protein interaction networks
Comparative analysis of viral protein interaction networks in Hepatitis B Virus and Hepatitis C Virus infected HCC ☆
Weilan Yuan a ,b ,1,Tao Huang b ,d ,1,Jian Yu b ,Lingyao Zeng a ,b ,Baofeng Lian b ,e ,Qinwen He b ,Yixue Li a ,b ,d ,Xiaoyan Zhang a ,?,Fengli Zhou c ,?,Lu Xie b ,?
a
School of Life Sciences and Technology,Tongji University,Shanghai 200092,PR China b
Shanghai Center for Bioinformation Technology,Shanghai 201203,PR China c
Fengli Zhou,Department of Respiration,The Third Af ?liated Hospital of Sun Yat-sen University,Guangzhou 510630,PR China d
Key Laboratory of Systems Biology,Shanghai Institutes for Biological Sciences,Chinese Academy of Sciences,Shanghai 200031,PR China e
School of Life Sciences and Biotechnology,Shanghai Jiao Tong University,Shanghai 200240,PR China
a b s t r a c t
a r t i c l e i n f o Article history:
Received 17December 2012
Received in revised form 9May 2013Accepted 4June 2013
Available online 14June 2013Keywords:HCC HBV HCV
Dysfunctional network Three-level network
Previously,the different mechanisms of HBV infection and HCV infection were studied experimentally.Mul-tiple studies also compared the differential network between HBV induced HCC and HCV induced HCC based on gene expression data.However network level comparison combining viral –human interaction network and dysfunctional protein interaction network for HBV and HCV –HCC has rarely been done before.In this work we did some pioneer job in construction of HBV/HCV viral dysfunctional network in HCC,in hope of in-vestigating viral infection impact on the change of genome expression and eventually,the development of HCC.We found that HBx,the main HBV viral protein,directly acted on the gene groups of cell cycle,which could perfectly explain the dominant cell proliferation effect shown in the dysfunctional network of HBV –HCC.On the other hand,multiple important HCV viral proteins including CORE,NS3and NS5A were found to target very important cancer related proteins such as TP53and SMAD3,but no direct targeting to major immune response or in ?ammation related proteins.Therefore the dominant activation of immune response and in ?ammation related pathways shown in dysfunctional network of HCV –HCC might not be a direct effect of HCV infection.They might have been an indirect demonstration of activated cancer promoting pathways.Similar approaches may as well be applied to other important virus infection caused human diseases to help elucidate the mechanisms of virus –host interaction,and even help with investigations on anti-virus based therapies.This article is part of a Special Issue entitled:Computational Proteomics,Systems Biology &Clinical Implications.
?2013Published by Elsevier B.V.
1.Introduction
Hepatocellular carcinoma (HCC)is a malignant tumor with high mortality worldwide.In addition to other risk factors,80%HCC is in-duced by the liver cirrhosis related to chronic infection of Hepatitis B Virus (HBV)or Hepatitis C Virus (HCV)[1].HBV-infected HCC pa-tients spread over Asia (except Japan)and sub-Saharan Africa while HCV-infected HCC patients are distributed in Japan,America and
European countries [2,3].Comparing the virus characteristics and the initial manifestation of infection between HBV and HCV,other than having a wider dissemination and longer incubation period,HBV is more resistant to the environment and can tolerate low tem-perature,dryness,ultraviolet and general chemical disinfectant.In contrast,HCV is more sensitive to general chemical disinfectant with limited dissemination and shorter incubation period.HCV-infected patients have symptoms of hepatalgia and general weakness in the short run and most of them easily develop to chronic hepatitis with a decline in immunity [4].Moreover,HCV-infected HCC occurs at a younger age with a higher alpha-Fetoprotein (AFP)concentration in serum and a larger tumor size [5].
Many researchers suggested that both HBV and HCV-infected HCC patients would experience active virus replication at early stage,and develop to chronic liver cell necrosis,liver ?brosis and liver cirrhosis [6].But HBV and HCV are different in genetic background.HBV belongs to Hepadnaviridae with double strands and replicates by reverse tran-scription.Part of HBV DNA integrates into host genome,causing
Biochimica et Biophysica Acta 1844(2014)271–279
Abbreviations:HCC,hepatocellular carcinoma;HBV,Hepatitis B Virus;HCV,Hepatitis C Virus
☆This article is part of a Special Issue entitled:Computational Proteomics,Systems Biology &Clinical Implications.
?Corresponding authors.Tel.:+862120283705;fax:862120283780.
E-mail addresses:1987_yuanweilan@https://www.wendangku.net/doc/cc2517930.html, (W.Yuan),huangtao@https://www.wendangku.net/doc/cc2517930.html, (T.Huang),shskyjian@https://www.wendangku.net/doc/cc2517930.html, (J.Yu),lenazly@https://www.wendangku.net/doc/cc2517930.html, (L.Zeng),b ?ian@https://www.wendangku.net/doc/cc2517930.html, (B.Lian),qwhe@https://www.wendangku.net/doc/cc2517930.html, (Q.He),yxli@https://www.wendangku.net/doc/cc2517930.html, (Y.Li),xyzhang@https://www.wendangku.net/doc/cc2517930.html,
(X.Zhang),zhoufengliz ?@https://www.wendangku.net/doc/cc2517930.html, (F.Zhou),luxie@https://www.wendangku.net/doc/cc2517930.html, ,xielu@https://www.wendangku.net/doc/cc2517930.html, (L.Xie).1
Contributed
equally.1570-9639/$–see front matter ?2013Published by Elsevier B.V.
https://www.wendangku.net/doc/cc2517930.html,/10.1016/j.bbapap.2013.06.002
Contents lists available at ScienceDirect
Biochimica et Biophysica Acta
j o u rn a l ho m e p a g e :w w w.e l s e v i e r.c o m /l o c a t e /b b a p a
p
chromosome deletion,genomic instability and carcinogenesis[7].In contrast,HCV is a kind of RNA virus that replicates in the cytoplasm of host hepatocyte and its viral proteins play crucial roles in HCC[8]. HCV destroys the stable structure of host cell genome through viral pro-teins.For example,the NS5A protein in HCV causes chromosome struc-ture change by disturbing cell cycle function[9].Despite all these,we do not have a comprehensive understanding of the molecular mechanism differences between HBV and HCV-infected HCC.
Comparative researches on HBV and HCV-infected HCC have come to more attention in recent years.It was shown that HBV and HCV-infected HCC had distinct differential gene expression patterns [10].HBV-associated HCC was enriched with up-regulated genes re-lated to signal transduction,transcription and metastasis,while HCV-associated HCC was enriched with up-regulated genes related to immune response and detoxi?cation function of liver[11].With the method of pathway-based class discrimination,immune-related pathways,cell cycle pathways and RNA metabolism pathways were signi?cantly present in HBV-infected HCC;Cancer-related and lipid metabolism pathways were enriched in HCV-infected HCC;and TGF-βpathway,MAPK and p53pathways were enriched in both the two types of HCCs[12].
Moreover,methods of comparing HBV and HCV-infected HCC are becoming increasingly diverse.To study the molecular mechanisms of complex diseases,interactive networks based on gene expression pattern were constructed and the important modules and genes were analyzed[13,14].Gene weighted co-expression network was applied to analyze the different gene expression modules of HBV and HCV-infected HCC under different clinical states.It was found that pro-apoptosis modules are carcinogenic in HBV-infected HCC while modules of in?ammation and anti-apoptosis are carcinogenic in HCV-infected HCC[15].
All the researches above only focused on changes of molecules in host cells between HBV and HCV-infected HCC.There was no consid-eration on how viral proteins of HBV and HCV might directly act on infected human hosts.HBV and HCV viral proteins play different roles in viral infection by binding to different biological proteins and acting on different modules and signal pathways[16].The main viral proteins of HBV include one core protein and four proteins encoded by ORFs of pre-C,pre-S,polymerase(P)and HBx.HBx is the most important viral protein in HBV and promotes hepatoma cell proliferation and in?ates protein levels in VEGF by increasing the IKKb gene expression level in mTOR signaling pathway[17]. HBx also plays a crucial role in the key proteins of cancer.For exam-ple,it directly inhibits the transcription of p53,causing inactivation of p53-dependent signaling pathways[18].
The main viral proteins of HCV are mainly divided into two groups: one group consists structural proteins(core,p7,glycoprotein E1,E2); the other group non-structural proteins(NS2,NS3,NS4B,NS5A and NS5B)[19].HCV infection mainly damages the immune system, causing a decline in immunity[20].HCV core protein up-regulates Wnt/beta-Catenin signaling,promoting cell growth of hepatocellular carcinoma cells[21].It also directly binds on receptor proteins in Raf/MAPK pathways and inhibits immune function[22].The increased expression of NS3protein enhanced cell growth and increased the activity of DNA binding transcription factors[23].NS5A protein interacted with partial sequence of p53,suppressing p53-mediated reverse transcription and apoptosis during HCV infection[24].It can be seen that NS3and NS5A have complicate roles in HCV-infected liver,but the roles of other HCV viral proteins are not so clear.
In this work we attempt to explain the different biological effects of HBV and HCV infection on carcinogenic progression by investigating the hepatitis viral protein interaction networks in HBV and HCV-infected HCC.The work was carried out in the following steps: Firstly,we collected gene expression data of liver cancer tissues infected by HBV and HCV and constructed dysfunctional networks of differentially-expressed genes.Secondly,we curated the interaction network data of HBV and HCV viral proteins targeting human proteins. Thirdly,we integrated these two networks to build the three-level hepatitis viral protein interaction network(viral protein–targeted human protein–proximal downstream targeted proteins).Finally key genes and important signaling pathways of two viral interaction net-works were extracted and comparatively analyzed.The work?ow is il-lustrated in Fig.1.
2.Material&methods
2.1.Gene expression pro?le
In this work,we used dataset GSE19665from Gene Expression Omnibus(GEO)database[25].The samples include5HBV-infected HCC tissues,5HBV-infected cancer-adjacent tissues,5HCV-infected HCC tissues and5HCV-infected cancer-adjacent tissues.The dataset is from Affymetrix Human Genome U133Plus2.0Array Platform in CEL format.The clinical information of samples in GSE19665is shown in Table1.We preprocessed dataset GSE19665with R“Affy”package[26].The expression levels of probes were normalized by Robust Multi-Array(RMA)[27]and duplicated probes for each gene were averaged.Totally19,798gene expression values were obtained.
R“samr”package[28]was used to identify the differentially-expressed genes between HBV-infected HCC cancer and matched adja-cent tissues,HCV-infected HCC cancer and matched adjacent tissues. The criteria of the differential expression were that the fold change is above1.5and the False Discovery Rate(FDR)is less than5%.Overall, the numbers of differentially-expressed genes in HBV-infected HCC and HCV-infected HCC versus respective adjacent tissues were2213 and2228.
2.2.Dysfunctional network
The differentially expressed genes in HBV-infected HCC and HCV-infected HCC were directly mapped to STRING database(version 8.2)and got pairwise interactions.STRING is the largest protein–protein interaction relationship database curated from biology experiments and prior knowledge[29].It weighs protein–protein interactions by calcu-lating con?dence score with a number of prediction algorithms[30,31]. The higher the con?dence score,the more certain the protein–protein interaction is.75%con?dence score is considered high,according to statement in the STRING database.In many studies,70%con?dence score has been used as a cut-off and proved to be reliable[32],so in this study only interactions with high con?dence score above70% were considered for further analysis.We used t-test to determine the differential genes between the disease samples and control samples, while Pearson's correlation test was used to measure the correlated expression of paired interactions in all samples.To score interactions of gene pairs in disease stages,Liu et al.and Huang et al.[33,34] summarized the below equation to combine the information from noncancerous and HCC livers:
Score e x;y
eT
eT??2log e p x
eTtlog e p cor x;yeT
tlog e p y
h i
:e1T
In this equation,p x and p y are p values of t-test which respectively represent differential expressions of node x and node y in a pair of inter-action.p cor(x,y)is the p value between x and y in Pearson's product–mo-ment correlation test.The equation provided the con?dence score of p values of each interaction.This method has been widely applied in sta-tistical meta-analysis to combine p values from different datasets[35].
After pairwise interaction was scored,the signi?cance of each score was measured by FDR.The FDR was de?ned as follows:First,both the real score(between x and y)and the pseudo-scores(between x and the rest of proteins in the network)were calculated by Eq.(1).Then, the percent of pseudo-scores larger than the real score was taken as
272W.Yuan et al./Biochimica et Biophysica Acta1844(2014)271–279
the FDR for the interaction of nodes x and y .A pairwise interaction would be de ?ned as a dysfunctional interaction only when the FDR of this interaction was less than 0.05.Finally,the dysfunctional interac-tions were summarized to dysfunctional networks by using “igraph ”package in R environment.
2.3.Virus –human interactome
The virus –human interaction datasets include HBV –human interactome and HCV –human interactome.HBV –human interactome comes from two resources:HBV –HIM MAP and text mining from lit-erature.HBV –HIM MAP is a database of interactions between HBV viral proteins and human proteins con ?rmed by experiments from literature of NCBI PubMed before 2008[36].Data after 2008were added by ?rst text mining in PubMed,and then manual curation.After removing redundant information,we collected 147pairs of HBV –human interactions.HCV –human interactome mainly comes from 481interaction pairs between HCV viral proteins and human proteins from de Chassey's yeast-two-hybrid experiments in 2008[37],which are the most comprehensive interaction datasets of HCV viral proteins and human proteins available.
2.4.Three-level network about hepatitis virus
We combined the virus –human interaction network and dysfunc-tional network of HCC to make a full virus-dysfunctional network,by anchoring the overlapped proteins of these two networks.Then we extracted HBV or HCV viral proteins,their directly targeted human proteins,and the proximal interacted proteins of these virus targeted proteins,to construct the simpli ?ed three-level hepatitis virus inter-action networks in HBV HCC and HCV HCC respectively.These two virus interaction networks were comparatively analyzed to identify key targeted proteins of viral proteins and the functional modules,in the purpose of exploring the different mechanisms of HBV and HCV acting on HCC development.3.Results
https://www.wendangku.net/doc/cc2517930.html,parative analysis of the dysfunctional networks of HBV-infected HCC and HCV-infected HCC
3.1.1.Dysfunctional networks
We constructed two dysfunctional networks:dysfunctional net-work of HBV-infected HCC with 359nodes and 625edges as well
as
Fig.1.Schematic overview of constructing dysfunctional networks and three-level networks,analyzing different functional modules in HBV-infected HCC and HCV-infected HCC.
Table 1
The clinical information of samples in the GSE19665dataset.
Case number
Virus type Tissue
Disease state Tissue pathology
Gender Age GSM490987189N HBV Surrounding noncancerous liver Control Chronic hepatitis
M 68GSM490988189T HBV Liver cancer
Cancer Moderately differentiated M 68GSM490989358N HBV Surrounding noncancerous liver Control Chronic hepatitis
M 52GSM490990358T HBV Liver cancer
Cancer Moderately differentiated M 52GSM490995363N HBV Surrounding noncancerous liver Control Chronic hepatitis M 64GSM490996363T HBV Liver cancer
Cancer Poorly differential M 64GSM491003375N HBV Surrounding noncancerous liver Control Liver cirrhosis
M 68GSM491004375T HBV Liver cancer
Cancer Moderately differentiated M 68GSM491005376N HBV Surrounding noncancerous liver Control Chronic hepatitis
M 57GSM491006376T HBV Liver cancer
Cancer Moderately-poorly differentiated M 57GSM490991359N HCV Surrounding noncancerous liver Control Liver cirrhosis
M 51GSM490992359T HCV Liver cancer
Cancer Moderately differentiated M 51GSM490993360N HCV Surrounding noncancerous liver Control Chronic hepatitis
M 66GSM490994360T HCV Liver cancer
Cancer Moderately differentiated M 66GSM490997364N HCV Surrounding noncancerous liver Control Liver cirrhosis
M 74GSM490998364T HCV Liver cancer
Cancer Moderately differentiated M 74GSM490999365N HCV Surrounding noncancerous liver Control Liver cirrhosis
F 71GSM491000365T HCV Liver cancer
Cancer Moderately differentiated F 71GSM491001367N HCV Surrounding noncancerous liver Control Liver cirrhosis
M 72GSM491002
367T
HCV
Liver cancer
Cancer
Moderately differentiated
M
72
273
W.Yuan et al./Biochimica et Biophysica Acta 1844(2014)271–279
dysfunctional network of HCV-infected HCC with391nodes and 663edges;detailed interactions in these dysfunctional networks can be provided upon request to authors.Fig.2A and B shows the dys-functional networks of HBV-infected HCC and HCV-infected HCC re-spectively.Fig.2C shows the combined dysfunctional network in HBV-infected HCC or HCV-infected HCC.We found that there are 131overlapping proteins and93overlapping interactions in two dysfunctional networks by using the VennDiagram package in R environment.
To verify whether the dysfunctional networks are mechanically related to HCC,?rst we enriched the proteins in each network to a validated dataset of335HCC-related proteins[38],then we used Fisher's exact test to detect whether the enrichment was signi?cant (see in Table S1).56nodes in the dysfunctional network of HBV-infected HCC belonged to HCC-related proteins(p-values of Fisher's exact test3.84E?03)while58nodes in the dysfunctional network of HCV-infected HCC belonged to HCC-related proteins (p-values of Fisher's exact test1.034E?04).Nodes of these two dys-functional networks were both signi?cantly enriched to the known HCC-related protein,suggesting that the dysfunctional networks are functionally related to HCC and therefore can be used to study the molecular mechanisms of HCC.
To verify if the dysfunctional network is in accordance with the scale-free feature of biological network,we drew the curve of node degree distribution and calculated the power-law-?t value of node degree in R“igraph”package(see in Fig.S1).The degree of a node was de?ned as the number of its direct neighbors.The node degree distribution summarized all the node degrees of a biological network. The power-law-?t value is a power-law exponent for node degree distribution?tting power-law distribution in a network.When the power-law-?t value of a network is higher than1.5,it suggests that the network is close to scale-free.Scale-free is an important feature of a biological network,which means that there are many nodes with few connections and a few nodes with many connections(hub nodes)[39].When there are hub nodes there are functionally key reg-ulating proteins in a PPI network that can be further studied.In our study,we compared the hub proteins'functions between HBV-infected HCC and HCV-infected HCC dysfunctional networks.
3.1.2.Differences in dysfunctional networks
Next,we speci?cally compared the topological feature difference of dysfunctional networks between HBV-infected and HCV-infected HCC.The shortest path length is an important topological feature for measuring node connection in network by using Dijkstra's algo-rithm[40].We calculated clustering coef?cient and average shortest path length,then used Wilcox test to verify their differences between the dysfunctional networks of HBV–HCC and HCV–HCC.The dysfunc-tional network of HBV–HCC has a higher clustering coef?cient(0.383 vs0.320,p=0.042)and a shorter average shortest path(5.871vs 7.362,p=2.2e?16)than HCV–HCC.It means that the dysfunctional network of HBV-infected HCC had more direct-acting neighboring proteins than dysfunctional network of HCV-infected HCC.We may conclude that the topological features of these dysfunctional net-works are different and the dysfunctional network of HBV–HCC is more modular than HCV–HCC.
In addition,we analyzed the functional difference between the hub nodes of two dysfunctional networks(detail in Table S2).Hub nodes are those with degrees above20,ranking in top2%of all nodes.The numbers of hub proteins in two dysfunctional networks are basically the same,but there are obvious biological function dif-ferences between them.In dysfunctional network of HBV-infected HCC,?ve out of six hub proteins are directly involved in cell cycle ac-tivities(CCNB1,CDKN3,PRC1,BIRC5and MKI67),except MELK.In dysfunctional network of HCV-infected HCC,the7hub proteins can be divided into two kinds:proteins related to immunity(MMP2, KIF4A CCL19and ITGA9)and proteins of cell cycle(CCNB1,
TOP2A
Fig.2.Dysfunctional networks in HBV-infected HCC and HCV-infected HCC.(A)The dysfunctional network of HBV-infected HCC.(B)The dysfunctional network of HCV-infected HCC.The green nodes in dysfunctional network represent the proteins.The black edges represent interactions.(C)The combined dysfunctional network in HBV-infected HCC or HCV-infected HCC.The light green and pink edges indicate dysfunctional interactions in HBV-infected and HCV-infected HCC,respectively.The black edges are dysfunctional inter-actions existing in both types of HCCs.
274W.Yuan et al./Biochimica et Biophysica Acta1844(2014)271–279
and KIF2A).Only CCNB1is present in both networks.In dysfunctional network of HBV–HCC,many of the cell cycle related proteins,includ-ing the?ve hub proteins are elevated in gene expression,suggesting that the promotion of cell cycle-related gene expression may induce HCC.While in HCV-infected HCC dysfunctional network,the expres-sion of genes related to immune response is downregulated,and the expression of cell cycle genes is elevated.
We also studied KEGG pathway enrichments of nodes in two dysfunctional networks by using the functional annotation clustering online software DAVID[41].Pathways with FDR less than0.05and ad-justed p-values less than1.0E?05were analyzed(detail in Table S3). It was found that many genes were mapped to the pathways of hsa04110(cell cycle)and hsa03030(DNA replication)in dysfunctional
network of HBV-infected HCC.And genes were enriched in the KEGG pathways of immune response,in?ammation and metabolism in dys-functional network of HCV-infected HCC.Immune response and in-?ammation pathways include hsa04510(focal adhesion),hsa04512 (ECM–receptor interaction)and hsa04062(chemokine signaling path-way),and metabolism pathways include hsa00590(arachidonic acid metabolism)and hsa04610(complement and coagulation cascades). The results show that there are functional differences between the pro-tein nodes in HBV and HCV-infected HCC dysfunctional networks.We summarized different characteristics of dysfunctional networks of HBV/HCV-infected HCC in Table2.
https://www.wendangku.net/doc/cc2517930.html,parative analysis of HBV/HCV–human interaction network
We brought in virus–human protein interactions to further analyze how viral proteins of hepatitis might disturb the dysfunctional net-works of HBV/HCV-infected HCC.Virus–human interactions include HBV/HCV viral proteins and their targeted proteins(detailed virus–human interactions can be provided upon request).HBV viral proteins in147HBV–human interactions are HBx,PreS,HBcAg,HBsAg and Pre, while HCV viral proteins in481HCV–human interactions are CORE, NS5A,NS5B,NS4A,NS4B,E1,E2and F.Results show that the amount of HBV targeted genes is less than HCV targeted genes.To get a clear view of the functions of HBV/HCV targeted genes,we performed KEGG pathway enrichment analysis.HBV targeted proteins took part in many pathways while HCV targeted proteins only took part in focal adhesion and pathways in cancer(detail in Table S4).In addition, there are common targeted proteins of HBV/HCV viral proteins and most of them are enriched in cancer pathways(detail in Table S5).Fur-ther,a hypergeometric test indicated that speci?cally overlapped targeted proteins are between those of HBV HBx protein and HCV CORE,NS3and NS5A proteins(see in Table3).
These results suggest that HBV targeted proteins have more various functions than HCV targeted proteins.However,HBx protein of HBV and CORE,NS3and NS5A proteins of HCV share common targeted gene proteins which are mostly related to cancer.The results of virus–human interactions demonstrated that HBV and HCV share cancer re-lated common functions,but it is also justi?ed that there are differences between the ways HBV/HCV viral proteins induce HCC,since most of the viral proteins act on proteins with different https://www.wendangku.net/doc/cc2517930.html,parative analysis of three-level viral dysfunctional networks
in HCC
In the above comparative analyses the results show that HBV-infected HCC and HCV-infected HCC demonstrate different dysfunc-tional protein networks,and HBV and HCV viral proteins target differ-ent functional protein groups.We hope to see how the differences in dysfunctional network of HCCs might relate to different viral infec-tions from HBV or https://www.wendangku.net/doc/cc2517930.html,bining virus–human interactions and dysfunctional networks based on overlapping anchoring proteins, we constructed the three-level viral-dysfunctional networks(see in Fig.3).The?rst level consists HBV or HCV viral proteins,the second level their targeted proteins which are present in dysfunctional net-works of the HBV or HCV infected HCC,and only the proximal interacted proteins of these targeted proteins are kept as the third level(see Fig.4).
In the three-level network of HBV–HCC,only HBx of all HBV pro-teins was found to target multiple proteins which were differentially expressed in HBV–HCC versus adjacent tissue.Among them CCNB1 and CCNA2are the two major hubs,with more one-step down interacted proteins https://www.wendangku.net/doc/cc2517930.html,NB1and CCNA2are the key pro-teins for G phase in cell mitotic cycle.They both had been reported to be related to HCC with HBV infection.Increased level of CCNB1in HBV–HCC patients after surgery was de?ned as a sign of tumor recur-rence[42].In animal model CCNA2was found to form fusion protein with HBV surface antigens in the endoplasmic reticulum of the cell and could cause cell?brosis[43].Many of the downstream proteins of CCNB1and CCNA2are key proteins in cell cycle pathways,includ-ing CDC6,TTK,CDC20,ESPL1,CDK7,MCM2,PTTG1,CDC25C,MCM4, CCND1,CDC45,CCNB2,MAD2L1,PCNA,and BUB1B(Fig.4A,detail in Table S6),and they are all differentially expressed in HBV–HCC since they are present in the dysfunctional network.Therefore it might be hypothesized that the gene expression level change caused disturbance of the normal cell cycle and enhanced immortalization of hepatoma cells in HBV–HCC might at least in part source back to HBx action of HBV infection.
On the other hand,in the three-level network of HCV–HCC,multi-ple HCV viral proteins including CORE,NS2,NS3,NS5A,NS5B and F, were found to have targeted human proteins that were present in the dysfunctional network of HCV–HCC.Out of them CORE,NS3and NS5A interacted with more downstream proteins.These three viral
Table2
Characteristics of the dysfunctional network of HBV/HCV-infected HCC.
Nodes Edges The KEGG enrichment of
dysfunctional nodes Hub nodes Average shortest
path length
Clustering
coef?cient
Dysfunctional network of HBV-infected HCC 359625Cell cycle,DNA replication MKI67,BIRC5,CCNB1,
CDKN3,MELK,PRC1
5.8710.383
Dysfunctional network of HCV-infected HCC 391663Focal adhesion,ECM–receptor
interaction,chemokine
signaling pathway,complement and coagulation
cascades,arachidonic acid metabolism
TOP2A,KIF20A,CCNB1,
MMP2,CCL19,KIF4A,ITGA9
7.3620.320
Table3
Overlapped human proteins targeted by both HBV protein HBx and HCV viral proteins.
The HCV viral
proteins
The number of
overlapped proteins
targeted by HBx
P value of
hypergeometric
test
Overlapped proteins
CORE5 2.23E?03CDKN1A,CREBBP,FAS,
TNFRSF1A,TBP
NS34 2.00E?04FN1,JUN,STAT3,TP53
NS5A7 2.65E?18AKT1,BAX,GRB2,JAK1,
RAF1,SRC,TP53
HBV viral proteins:HBx.
HCV viral proteins:CORE,NS3,NS5A.
275 W.Yuan et al./Biochimica et Biophysica Acta1844(2014)271–279
proteins all interacted with TP53which is the key gene related to can-cer.The expression of TP53was downregulated in the dysfunctional network of HCV-infected HCC.HCV NS5A protein was reported to in-hibit apoptosis and increase cell diversion by accelerating the degra-dation of TP53.So NS5A is regarded as an important viral protein of HCV infection-induced HCC [44].Our own previous work identi ?ed CORE as an important viral mediator in the process of HCV induced development of liver cirrhosis to HCC.In Fig.4B it can also be seen that CORE and NS3both interact with SMAD3in the TGF-beta path-way.HCV core variants isolated from liver tumor but not from adja-cent non-tumor tissue were found before to interact with SMAD3[45].SMAD3is a key factor in the TGF-beta pathway.The TGF-beta pathway plays a double-edge role in cancer development,either tumor promoting or tumor inhibitory.For example,it was suggested that chronic in ?ammation caused by HCV infection switched the TGF-beta pathway from inhibiting tumor to promoting liver ?brosis,and increased the risk of HCC [22].Therefore,our HCV viral interac-tion network veri ?ed and provided more evidence for NS5A,CORE and NS3as very important viral protein factors in HCC development (detail in Table S8).
To get fuller views of speci ?c viral protein functions,we extended certain parts of the three-level dysfunctional networks of HBV –HCC
and HCV –HCC,and mapped proteins to speci ?c pathways in KEGG.As shown in Fig.5A,we found that HBV protein HBx can regulate cell cycle pathways by acting on key protein CCND1,then directly or indi-rectly interacting with other members of cell cycle and growth related proteins,such as PTTG1,CDC6,CDC20,CDC23,and PCNA.Through di-rectly regulating these signaling proteins,the impact of HBx may extend to other downstream pathways such as Wnt/beta-catenin.As for HCV –HCC,in Fig.5B it can be seen that multiple HCV viral proteins such as CORE,NS3,and NS5A act directly or indirectly (through TP53)on a key protein in the TGF-βpathway,THBS1[54],and through it the effects of HCV proteins are extended to other key members of the TGF-βpath-way,such as SMAD3and TGFB1proteins.3.4.Validation in extended datasets
Because sample size in our dataset is small,we would like to vali-date our discoveries in larger dataset with more clinical samples.To our best effort,we could not ?nd similar matched comparison studies of HBV –HCC and HCV –HCC samples performed on the same experi-mental platform.We have to reconcile on separate datasets of HBV –HCC versus controls,and HCV –HCC versus controls.We downloaded the expression pro ?les of HBV –HCC (GSE3500)and HCV –
HCC
Fig.3.The total viral-dysfunctional networks of HBV-infected HCC and HCV-infected HCC.(A)The total HBV-dysfunctional network combining the HBV-interactome and dysfunc-tional network of HBV-infected HCC.(B)The total HCV-dysfunctional network combining the HCV-interactome and dysfunctional network of HCV-infected HCC.The green nodes are the proteins in the dysfunctional network;the red nodes are the viral proteins.The gray edges represent the dysfunctional interactions;the yellow edges represent interactions between viral proteins and human
proteins.
Fig.4.The three-level networks of HBV-infected HCC and HCV-infected HCC.(A)The three-level network of HBV-infected HCC.(B)The three-level network of HCV-infected HCC.The red nodes are the viral proteins and the green nodes are the human proteins targeted by viral proteins.The yellow nodes are the ?rst layer interacted proteins of viral targeted proteins.
276W.Yuan et al./Biochimica et Biophysica Acta 1844(2014)271–279
(GSE14323)in the GEO database.We selected 88HBV –HCC patient samples and 54HBV-infected patient liver samples as controls from GSE3500.Likewise,47HCV –HCC patient samples and 41HCV-infected patient cirrhotic liver samples as controls were selected from GSE14323.Then for these two datasets of curated HBV –HCC vs.control and HCV –HCC vs.control,we performed similar construc-tion of dysfunctional HBV –HCC and HCV –HCC networks,and then we constructed the ?nal three-level dysfunctional networks of HBV and HCV viral protein –target protein-interacting proteins as well respec-tively,just as described previously in our own dataset analyses.Then we tried to verify what we observed in Fig.5in these three-level dysfunctional networks extracted from the enlarged datasets.As shown in Fig.S2A,HBx also works on cell cycle pathways by direct-ly targeting CCNA2and CDKNA1.As for HCV –HCC shown in Fig.S2B,the impact of HCV viral proteins on TGF-βpathway can also be observed.
4.Discussion
Previously,the different mechanisms of HBV infection and HCV infection were studied experimentally.W-L Tsai [16]summarized the difference between HBV and HCV infections acting on viral hepatocarcinogenesis.Multiple studies also compared the differential network between HBV induced HCC and HCV induced HCC based on gene expression data [46].However network level comparisons com-bining viral –human interaction network and dysfunctional protein interaction network for HBV and HCV HCC have rarely been done before.In this work we did some pioneer job in construction of HBV/HCV viral dysfunctional network in HCC,in hope of investigating
viral infection impact on the change of genome expression and even-tually,the development of HCC.
HBV and HCV belong to different types of virus and work differently on hosting cells.The integration of HBV can change the genetic struc-ture of host genes,causing such as chromosome de ?ciency and translo-cation and abnormal DNA replication [47].Cyclin-A2is a common locus of HBV gene integration.Therefore HBV infection damages the stability of host genome.It was experimentally shown that part of HBV DNA in-tegrating to host genomes would cause DNA replication,and if the inte-gration happens at the locus of regulatory genes for cell proliferation,it would change the normal cell regulation mode,promote cell growth,and ?nally cause liver cancer [48].HCV infection on the other hand barely acts on host genome.It was hypothesized that HCV-induced in-?ammation promotes hepatocarcinogenesis by causing oxidative DNA damage [49].Metabolic disorders caused by HCV infection may also play important roles in inducing HCC.Those HCV-infected cirrhosis pa-tients with high concentration of adiponectin in serum are more likely to develop to liver cancer [50].It was also found that arachidonic acid metabolism had a direct relationship with HCV-infected cirrhosis [51].The different genetic properties of HBV and HCV may be the reason why their effects on inducing HCC are multifarious and complicate,and dif ?cult to be compared.
Dysfunctional networks of HBV and HCV-infected HCC were found to be functionally different by topological analysis and KEGG pathway enrichments.Most genes in the dysfunctional network of HBV-infected HCC are related to cell proliferation,while in the dysfunctional network of HCV-infected HCC most genes are related to immune and in-?ammatory responses.Previous microarray study also found that the cell proliferation genes of metastasis were predominantly expressed in HBV –HCC [52],while in ?ammatory phenotypes were
predominant
Fig.5.The speci ?c pathways associated with HBV or HCV proteins.(A)The role of HBV virus protein in cell cycle pathway.The red node is HBx protein.The green node is the di-rectly associated key protein to the cell cycle pathway.The blue nodes are other members in cell cycle related or downstream pathways.(B)The role of HCV virus proteins in TGF-beta pathway.The red nodes are HCV viral proteins.The green nodes are associated key proteins to the TGF-beta pathway.The blue nodes are other related members in TGF-beta pathway.
277
W.Yuan et al./Biochimica et Biophysica Acta 1844(2014)271–279
in HCV–HCC[53].It is justi?ed to think that such differences might have originated from different viral infection sources,however,the dysfunc-tional networks on their own can't tell for certain what viral proteins acted on what part of the network features.
Only through the construction of the three-level networks that we can directly correlate viral proteins to target proteins present in dys-functional networks and then study further downstream effects.We found that HBx,the main HBV viral protein,directly acted on the gene groups of cell cycle,which could perfectly explain the dominant cell proliferation effect shown in the dysfunctional network of HBV–HCC.On the other hand,multiple important HCV viral proteins in-cluding CORE,NS3and NS5A were found to target very important cancer related proteins such as TP53and SMAD3,but no direct targeting to major immune response or in?ammation related pro-teins.Therefore the dominant activation of immune response and in-?ammation related pathways shown in HCV dysfunctional network might not be a direct effect of HCV infection.They might have been an indirect demonstration of activated cancer promoting pathways.
Although no perfect extended datasets can be obtained at current stage,we did try to validate our major?ndings in curated datasets of HBV–HCC samples versus controls,and HCV–HCC samples versus controls from other resources.The results show that,the HBV HBx protein's action on cell cycle pathway can be con?rmed.The HCV viral proteins'action on immune pathways may very likely be caused by the viral proteins'impact on the TGF-βpathway.
There are limitations to this work:the patient sample size is small, and the virus–human interaction data may not be wholesome.Both are due to limited data availability under current times.Therefore fu-ture directions include collecting more virus–human interaction rela-tionships and testify our approach on more available gene or protein expression data on HCC patients with virus infection information.In the future when more patient data with clinical information are avail-able,network feature analyses should be carried out in correlation with clinical phenotypes,not only to decipher the virus–human inter-action module,but to see how such modularity would affect human disease phenotype.Such approaches may as well be applied to other important virus infection caused human diseases such as HIV infection,HPV infection and EBV infection related diseases,and help in elucidating the mechanisms of virus–host interaction,and even help with investigations on anti-virus based therapy to the diseases.
5.Conclusion
The different genetic properties of HBV and HCV may be the reason why their effects on inducing HCC are multifarious and complicate, and traditional dysfunctional networks can't make out what viral pro-teins acted on what part of the host gene features.In this work we performed network level comparisons combining viral–human inter-action network and dysfunctional protein interaction network for HBV and HCV–HCC,in hope of investigating viral infection impact on the change of genome expression and eventually,the development of HCC.We found that HBx,the main HBV viral protein,directly acted on the gene groups of cell cycle,which could perfectly explain the dominant cell proliferation effect shown in the dysfunctional net-work of HBV–HCC.On the other hand,multiple important HCV viral proteins including CORE,NS3and NS5A were found to target very im-portant cancer related proteins such as TP53and SMAD3,but no direct targeting to major immune response or in?ammation related pro-teins.Therefore the dominant activation of immune response and in-?ammation related pathways shown in HCV dysfunctional network might not be a direct effect of HCV infection.They might have been an indirect demonstration of activated cancer promoting pathways. Similar approaches may as well be applied to other important virus in-fection caused human diseases to help elucidate the mechanisms of virus–host interaction,and even help with investigations on anti-virus based therapies.
Supplementary data to this article can be found online at http:// https://www.wendangku.net/doc/cc2517930.html,/10.1016/j.bbapap.2013.06.002.
Acknowledgements
This work was funded by the Key Infectious Disease Project (2012ZX10002012-014),the National Natural Science Foundation of China(31070752),and in part supported by the National Hi-Tech Program2012AA020201,and the National Key Basic Research Pro-grams2010CB912702and2011CB910204.
References
[1]J.F.Perz,G.L.Armstrong,L.A.Farrington,Y.J.F.Hutin,B.P.Bell,The contributions of
hepatitis B virus and hepatitis C virus infections to cirrhosis and primary liver cancer worldwide,J.Hepatol.45(2006)529–538.
[2]H.B.El-Serag,Epidemiology of viral hepatitis and hepatocellular carcinoma,
Gastroenterology142(2012)1264–1273,(e1261).
[3] A.P.Venook,C.Papandreou,J.Furuse,https://www.wendangku.net/doc/cc2517930.html,dron de Guevara,The incidence and epi-
demiology of hepatocellular carcinoma:a global and regional perspective,Oncolo-gist15(2010)5–13.
[4]S.E.Buskin,Hepatitis B and C infection and liver disease trends among human
immunode?ciency virus-infected individuals,World J.Gastroenterol.17(2011) 1807–1816.
[5]S.P.Hiotis,N.N.Rahbari,G.A.Villanueva,E.Klegar,W.Luan,Q.Wang,H.T.Yee,
Hepatitis B vs.hepatitis C infection on viral hepatitis-associated hepatocellular carcinoma,BMC Gastroenterol.12(2012)7.
[6]H.B.El-Serag,K.L.Rudolph,Hepatocellular carcinoma:epidemiology and molecu-
lar carcinogenesis,Gastroenterology132(2007)2557–2576.
[7] C.Bréchot,Pathogenesis of hepatitis B virus-related hepatocellular carcinoma:
old and new paradigms,Gastroenterology127(2004)S56–S61.
[8] D.R.McGivern,S.M.Lemon,Virus-speci?c mechanisms of carcinogenesis in
hepatitis C virus associated liver cancer,Oncogene30(2011)1969–1983.
[9]K.-H.Baek,H.-Y.Park,C.-M.Kang,S.-J.Kim,S.-J.Jeong,E.-K.Hong,J.-W.Park,Y.-C.
Sung,T.Suzuki,C.-M.Kim,C.-W.Lee,Overexpression of hepatitis C virus NS5A protein induces chromosome instability via mitotic cell cycle dysregulation, J.Mol.Biol.359(2006)22–34.
[10]S.Y.Yoon,J.-M.Kim,J.-H.Oh,Y.-J.Jeon,D.-S.Lee,J.H.Kim,j.y.Choi,B.M.Ahn,S.S.
Kim,H.-S.Yoo,Y.S.Kim,N.-S.Kim,Gene expression pro?ling of human HBV and HCV-associated hepatocellular carcinoma cells using expressed sequence tags, Int.J.Oncol.29(2006)315–327.
[11]N.Iizuka,O.Masaaki,Y.-O.Hisafumi,N.Mori,T.Tamesa,T.Okada,N.Takemoto,A.
Tangoku,K.Hamada,H.Nakayama,M.Takanobu,S.Uchimura,Y.Hamamoto, Comparison of gene expression pro?les between hepatitis B virus-and hepatitis
C virus-infected hepatocellular carcinoma by oligonucleotide microarray data
on the basis of a supervised learning method,Cancer Res.62(2002)3939–3944.
[12]S.Y.Lee,K.H.Song,I.Koo,K.-H.Lee,K.-S.Suh,B.-Y.Kim,Comparison of pathways
associated with hepatitis B-and C-infected hepatocellular carcinoma using pathway-based class discrimination method,Genomics99(2012)347–354. [13]L.Chen,R.Liu,Z.-P.Liu,M.Li,K.Aihara,Detecting early-warning signals for sud-
den deterioration of complex diseases by dynamical network biomarkers,Sci.
Rep.2(2012)342–350.
[14]K.-Q.Liu,Z.-P.Liu,J.-K.Hao,L.Chen,X.-M.Zhao,Identifying dysregulated pathways
in cancers from pathway interaction networks,BMC Bioinformatics13(2012)11.
[15] D.He,Z.-P.Liu,M.Honda,S.Kaneko,L.Chen,Coexpression network analysis in
chronic hepatitis B and C hepatic lesion reveals distinct patterns of disease pro-gression to hepatocellular carcinoma,J.Mol.Cell Biol.32(2012)32.
[16]W.L.Tsai,R.T.Chung,Viral hepatocarcinogenesis,Oncogene29(2010)2309–2324.
[17] C.-J.Yen,Yih-Jyh Lin,C.-S.Yen,H.-W.Tsai,T.-F.Tsai,K.-Y.Chang,W.-C.Huang,
P.-W.Lin,C.-W.Chiang,T.-T.Chang,Hepatitis B virus X protein upregulates mTOR signaling through IKKb to increase cell proliferation and VEGF production in hepatocellular carcinoma,PLoS One7(2012).
[18]S.P.Hussain,J.Schwank,F.Staib,X.W.Wang,C.C.Harris,TP53mutations and
hepatocellular carcinoma:insights into the etiology and pathogenesis of liver cancer,Oncogene26(2007)2166–2176.
[19] A.Banerjee,R.B.Ray,R.Ray,Oncogenic potential of hepatitis C virus proteins,
Viruses2(2010)2108–2133.
[20]G.Castello,S.Scala,G.Palmieri,S.A.Curley,F.Izzo,HCV-related hepatocellular
carcinoma:from chronic in?ammation to cancer,Clin.Immunol.134(2010) 237–250.
[21] B.C.B.Ko,J.Liu,X.Ding,J.Tang,Y.Cao,P.Hu,F.Zhou,X.Shan,X.Cai,Q.Chen,N.
Ling,B.Zhang,Y.Bi,K.Chen,H.Ren,A.Huang,T.-C.He,N.Tang,Enhancement of canonical Wnt/β-catenin signaling activity by HCV core protein promotes cell growth of hepatocellular carcinoma cells,PLoS One6(2011)e27496.
[22]K.Matsuzaki,M.Murata,K.Yoshida,G.Sekimoto,Y.Uemura,N.Sakaida,M.Kaibori,
Y.Kamiyama,M.Nishizawa,J.Fujisawa,K.Okazaki,T.Seki,Chronic in?ammation as-sociated with hepatitis C virus infection perturbs hepatic transforming growth factor βsignaling,promoting cirrhosis and hepatocellular carcinoma,Hepatology46 (2007)48–57.
[23]M.Hassan,H.Ghozlan,O.Abdel-Kader,Activation of c-Jun NH2-terminal kinase
(JNK)signaling pathway is essential for the stimulation of hepatitis C virus(HCV) non-structural protein3(NS3)-mediated cell growth,Virology333(2005)324–336.
278W.Yuan et al./Biochimica et Biophysica Acta1844(2014)271–279
[24]Keng-Hsin Lan,Meei-Ling Sheu,Shinn-Jang Hwang,Sang-Hue Yen,Shiow-Yi
Chen,Wu.Jaw-Ching,Yuan-Jan Wang,Naoya Kato,Masao Omata,Full-Young Chang,Shou-Dong Lee,HCV NS5A interacts with p53and inhibits p53-mediated apoptosis,Oncogene2(1)(2002)11.
[25]Y.-B.Deng,G.Nagae,Y.Midorikawa,K.Yagi,S.Tsutsumi,S.Yamamoto,K.
Hasegawa,N.Kokudo,H.Aburatani,A.Kaneda,Identi?cation of genes preferen-tially methylated in hepatitis C virus-related hepatocellular carcinoma,Cancer Sci.101(2010)1501–1510.
[26]L.Gautier,L.Cope, B.M.Bolstad,R.A.Irizarry,affy—analysis of Affymetrix
GeneChip data at the probe level,Bioinformatics20(2004)307–315.
[27]R.A.Irizarry,B.Hobbs,F.Collin,T.P.Speed,Exploration,normalization,and summaries
of high density oligonucleotide array probe level data,Biostatistics4(2003)249–264.
[28]V.G.Tusher,R.Tibshirani,G.Chu,Signi?cance analysis of microarrays applied to
the ionizing radiation response,Proc.Natl.Acad.Sci.98(2001)5116–5121. [29] D.Szklarczyk,A.Franceschini,M.Kuhn,M.Simonovic,A.Roth,P.Minguez,T.
Doerks,M.Stark,J.Muller,P.Bork,L.J.Jensen,C.von Mering,The STRING database in2011:functional interaction networks of proteins,globally integrated and scored,Nucleic Acids Res.39(2011)D561–D568.
[30] E.D.Harrington,L.J.Jensen,P.Bork,Predicting biological networks from genomic
data,FEBS Lett.582(2008)1251–1258.
[31]L.Skrabanek,H.K.Saini,G.D.Bader,A.J.Enright,Computational prediction of pro-
tein–protein interactions,Mol.Biotechnol.38(2008)1–17.
[32]H.-C.Lu,F.-P.Chou,K.-T.Yeh,Y.-S.Chang,N.C.Hsu,J.-G.Chang,Expression of protein
kinase C family in human hepatocellular carcinoma,Pathol.Oncol.Res.16(2010) 385–391.
[33]Zhi-Ping Liu,Yong Wang,Tieqiao Wen,Xiang-Sun Zhang,W.Xia,L.Chen,Dynam-
ically dysfunctional protein interactions in the development of Alzheimer's disease,IEEE6(2009).
[34]T.Huang,L.Liu,Q.Liu,G.Ding,Y.Tan,Z.Tu,Y.Li,H.Dai,L.Xie,The role of
hepatitis C virus in the dynamic protein interaction networks of hepatocellular cirrhosis to carcinoma,https://www.wendangku.net/doc/cc2517930.html,put.Biol.Drug Des.4(2011)14.
[35] D.Hwang,A.G.Rust,S.Ramsey,J.J.Smith,D.M.Leslie,A.D.Weston,P.de Atauri,
J.D.Aitchison,L.Hood,A.F.Siegel,H.Bolouri,A data integration methodology for systems biology,Proc.Natl.Acad.Sci.U.S.A.102(2005)17296–17301. [36] F.Li,Y.Peng,W.Xu,G.Yu,P.Li,X.Bo,S.Wang,Towards a comprehensive HBV–
human interaction map,IEEE Comput.Soc.(2009)311–314.
[37] B.de Chassey,V.Navratil,L.Tafforeau,M.S.Hiet,A.Aublin-Gex,S.Agaugué,G.
Meiffren,F.Pradezynski,B.F.Faria,T.Chantier,M.Le Breton,J.Pellet,N.Davoust, P.E.Mangeot,A.Chaboud,F.Penin,Y.Jacob,P.O.Vidalain,M.Vidal,P.André,C.
Rabourdin-Combe,V.Lotteau,Hepatitis C virus infection protein network,Mol.
Syst.Biol.4(2008).
[38]L.Zeng,J.Yu,T.Huang,H.Jia,Q.Dong,F.He,W.Yuan,L.Qin,Y.Li,L.Xie,Differ-
ential combinatorial regulatory network analysis related to venous metastasis of hepatocellular carcinoma,BMC Bioinformatics(2012)2012.
[39]R.Khanin,E.Wit,How scale-free are biological networks,https://www.wendangku.net/doc/cc2517930.html,put.Biol.13(2006)
810–818.[40] E.W.Dijkstra,A note on two problems in connexion with graphs,Numer.Math.
(1959)3.
[41]G.D.Jr,B.T.Sherman,D.A.Hosack,J.Yang,W.Gao,https://www.wendangku.net/doc/cc2517930.html,ne,R.A.Lempicki,
DAVID:database for annotation visualization and integrated discovery,Genome Biol.4(2003)3.
[42]L.Weng,J.Du,Q.Zhou,B.Cheng,J.Li,D.Zhang,C.Ling,Identi?cation of cyclin B1
and Sec62as biomarkers for recurrence in patients with HBV-related hepatocel-lular carcinoma after surgical resection,Mol.Cancer11(2012)39.
[43]J.Faivre,M.Frank-Vaillant,R.Poulhe,H.Mouly, C.Jessus, C.Bréchot,J.
Sobczak-Thépot,Centrosome overduplication,increased ploidy and transforma-tion in cells expressing endoplasmic reticulum-associated cyclin A2,Oncogene 21(2002).
[44]https://www.wendangku.net/doc/cc2517930.html,n,M.-L.Sheu,S.-J.Hwang,S.-H.Yen,S.-Y.Chen,J.-C.Wu,Y.-J.Wang,N.
Kato,M.Omata,F.-Y.Chang,S.-D.Lee,HCV NS5A interacts with p53and inhibits p53-mediated apoptosis,Oncogene21(2002)11.
[45]N.Pavio,S.Battaglia,D.Boucreux,B.Arnulf,R.Sobesky,O.Hermine,C.Brechot,
Hepatitis C virus core variants isolated from liver tumor but not from adjacent non-tumor tissue interact with Smad3and inhibit the TGF-βpathway,Oncogene 24(2005)6119–6132.
[46] C.Ozen,G.Yildiz,A.Dagcan,D.Cevik,A.Ors,U.Keles,H.Topel,M.Ozturk,Genetics
and epigenetics of liver cancer,New Biotechnol.(2013).
[47]R.Bonilla Guerrero,L.R.Roberts,The role of hepatitis B virus integrations in
the pathogenesis of human hepatocellular carcinoma,J.Hepatol.42(2005) 760–777.
[48]Y.Murakami,Large scaled analysis of hepatitis B virus(HBV)DNA integration in
HBV related hepatocellular carcinomas,Gut54(2005)1162–1168.
[49] A.Maki,H.Kono,M.Gupta,M.Asakawa,T.Suzuki,M.Matsuda,H.Fujii,I.Rusyn,Pre-
dictive power of biomarkers of oxidative stress and in?ammation in patients with hepatitis C virus-associated hepatocellular carcinoma,Ann.Surg.Oncol.14(2007) 1182–1190.
[50]M.A.Khattab,M.Eslam,Y.I.Mousa,N.Ela-adawy,S.Fathy,M.Shatat,H.
Abd-Aalhaim,M.M.Kamal,M.A.Sharawe,Association between metabolic abnormal-ities and hepatitis C-related hepatocellular carcinoma,Ann.Hepatol.11(2012)7.
[51]M.Okita,K.Tomioka,Y.Ota,T.Sasagawa,T.Osawa,N.Sakai,M.Kawaguchi,T.
Itoshima,Arachidonic acid in mononuclear cells and its clinical signi?cance in HCV cirrhotic patients,Nutrition19(2003)727–732.
[52]Norio Iizuka,Masaaki Oka,Hisafumi Yamada-Okabe,Comparison of gene expression
pro?les between hepatitis B carcinoma virus-and hepatitis C virus-infected HCC oligonucleotide microarray data on the basis of a supervised learning method, Cancer Res.62(2002).
[53]M.Honda,T.Yamashita,T.Ueda,H.Takatori,R.Nishino,S.Kaneko,Different
signaling pathways in the livers of patients with chronic hepatitis B or chronic hepatitis C,Hepatology44(2006)1122–1138.
[54]L.D.Presser, A.Haskett,G.Waris,Hepatitis C virus-induced furin and
thrombospondin-1activate TGF-beta1:role of TGF-beta1in HCV replication, Virology412(2011)284–296.
279
W.Yuan et al./Biochimica et Biophysica Acta1844(2014)271–279
蛋白质的生理功能
蛋白质的生理功能 1、构造人的身体：蛋白质是一切生命的物质基础，是肌体细胞的重要组成部分，是人体组织更新和修补的主要原料。人体的每个组织：毛发、皮肤、肌肉、骨骼、内脏、大脑、血液、神经、内分泌等都是由蛋白质组成，所以说饮食造就人本身。蛋白质对人的生长发育非常重要。比如大脑发育的特点是一次性完成细胞增殖，人的大脑细胞的增长有二个高峰期。第一个是胎儿三个月的时候；第二个是出生后到一岁，特别是0---6个月的婴儿是大脑细胞猛烈增长的时期。到一岁大脑细胞增殖基本完成，其数量已达成人的9/10。所以0到1岁儿童对蛋白质的摄入要求很有特色，对儿童的智力发展尤关重要。 2、修补人体组织：人的身体由百兆亿个细胞组成，细胞可以说是生命的最小单位，它们处于永不停息的衰老、死亡、新生的新陈代谢过程中。例如年轻人的表皮28天更新一次，而胃黏膜两三天就要全部更新。所以一个人如果蛋白质的摄入、吸收、利用都很好，那么皮肤就是光泽而又有弹性的。反之，人则经常处于亚健康状态。组织受损后，包括外伤，不能得到及时和高质量的修补，便会加速机体衰退。 3、维持肌体正常的新陈代谢和各类物质在体内的输送。载体蛋白对维持人体的正常生命活动是至关重要的。可以在体内运载各种物质。比如血红蛋白—输送氧（红血球更新速率250万/秒）、脂蛋白—输送脂肪、细胞膜上的受体还有转运蛋白等。 4、白蛋白：维持机体内的渗透压的平衡及体液平衡。 5、维持体液的酸碱平衡。 6、免疫细胞和免疫蛋白：有白细胞、淋巴细胞、巨噬细胞、抗体（免疫球蛋白）、补体、干扰素等。七天更新一次。当蛋白质充足时，这个部队就很强，在需要时，数小时内可以增加100倍。 7、构成人体必需的催化和调节功能的各种酶。我们身体有数千种酶，每一种只能参与一种生化反应。人体细胞里每分钟要进行一百多次生化反应。酶有促进食物的消化、吸收、利用的作用。相应的酶充足，反应就会顺利、快捷的进行，我们就会精力充沛，不易生病。否则，反应就变慢或者被阻断。 8、激素的主要原料。具有调节体内各器官的生理活性。胰岛素是由51个氨基酸分子合成。生长素是由191个氨基酸分子合成。 9、提供热能。蛋白质和健康蛋白质是荷兰科学家格里特在1838年发现的。他观察到有生命的东西离开了蛋白质就不能生存。蛋白质是生物体内一种极重要的高分子有机物，占人体干重的54%。蛋白质主要由氨基酸组成，因氨基酸的组合排列不同而组成各种类型的蛋白质。人体中估计有10万种以上的蛋白质。生命是物质运动的高级形式，这种运动方式是通过蛋白质来实现的，所以蛋白质有极其重要的生物学意义。人体的生长、发育、运动、遗传、繁殖等一切生命活动都离不开蛋白质。生命运动需要蛋白质，也离不开蛋白质。人体内的一些生理活性物质如胺类、神经递质、多肽类激素、抗体、酶、核蛋白以及细胞膜上、血液中起“载体”作用的蛋白都离不开蛋白质，它对调节生理功能，维持新陈代谢起着极其重要的作用。人体运动系统中肌肉的成分以及肌肉在收缩、作功、完成动作过程中的代谢无不与蛋白质有关，离开了蛋白质，体育锻炼就无从谈起。在生物学中，蛋白质被解释为是由氨基酸借肽键联接起来形成的多肽，然后由多肽连接起来形成的物质。通俗易懂些说，它就是构成人体组织器官的支架和主要物质。 蛋白质能供给能量。这不是蛋白质的主要功能，我们不能拿“肉”当“柴”烧。但在能量缺乏时，蛋白质也必须用于产生能量。另外，从食物中摄取的蛋白质，有些不符合人体需要，或者摄取数量过多，也会被氧化分解，释放能量。
蛋白质的主要生理功能和作用
蛋白质的主要生理功能和作用 张世林外语学院日语14.1 学号：201407030120 摘要本文阐述了蛋白质的定义概念、组成特点、结构性质、生理功能以及作用。 关键词历史定义组成特点结构性质功能 正文： 在18世纪，安东尼奥·弗朗索瓦（Antoine Fourcroy）和其他一些研究者发现蛋白质是一类独特的生物分子，他们发现用酸处理一些分子能够使其凝结或絮凝。当时他们注意到的例子有来自蛋清、血液、血清白蛋白、纤维素和小麦面筋里的蛋白质。荷兰化学家格利特·马尔德（Gerhardus Johannes Mulder）对一般的蛋白质进行元素分析发现几乎所有的蛋白质都有相同的实验公式。用“蛋白质”这一名词来描述这类分子是由Mulder的合作者永斯·贝采利乌斯于1838年提出。Mulder随后鉴定出蛋白质的降解产物，并发现其中含有为氨基酸的亮氨酸，并且得到它（非常接近正确值）的分子量为131Da。 对于早期的生物化学家来说，研究蛋白质的困难在于难以纯化大量的蛋白质以用于研究。因此，早期的研究工作集中于能够容易地纯化的蛋白质，如血液、蛋清、各种毒素中的蛋白质以及消化性和代谢酶（获取自屠宰场）。1950年代后期，Armour Hot Dog Co.公司纯化了一公斤纯的牛胰腺中的核糖核酸酶A，并免费提供给全世界科学家使用。
这一构想最早是由威廉·阿斯特伯里于1933年提出。随后，Walter Kauzman在总结自己对变性的研究成果和之前Kaj Linderstrom-Lang的研究工作的基础上，提出了蛋白质折叠是由疏水相互作用所介导的。1949年，弗雷德里克·桑格首次正确地测定了胰岛素的氨基酸序列，并验证了蛋白质是由氨基酸所形成的线性（不具有分叉或其他形式）多聚体。原子分辨率的蛋白质结构首先在1960年代通过X射线晶体学获得解析；到了1980年代，NMR也被应用于蛋白质结构的解析；近年来，冷冻电子显微学被广泛用于对于超大分子复合体的结构进行解析。截至到2008年2月，蛋白质数据库中已存有接近50,000个原子分辨率的蛋白质及其相关复合物的三维结构的坐标。 蛋白质是一种复杂的有机化合物，旧称“朊（ruǎn)”。氨基酸是组成蛋白质的基本单位，氨基酸通过脱水缩合连成肽链。蛋白质是由一条或多条多肽链组成的生物大分子，每一条多肽链有二十至数百个氨基酸残基（-R）不等；各种氨基酸残基按一定的顺序排列。蛋白质的氨基酸序列是由对应基因所编码。除了遗传密码所编码的20种基本氨基酸，在蛋白质中，某些氨基酸残基还可以被翻译后修饰而发生化学结构的变化，从而对蛋白质进行激活或调控。多个蛋白质可以一起，往往是通过结合在一起形成稳定的蛋白质复合物，折叠或螺旋构成一定的空间结构，从而发挥某一特定功能。合成多肽的细胞器是细胞质中
蛋白质的生理作用.
《食品化学与健康》电子教材 蛋白质的生理作用 一、是人体最重要的组成成分 人体中所有重要组织都有蛋白质参与如神经、肌肉、内脏、血液等都含有蛋白质。蛋白质是构成细胞和组织的“建筑材料”，在人体细胞中的含量仅次于水，占细胞干重的50%以上。一切生物膜，如细胞膜、细胞内各种细胞器的膜，几乎都是由蛋白质和脂类等物质组成。蛋白质是生命活动的重要物质基础。在体内多种重要生理活性物质的成分是蛋白质，蛋白质参与调节生理功能，如构成细胞核的核蛋白能影响细胞功能；促进食物消化、吸收和利用作用的是酶蛋白；维持机体免疫功能作用的是免疫蛋白；具有调节肌肉收缩的功能的是肌球蛋白；具有运送营养素的作用的是血液中的脂蛋白、运铁蛋白、视黄醇结合蛋白质；具有携带、运送氧气功能的是血红蛋白；具有调节渗透压、维持体液平衡的作用(肝癌) 是白蛋白；由蛋白质或蛋白质衍生物构成的某些激素，如垂体激素、甲状腺激素、胰岛素及肾上腺素等等都是机体的重要调节物质。蛋白质能向机体提供能量，大约占总热能的14%，每克蛋白质在体内代谢，能产生4千卡左右的能量。 二、蛋白质的生理作用表现为 1.参与生理活动和劳动做功 心脏跳动、呼吸运动、胃肠蠕动以及日常各种劳动做功等，都离不开肌肉的收缩，而骨肉的收缩又离不开具有骨肉收缩功能的蛋白质。 2.参与氧和二氧化碳的运输 在生命活动中，将氧气供给全身组织，同时将新陈代谢所产生的二氧化碳排出体外的运输工具就是血红蛋白。血红蛋白是红细胞的主要成分，也是红细胞行使其功能的物质基础。 3.参与维持人体的渗透压
血浆中有多种蛋白质，对维持血液的渗透压、维持细胞内外的压力平衡起着重要作用。 4.具有防御功能 血浆中含有的抗体，主要是丙种球蛋白，这是一种具有防御功能的蛋白质。 5.参与调节人体内物质的代谢 在物质代谢中，都需要酶系统的催化或调节，而酶的本质就是蛋白质。在调节代谢过程中，蛋白质以酶和激素的形式出现，发挥了生命活动中“指挥员”的作用。
蛋白质结构预测在线软件
蛋白质预测在线分析常用软件推荐 蛋白质预测分析网址集锦 物理性质预测： Compute PI/MW http://expaxy.hcuge.ch/ch2d/pi-tool.html Peptidemasshttp://expaxy.hcuge.ch/sprot/peptide-mass.html TGREASE ftp://https://www.wendangku.net/doc/cc2517930.html,/pub/fasta/ SAPS http://ulrec3.unil.ch/software/SAPS_form.html 基于组成的蛋白质识别预测 AACompIdent http://expaxy.hcuge.ch ... htmlAACompSim http://expaxy.hcuge.ch/ch2d/aacsim.html PROPSEARCH http://www.e mbl-heidelberg.de/prs.html 二级结构和折叠类预测 nnpredict https://www.wendangku.net/doc/cc2517930.html,/~nomi/nnpredict Predictprotein http://www.embl-heidel ... protein/SOPMA http://www.ibcp.fr/predict.html SSPRED http://www.embl-heidel ... prd_info.html 特殊结构或结构预测 COILS http://ulrec3.unil.ch/ ... ILS_form.html MacStripe https://www.wendangku.net/doc/cc2517930.html,/ ... acstripe.html 与核酸序列一样，蛋白质序列的检索往往是进行相关分析的第一步，由于数据库和网络技校术的发展，蛋白序列的检索是十分方便，将蛋白质序列数据库下载到本地检索和通过国际互联网进行检索均是可行的。 由NCBI检索蛋白质序列 可联网到：“http://www.ncbi.nlm.ni ... gi?db=protein”进行检索。 利用SRS系统从EMBL检索蛋白质序列 联网到：https://www.wendangku.net/doc/cc2517930.html,/”，可利用EMBL的SRS系统进行蛋白质序列的检索。 通过EMAIL进行序列检索 当网络不是很畅通时或并不急于得到较多数量的蛋白质序列时，可采用EMAIL方式进行序列检索。 蛋白质基本性质分析 蛋白质序列的基本性质分析是蛋白质序列分析的基本方面，一般包括蛋白质的氨基酸组成，分子质量，等电点，亲水性，和疏水性、信号肽，跨膜区及结构功能域的分析等到。蛋白质的很多功能特征可直接由分析其序列而获得。例如，疏水性图谱可通知来预测跨膜螺旋。同时，也有很多短片段被细胞用来将目的蛋白质向特定细胞器进行转移的靶标（其中最典型的
蛋白质的营养生理作用
“蛋白质”一词，源于希腊字“Proteios”，其意是“最初的”、“第一重要的”；蛋白质是细胞的重要组成成份，在生命过程中起着重要的作用, 涉及动物代谢的大部分与生命攸关的化学反应。不同种类动物都有自己特定的、多种不同的蛋白质。在器官、体液和其它组织中，没有两种蛋白质的生理功能是完全一样的。这些差异是由于组成蛋白质的氨基酸种类、数量和结合方式不同的必然结果。 动物在组织器官的生长和更新过程中，必须从食物中不断获取蛋白质等含氮物质。因此，把食物中的含氮化合物转变为机体蛋白质是一个重要的营养过程。 蛋白质在动物的生命活动中的重要营养作用： (一)蛋白质是构建机体组织细胞的主要原料 动物的肌肉、神经、结缔组织、腺体、精液、皮肤、血液、毛发、角、喙等都以蛋白质为主要成份，起着传导、运输、支持、保护、连接、运动等多种功能。肌肉、肝、脾等组织器官的干物质含蛋白质80%以上。蛋白质也是乳、蛋、毛的主要组成成份。除反刍动物外，食物蛋白质几乎是唯一可用以形成动物体蛋白质的氮来源。 (二)蛋白质是机体内功能物质的主要成份 在动物的生命和代谢活动中起催化作用的酶、某些起调节作用的激素、具有免疫和防御机能的抗体（免疫球蛋白）都是以蛋白质为主要成分。另外，蛋白质对维持体内的渗透压和水分的正常分布，也起着重要的作用。 (三) 蛋白质是组织更新、修补的主要原料 在动物的新陈代谢过程中，组织和器官的蛋白质的更新、损伤组织的修补都需要蛋白质。据同位素测定，全身蛋白质6-7个月可更新一半。 (四)蛋白质可供能和转化为糖、脂肪 在机体能量供应不足时，蛋白质也可分解供能，维持机体的代谢活动。当摄入蛋白质过多或氨基酸不平衡时，多余的部分也可能转化成糖、脂肪或分解产热。正常条件下，鱼等水生动物体内亦有相当数量的蛋白质参与供能作用。 “蛋白质”一词，源于希腊字“Proteios”，其意是“最初的”、“第一重要的”；蛋白质是细胞的重要组成成份，在生命过程中起着重要的作用, 涉及动物代谢的大部分与生命攸关的化学反应。不同种类动物都有自己特定的、多种不同的蛋白质。在器官、体液和其它组织中，没有两种蛋白质的生理功能是完全一样的。这些差异是由于组成蛋白质的氨基酸种类、数量和结合方式不同的必然结果。 动物在组织器官的生长和更新过程中，必须从食物中不断获取蛋白质等含氮物质。因此，把食物中的含氮化合物转变为机体蛋白质是一个重要的营养过程。 蛋白质在动物的生命活动中的重要营养作用： (一)蛋白质是构建机体组织细胞的主要原料 动物的肌肉、神经、结缔组织、腺体、精液、皮肤、血液、毛发、角、喙等都以蛋白质为主要成份，起着传导、运输、支持、保护、连接、运动等多种功能。肌肉、肝、脾等组织器官的干物质含蛋白质80%以上。蛋白质也是乳、蛋、毛的主要组成成份。除反刍动物外，食物蛋白质几乎是唯一可用以形成动物体蛋白质的氮来源。 (二)蛋白质是机体内功能物质的主要成份 在动物的生命和代谢活动中起催化作用的酶、某些起调节作用的激素、具有免疫和防御机能的抗体（免疫球蛋白）都是以蛋白质为主要成分。另外，蛋白质对维持体内的渗透压和水分的正常分布，也起着重要的作用。 (三) 蛋白质是组织更新、修补的主要原料 在动物的新陈代谢过程中，组织和器官的蛋白质的更新、损伤组织的修补都需要蛋白质。据同位素测定，全身蛋白质6-7个月可更新一半。
蛋白质结构预测和序列分析软件
蛋白质结构预测和序列分析软件蛋白质数据库及蛋白质序列分析 第一节、蛋白质数据库介绍 一、蛋白质一级数据库 1、 SWISS-PROT 数据库 SWISS-PROT和PIR是国际上二个主要的蛋白质序列数据 库，目前这二个数据库在EMBL和GenBank数据库上均建 立了镜像 (mirror) 站点。 SWISS-PROT数据库包括了从EMBL翻译而来的蛋白质序 列，这些序列经过检验和注释。该数据库主要由日内瓦大 学医学生物化学系和欧洲生物信息学研究所(EBI)合作维 护。SWISS-PROT的序列数量呈直线增长。 2、TrEMBL数据库： SWISS-PROT的数据存在一个滞后问题，即 进行注释需要时间。一大批含有开放阅读 了解决这一问题，TrEMBL(Translated E 白质数据库，它包括了所有EMBL库中的 质序列数据源，但这势必导致其注释质量 3、PIR数据库： PIR数据库的数据最初是由美国国家生物医学研究基金 会（National Biomedical Research Foundation, NBRF） 收集的蛋白质序列，主要翻译自GenBank的DNA序列。 1988年，美国的NBRF、日本的JIPID（the Japanese International Protein Sequence Database日本国家蛋 白质信息数据库）、德国的MIPS（Munich Information Centre for Protein Sequences摹尼黑蛋白质序列信息 中心）合作，共同收集和维护PIR数据库。PIR根据注释 程度(质量)分为4个等级。 4、 ExPASy数据库： 目前，瑞士生物信息学研究所（Swiss I 质分析专家系统（Expert protein anal 据库。 网址：https://www.wendangku.net/doc/cc2517930.html, 我国的北京大学生物信息中心(www.cbi.
蛋白质的作用及功能
（1）氨基酸、蛋白质的生理功能 蛋白质是人体必需的主要营养物质。蛋白质的分解产物是氨基酸；氨基酸的重要生理功能之一是作为蛋白质、多肽合成的原料，是蛋白质或多肽的基本组成单位。 蛋白质的生理功能： ①维持组织的生长、更新和修复：膳食中必须提供足够质和量的蛋白质，才能维持组织、细胞的生长、更新和修复。 ②参与多种重要的生理功能：如催化功能、调节功能、运输功能、储存功能、保护功能和维持体液胶体渗透压（如清蛋白）等。 ③氧化供能：体内蛋白质、多肽分解成氨基酸后，产生（17.19kJ/g）能量，成人每日约有18%的能量来自蛋白质。 ④转变为糖类和脂肪。 （2）营养必需氨基酸的概念和种类 体内需要而不能自身合成、或合成量不能满足机体需要，必须由食物供应的氨基酸称为营养必需氨基酸。营养必需氨基酸包括赖氨酸、色氨酸、苯丙氨酸、甲硫氨酸、苏氨酸、亮氨酸、异亮氨酸和缬氨酸。 蛋白质的功能和对人体的作用 人体的所有组织器官都会有蛋白质，蛋白质是生命的物质基础。蛋白质是人体的主要“建筑材料”。婴幼儿靠它形成肌肉、血液、骨骼、神经、毛发等；成年人需要它更新组织，修补损伤、老化的机体。没有蛋白质的供给，人就不可能从3～4千克的新生儿长成50～60千克重的成年人，所以说蛋白质是人体生命得以延续的主要物质基础。它在人体内的功能共有6 个方面： ◎ 结构功能与催化调节功能 蛋白质是构成体内各组织的主要成分，蛋白质在人体内的主要功能是构成组织和修补组织。人的大脑、神经、肌肉、内脏、血液、皮肤乃至指甲、头发等都是以蛋白质为主要成分构成的。人体发育成长后，随着机体内新陈代谢的不断进行，部分蛋白质分解，组织衰老更新以及损伤后的组织修补等都需要不断补充蛋白质。所以，人每天都要补充一定量的蛋白质，以满足身体的正常需要。人体内的化学变化几乎都是在酶的催化下不断进行的。激素对代谢的调节作用也具有重要意义，而酶和激素都直接或间接来自于蛋白质。 ◎ 防御功能与运动功能 机体抵抗力的强弱，取决于抵抗疾病的抗体的多少，抗体的生成与蛋白质有密切关系。近年来被誉为抑制病毒的法宝和抗癌生力军的干扰素，也是一种复合蛋白质（糖和蛋白质结合而成）。肌肉收缩依赖于肌球蛋白和肌动蛋白，有肌肉收缩才有躯体运动、呼吸、消化及血液循环等生理活动。 ◎ 供给热能与运输和存储功能 人体每日需要的能量，主要来自于糖类及脂肪。当蛋白质的量超过人体的需要，或者饮食中的糖类、脂肪供给不足时，蛋白质亦可作为热量的来源。另外，在人体新陈代谢过程中，被更新的组织蛋白亦可氧化产生热能，供给人体的需要。不论是营养素的吸收、运输和储存以及其他物质的运输和储存，都有特殊蛋白质作为载体。如氧和二氧化碳在血液中的运输、脂类的运输、铁的运输和储存都与蛋白质有密切的关系。
蛋白质结构预测在线软件
蛋白质预测分析网址集锦? 物理性质预测：? Compute PI/MW?? ?? SAPS?? 基于组成的蛋白质识别预测? AACompIdent???PROPSEARCH?? 二级结构和折叠类预测? nnpredict?? Predictprotein??? SSPRED?? 特殊结构或结构预测? COILS?? MacStripe?? 与核酸序列一样，蛋白质序列的检索往往是进行相关分析的第一步，由于数据库和网络技校术的发展，蛋白序列的检索是十分方便，将蛋白质序列数据库下载到本地检索和通过国际互联网进行检索均是可行的。? 由NCBI检索蛋白质序列? 可联网到：“”进行检索。? 利用SRS系统从EMBL检索蛋白质序列? 联网到：”，可利用EMBL的SRS系统进行蛋白质序列的检索。? 通过EMAIL进行序列检索?
当网络不是很畅通时或并不急于得到较多数量的蛋白质序列时，可采用EMAIL方式进行序列检索。? 蛋白质基本性质分析? 蛋白质序列的基本性质分析是蛋白质序列分析的基本方面，一般包括蛋白质的氨基酸组成，分子质量，等电点，亲水性，和疏水性、信号肽，跨膜区及结构功能域的分析等到。蛋白质的很多功能特征可直接由分析其序列而获得。例如，疏水性图谱可通知来预测跨膜螺旋。同时，也有很多短片段被细胞用来将目的蛋白质向特定细胞器进行转移的靶标（其中最典型的例子是在羧基端含有KDEL序列特征的蛋白质将被引向内质网。WEB中有很多此类资源用于帮助预测蛋白质的功能。? 疏水性分析? 位于ExPASy的ProtScale程序（?）可被用来计算蛋白质的疏水性图谱。该网站充许用户计算蛋白质的50余种不同属性，并为每一种氨基酸输出相应的分值。输入的数据可为蛋白质序列或SWISSPROT数据库的序列接受号。需要调整的只是计算窗口的大小（n）该参数用于估计每种氨基酸残基的平均显示尺度。? 进行蛋白质的亲/疏水性分析时，也可用一些windows下的软件如，bioedit,dnamana等。? 跨膜区分析? 有多种预测跨膜螺旋的方法，最简单的是直接，观察以20个氨基酸为单位的疏水性氨基酸残基的分布区域，但同时还有多种更加复杂的、精确的算法能够预测跨膜螺旋的具体位置和它们的膜向性。这些技术主要是基于对已知
8.1 蛋白质的生理功能和营养价值
8.1 蛋白质的生理功能和营养价值维持细胞组织的生长、更新和修补 参与体内多种重要的生理活动 氧化供能 蛋白质 生理功能physiological function of protein 催化（酶）、调节（激素）、免疫（抗原及抗体）、运动（肌肉）、物质转运（载体）、凝血（凝血系统）等。 人体每日18%能量由蛋白质提供。 蛋白质占人体重量的16%~20%。 不可替代
?体内蛋白质的代谢状况可用氮平衡描述 氮平衡(nitrogen balance)指每日氮的摄入量与排出量之间的关系。粪便尿液 氮正平衡氮负平衡正常成人 饥饿、严重烧伤、出血、消耗性疾病患者 氮总平衡 儿童、孕妇、恢复期病人
食物蛋白质的营养需求 ?量：蛋白质的生理需要量 ●成人每日蛋白质最低生理需 要量为30g-50g ●我国营养学会推荐成人每日 蛋白质需要量为80g ?质：蛋白质的营养价值 营养必需氨基酸（nutritional essential amino acid）指体内需要而又不能自身合成，必须由食物供给的氨基酸。 通常认为有8种必需氨基酸，分别是亮氨酸、异亮氨酸、苏氨酸、缬氨酸、赖氨酸、甲硫氨酸、苯丙氨酸和色氨酸。现在认为，组氨酸也是一种必需氨基酸。
蛋白质的营养价值（nutrition value ） 蛋白质的营养价值是指食物蛋白质在体内的利用率，取决于必需氨基酸的数量、种类、量质比。 动物性蛋白质所含必需氨基酸的种类和比例与人体需要接近，所以营养 价值较高。 蛋白质的互补作用指营养价值较低的蛋白质混合食用，其必需氨基酸可以互相补充而提高营养价值。 Complementary 鱼和豆腐同吃，提高营养价值
蛋白质结构预测方法综述
蛋白质结构预测方法综述 卜东波陈翔王志勇 《计算机不能做什么？》是一本好书，其中文版序言也堪称佳构。在这篇十余页的短文中，马希文教授总结了使用计算机解决实际问题的三步曲，即首先进行形式化，将领域相关的实际问题抽象转化成一个数学问题；然后分析问题的可计算性；最后进行算法设计，分析算法的时间和空间复杂度，寻找最优算法。 蛋白质空间结构预测是很有生物学意义的问题，迄今亦有很多的工作。有意思的是，其中一些典型工作恰恰是上述三步曲的绝好示例，本文即沿着这一路线作一总结，介绍于后。 1 背景知识 生物细胞种有许多蛋白质（由20余种氨基酸所形成的长链），这些大分子对于完成生物功能是至关重要的。蛋白质的空间结构往往决定了其功能，因此，如何揭示蛋白质的结构是非常重要的工作。 生物学界常常将蛋白质的结构分为4个层次：一级结构，也就是组成蛋白质的氨基酸序列；二级结构，即骨架原子间的相互作用形成的局部结构，比如alpha螺旋,beta片层和loop区等；三级结构，即二级结构在更大范围内的堆积形成的空间结构；四级结构主要描述不同亚基之间的相互作用。 经过多年努力，结构测定的实验方法得到了很好的发展，比较常用的有核磁共振和X光晶体衍射两种。然而由于实验测定比较耗时和昂贵，对于某些不易结晶的蛋白质来说不适用。相比之下，测定蛋白质氨基酸序列则比较容易。因此如果能够从一级序列推断出空间结构则是非常有意义的工作。这也就是下面的蛋白质折叠问题： 1蛋白质折叠问题（Protein Folding Problem） 输入: 蛋白质的氨基酸序列
输出: 蛋白质的空间结构 蛋白质结构预测的可行性是有坚实依据的。因为一般而言，蛋白质的空间结构是由其一级结构确定的。生化实验表明：如果在体外无任何其他物质存在的条件下，使得蛋白质去折叠，然后复性，蛋白质将立刻重新折叠回原来的空间结构，整个过程在不到1秒种内即可完成。因此有理由认为对于大部分蛋白质而言，其空间结构信息已经完全蕴涵于氨基酸序列中。从物理学的角度讲，系统的稳定状态通常是能量最小的状态，这也是蛋白质预测工作的理论基础。 2 蛋白质结构预测方法 蛋白质结构预测的方法可以分为三种： 同源性（Homology ）方法：这类方法的理论依据是如果两个蛋白质的序列比较相似，则其结构也有很大可能比较相似。有工作表明，如果序列相似性高于75％，则可以使用这种方法进行粗略的预测。这类方法的优点是准确度高，缺点是只能处理和模板库中蛋白质序列相似性较高的情况。 从头计算（Ab initio ） 方法：这类方法的依据是热力学理论，即求蛋白质能量最小的状态。生物学家和物理学家等认为从原理上讲这是影响蛋白质结构的本质因素。然而由于巨大的计算量，这种方法并不实用，目前只能计算几个氨基酸形成的结构。IBM 开发的Blue Gene 超级计算机，就是要解决这个问题。 穿线法(Threading )方法：由于Ab Initio 方法目前只有理论上的意义，Homology 方法受限于待求蛋白质必需和已知模板库中某个蛋白质有较高的序列相似性，对于其他大部分蛋白质来说，有必要寻求新的方法。Threading 就此应运而生。 以上三种方法中，Ab Initio 方法不依赖于已知结构，其余两种则需要已知结构的协助。通常将蛋白质序列和其真实三级结构组织成模板库，待预测三级结构的蛋白质序列，则称之为查询序列(query sequence)。 3 蛋白质结构预测的Threading 方法 Threading 方法有三个代表性的工作：Eisenburg 基于环境串的工作、Xu Ying 的Prospetor 和Xu Jinbo 、Li Ming 的RAPTOR 。 Threading 的方法：首先取出一条模版和查询序列作序列比对(Alignment)，并将模版蛋白质与查询序列匹配上的残基的空间坐标赋给查询序列上相应的残基。比对的过程是在我们设计的一个能量函数指导下进行的。根据比对结果和得到的查询序列的空间坐标，通过我们设计的能量函数，得到一个能量值。将这个操作应用到所有的模版上，取能量值最低的那条模版产生的查询序列的空间坐标为我们的预测结果。 需要指出的是，此处的能量函数却不再是热力学意义上的能量函数。它实质上是概率的负对数，即 ，我们用统计意义上的能量来代替真实的分子能量，这两者有大致相同的形式。 p E log ?=如果沿着马希文教授的观点看上述工作 ，则更有意思：Eisenburg 指出如果仅仅停留在简单地使用每个原子的空间坐标(x,y,z)来形式化表示蛋白质空间结构，则难以进一步深入研究。Eisenburg 创造性地使用环境串表示结构，从而将结构预测问题转化成序列串和环境串之间的比对问题；其后，Xu Ying 作了进一步发展，将蛋白质序列表示成一系列核（core ）组成的序列，Core 和Core 之间存在相互作用。因此结构就表示成Core 的空间坐标，以及Core 之间的相互作用。在这种表示方法的基础上，Xu Ying 开发了一种求最优匹配的动态规划算法，得到了很好的结果。但是由于其较高的复杂度，在Prospetor2上不得不作了一些简化；Xu Jinbo 和Li Ming 很漂亮地解决了这个问题，将求最优匹配的过程表示成一个整数规划问题，并且证明了一些常用
蛋白质结构预测网址
蛋白质结构预测网址 物理性质预测： Compute PI/MW Peptidemass TGREASE SAPS 基于组成的蛋白质识别预测 AACompIdent PROPSEARCH 二级结构和折叠类预测 nnpredict Predictprotein SSPRED 特殊结构或结构预测 COILS MacStripe 与核酸序列一样，蛋白质序列的检索往往是进行相关分析的第一步，由于数据库和网络技校术的发展，蛋白序列的检索是十分方便，将蛋白质序列数据库下载到本地检索和通过国际互联网进行检索均是可行的。 由NCBI检索蛋白质序列 可联网到：“”进行检索。 利用SRS系统从EMBL检索蛋白质序列 联网到：”，可利用EMBL的SRS系统进行蛋白质序列的检索。 通过EMAIL进行序列检索 当网络不是很畅通时或并不急于得到较多数量的蛋白质序列时，可采用EMAIL方式进行序列检索。 蛋白质基本性质分析 蛋白质序列的基本性质分析是蛋白质序列分析的基本方面，一般包括蛋白质的氨基酸组成，分子质量，等电点，亲水性，和疏水性、信号肽，跨膜区及结构功能域的分析等到。蛋白质的很多功能特征可直接由分析其序列而获得。例如，疏水性图谱可通知来预测跨膜螺旋。同时，也有很多短片段被细胞用来将目的蛋白质向特定细胞器进行转移的靶标（其中最典型的例子是在羧基端含有KDEL序列特征的蛋白质将被引向内质网。WEB中有很多此类资源用于帮助预测蛋白质的功能。 疏水性分析 位于ExPASy的ProtScale程序（）可被用来计算蛋白质的疏水性图谱。该网站充许用户计算蛋白质的50余种不同属性，并为每一种氨基酸输出相应的分值。输入的数据可为蛋白质序列或SWISSPROT数据库的序列接受号。需要调整的只是计算窗口的大小（n）该参数用于估计每种氨基酸残基的平均显示尺度。 进行蛋白质的亲/疏水性分析时，也可用一些windows下的软件如， bioedit,dnamana等。 跨膜区分析 有多种预测跨膜螺旋的方法，最简单的是直接，观察以20个氨基酸为单位的疏水性氨基酸残基的分布区域，但同时还有多种更加复杂的、精确的算法能够预测跨膜螺旋的具体位置和它们的膜向性。这些技术主要是基于对已知跨膜螺旋的研究而得到的。自然存在的跨膜螺旋Tmbase 数据库，可通过匿名FTP获得()，参见表一
蛋白质结构预测
实习 5 ：蛋白质结构预测 学号20090***** 姓名****** 专业年级生命生技**** 实验时间2012.6.21 提交报告时间2012.6.21 实验目的： 1.学会使用GOR和HNN方法预测蛋白质二级结构 2.学会使用SWISS-MODEL进行蛋白质高级结构预测 实验内容： 1.分别用GOR和HNN方法预测蛋白质序列的二级结构，并对比异同性。 2.利用SWISS-MODEL进行蛋白质的三级结构预测，并对预测结果进行解释。 作业： 1. 搜索一条你感兴趣的蛋白质序列，分别用GOR和HNN进行二级结构预测，解释预测结果，分析两个方法结果有何异同。 答：所选用蛋白质序列为>>gi|390408302|gb|AFL70986.1| gag protein, partial [Human immunodeficiency virus] （1）GOR预测结果： 图1 图1是每个氨基酸在序列中所处的状态，可以看出序列的二级结构预测结果为： 1到9位个氨基酸为无规卷曲，10到33位氨基酸为α螺旋,34到37位为β折叠，38到45位为无规卷曲，46到49位为α螺旋，50到53位为无规卷曲，54到65为α螺旋，66到72位为无规卷曲，73到95位为α螺旋，96到101位为无规卷曲，102到108为β折叠，109到115位为无规卷曲，117位为β折叠。 图2 图2为各种结构在序列中所占的比例，其中Alpha helix占53.85%，Extended strand占11.11%，Random coil占35.04%，无他二级结构。
图3 图3为各个氨基酸在序列中的状态以及二级结构在全序列中二级结构分布情况。 （2）HNN预测： 图4 图4是每个氨基酸在序列中所处的状态，可以看出序列的二级结构预测结果为： 1到6位个氨基酸为无规卷曲，7到34位氨基酸为α螺旋,35到37位为β折叠，38位为α螺旋，39到44位为无规卷曲，45到49位为α螺旋，50到55位为无规卷曲，56到65为α螺旋，66到71位为无规卷曲，72到83位为α螺旋，84到86位为无规卷曲，87到95位为α螺旋，96到102为无规卷曲，103到108位为β折叠，108到117位为无规卷曲。 图5 图5为各种结构在序列中所占的比例，其中Alpha helix占55.56%，Extended strand占7.69%，Random coil占36.75%，无他二级结构。
13第二节 蛋白质的生理功能
第二节蛋白质的生理功能 一、构成和修复组织 蛋白质是构成机体组织、器官的重要成分，人体各组织、器官无一不含蛋白质。在 人体的瘦组织中，如肌肉组织和心、肝、肾等器官均含有大量蛋白质；骨骼、牙齿、乃至指、趾也含有大量蛋白质；细胞中，除水分外，蛋白质约占细胞内物质的80％。因此，构成机体组织、器官的成分是蛋白质最重要的生理功能。身体的生长发育可视为蛋白质的不断积累过程。蛋白质对生长发育期的儿童尤为重要。 人体内各种组织细胞的蛋白质始终在不断更新。例如，人血浆蛋白质的半寿期约为10 天，肝中大部分蛋白质的半寿期为1～8 天，某些蛋白质的半寿期很短，只有数秒钟。只有摄入足够的蛋白质方能维持组织的更新。身体受伤后也需要蛋白质作为修复材料。 二、调节生理功能 机体生命活动之所以能够有条不紊的进行，有赖于多种生理活性物质的调节。而蛋白质在体内是构成多种重要生理活性物质的成分，参与调节生理功能。如核蛋白构成细胞核并影响细胞功能；酶蛋白具有促进食物消化、吸收和利用的作用；免疫蛋白具有维持机体免疫功能的作用；收缩蛋白，如肌球蛋白具有调节肌肉收缩的功能；血液中的脂蛋白、运铁蛋白、视黄醇结合蛋白具有运送营养素的作用；血红蛋白具有携带、运送氧的功能；白蛋白具有调节渗透压、维持体液平衡的功能；由蛋白质或蛋白质衍生物构成的某些激素，如垂体激素、甲状腺素、胰岛素及肾上腺素等等都是机体的重要调节物质。 三、供给能量 蛋白质在体内降解成氨基酸后，经脱氨基作用生成的仅一酮酸，可以直接或间接经 三羧酸循环氧化分解，同时释放能量，是人体能量来源之一。但是，蛋白质的这种功能 可以由碳水化合物、脂肪所代替。因此，供给能量是蛋白质的次要功能。
蛋白质功能-结构-相互作用预测网站工具合集
蛋白质组学 蛋白质是生物体的重要组成部分，参与几乎所有生理和细胞代谢过程。此外，与基因组学和转录组学比较，对一个细胞或组织中表达的所有蛋白质，及其修饰和相互作用的大规模研究称为蛋白质组学。 蛋白质组学通常被认为是在基因组学和转录组学之后，生物系统研究的下一步。然而，蛋白质组的研究远比基因组学复杂，这是由于蛋白质内在的复杂特点，如蛋白质各种各样的翻译后修饰所决定的。并且，研究基因组学的技术要比研究蛋白质组学的技术强得多，虽然在蛋白质组学研究中，质谱技术的研究已取得了一些进展。 尽管存在方法上的挑战，蛋白质组学正在迅速发展，并且对癌症的临床诊断和疾病治疗做出了重要贡献。几项研究鉴定出了一些蛋白质在乳腺癌、卵巢癌、前列腺癌和食道癌中表达变化。例如，通过蛋白质组学技术，人们可以在患者血液中明确鉴定出肿瘤标志物。表1列出了更多的蛋白质组学技术用于研究癌症的例子。 另外，高尔基体功能复杂。最新研究表明，它除了参与蛋白加工外，还能参与细胞分化及细胞间信号传导的过程，并在凋亡中扮演重要角色，其功能障碍也许和肿瘤的发生、发展有某种联系。根据人类基因组研究，约1000多种人类高尔基体蛋白质中仅有500~600种得到了鉴定，建立一条关于高尔基体蛋白质组成的技术路线将有助于其功能的深入研究。 蛋白质组学是一种有效的研究方法，特别是随着亚细胞器蛋白质组学技术的迅猛发展，使高尔基体的全面研究变为可能。因此研究人员希望能以胃癌细胞中的高尔基体为研究对象，通过亚细胞器蛋白质组学方法，建立胃癌细胞中高尔基体的蛋白质组方法学。 研究人员采用蔗糖密度梯度的超速离心方法分离纯化高尔基体，双向凝胶电泳（2-DE）分离高尔基体蛋白质，用ImageMaster 2D软件分析所得图谱，基质辅助激光解吸离子化飞行时间质谱（MALDI-TOF MS）鉴定蛋白质点等一系列亚细胞器蛋白质组学方法建立了胃癌细胞内高尔基体的蛋白图谱。 最后，人们根据分离出的纯度较高的高尔基体建立了分辨率和重复性均较好的双向电泳图谱，运用质谱技术鉴定出12个蛋白质，包括蛋白合成相关蛋白、膜融合蛋白、调节蛋白、凋亡相关蛋白、运输蛋白和细胞增殖分化相关蛋白。通过亚细胞器分离纯化、双向电泳的蛋白分离及MALDI-TOF MS蛋白鉴定分析，研究人员首次成功建立了胃癌细胞SGC7901中高尔基体的蛋白质组学技术路线。 3.1 蛋白质功能预测工具 也许生物信息学方法在癌症研究中最常用的就是基因功能预测方法，但是这些数据库只存储了基因组的大约一半基因的功能。为了在微阵列资料基础上完成功能性的富集分析，基因簇的功能注解是非常重要的。近几年生物学家研发了一些基因功能预测的方法，这些方法旨在超越传统的BLAST搜索来预测基因的功能。基因功能预测可以以氨基酸序列、三级结构、与之相互作用的配体、相互作用过程或基因的表达方式为基础。其中最重要的是基于氨基酸序列的分析，因为这种方法适合于微阵列分析的全部基因。 在表3中，前三项列举了三种同源搜索方法。FASTA方法虽然应用还不太广泛，但它要优于BLAST，或者至少相当。FASTA程序是第一个使用的数据库相似性搜索程序。为了达到较高的敏感程度，程序引用取代矩阵实行局部比对以获得最佳搜索。美国弗吉尼亚大学可以提供这项程序的地方版本，当然数据库搜索结果依赖于要搜索的数据库序列。如果最近的序列数据库版本在弗吉尼亚大学不能获得，那么就最好试一下京都大学（Kyoto University）的KEGG站点。PSI-BLAST（位点特异性反复BLAST）是BLAST的转化版本，PSI-BLAST的特色是每次用profile 搜索数据库后再利用搜索的结果重新构建profile，然后用新的profile再次搜索数据库，如此反复直至没有新的结果产生为止。PSI-BLAST先用带空位的BLAST搜索数据库，将获得的序列通过多序列比对来构建第一个profile。PSI-BLAST自然地拓展了BLAST方法，能寻找蛋白质序列中的隐含模式，有研究表明这种方法可以有效地找到很多序列差异较大而结构功能相似的相关蛋白，所以它比BLAST和FASTA有更好的敏感性。PSI-BLAST服务可以
蛋白质结构与功能的生物信息学研究
实验名称：蛋白质结构与功能的生物信息学研究 实验目的：1.掌握运用BLAST工具对指定蛋白质的氨基酸序列同源性搜索的方法。 2.掌握用不同的工具分析蛋白质的氨基酸序列的基本性质 3掌握蛋白质的氨基酸序列进行三维结构的分析 4.熟悉对蛋白质的氨基酸序列所代表蛋白的修饰情况、所参与的 代谢途径、相互作用的蛋白，以及与疾病的相关性的分析。实验方法和流程： 一、同源性搜索 同源性从分子水平讲则是指两个核酸分子的核苷酸序列或两个蛋白质分子的氨基酸序列间的相似程度。BLAST工具能对生物不同蛋白质的氨基酸序列或不同的基因的DNA序列极性比对，并从相应数据库中找到相同或相似序列。对指定的蛋白质的氨基酸序列进行同源性搜索步骤如下： ↓ 登录网址https://www.wendangku.net/doc/cc2517930.html,/blast/ ↓ 输入序列后,运行blast工具 ↓ 序列比对的图形结果显示
序列比对的图形结果：用相似性区段（Hit）覆盖输入序列的范围判断两个序列 的相似性。如果图形中包含低得分的颜色（主要是红色） 区段，表明两序列的并非完全匹配。 ↓ 匹配序列列表及得分
各序列得分 可选择不同的比对工具 备注: Clustal是一款用来对()的软件。可以用来发现特征序列，进行蛋白分类，证明序列间的同源性，帮助预测新序列二级结构与三级结构，确定PCR引物，以及 在分子进化分析方面均有很大帮助。Clustal包括Clustalx和Clustalw(前者是 图形化界面版本后者是命令界面)，是生物信息学常用的多序列比对工具。 该序列的比对结果有100条，按得分降序排列，其中最大得分2373，最小得分 分为1195. ↓ 详细的比对序列的排列情况 第一个匹配 序列 第一个序列的匹配率为100% Score表示打分矩阵计算出来的值，由搜索算法决定的，值越大说明匹配程度
蛋白质结构与功能的关系
蛋白质结构与功能的关系 专业：植物学 摘要：蛋白质特定的功能都是由其特定的构象所决定的，各种蛋白质特定的构象又与其一级结构密切相关。天然蛋白质的构象一旦发生变化，必然会影响到它的生物活性。由于蛋白质的构象的变化引起蛋白质功能变化，可能导致蛋白质构象紊乱症，当然也能引起生物体对环境的适应性增强。而分子模拟技术为蛋白质的研究提供了一种崭新的手段。在理论上解决了结构预测和功能分析以及蛋白质工程实施方面所面临的难题。它在蛋白质的结构预测和模建工作中占有举足轻重的地位，实现了生物技术与计算机技术的完美结合。 关键词：蛋白质的结构、功能；折叠/功能关系；蛋白质构象紊乱症；分子模拟技术；同源建模 RNase是由124个氨基酸残基组成的单肽链，分子中 8 个Cys的-SH构成4对二硫键，形成具有一定空间构象的蛋白质分子。在蛋白质变性剂和一些还原剂存在下，酶分子中的二硫键全部被还原，酶的空间结构破坏，肽链完全伸展，酶的催化活性完全丧失。当用透析的方法除去变性剂和巯基乙醇后，发现酶大部分活性恢复，所有的二硫键准确无误地恢复原来状态。若用其他的方法改变分子中二硫键的配对方式，酶完全丧失活性。这个实验表明，蛋白质的一级结构决定它的空间结构，而特定的空间结构是蛋白质具有生物活性的保证。前体与活性蛋白质一级结构的关系，由108个氨基酸残基构成的前胰岛素原，在合成的时候完全没有活性，当切去N-端的24个氨基酸信号肽，形成84个氨基酸的胰岛素原，胰岛素原也没活性，在包装分泌时，A、B链之间的33个氨基酸残基被切除，才形成具有活性的胰岛素。 功能不同的蛋白质总是有着不同的序列；种属来源不同而功能相同的蛋白质的一级结构，可能有某些差异，但与功能相关的结构也总是相同。若一级结构变化，蛋白质的功能可能发生很大的变化。蛋白质特定的功能都是由其特定的构象所决定的，各种蛋白质特定的构象又与其一级结构密切相关。天然蛋白质的构象一旦发生变化，必然会影响到它的生物活性。由于蛋白质的构象的变化引起蛋白质功能变化，可能导致蛋白质构象紊乱症，当然也能引起生物体对环境的适应性增强。 虽然蛋白质结构与生物功能的关系比序列与功能的关系更加紧密，但结构与功能的这种关联亦若隐若现，并不能排除折叠差别悬殊的蛋白质执行相似的功能，折叠相似的蛋白质执行差别悬殊功能的现象的存在。无奈，该领域仍不得不将100多年前Fisher提出的“锁一钥