 

Translocation of an HIV-1 peptide across a lipid bilayer (Garcia)

Cell penetrating peptides (CPP) are capable of translocating through the cell membrane and deliver proteins, siRNA or other cargo into the cell. These peptides have great potential for drug delivery and biological research applications. The mechanism by which these peptides cross the bilayer has remained elusive. Computer simulations of the crossing of the HIV-1 Tat peptide CPP revealed that charged amino acid side chains interact with the lipid bilayer, causing a distortion of the bilayer and opening a transient pore. Understanding of the crossing mechanism may lead to the design of better CPPs and antimicrobial peptides.





Figure 3. The HIV-1 Tat protein in an important protein in the transcription activation of the HIV-1 virus. This protein is capable of translocating across the cell membrane.




Figure 4.  A short 11 amino acid sequence, Tat peptide, is responsible for this property. Surprisingly, this peptide has a net charge of +8 and is capable of crossing the hydrophobic lipid bilayer. 




Figure 5. Molecular Dynamics (MD) simulations of the HIV-1 Tat peptide bound to a lipid bilayer (left) suggest a new mechanism for CPP translocation. The MD simulations were performed in RPI CCNI supercomputers (right). 

The mechanism for Tat peptide translocation through cell membrane is summarized as follows:


garcia_frame1   garcia_frame2   garcia_frame3   garcia_frame4


  1. Tat peptides bind underneath the lipid bilayer surface, at the boundary between phosphate groups and carbon chains 
  2. High concentration of peptides causes phosphate groups to crowd around peptides, thus creating a charge depleted zone on the lipid surface
  3. Tat peptides around the depletion zone will reach for phosphates on the other face of the bilayer, thinning the bilayer.
  4. The peptide charged side chains bind to phosphates on opposite layer and start translocation.

Herce and Garcia, PNAS 104, 20805 (2007)

Encapsulation of Proteins Inside AOT Reverse Micelles (Garcia)

We study the structure and dynamics of proteins and peptides encapsulated in reverse micelles (RM). The highly electrostricted environment of the RM interior is treated as a model of the crowded cell environment. In these studies we performed extensive MD simulations in order to understand the nature of the interactions of the protein with RMs, the location of the protein inside RM under various salt conditions, and to identify the source of the altered protein dynamics in the confined environment. The reverse micelles are self assembled in presence of the folded protein, and the dynamics and hydration of the protein are compared with the corresponding protein in bulk water.

We have performed studies of the encapsulation of alpha helices [Tian and Garcia, BJ 2009] and ubiquitin. Here we briefly describe our results for ubiquitin. It is found that the protein in the RM is partially dehydrated relative to bulk conditions. We find that the protein dynamics are similar in both environments, and differences are restricted to specific regions of the protein. Interestingly, we found that the protein prefers to be close to the RM interface at low excess salt and for an idealized neutral head group RM. However, at high excess salt, the protein prefers the water phase of the RM. This result is consistent with free-energy calculations of the position of ubiquitin relative to the center of mass of the RM calculated using the adaptive biasing force algorithm [Henin et al., JCP (2004)]. In unbiased simulations it is found that ubiquitin binds to the AOT surfactants through the hydrophobic patch formed by amino acid chains Leu 8, Ile 44 and Val 70, and which is the same patch region that interacts with ubiquitin-binding-domains in-vivo. In turn, this hydrophobic patch is surrounded by positively charged amino acids Lys 6, Arg 42, Lys 48, His 68, Arg 72 and Arg 74. The positioning of ubiquitin inside a RM is found to be closely related to the behavior of water in the RM environment.




Figure 6. Schematic picture of the positioning of ubiquitin inside a reverse micelle (left). At the center we show the AOT headgroups at the RM interface. At right we show the water an protein content of the RM-protein system. Ubiquitin binds to the RM through its hydrophobic patch (indicated by green amino acids), which is in turn surrounded by positively charged amino acids (in red). Negatively charged amino acids are colored in blue, and water is gray.

Participants: Jianhui Tian and Angel E Garcia