3D structure of Lysine deficient protein kinase present in humans(WNK4)

WNK, s are the protein kinase super family with no lysine residue and are newly identifiable proteins. These proteins have unique replacement of lysine that is involved in the process of ATP binding and phosphorylation that involves phosphoryl transfer . WNK, s occurs in multicellular organisms and their homologs are not present in unicellular organisms. Serine/ threonine protein kinase is an important member of the WNK family and is encoded by WNK4 gene. WNK4 is also known as PHA2B and PRKWNK4. The orthologs of WNK4 protein are mouse (Mus musculus), chicken (Gallus gallus) and lizard ( Anolis carolinensis). The paralogs of WNK4 protein are WNK1, WNK2, and WNK3. WNK4 proteins play an important role in kidneys. These proteins play an important role in Na+ and K+ ions homeostasis by controlling these ion channels in the cell membrane . These channels are involved in transportation of Na+ and K+ ions into and out of the cell, present mainly in kidneys. Na+ channels help in maintaining the  balance of Na+ and their transport to kidney cells and then further into blood, while K+ channels , help in transport of excess K+ by a process known as secretion. WNK4 protein is present mainly in the distal convoluted tubule of kidney and it is part of tight junction. It is mainly expressed in kidney, colon and skin . Mutation in this protein causes inherited hypertension and hyperkelemia (elevated level of K+) disorder known as pseudohypoaldosteronium type II also known as Gordon’s syndrome .

Sequence retreival from NCBI

Protein sequence of WNK lysine deficient protein kinase 4 in humans is retrieved from NCBI (National center for biotechnology information) in FASTA format.TheAccession number of the protein present in NCBI is AAI36665 and total consisting of 1423 total amino acids.

3D structure prediction by SWISS MODEL
To achieve the 3D structure efficient computer programs are required. A huge gap between the protein structures and available protein sequences developed the need of automated, stable and reliable modeling methods . The idea of an internet based automated modeling facility with integrated expert knowledge was first implemented 10 years ago by Peitsch and co-workers and formed the starting point for the SWISS-MODEL server . Swiss model is reachable on World Wide Web.Three models(template sequences were identified from the swiss model but on the  basis of highest sequence similarity model 2 was chosen.
On the basis of sequence identity model 2 was best having sequence identity score of 80.49% and this template sequence was lysine deficient domain of WNK1 present in ratus norvagicus.

Chimera results
for 3D structure visualization chimera tool was used and got the following results  for all the three models.

3D structure of the best model on the basis of RMSD value(Model 3)

Models	RMSD Value		Sequence length
Model  1	0.062		264 269
Model  2	0.062		263 263
Model  3	0.061		271 278

(Table showing RMSD value)

SAVES server Results

SAVES server is a structure analysis and verification tool consisting of six different programs to evaluate the 3D structure of protein whose structure is not predicted yet. To evaluate the 3D structure protein i used following programs.Ramachandaran plot

ERRAT

VERIFY3D

Results  for the best models are given below.

Ramachandaran plot for best model

ERRAT results for best model

Main chain parameter results

3D structure of lysine deficient protein kinase (WNK4 in humans)

Conclusion

The 3D structure prediction is very important to know about the molecular functions of protein. Structural analysis of lysine deficient protein kianse was done using Swiss Model. .  Further experimentation and analysis of structure is needed to study the molecular function of the protein in more detail. The study of the bond lengths, bond energies, bonding interactions and information about the 3D structure of lysine deficient protein kinase is needed to be studied in more detail to get more information about its function.

Homology modelling Servers

 Swiss Model

Homology modeling of proteins involves four steps: template selection, target-template alignment, model building, and model evaluation. To achieve the 3D structure efficient computer programs are required. A huge gap between the protein structures and available protein sequences developed the need of automated, stable and reliable modelling methods . The idea of an internet based automated modelling facility with integrated expert knowledge was first implemented 10 years ago by Peitsch and co-workers and formed the starting point for the SWISS-MODEL server. Swiss model is reachable on World Wide Web.To get the Swiss Model results first step is to submit the amino acid sequence of protein whose 3D structure is not predicted yet. After submitting the amino acid sequence the BLAST search will be run against the sequences of known protein structures from the EXPDB which is the library of Swiss Model. After this procedure choose the three best models (target sequences) for the protein whose structure is not predicted yet. The 3D template is chosen on the basis of sequence identity with the target protein. For better results the sequence identity must be greater than 50%. Models having highest sequence similarity are chosen as template for our target protein. Download the target sequence for best three models in in PDB format. Swiss model provides you the four lettered code to find the template sequence for targets being chosen on the basis of sequence similarity. By entering this code in protein data bank (PDB) we will get the template protein sequences for the target.

CPH Server

CPH modeling server is a protein homology modeling tool. It is used for the 3D structure prediction of the target protein whose structure is not predicted yet. To find the template of our protein of interest first step is the submission of query sequence in the form of FATA format. For this purpose the sequence is pasted in the text field given.  After submitting sequence, the searching for the template is being done by BLAST and at the end website updates the results. In CPH the template recognition is based on profile profile alignment.

Hits are obtained from PDB using profile matrix that is generated by psi BLAST against a UNIPROT database. After the hits were in PDB data base, the PDB entry (4 lettered codes) and chain identifier are listed for the template to construct the three dimensional structure . The sequence identity is also given for the template of our target protein. The results from the CPH Model are given in the (Figure 2.1.2).

Besides accuracy the most important feature of the CPH models is its speed and it will give results in hardly 20 to 25 minutes.