CMD1

There was no available Chime file depicting the three dimensional structure of yeast calmodulin. Below is a Chime file for calmodulin in Rattus rattus, whose calmodulin sequence is 60% identical and 34% similar to yeast calmodulin (SGD, 2002; <Alignment of CMD1/YBR109C and PDB 3CLN> Caution: this link will take several minutes to load.)

Figure 2: 3D structure of rat calmodulin in Chime format. (PDB, 2002; <Structure Explorer: 3CLN>) Click here to download the Chime plug-in.

First discovered in 1970, calmodulin is ubiquitously expressed in vertebrates, where it regulates a number of proteins and processes. According to Ohya and Botstein, calmodulin interacts with over 20 proteins, including "several metabolic enzymes, protein kinases, a protein phosphatase, ion transporters, receptors, motor proteins, and cytoskeletal components" (1994). Stanford's Saccharmomyces Genome Database summarizes calmodulin's essential cellular funcions in terms of the hierarchy set forth by the Gene Ontology Consortium: biological process, biological function, and cellular component. Using these categories as a starting point, I have assembled a brief description of the why, what, and where of calmodulin in S. cerevisiae

Biological Process:

One process in which calmodulin is involved is mitosis. The spindle pole body (SPB) is an organelle responsible for the nucleation of microtubules. This is a function equivalent to that of the centrosome in animal cells. One component of the SPB is a protein called Spc110p (aka Nuf1p), which calmodulin binds to at the central plaque of the SPB. This apparently anchors Spc110p to the spindle pole during mitosis. Calmodulin and Spc110p mutants both exhibit defective spindle formation and loss of microtuble attachment to the spindle pole (Cyert, 2001). Calmodulin also performs a role in budding, by binding Myo2p, a type of myosin that is thought to have play a role in polarized growth by transporting secretory vesicles to the bud tip. Mutations in CMD1 and MYO2 both lead to similar defects in bud emergenc. Calmodulin is also required for endocytosis. It is known to interact with Myo5p and Arc35p. Yeast carrying the mutant allele cmd1-226 have been shown to display defects in endocytosis, while not interacting with Myo5p. Interestingly enough, the previous three functions are not dependent upon calmodulin's ability to bind Ca2+ (Cyert, 2001). Finally, calmodulin is involved in Ca2+-dependent signalling pathways that are triggered by stress and environmental changes. The target proteins for these interactions are calcineurin, a Ca2+ calmodulin-dependent protein phosphatase, Cmk1p, and Cmk2p, both of which are calmodulin-dependent protein kinases. Whereas the function of Cmk1p and Cmk2p are not well understood in terms of signalling pathways, calcineurin is known, in yeast, to be primarily involved in stress response, regulation of Ca2+ homeostasis, and cell cycle regulation (Cyert, 2001).

Biological Function:

Calmodulin changes conformation when bound to Ca2+, allowing it to bind to other target enzymes. Each calmodulin contains 4 structures called EF-hands, that allow Ca2+ binding to occur. However, yeast calmodulin can only bind three Ca2+ ions at a time, whereas vertebrate calmodulin can bind four ions. Although Ca2+ binding is an important function of calmodulin, calmodulin's essential role in yeast cells is not dependent upon its ability to bind Ca2+. Yeast cells that have been mutated to express Ca2+-binding-defective calmodulin show "minimal disruptions in growth and morphology" (Cyert, 2001).

Figure 3: P. tetraurelia calmodulin molecule shown with bound calcium ions, depicted as grey spheres. The image is a screen capture taken from a Millenium STING Chime applet. PDB ID = 1EXR (PDB, 2002; <Structure Explorer>).

Cellular Components:

SGD lists the locations of calmodulin activity to be incipient bud sites, the cytoplasm, the central plaque of the SPB, bud tips, bud necks, and schmoo tips, which is a region of polarized growth (2002).

YBR108W:

YBR108W is an ORF located on yeast chromosome 2, near the genes ALG1 and PHO88. I chose it because of its relative proximity to my favorite yeast gene, CMD1, as you can see in the map below.

Figure 4: Map showing features around YBR108W on chromosome 2. (SGD, 2002; <Chromosomal Features Map>).

YBR108W's sequence is 2547 kb long (SGD, 2002; <Sequence in FASTA format>). Using the Mammalian homolog option at SGD, I retrieved the following list of alignments, most of which are less than 50% identical to the portion of YBR108W that they are aligned (2002; <SGD BLAST Mammal>).

Figure 5: Screen capture of mammalian homolog results. For the most part, these alignments did not have significant E values (SGD, 2002; <SGD BLAST Mammal>).

Using the sequence provided at SGD, I performed a BLASTn search (Request ID: 1033945276-08698-20824). The two main hits (shown as long red stripes below, are the NCBI entries for YBR108W). It seems interesting that several 20-30 kb sequences within YBR108W each produced a number of alignments (NCBI, 2002; <BLASTn>). This lead me to wonder if there might be any STS's found within the YBR108W sequence. However, I did not get any matches using Electronic PCR (NCBI, 2002; <UniSTS>).

Figure 6: Screen capture of BLASTn hits. The third line of blue bars from the left had alignments with the following sequence: acaaccacaacagcaacaacaa (SGD, 2002; <BLASTn>).

SGD predicts the following protein structure for YBR108W (2002; <YBR108W Single Page Format>).:

MGFWENNKDSITSGLKSAGKYGYQGTKYVAKTGYKASKKHYNNSKARRER
KSGKKNSSDEEYDSEDEMEYERKPTDIRSLKDPKSFPPPPLKPGQKTYTG
QQQQQMPNGQASYAFQGAYQGQPGAGSTEQSQYAQPQYNQYPQQQLQQGV
MPQQQQLQQGVVPQQPPIYGEQVPPYGSNSNATSYQSLPQQNQPQNAIPS
QVSLNSASQQSTGFVSQNLQYGTQSSNPAPSPSFQNGLQCHQQPQYVSHG
STNLGQSQFPSGQQQQPTTQFGQQVLPSPAQPQQQQQGQPLPPPRGQVIL
PAPGEPLSNGFGQQQQQQQQQQQPLNQNNALLPQMNVEGVSGMAAVQPVY
GQAMSSTTNMQDSNPSYGASPMQGQPPVGGQPPVPVRMQPQPPQPMQQGN
IYPIEPSLDSTGSTPHFEVTPFDPDAPAPKPKIDIPTVDVSSLPPPPTHR
DRGAVVHQEPAPSGKIQPNTTSSAASLPAKHSRTTTADNERNSGNKENDE
STSKSSILGHYDVDANIMPPPKPFRHGLDSVPSEHTTKNAPERAVPILPP
RNNVEPPPPPSRGNFERTESVLSTNAANVQEDPISNFLPPPKPFRHTETK
QNQNSKASPVEMKGEVLPGHPSEEDRNVEPSLVPQSKPQSQSQFRRAHME
TQPIQNFQPPPKPFRRSQSSNSSDSSYTIDGPEANHGRGRGRIAKHHDGD
EYNPKSENSTENGRLGDAPNSFIRKRAPTPPAPSRSEKLHEGTITSEVDS
SKDANKYEKSIPPVTSSIQAQQSTKKAPPPVVKPKPRNFSLKANEYPKEL
TREATGQDEVLNSITNELSHIKLRKTNVNLEKLGGSKKVKTLALFPQI

This sequence contains 848 residues. I performed a BLASTp search using this sequence, and received the alignment shown in Figure 7. The only significant hit was the entry for the protein sequence predicted for YBR108W. The second largest hit was for Zinc metalloprotease in Streptococcus pneumoniae (NCBI, 2002; <BLASTp>; Request ID: 1033962783-025846-19786).

Figure 7: Screen capture of alignments retrieved from BLASTp search of YBR108W's predicted protein sequence (NCBI, 2002; <BLASTp>).

I submitted the protein sequence to PREDATOR for prediction of secondary structure (PBIL, 2002; <PREDATOR NPS@>). It shows that the predicted structure is over 85% random coil, and about 10% alpha helix.

Figure 8: Screen capture received from PREDATOR analysis of secondary structure (PBIL, 2002; <PREDATOR NPS@>).

My next step was to enter the predicted amino acid sequence into a web program that produces a Kyte-Doolittle hydropathy plot. The graph shown in Figure 9 is the plot that the sequence produced. According to interpretation information provided on the page, strong negative peaks are an indicator of a globular protein (Johnson, et al. 2002; <Kyte-Doolittle Entry Form>).

Figure 9: Screen Capture of Kyte-Doolittle Plot produced from predicted amino acid sequence (Johnson, et al., 2002).

Since my BLAST searches failed to reveal any close matches, I have chosen Kyte-Doolittle analysis as the source of my speculation as to the nature of YBR108W. Since the Kyte-Doolittle plot indicated the presence of a globular protein, I took amino acids sequences for the proteins hemoglobin (NCBI, 2002; <Hemoglobin sequence>) and calcineurin b subunit (NCBI, 2002; <Calcineurin sequence>), which are of known globular structure, and produced Kyte-Doolittle plots, just to provide bases for comparison. There is a lot of similarity between the shapes of the Kyte-Doolittle plots of hemoglobin and calcineurin b and YBR108W's hypothetical protein (Fig. 9). Upon this evidence, I conclude that the protein which YBR108W encodes is quite possibly a globular protein. However, at this point I am unable to make any further claims.

Figure 10: Kyte-Doolittle plots for calcineurin and hemoglobin I, two globular proteins. The Kyte-Doolittle plot shows negative peaks, typical of globular proteins (Johnson et al, 2002).

References:

CMD1: My Favorite Yeast Gene