Home About us Editorial board Ahead of print Current issue Search Archives Submit article Instructions Subscribe Contacts Login 

 Table of Contents  
Year : 2016  |  Volume : 12  |  Issue : 2  |  Page : 650-656

Application of intelligent algorithm in the optimization of novel protein regulatory pathway: Mechanism of action of gastric carcinoma protein p42.3

1 Department of Radiotherapy, People's Hospital of Zhengzhou, Zhengzhou, Henan Province, China
2 Zhengzhou Central Hospital, Zhengzhou, Henan Province, China
3 Department of Pharmacology, School of Basic Medicine, Zhengzhou University, ZhengZhou, Henan Province, China
4 Department of Oncology, The First Affiliated Hospital, College of Medicine of Xi'an Jiaotong University, Xi'an, Shaanxi Province, China

Date of Web Publication25-Jul-2016

Correspondence Address:
Xing’an Liu
People's Hospital of Zhengzhou, Henan Province 450003
Login to access the Email id

Source of Support: National Natural Science Foundation of China (60971110), Science and Technology Corporation Project of Henan Province (122106000042) and Open Science and Technology Corporation Project of Henan Province in 2013(132106000064). Presentation at a Meeting,, Conflict of Interest: None

DOI: 10.4103/0973-1482.151856

Rights and Permissions
 > Abstract 

Aims: This purpose of the study was to optimize the regulatory mechanism of p42.3 novel protein molecule in gastric cancer and also verified it by the use of intelligent algorithms.
Subjects and Methods: Threading method was employed to analyze structural domain characteristics of p42.3 protein. Referential proteins were gathered and formed by domain homology and function similarity. Afterwards, the possible regulatory network of p42.3 was established by analyzing the acting pathways of the referential proteins. Spherical polar coordinates stratification and stratified multi-parameter weight were used for calculation of the similarity between the referential proteins and p42.3 protein, the result of which was taken as the prior probability of the initial node in Bayes network, thus the probability of occurrence of each pathway was figured out by using conditional probability formula, and the one with the biggest probability was considered as the possible pathway of p42.3. At last, molecular biological experiments were conducted to verify it.
Results: The acting pathway with the maximum probability predicted by Bayesian probability optimizing calculation was “S100A11” – RAGE – P38 – MAPK – Microtubule–associated protein – Spindle protein-Centromere protein – Cell proliferation” which was the most likely acting pathway participated by p42.3, and has been validated by biological experiments.
Conclusions: By the theoretical analysis and experimental verification, this study confirmed that assumptions that p42.3 protein was related to the occurrence and development of gastric carcinoma, predicted and verified the acting pathways of p42.3, which will provide a new research direction of the relationship between p42.3 and gastric cancer, as well as the target therapy of gastric cancer. The algorithm in predicting the acting pathway of the protein also offers a new thought in studying new functional proteins.

Keywords: Bayesian probability optimization, intelligent algorithms, mechanism of action, novel protein, p42.3, spherical polar coordinates

How to cite this article:
Liu X, Hao Y, Fan T, Nan K. Application of intelligent algorithm in the optimization of novel protein regulatory pathway: Mechanism of action of gastric carcinoma protein p42.3. J Can Res Ther 2016;12:650-6

How to cite this URL:
Liu X, Hao Y, Fan T, Nan K. Application of intelligent algorithm in the optimization of novel protein regulatory pathway: Mechanism of action of gastric carcinoma protein p42.3. J Can Res Ther [serial online] 2016 [cited 2020 Jul 14];12:650-6. Available from: http://www.cancerjournal.net/text.asp?2016/12/2/650/151856

 > Introduction Top

The occurrence and development of gastric cancer is a complex molecular network regulation process involving multiple factors, stages and steps. Finding the key molecules and biomarkers to predict gastric cancer risk, aid in early diagnosis and predict therapeutic outcome, is currently an intensely researched topic. Studies have demonstrated that a variety of tumor suppressor genes such as phosphatase and tensin homolog (PTEN), p16, p21, Smad4, Fas and RECK, and oncogenes, such as Ras, c-myc, and matrix metalloproteinases (MMPs) have been linked to the occurrence of gastric carcinoma.[1],[2],[3],[4],[5],[6],[7] p42.3 is a novel gene, cloned by the combination of cell synchronization, messenger ribonucleic acid (mRNA) differential display technology with bioinformatics methods.[8] A series of studies show that p42.3 is closely related to the occurrence and development of gastric carcinoma. Its gene silencing changes the expression of two key genes, Checkpoint kinase 2 (CHK2) and cyclin B1, involved in cell cycle regulation, which consequently blocks the cell cycle and inhibits cell proliferation.[9] p42.3 gene characterized oncogenes and tumor markers, is most likely one of the early molecular events in the development from gastric mucosal lesion to gastric carcinoma.[10] Previous studies in our laboratory showed that the regulatory pathway of p42.3 involved in the occurrence and development of gastric cancer may be Ras – Raf-1 – MEK – MAPK kinase – MAPK – microtubule-associated protein – spindle protein – centromere protein – cell proliferation.[11] But the result has not been supported by sufficient theories or experiments. In this study, through improvement of the similarity algorithm between the referential protein and p42.3 protein, the relevant regulatory network of p42.3 has been optimized, the possible pathway being modulated correspondingly. All of the above studies have been verified by molecular biological experiments.

 > Subjects and Methods Top

Six gastric carcinoma cell lines BGC823, MGC803, SGC7901, AGS, N87, and GES1 were provided by the Beijing Cancer Hospital. Culture conditions: They and were cultivated in Dulbecco's Modified Eagle's medium (DMEM) culture media with 5% fetal bovine serum in a 5% CO2 cell culture box at 37°C.

The complementary deoxyribonucleic acid (cDNA) reverse transcription kit and polymerase chain reaction (PCR) amplifier were purchased from Thermo fisher scientific Company (UK) and Eppendorf Company (Germany), respectively.

Structural features of p42.3 Protein molecule

The amino acid sequence [Accession Number: NP_848543] of p42.3 was obtained from the NCBI protein database. Duo to its low homology, the spatial structure of protein was predicted using the threading prediction tool Phyre.[12] Afterwards, association with cell proliferation in function was set as the confined condition, and proteins with any one of the domains were searched and composed of the referential protein data set.

Similarity calculation of the referential protein and p42.3 Protein

Selection of the parameters

After a comprehensive analysis, a total of eleven parameters as measures of protein similarity were selected: Protein spatial density (denoted as S11), atom number within molecules (S1), amino acid number within molecules (S2), the number of amino acid type within molecules (S3), the proportion of C atom number (S4), the proportion of N atom (S5), the proportion of O atom (S6), the number of P atom (S7), the number of S atom (S9), the location of element P (S8) and the location of element S (S10).

Similarity calculation of the spatial density of protein

Textread function in Matlab was used to read coordinates of each atom in the protein structure file (.pdb file) and Euclidean coordinates were unified the spatial coordinates with the geometrical center of the protein as the origin. The distance of each atom to this origin was then calculated. According to the distances, the protein was divided into layers and the number of atoms of two proteins in corresponding layer was recorded for comparison. In this algorithm, ten was selected as the number of layers. According to the length of radius, the protein was divided into ten spherical shells with its hub as the center of sphere so that the protein was divided into ten even layers. For most atoms the distance to the center of the protein was within 0–80, some were within 80–100, and a few were beyond 100. Therefore, the 10 layers were divided as follows: The first layer 0–10; the second layer 10–20; the third layer 20–30; the fourth layer 30–40; the fifth layer 40–50; the sixth layer 50–60; the seventh layer 60–70; the eighth layer 70–80; the ninth layer 80–100; the tenth layer beyond 100. Parameters such as the number of atoms, amino acid and amino acid type etc. in each layer were counted for comparison of two proteins, and stored in data1 and data2, respectively. The similarity of each parameter was then compared using the following formula:

In this formula, sim represents a ten-dimensional vector that has stored the similarity of each layer. After calculating the similarity of each layer, the similarity of the two objects was obtained using weighted average method. Theoretically, if one layer contains more atoms, it may be considered that the weight of the layer is larger. Based on this assumption, the weighted value of each layer was calculated by formula (2). Thus, the weight of protein in each took the average value of various proteins in each layer:

where n1 is the total number of atoms of the first protein, n2 is that of atoms of the second protein and li is the number of atoms in the i th layer.

Therefore, in accordance with the following formula (3), the similarity of certain parameter of the two proteins could be obtained.

Similarity of the total number of atoms, number and type of amino acids

According to the above spherical polar coordinates layered approach, the number of atoms, and the number and type of amino acid in the two proteins were calculated in matrix laboratory (MATLAB). The total number of atoms of the two proteins were denoted as n1 and n2 respectively, thus the similarity of atom numbers was: . In the same way, similarities in the number and type of amino acids can be also obtained.

Similarity of each element

Primary analysis of element C, N, O, P, and S was conducted in this study. Firstly, the proportion of C, N, and O in each protein was calculated. Then, the similarity between proteins was calculated for C, N, and O, according to the formula: . Where p1 is the proportion of element in p42.3; p2 is the ratio of the element in the comparative protein with p42.3. Furthermore, in the protein molecule, the number of element P and S is usually small (with an average of no more than 5), but they play a vital role in the function of protein. Thus we took the number and location of P and S atoms as the standard in calculating the similarity of P and S atom, instead of calculating the ration of the number of atoms in the protein. The following formula was used of calculating the similarity in the number of P and S:

where n1 is the number of P / S atom in p42.3; n2 is the number of P / S atom in the protein being compared.

In the calculation of similarity of P and S position, we made the following agreement, if P / S elemens of the two proteins were in the same layer, the similarity was taken as 1.0; if they were in neighboring layers, the similarity was 0.5; the similarity in other situations was 0. Therefore, the similarity parameter of each element in two proteins was achieved.

Calculation of the Value of Each Parameter

On the basis of each parameter similarity of two proteins being calculated, the overall similarity was the weighted summation method was utilized to calculate the overall similarity. Weights setting: Collecting the data of 100 pairs of similar protein, each parameter similarity of each pair of protein (S1 to S11) was figured out according to the method described above. Then, the homology of each pair of protein searched by BLASTp was taken as the overall similarity (S). A similarity data vector of 1*12 can be obtained for each pair of protein [S1, S2, S3, S4, S5, S6, S7, S8, S9, S10, S11, S]. Data of the 100 pairs of protein similarity was used for regression analysis, thus the relational model between S and S1-S11 can be achieved. The regression analysis results are shown in [Table 1].
Table 1: Coefficients results of regression analysis

Click here to view

Hence, the overall similarity of any pair of protein can be calculated according to the formula (5):

S = -0.561 * S1 + 0.752 * S2 + 0.073 * S3-3.105 * S4 + 1.835 * S5 + 0.727 * S6-0.037 * S7-0.062 * S8-0.327 * S9-0.012 * S10 + 0.945 * S11 + 0.843

wherein, Si represents the similarity of each parameter, i = 1, 2…11. According to this formula, similarity of the referential protein and p42.3 was worked out.

Construction and Optimization of a Bayes' Regulatory Network

The above study mentioned that each referential protein was screened out based on the relevance of malignant cell proliferation in terms of function. Therefore, with this protein as the initial point and cell proliferation as the ending point, the mode of action and path node of each referential protein was collected by literature synthesis search method. There are crosses between different pathways, thus the primary regulatory networks can be drawn. Then the similarity of each referential protein and p42.3 was set as the initial prior probability and the probability of occurrence in each node was calculated using conditional probability, see formula:

Bayesian networks is a Directed Acyclic Graphs (DAG) describing the joint probability distribution of a finite set of variables U = {X1, X2, Xn}. Bayesian networks can be represented by elements pair B = (G, θ), where G is a DAG in which the nodes represent corresponding random variables {X1, X2, Xn} that could symbolize expression vectors consisting of gene expression profiling data, while θ symbolizes the conditional probability of each variable. DAG showed the following conditional independent relationships, that is the Markov assumption: Each variable Xi was independent from its non-son node under the premise that it was the parent node in graph. Based on the assumption of independence, the Bayesian network, G had a joint probability distribution for specific set U as follows:

where Pa (Xi) represents the parent node of Xi. In order to ensure the above joint probability distribution, all the conditional probability in formula (6) needed to be confirmed.

In the Bayes network in this paper, after calculating the data of probability of occurrence in each node and pathway of the primary network by conditional probability, the Bayes theorem was used to backstep the probability of protein in acting their roles in each node, so the most plausible regulatory pathway of p42.3 protein was determined.

Molecular Biological Verification of the Most Likely Path

The six cell lines BGC823, MGC803, SGC7901, AGS, N87 and GES1 were cultured according to the conditions described above. The total mRNA was extracted using Trizol method and cDNA was synthesized by reverse transcription. Design of primers was conducted according to the known gene sequence of the referential proteins and p42.3. With β-actin as contrast, the real-time (RT)-PCR amplication was conducted, respectively. The expressions of different proteins in different cell lines were figured out by Agarose gel electrophoresis detection. Meanwhile, expression of any node in the network can be treated as reference of pathway verification.

 > Results and Discussion Top

Structural Features of the EF-Hand and CC-domain of p42.3

After predicting the spatial structure of p42.3 protein using the threading prediction tool Phyre, it was found that p42.3 protein molecule contained two functional domains: The EF-hand and CC-domain [Figure 1]. The protein data set that had relatively high homology with EF-hand and CC-domain structure of p42.3 molecule was collected from the existing studies, and part with EF-hand structure is shown in [Table 2]. Then proteins that were related to cell proliferation in terms of the function were screened out as the referential protein.
Figure 1: Structural modelling results of p42.3

Click here to view
Table 2: EF-hand structural set that is similar to local structure of p42.3 molecule

Click here to view

Similarity Calculation of the Referential Protein and p42.3

The spatial coordinates of protein were read using MATLAB software and the similarity of the referential protein and p42.3 was calculated [Table 3]. Wherein, the structural similarities between six referential proteins with related functions CENP-B, GCN4, S100A11, RAS, S100A1 as well as FKBP and p42.3 were all up to more than 40%.
Table 3: Protein dataset gained by using spherical coordinate spatial stratified similarity algorithm

Click here to view

Bayes' Regulatory Network

Association with the function of cell proliferation was set as the confined condition, and the acting pathways and nodes of each referential protein were worked out by literature collection. Because of the crosses of each path, a primary regulatory network was drawn [Figure 2]. The round nodes represent the referential proteins and they are the initial nodes. The rectangular boxes are the searched acting nodes by literature collection during the process. Arrows represent the direction of action. “+” and “-” above the arrow indicate a promotion or inhibition action.
Figure 2: Primary regulatory networks. Note: The arrows indicate the direction of action. “+” indicates a positive control; “-” indicates a reverse regulation

Click here to view

The similarity of the referential proteins and p42.3 was regarded as prior probability of the initial nodes and according to formula (4), the probability of occurrence in each of the downstream node was calculated until the final result of cell proliferation was obtained, which is the probability of occurrence of the pathway. With cell proliferation as the result and application of Bayes theorem, the possible reasons that led to that result was taken backward deduction until the original node [Figure 3]. By comparing the similarity of each pathway as well as each node, and combining with the comparison results of protein similarity, it can be primarily confirmed that the pathway “S100A11→ RAGE→ P38 →MAPK→ Microtubule-associated protein →Spindle protein→Centromere protein→Cell proliferation” was the most likely acting pathway of p42.3 protein playing its role in cell proliferation.
Figure 3: The most plausible acting pathway of p42.3 protein optimized by Bayes theorem. Note: The arrows indicate the direction of action. “+” Indicates a positive control; “-” indicates a reverse regulation

Click here to view

Molecular Biological Verification

In line with the analyses of p42.3 spatial structure and the Bayes' regulatory network, RT-PCR was adopted to detect the expression of the pathway starting protein S100A11 with the biggest positive weight, and the starting protein S100A2 of S100 family in the reverse acting pathway in different gastric carcinoma cell lines. In addition, comparison of the relevance to the expression of p42.3 protein was carried out. It was found that p42.3, S100A11 and S100A2 all had different levels of expression in the six gastric cancer cell lines. As can be seen from [Figure 4], the expression of S100A11 was very similar to the expression result of p42.3, but that of S100A2 showed great difference to p42.3. Binding the results of protein structural similarity analysis, it suggested that the regulatory pathway of p42.3 may be associated with S100A11, which is in an agreement with the result of optimized Bayes' regulatory network that p42.3 protein may be involved in the regulation of the pathway of S100A11.
Figure 4: Expressions of p42.3, S100A11, and S100A2 in the cell lines of gastric carcinoma

Click here to view

The occurrence and development of gastric cancer involve a large number of changes in the structure and function of related genes. The activation of oncogenes and inactivation of tumor suppressor genes is of particular importance. Currently, a large number of studies of molecular mechanisms of gastric cancer have tried to unravel the molecular regulatory mechanisms of gastric carcinoma with the intention of providing ideas in finding new molecular biomarkers and therapeutic drug target for the diagnosis of gastric cancer.

As a cyclin-dependent protein, p42.3 shows different expression levels at different stages of the cell cycle, with no or low expression in normal gastric mucosa, and high expression in gastric cancer tissues where it promotes cellular proliferation and tumor metastasis.[8] Changes in expression of p42.3 that occurs during the occurrence and development of gastric carcinoma suggest that it is expected to be a diagnostic and therapeutic target for gastric carcinoma. The study found that, p42.3 protein contained EF-hand and CC-domain two functional domains, in which the EF-hand domain of the N-terminal is common in oncoprotein S100 family. The EF-hand structure is a typical helix-loop-helix structural unit, where the two α helixes that are connected through a Ca 2+ chelate ring.[31] The majority of EF-hand domains appear in even number and form structural domain pairs, which are divided by connexin; or homologous or heterologous dimers such as the family of S100 proteins with two EF-hand structural domains.[32],[33] When the domain is an odd number, protein usually requires the formation of homologous or heterologous dimers whose activity depends on dimerization. CC-domain is a protein super-secondary structure, entwined by two to seven α helix (commonly two or four) to form the twist structure.[34] Many proteins containing coiled helical structures have important biological functions, such as the transcription factors in the regulation of gene expression. Some famous proteins containing coiled helical structures are oncoprotein and tropomyosin (a kind of muscle protein). In order to study the mechanism of action of p42.3, this study screened the referential proteins with high structural and functional homology with p42.3 protein using spherical polar coordinates stratified and multi-parameter calculation. Additionally, relevant gene regulatory pathway nodes between the referential proteins and occurrence of gastric cancer were gathered, and primary regulatory networks were drawn. By optimization series of Bayesian probability calculation, it is found that the most likely mechanism of action of p42.3 in the pathogenesis of gastric carcinoma is S100A11 → RAGE → P38 → MAPK → Microtubule-associated protein → Spindle protein → Centromere protein → Cell proliferation, which has been preliminary verified by experiments that p42.3 is in consistent with the S100A11 gene expression. The study of gene regulatory networks can be used to quantitatively mine information with regard to gene expression regulation. Through extracting and integrating this information, the knowledge of gene function and understanding of genetic networks to clarify disease pathogenesis can be improved. Through the research of gene regulatory networks, we can investigate gene functions from the overall framework. Genes affect each other and work together in complicated networks, which imply new features that are not revealed by examining the DNA sequence.

The S100 protein family consists of low molecular weight (10–12 kDa) calcium-binding proteins with a highly conserved amino acid sequence in vertebrate animals. S100 proteins have a high homology with calmodulin and other EF-type calcium binding proteins.[35] S100 portein family is found to closely relate to tumor from its biological functions, the specific expression and chromosomal localization.[36] S100A11 protein is overexpressed in breast cancer, prostate cancer, and non-small cell lung cancer, promoting tumor metastasis and invasion,[37],[38] but in the urinary bladder and kidney cancer, it is associated with tumor suppression.[39] It should be noted that up-regulation of S100A11 expression in gastric cancer cell lines appeared. Meanwhile, the expression of S100A2, which acts as a tumor suppressor in a variety of malignancies such as breast, liver, prostatic, and esophageal cancer, was significantly reduced in our gastric carcinoma cell lines.[40],[41],[42],[43] Studies have shown that S100A2 acts as a tumor suppressor in gastric cancer cells by inhibiting cell proliferation and invasion.[44] Based on the analysis of expression of S100A11 and S100A2 in gastric cancer both of which contained EF-Hand structure, it was verified that p42.3 could participate in the occurrence and development of gastric cancer from the perspectives of accordant and opposite to the p42.3 effect direction.

 > Conclusion Top

There are a variety of methods of protein structure comparison with both advantages and disadvantages. Most of them are analyses of the protein structures with homology template to infer its possible functions. This conventional comparison of spatial structures only analyzed the characteristics of proteins spatial structure, not allowing for other aspects of protein similarity. For example, element P and S are crucial in playing protein functions. What this study has used are the spherical polar coordinates stratified and multi-parameter weighted comprehensive comparison methods. Not only has the differences of spatial three-dimensional structure been compared, but also the elements, data, amino acid and other eleven important parameters of protein structure similarities have been taken into account. During the weighted summation of each parameter, weights used all came from diverse data training, not artificial subjective weighting, thus ensuring the accuracy of each parameter weight and more accurately calculating similarities of the two proteins, which is one of the characteristics of this article. The algorithm offers a new idea for the study on the structures, functions and regulatory mechanisms of novel proteins which lack of homology.

 > Acknowledgement Top

We thank workmates of the authors within the same department for their general support. In addition, this work was supported by the National Natural Science Foundation of China (60971110), Science and Technology Corporation Project of Henan Province (122106000042) and Open Science and Technology Corporation Project of Henan Province in 2013 (132106000064).

 > References Top

Zhang BG, Li JF, Yu BQ, Zhu ZG, Liu BY, Yan M. microRNA-21 promotes tumor proliferation and invasion in gastric cancer by targeting PTEN. Oncol Rep 2012;27:1019-26.  Back to cited text no. 1
Xiao B, Zhu ED, Li N, Lu DS, Li W, Li BS, et al. Increase miR-146a in gastric cancer directly targets SMAD4 and is involved in modulating cell proliferation and apoptosis. Oncol Rep 2012;27:559-66.  Back to cited text no. 2
Rippa E, La Monica G, Allocca R, Romano MF, De Palma M, Arcari P. Overexpression of gastrokine 1 in gastric cancer cells induces Fas-mediated apoptosis. J Cell Physiol 2011;226:2571-8.  Back to cited text no. 3
Du YY, Dai DQ, Yang Z. Role of RECK methylation in gastric cancer and its clinical significance. World J Gastroenterol 2010;16:904-8.  Back to cited text no. 4
Lam EK, Wang X, Shin VY, Zhang S, Morrison H, Sun J, et al. A microRNA contribution to aberrant Ras activation in gastric cancer. Am J Transl Res 2011;3:209-18.  Back to cited text no. 5
Zhang L, Hou Y, Ashktorab H, Gao LC, Xu YJ, Wu K, et al. The impact of C-MYC gene expression on gastric cancer cell. Mol Cell Biochem 2010;344:125-35.  Back to cited text no. 6
Holmberg C, Ghesquière B, Impens F, Gevaert K, Kumar JD, Cash N, et al. Mapping proteolytic processing in the secretome of gastric cancer-associated myofibroblasts reveals activation of MMP-1, MMP-2, and MMP-3. J Proteome Res 2013.  Back to cited text no. 7
Xu X, Li W, Fan X, Liang Y, Zhao Y, Zhang J, et al. Identification and characterization of a novel p42.3 gene astumors pecific and mitosis phase-dependent expression in gastric cancer. Oncogene 2007;26:7371-9.  Back to cited text no. 8
Cui Y, Su WY, Xing J, Wang YC, Wang P, Chen XY, et al. MiR-29a inhibits cell proliferation and induces cell cycle arrest through the downregulation of p42.3 in human gastric cancer. PLoS One 2011;6:e25872.  Back to cited text no. 9
Sun W. Expression change and biological significance of p42.3 in gastric mucosal lesion. Beijing: College of Life Sciences, Beijing Normal University; 2012.  Back to cited text no. 10
Zhang J, Lu C, Shang Z, Xing R, Shi L, Lv Y. p42.3 gene expression in gastric cancer cell and its protein regulatory network analysis. Theor Biol Med Model 2012;9:53.  Back to cited text no. 11
Kelley LA, Sternberg MJE. Protein structure prediction on the Web: A case study using the Phyre server. Nature protocols 2009;4:363-371.  Back to cited text no. 12
Santonico E, Panni S, Falconi M, Castagnoli L, Cesareni G. Binding to DPF-motif by the POB1 EH domain is responsible for POB1-Eps15 interaction. BMC Biochemistry 2007;8:29.  Back to cited text no. 13
Koshiba S, Kigawa T, Iwahara J, Kikuchi A, Yokoyama S. Solution structure of the Eps15 homology domain of a human POB1 (partner of RalBP1). FEBS Lett 1999;442:138-42.  Back to cited text no. 14
Harper JF, Sussman MR, Schaller GE, Putnam-Evans C, Charbonneau H, Harmon AC. A calcium-dependent protein kinase with a regulatory domain similar to calmodulin. Science 1991;252:951-4.  Back to cited text no. 15
Weljie AM, Vogel HJ. Unexpected structure of the Ca (2+)-regulatory region from soybean calcium-dependent protein kinase-alpha. J Biol Chem 2004;279:35494-502.  Back to cited text no. 16
Nakamura S, Takemura T, Tan L, Nagata Y, Yokota D, Hirano I, et al. Small GTPase RAB45-mediated p38 activation in apoptosis of chronic myeloid leukemia progenitor cells. Carcinogenesis 2011;32:1758-72.  Back to cited text no. 17
Shintani M, Tada M, Kobayashi T, Kajiho H, Kontani K, Katada T. Characterization of Rab45/RASEF containing EF-hand domain and a coiled-coil motif as a self-associating GTPase. 19 Biochem Biophys Res Commun 2007;357:661-7.  Back to cited text no. 18
Babu YS, Bugg CE, Cook WJ. Structure of calmodulin refined at 2.2 A resolution. J Mol Biol 1988;204:191-204.  Back to cited text no. 19
Koller M, Schnyder B, Strehler EE. Structural organization of the human CaMIII calmodulin gene. Biochim Biophys Acta 1990;1087:180-9.  Back to cited text no. 20
Koller M, Strehler EE. Functional analysis of the promoters of the human CaMIII calmodulin gene and of the intronless gene coding for a calmodulin- like protein. Biochim Biophys Acta 1993;1163:1-9.  Back to cited text no. 21
Yuasa HJ, Cox JA, Takagi T. Genomic structure of the amphioxus calcium vector protein. J Biochem 1999;126:572-7.  Back to cited text no. 22
Théret I, Baladi S, Cox JA, Sakamoto H, Craescu CT. Sequential calcium binding to the regulatory domain of calcium vector protein reveals functional asymmetry and a novel mode of structural rearrangement. Biochemistry 2000;39:7920-6.  Back to cited text no. 23
Jia J, Tarabykina S, Hansen C, Berchtold M, Cygler M. Structure of apoptosis-linked protein ALG-2: Insights into Ca2+-induced changes in penta-EF-hand proteins. Structure 2001;9:267-75.  Back to cited text no. 24
Song J, Zhao Q, Thao S, Frederick RO, Markley JL. Solution structure of a calmodulin-like calcium-binding domain from Arabidopsis thaliana. J Biomol NMR 2004;30:451-6.  Back to cited text no. 25
McCormack E. Calmodulins and related potential calcium sensors of Arabidopsis. New Phytol 2003;159:585-98.  Back to cited text no. 26
Berchtold MW, Heizmann CW, Wilson KJ. Primary structure of parvalbumin from rat skeletal muscle. Eur J Biochem 1982;127:381-9.  Back to cited text no. 27
Williams TC, Corson DC, Oikawa K, McCubbin WD, Kay CM, Sykes BD. 1H NMR spectroscopic studies of calcium-binding proteins. 3. Solution conformations of rat apo-alpha-parvalbumin and metal-bound rat alpha-parvalbumin. Biochemistry 1986;25:1835-46.  Back to cited text no. 28
Watterson DM, Sharief F, Vanaman TC. The complete amino acid sequence of the Ca2+-dependent modulator protein (calmodulin) of bovine brain. J Biol Chem 1980;255:962-75.  Back to cited text no. 29
Ogura K, Kumeta H, Takahasi K, Kobashigawa Y, Yoshida R, Itoh H, et al. Solution structures of yeast Saccharomyces cerevisiae calmodulin in calcium-and target peptide-bound states reveal similarities and differences to vertebrate calmodulin. Genes Cells 2012;17:159-72.  Back to cited text no. 30
Gifford JL, Walsh MP, Vogel HJ. Structures and metal-ion-binding properties of the Ca2+-binding helix-loop-helix EF-hand motifs. Biochem J 2007;405:199-221.  Back to cited text no. 31
Drohat AC, Baldisseri DM, Rustandi RR, Weber DJ. Solution structure of calcium-bound rat S100B(betabeta) as determined by nuclear magnetic resonance spectroscopy. Biochemistry 1998;37:2729-40.  Back to cited text no. 32
Hessian PA, Fisher L. The heterodimeric complex of MRI-8 (S100A8) and MRI-14 (S100A9). Antibody recognition, epitope definition and the implications for structure. Eur J Biochem 2001;268:353-63.  Back to cited text no. 33
Liu J, Zheng Q, Deng Y, Cheng CS, Kallenbach NR, Lu M. A seven-helix coiled coil [M]. Proc Natl Acad Sci U S A 2006;103:15457-62.  Back to cited text no. 34
Moroz OV, Antson AA, Murshudov GN, Maitland NJ, Dodson GG, Wilson KS, et al. The three dimensional structure of human S100A12. Acta Crystallogr D Biol Crystallogr 2001;57:20-9.  Back to cited text no. 35
Sakaguchi M, Sonegawa H, Murata H, Kitazoe M, Futami JI, Kataoka K, et al. S100A11, an dual mediator for growth regulation of human keratinocytes. Mol Biol Cell 2008;19:78-85.  Back to cited text no. 36
Hiratsuka S, Watanable A, Aburatani H, Maru Y. Tumor-mediated upregulation of chemoattractants and recruitment of myeloid cells predetermines lung metastasis. Nat Cell Biol 2006;8:1369-75.  Back to cited text no. 37
Liu XG, Wang XP, Li WF, Yang S, Zhou X, Li SJ, et al. Ca2+-binding protein S100A11: A novel diagnostic marker for breast carcinoma. Oncol Rep 2010;23:1301-8.  Back to cited text no. 38
Yao R, Davidson DD, Lopez-Beltran A, MacLennan GT, Montrironi R, Cheng L. The S100 proteins for screening and prognostic grading of bladder cancer. Histol Histopathol 2007;22:1025-32.  Back to cited text no. 39
Lee SW, Tomasetto C, Swisshelm K, Keyomarsi K, Sager R. Down-regulation of a member of the S100 gene family in mammary carcinoma cells and reexpression by azadeoxycytidine treatment. Proc Natl Acad Sci U S A 1992;89:2505-8.  Back to cited text no. 40
Feng G, Xu X, Youssef EM, Lotan R. Diminished expression of S100A2, a putative tumor suppressor, at early stage of human lung carcinogenesis. Cancer Res 2001;6:7999-8004.  Back to cited text no. 41
Kyriazanos ID, Tachibana M, Dhar DK, Shibakita M, Ono T, Kohno H, et al. Expression and prognostic significance of S100A2 protein in squamous cell carcinoma of the esophagus. Oncol Rep 2002;9:503-10.  Back to cited text no. 42
Gupta S, Hussain T, Maclennan GT, Fu P, Patel J, Mukhtar H. Differential expression of S100A2 and S100A4 during progression of human prostate adenocarcinoma. J Clin Oncol 2003;21:106-12.  Back to cited text no. 43
Su XH, You ZY, Zheng ZC. Effects of gene silencing in S100A2 and E-cadherin by RNA interference on gastric cancer cell proliferation and invasiveness. J China Med Univ 2012;41:297-306.  Back to cited text no. 44


  [Figure 1], [Figure 2], [Figure 3], [Figure 4]

  [Table 1], [Table 2], [Table 3]


Similar in PUBMED
   Search Pubmed for
   Search in Google Scholar for
 Related articles
Access Statistics
Email Alert *
Add to My List *
* Registration required (free)

  >Abstract>Introduction>Subjects and Methods>Results and Disc...>Conclusion>Acknowledgement>Article Figures>Article Tables
  In this article

 Article Access Statistics
    PDF Downloaded72    
    Comments [Add]    

Recommend this journal