This web page was produced as an assignment for Genetics 677, an undergraduate course at UW-Madison

Domains in CDH23 Homo sapien protein

Picture
Using SMART, cadherin repeats (CA) were the only domain found in the human CDH23 protein (final cut product) with statistical significance. The threshold was preset by SMART, and other domains were listed, but neglected for our study. The end E-value was 3.78e-11 for cadherin domains (typical significance threshold: E<1*10^-5). The figure above indicates the residue locations of each cadherin domain, as well as the likelihood of that region to be the domain in question (E-value). The E-value is a measurement that determines the chance of background noise, or random alignment, for two blasted sequences. Below, a protein domain map from SMART is displayed. These 4 cadherin domains were confirmed by ExPASy Prosite Scan.

Picture
This diagram roughly maps the cadherin repeats within the 530 amino acid protein. 'Low compositional complexity' is highlighted in pink and was determined by the SEG program

Background info on the Cadherin domain

Cadherin proteins are glycoproteins (have lipids incorporated into their composition) which are calcium ion dependent in mediating cell-cell adhesion. Cadherin domains often exists in multiple repeats in cadherin proteins in extracellular space. They are thought to chaperone contact between cells upon calcium binding. Gene ontology of this protein, as identified from its cadherin domains, include the following:

- GO process:homophilic cell adhesion (GO:0007156)
- GO component:membrane (GO:0016020)
- GO function:calcium ion binding (GO:0005509)

These will be discussed in detail in the protein ontology page. (EMBL-EBI GO Finder)

Domains in the Homo sapien cadherin-like 23 protein


Picture
































A total of 26 cadherin repeat domains were found within the human cadherin-like 23 protein that we have been using to compare against other model organisms. Using hydrophobic analysis a transmembrane segment was also found at the end of the sequence. This may indicate the glyco- attachment site of this glycoprotein, suggesting that all cadherin repeat domains of the protein are localized to the extracellular space in vivo.

All detected cadherin repeats met the significance threshold (stated above) for attaining significance as a domain.

Below we have analyzed 4 isoforms of this particular protein, to determine which, if any, domains are missing.
Picture
example of a cadherin repeat domain found in the sequence analyzed.
Picture
Another illustration of a cadherin domain. Pfam indicates 24 cadherin domain hits, whereas SMART indicates 26. The difference is likely to be attributed to algorithm discrepancies.

Protein interactions within cdh23

Picture
A network interaction between cadherin 23 and other related proteins was made using STRING. USH1C, or harmonin, has been identified as a potential partner in function with CDH23. USH1C is known as the "Usher Syndrome type 1-C" protein, because its loss of function is known to cuase this subtype. USH1C is suggested to be involved in protein-protein interaction.

I would suggest that studies investigating the possible function of USH1C in binding mutiple cells' CDH23 together be undergone. See protein network interactions page for more info.

Isoforms of CDH23 Protein and Domain Variances

We used the SMART domain identifier program to further analyze discrepancies in domains between wild type cdh23 and isoforms. We were able  to find discrepancies in cadherin domain numbers among isoforms:

-Isoform 3: 12 cadherin domains
-Isoform 4: 9 cadherin domains
-Isoform 5: 3 cadherin domains
-Isoform 6: 5 domains

Further research could identify specific cadherin domains missing from the original protein, and a comparison of pathogenic isoforms could shed light on whether a domain is necessary for normal function, or if a certain range of domains seem to be correlated with a pathogenic defect.
References:

SMART: http://smart.embl-heidelberg.de/

Expasy: http://ca.expasy.org/prosite/

EMBL-EBI GO finder: http://www.ebi.ac.uk/QuickGO/GProtein?ac=A5D6V9

String: http://string-db.org/newstring_cgi/show_input_page.pl?UserId=122XyLCvXqZ_&sessionId=iqPHGxezn071
Ben Hofeld, [email protected], last updated: 5.15.2010, Link to course page:www.gen677.weebly.com