UNCORRECTED PROOF ORIGINAL ARTICLE Gene Expression Profiling Identifies WNT7A As a Possible Candidate Gene for Decreased Cancer Risk in Fragile X Syndrome Patients Mo ´nica Alejandra Rosales-Reynoso, a Alejandra Berenice Ochoa-Herna ´ndez, b Adriana Aguilar-Lemarroy, c Luis Felipe Jave-Sua ´rez, c Rogelio Troyo-Sanroma ´n, d and Patricio Barros-Nu ´n ˜ez b a Divisio ´n de Medicina Molecular, b Divisio ´n de Gene ´tica, c Divisio ´n de Inmunologı ´a, CIBO, Instituto Mexicano del Seguro Social, Guadalajara, Jalisco, Me ´xico, d Departamento de Fisiologı ´a, CUCS, Universidad de Guadalajara, Guadalajara, Jalisco, Me ´xico Received for publication September 19, 2009; accepted January 25, 2010 (ARCMED-D-09-00440). Background and Aims. Although sporadic cases of cancer in patients with fragile X syndrome (FXS) have been reported, extensive studies carried out in Denmark and Finland concluded that cancer incidence in these patients is lower than in the general population. On the other hand, the FMR1 protein, which is involved in the translation process, is absent in FXS patients. Hence, it is reasonable to assume that these patients exhibit an abnormal expression of some proteins involved in regulating tumor suppressor genes and/or onco- genes, thus explaining its decreased cancer frequency. We undertook this study to analyze the expression of oncogenes and tumor suppressor genes in fragile X syndrome patients. Methods. Molecular analysis of the FMR1 gene was achieved in 10 male patients and controls. Total RNA from peripheral blood was used to evaluate expression of oncogenes and tumor suppressor genes included in a 10,000 gene microarray library. Quantitative real-time PCR was utilized to confirm genes with differential expression. Results. Among 27 genes showing increased expression in FXS patients, only eight genes exhibited upregulation in at least 50% of them. Among these, ARMCX2 and PPP2R5C genes are tumor suppressor related. Likewise, 23/65 genes showed decreased expression in O50% of patients. Among them, WNT7A gene is a ligand of the b-catenin pathway, which is widely related to oncogenic processes. Decreased expression of WNT7A was confirmed by quantitative RT-PCR. Expression of c-Myc, c-Jun, cyclin-D and PPARd genes, as target of the b-catenin pathway, was moderately reduced in FXS patients. Conclusions. Results suggest that this diminished expression of the WNT7A gene may be related to a supposed protection of FXS patients to develop cancer. Ó 2010 IMSS. Published by Elsevier Inc. Key Words: Fragile X syndrome, Tumor suppressor genes, Oncogenes, WNT7A, Cancer risk. Introduction Fragile X syndrome (FXS) is the most common inherited form of mental retardation and the second leading cause of mental retardation after Down syndrome (1e2). The estimated prev- alence of mental retardation in Western countries is 2e3%; of these, 25e35% may have a genetic background. Among the genetic causes, almost one third are probably due to mutations on the X chromosome (X-linked mental retardation) (3). The disease is associated with the expansion of CGG trinucleotide repeats at the 5 0 -untranslated region of the FMR1 gene at a fragile site of the X chromosome (FRAXA). Affected indi- viduals have O200 repeats, which results in hypermethylation of the promoter region, repressed transcription at the FMR1 gene, and clinical expression of the disease. Premutation carriers have 55e200 repeats (4e6). The FMR1 gene was cloned and sequenced in 1991. It is abundantly expressed during early embryonic development ARCMED1450_proof 20-3-2010 2-44-27 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 Address reprint requests to: Dr. Patricio Barros-Nu ´n ˜ez, Divisio ´n de Gene ´tica, AP 1-3838, Centro de Investigacio ´n Biome ´dica de Occidente, IMSS, Sierra Mojada 800, Col. Independencia, CP 44340, Guadalajara, Jalisco, Me ´xico; Phone: (þ52) (33) 3668-3000 ext. 31930; FAX: ---; E-mail: [email protected]0188-4409/10 $esee front matter. Copyright Ó 2010 IMSS. Published by Elsevier Inc. doi: 10.1016/j.arcmed.2010.03.001 ARTICLE IN PRESS Archives of Medical Research - (2010) -
12
Embed
Gene Expression Profiling Identifies WNT7A As a Possible Candidate Gene for Decreased Cancer Risk in Fragile X Syndrome Patients
This document is posted to help you gain knowledge. Please leave a comment to let me know what you think about it! Share it to your friends and learn new things together.
Transcript
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
ARTICLE IN PRESS
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
Archives of Medical Research - (2010) -
OF
ORIGINAL ARTICLE
Gene Expression Profiling Identifies WNT7A As a Possible CandidateGene for Decreased Cancer Risk in Fragile X Syndrome Patients
Adriana Aguilar-Lemarroy,c Luis Felipe Jave-Suarez,c Rogelio Troyo-Sanroman,d
and Patricio Barros-Nunezb
aDivision de Medicina Molecular, bDivision de Genetica, cDivision de Inmunologıa, CIBO, Instituto Mexicano del Seguro Social, Guadalajara, Jalisco,
Mexico, dDepartamento de Fisiologıa, CUCS, Universidad de Guadalajara, Guadalajara, Jalisco, Mexico
Received for publication September 19, 2009; accepted January 25, 2010 (ARCMED-D-09-00440).
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
Address reprint r
Genetica, AP 1-3838
IMSS, Sierra Mojada
Jalisco, Mexico; Ph
---; E-mail: pbar
0188-4409/10 $eseedoi: 10.1016/j.arcm
RRECTEDPRO
Background and Aims. Although sporadic cases of cancer in patients with fragile Xsyndrome (FXS) have been reported, extensive studies carried out in Denmark and Finlandconcluded that cancer incidence in these patients is lower than in the general population. Onthe other hand, the FMR1 protein, which is involved in the translation process, is absent inFXS patients. Hence, it is reasonable to assume that these patients exhibit an abnormalexpression of some proteins involved in regulating tumor suppressor genes and/or onco-genes, thus explaining its decreased cancer frequency. We undertook this study to analyzethe expression of oncogenes and tumor suppressor genes in fragile X syndrome patients.
Methods. Molecular analysis of the FMR1 gene was achieved in 10 male patients andcontrols. Total RNA from peripheral blood was used to evaluate expression of oncogenesand tumor suppressor genes included in a 10,000 gene microarray library. Quantitativereal-time PCR was utilized to confirm genes with differential expression.
Results. Among 27 genes showing increased expression in FXS patients, only eight genesexhibited upregulation in at least 50% of them. Among these, ARMCX2 and PPP2R5Cgenes are tumor suppressor related. Likewise, 23/65 genes showed decreased expressionin O50% of patients. Among them, WNT7A gene is a ligand of the b-catenin pathway,which is widely related to oncogenic processes. Decreased expression of WNT7A wasconfirmed by quantitative RT-PCR. Expression of c-Myc, c-Jun, cyclin-D and PPARdgenes, as target of the b-catenin pathway, was moderately reduced in FXS patients.
Conclusions. Results suggest that this diminished expression of the WNT7A gene maybe related to a supposed protection of FXS patients to develop cancer. � 2010 IMSS.Published by Elsevier Inc.
Key Words: Fragile X syndrome, Tumor suppressor genes, Oncogenes, WNT7A, Cancer risk.
109 O
110
111
112
113
114
115
116
117
NCIntroduction
Fragile X syndrome (FXS) is the most common inherited formof mental retardation and the second leading cause of mentalretardation after Down syndrome (1e2). The estimated prev-alence of mental retardation in Western countries is 2e3%; of
U
ARCMED1450_proof �
118
119
120
121
122
123
124
equests to: Dr. Patricio Barros-Nunez, Division de
front matter. Copyright � 2010 IMSS. Published by Elseved.2010.03.001
these, 25e35% may have a genetic background. Among thegenetic causes, almost one third are probably due to mutationson the X chromosome (X-linked mental retardation) (3). Thedisease is associated with the expansion of CGG trinucleotiderepeats at the 50-untranslated region of the FMR1 gene ata fragile site of the X chromosome (FRAXA). Affected indi-viduals have O200 repeats, which results in hypermethylationof the promoter region, repressed transcription at the FMR1gene, and clinical expression of the disease. Premutationcarriers have 55e200 repeats (4e6).
The FMR1 gene was cloned and sequenced in 1991. It isabundantly expressed during early embryonic development
in multiple tissues including brain and testes. The productof the FMR1 gene, the fragile X mental retardation protein(FMRP), is an RNA binding protein that is a component ofmessenger ribonucleoproteins (mRNPs) that regulate trans-lation and possibly RNA stability, which may have animpact on cellular mRNA levels. Inaccurate mRNA pro-cessing of some genes in these individuals may result inclinical manifestations of FXS or affect regulatory proteinswithin biochemical networks affecting the genome-widetranscription (5).
It is known that chromosomal fragile sites are specificloci preferentially exhibiting gaps and breaks on metaphasechromosomes resulting from partial inhibition of DNAsynthesis. Fragile sites represent intriguing components ofthe chromosome structure. Several authors suggest thatfragile sites taken significance as regions of the genome thatare particularly sensitive to replication stress and that arefrequently rearranged in tumor cells and cancer (7), corrob-orating the proposition made in 1986 by Le Beau (8).
In 1995, Panzer et al. (9) hypothesized that trinucleotideexpansions in FRAXA would also increase its cancersusceptibility. With this presumption many authors haveexplored cancer in FXS patients; nevertheless, only a verysmall group of individuals showing such association hasbeen reported: benign testicular tumor (10), seminomaand colon adenocarcinoma (11), malignant ganglioglioma(12), acute lymphoblastic leukemia (13e14), prostatecancer, glioma, meningioma (15), nephroblastoma (16),myelodysplastic syndrome (17), lung tumor (18), colorectalcancer (19), nasopharyngeal carcinoma (20), glioblastoma(21) and hepatic tumors (22). Conversely, in an extensiveand multicenter study conducted by Schultz-Pedersenet al. (23), assessing the occurrence of cancer in a largecohort of 223 FXS patients from the Danish Cytogeneticand Cancer Registries, they found a significant decreasedrisk of cancer in these patients in comparison with thegeneral population. With this same objective, in a morerecent study carried out by Sund et al. and based on anextensive sample of a Finnish population, a diminished(although nonsignificant) cancer frequency among FXSpatients was observed (24).
With the exception of Huntington disease and Downsyndrome, no genetic diseases have been reported witha decreased risk of cancer. For patients with Down syndrome,the number of solid tumors is actually significantly lower;however, the risk for leukemia is increased (25e27).
Recently, Siu and Jin (28) reported that stimulation ofcyclic adenosine monophosphate (cAMP) may either inhibitor selectively promote development of tumors depending onthe cell type. It is also known that cAMP and the responseelement binding protein (CREB) are actively involved inFMR1 transcriptional activity (29). However, it is probablethat cAMP or CREB alone may not be sufficient to inducecancer without simultaneous activation of other oncogenesor inactivation of tumor suppressor genes (30,31).
ARCMED1450_proof �
ROOF
Finally, upregulation of the Ras signaling pathway,a common cause of cancer, has recently been associatedin FXS patients (32). On the other hand, aberrant Rassignaling prevents spine maturation (33) and is linked todevelopmental disorders with facial dysmorphism (34),which are some of the most prominent clinical features inFXS (32).
Abnormal translation processing of a number of genes inFXS patients may explain not only its phenotype andclinical manifestations but the malfunction of metabolicregulatory networks. Therefore, it is reasonable to assumethat FXS patients may exhibit a differential expression ofsome proteins, particularly those related to regulation oftumor suppressor genes and/or oncogenes, which wouldexplain the decreased cancer frequency in these patients.
In order to examine such a diminished incidence ofcancer in FXS patients, we evaluate the expression of onco-genes and tumor suppressor genes using microarrayanalysis and confirmed the differential expression usingreal-time quantitative reverse transcription polymerasechain reaction.
EDMaterials and Methods
To establish the expression of oncogenes and tumorsuppressor genes in FXS patients, a comparison withnormal males was carried out.
Subjects
Ten male patients without a familial history of cancer andcarrying a complete mutation of the FMR1 gene werecompared with a control group comprised of 10 similarlyaged males (�3 years) with first-degree consanguinity.Control subjects were without malignancies or mental retar-dation and demonstrated a normal result from the molecularstudy for the FMR1 gene. FXS patients were clinicallydiagnosed at the Pediatric Hospital of the Centro MedicoNacional de Occidente, Mexican Institute of Social Secu-rity (IMSS), Guadalajara, Mexico. The study was approvedby the local institutional review board and consent proce-dures were followed accordingly (Register: 2005e108).For inclusion as patients or controls, all subjects weremolecularly analyzed for establishing the number of CGGrepeats and the methylation status of the FMR1 gene.
DNA Modification and PCR Amplification
For both groups, DNA was extracted from peripheral bloodlymphocytes using standard methods (35). DNA modifica-tion using sodium bisulfite treatment was achieved as re-ported by Panagopoulos et al. (36). PCR amplification ofthe repetitive sequence CGG was performed with primersFR526R (GGG AGT TTG TTT TTG AGA GGT GGG)and FR754F (CAA CCT CAA TCA AAC ACT CAA
20-3-2010 2-44-27
T
ARTICLE IN PRESS
3Reduced expression of Wnt7a in Fragile X syndrome patients
CTC CA), which were designed for amplifying unmethy-lated sequences on the antisense strand. The CpG island,located adjacent to the promoter of the FMR1 gene, wasamplified using primers FR611R (CGT CGT CGC GTTGTC GTA C) and FR690F (AAC GAC GAA CCG ACGACG). In this case, primers were designed for amplifyingonly methylated sequences. Individuals carrying a fullmutation showed methylation of this region and, conse-quently, inactivation of the FMR1 gene.
RNA Isolation and cDNA Preparation
Total RNA from peripheral blood was obtained frompatients and controls using the Pax gene blood RNA SystemKit (cat. #762134, Qiagen, Hilden, Germany) and thenfluorescently labeled with dUTP-Cy3 and dUTP-Cy5 byretrotranscription.
Microarrays
Slides were elaborated using the Human Genome Microar-ray library (MWG Biotech H10K_DB) including 10,000genes and open reading frames. Target preparation, hybrid-ization, and initial data collection were done according tothe Physiology Laboratory of the Universidad NacionalAutonoma de Mexico URL: (http://microarrays.ifc.unam.mx). Total RNA (5 mg) from patients and controls waslabeled with anchored oligo (dT) primer and incubatedwith the first-strand reaction kit component at 70�C for10 min. The second-strand reaction components were thenadded, followed by an additional 3 h incubation at 46�C andincorporation of Alexa 555 for FXS patients and Alexa 647for controls (SuperScript Plus Direct cDNA Labeling KitNucleotide Module (Invitrogen, Carlsbad, CA). cDNAwas then concentrated and purified (QIAquick PCR Purifi-cation Kit, Qiagen). The hybridization reaction was in-jected into a hybridization chamber containing the humanmicroarray and incubated with shaking (1500 rpm) for 16h at 42�C. Microarray slides were washed four times ina washing buffer followed by a final rinse in 0.1 SSC/0.05% SDS and dried by centrifugation. Two independenthybridizations (both with dye swapping) were performed,completing four hybridizations per experimental condition.
DNA Microarray Analysis
Spot detection, mean signals, mean local background inten-sities, image segmentation, and signal quantification weredetermined using the Array-Pro Analyzer 4.0 software formicroarray images (Media Cybernetics, L.P., Silver Spring,MD). Analysis of microarray data was performed with thegenArise software developed in the Computing Unit of theCellular Physiology Institute at the Universidad NacionalAutonoma de Mexico URL:(http://www.ifc.unam.mx/genarise/). This software identifies differentially expressedgenes by calculating an intensity-dependent z-score. It uses
ARCMED1450_proof �
EDPROOF
a sliding window algorithm to calculate the mean and stan-dard deviation (SD) within a window surrounding each datapoint and defines a z-score where z measures the number ofSDs that a data point is from the mean.
zi 5 ½Ri$meanðRÞ�=sdðRÞ
where zi is the z-score for each element, mean(R) is themean log ratio, Ri is the log ratio for each element, andsd(R) is the standard deviation of the log ratio. With thiscriterion, the elements in all experiments with a z-scoreof O2 SDs were considered significantly differentially ex-pressed genes.
We later used the Encyclopedia KEGG (Kanehisa Labo-ratories, Bioinformatics Center of Kyoto University and theHuman Genome Center of the University of Tokyo, http://www.genome.jp/kegg/) to examine the database ofgenetic and molecular pathways as previously describedin similar studies (37). KEGG is a database of biologicalsystems consisting of genetic building blocks of genesand proteins (KEGG GENES), chemical building blocksof both endogenous and exogenous substances (KEGGLIGAND), molecular wiring diagrams of interaction andreaction networks (KEGG PATHWAY), and hierarchiesand relationships of various biological objects (KEGGBRITE). KEGG provides a reference knowledge base forlinking genomes to biological systems and also to environ-ments by the processes of PATHWAY mapping and BRITEmapping.
Quantitative Real-time PCR
Cancer-related genes identified as significantly differen-tially expressed by microarray analysis were analyzed byquantitative real-time PCR (RT-PCR); each quantitativePCR was performed in duplicate. cDNA synthesis was per-formed using 5 mg of total RNA primed with oligo(dT)using the SuperScript III First-Strand Synthesis Systemfor RT-PCR (cat. #18080e051 Invitrogen) according tothe manufacturer’s recommendations. PCR primers weredesigned using the Oligo v6 software. Gene sequences wereobtained from the GenBank Nucleotide Database of theNCBI website. Primer pairs were as follows: 1) for Wnt7a,forward primer: 50-CAA AGA GAA GCA AGG CCA GTACCA-30, reverse primer: 50-GTA GCC CAG CTC CCGAAA CTG T-30; 2) for PPP2R5C, forward primer: 50-ACA GAG CAT CAT AAT GGC ATA G-30 and reverseprimer: 50-CAT TAC TTC TTT TGG ACT GTG AGT-03);for L32 ribosomal protein, forward primer: 50-GCA TTGACA ACA GGG TTC GTA G-30, reverse primer: 50-ATTTAA ACA GAA AAC GTG CAC A-30. ConventionalPCR reactions were performed using Taq DNA polymerase(cat. #11146173001, Roche Applied Science, Mannheim,Germany) and deoxynucleoside triphosphates (cat.#1969064, Roche Applied Science) in a PX2 ThermalCycler (Thermo Electron Corporation, Pittsburgh, PA).
All reactions were done in a final volume of 25 ml and 60�Cof annealing temperature for 40 cycles. PCR products wereresolved in 1.5% agarose gels containing 0.1 mg/mlethidium bromide (Sigma Aldrich, Munich, Germany) visu-alized under UV light and documented with a DigiDoc-ItSystem (UVP, Upland, CA).
RT-PCR reactions were achieved using the LightCycler-FastStart DNA MasterPLUS SYBR Green I Kit (cat.#03515885001, Roche Applied Science) in a LightCycler1.5 System (Roche Diagnostics GmbH, Mannheim,Germany). PCR amplification conditions were 95�C for10 sec, annealing temperature of 60�C for 10 sec and exten-sion time of 15 sec at 72�C for 45 cycles. Melting curveswere done by temperature increases of 0.1�C/sec for60�C to 98�C. Efficiency and error of the reaction werecalculated by using a standard curve with four serial dilu-tion points of a cDNA mixture. Relative-quantificationanalysis of gene expression was done with the LightCyclerSoftware v4 using the expression of L32 ribosomal proteinas constitutive reference gene. The evaluation of relativedifferences of PCR products among patients and controlswas carried out using the DDCP method. The mean valueof the control group was adjusted to 1 and the expressionvalues for each patient are presented as percentage withregard to the control group.
437438439440441442
OR
Statistical Analysis
Statistical difference between groups was evaluated by theStudent t-test; p ! 0.05 was considered statisticallysignificant.
C 443
444445446447448449450451452453454
UNResults
A human genome microarray containing 10,000 probes wasused to analyze gene expression in peripheral blood fromFXS patients and normal males. The following criteria wereused to define genes with measurable levels of expression:(1) a minimum signal intensity of 100 in at least one of thetwo probes, (2) signal-to-background ratios O2.3 for bothprobes, and (3) signal size O40% of the spotting area forboth probes. Signal intensities detected in the RNA samples
ARCMED1450_proof �
EDPROOFfrom affected and control individuals allowed us to calcu-
late the mean, SD and gene differential expression valuesfor all genes in each FXS patient using the genArise soft-ware. Genes up- and downregulated in at least two SDswere analyzed.
FMR1 gene expression analyzed by microarray wasreduced or absent in all FXS patients. In comparison, expres-sion of this gene in the pool of males from the control groupshowed expression close to the median values.
Only 27 genes showed statistically significant increasedexpression (O2 SD) in FXS patients with regard to controls(Suppl. Table 3). Of these, only eight genes had this signif-icant increased expression in five or more patients (Table1). In particular, the ARMCX2 and PPP2R5C genes, whichwere found upregulated in FXS patients, are also related totumor-suppressor genes (38e41).
Likewise, 65 genes showed a statistically significantdiminished expression in FXS patients in regard to controls(Suppl. Table 4), but only 23 genes showed this statisticaldifference in at least five individuals (Table 2). The mostrelevant finding was the downregulation of the WNT7Agene, which has been related to a number of oncogenicprocesses (42). The database of genetic and molecularsignaling pathways of the genes found in the context ofKEGG biological pathways shows the importance of theWNT7A and PPP2RC5 genes within the canonic b-cateninpathway (Figure 1).
With these findings, we intentionally searched for theexpression of target genes of the b-catenin pathway. Unfortu-nately, not all target genes for this pathway were included inthe library of genes used in this study; nevertheless, c-Myc,c-Jun, cyclin-D and PPARd genes showed a moderatelyreduced expression (1.5 SD) in patients with regard to normalsubjects.
In order to validate the most interesting observations(upregulation of PPP2RC5 and downregulation ofWNT7A), a quantitative real-time PCR procedure usingintercalation of SYBR-Green was done. For this approach,only one specific fragment should be obtained from theamplification process as depicted in (Figure 2a). Tempera-ture melting point analysis of the PCR showed one peakthat represents the amplification of only one product.Additionally, conventional PCR analysis confirmed the
20-3-2010 2-44-27
TPROOF
Table 2. Downregulated transcripts found in fragile X syndrome patients
Gene Genbank number Chromosome Number of patients Description
CADPS NM_604667 3 7 Calcium-dependent activator protein
specificity of our assay. To calculate the efficiency and errorof the PCR reaction, amplification of serial dilutions ofa cDNA mix were done, obtaining an error of 0.000451and an efficiency of 1.838 (Figure 2b). To avoid variationsin the genetic expression related to RNA extraction andcDNA synthesis, we used the same RNA samples as formicroarrays. Only nine FXS patient samples showedacceptable conditions for this assay. We used these samplesto analyze the expression of WNT7A, PPP2R5C and L32ribosomal protein. This last gene was used as referencefor the relative quantification. For this analysis, values fromthe pool of controls (healthy males) were set as 1 and thenormalized ratio was calculated by pairing each patientwith the pool of controls. As can be seen in Figure 3,
UNCOR
ARCMED1450_proof �
Figure 1. Schematic representation of the Wnt signaling pathway showing the p
clopedia of Genes and Genomes, KEEG).
EDrelative expression of WNT7A was lower in FXS patients
than in control individuals, with a statistically significantdifference (!0.005) (t-test). Setting the value of the indi-viduals control as 100%, WNT7A expression in FXSpatients ranged between 21 and 76% (Figure 3b).
Expression of the PPP2RC5 gene in FSX patients was notsignificant when compared with controls (data not shown).
Discussion
For almost two decades, dynamic and highly polymorphicmutations caused by trinucleotide expansion and producinghuman illnesses have been described. To date, more than
20-3-2010 2-44-27
articipation of Wnt7a and PPP2R5C (PP2A) (modified from Kyoto Ency-
Figure 2. Efficiency and specificity of PCR amplification. (a) Temperature melting analysis shows amplification of only one specific peak with Wnt7a
primers. Corroboration of this observation was also done by running amplification products on 1.5% agarose gel obtaining a 277-bp fragment. (b) Efficiency
and error of the PCR amplifications were assessed using standard curves generated by decreasing amounts of cDNAs (1:1, 1:2, 1:4 and 1:8, respectively)
obtaining an adequate linearity in all cases.
ARTICLE IN PRESS
6 Rosales-Reynoso et al./ Archives of Medical Research - (2010) -
one dozen of these kinds of diseases are known and, amongthem, FXS is the most important and common inheritablecause of mental retardation, which is associated with thecytogenetic expression of a fragile site on Xq27.3(FRAXA). Chromosomal instability syndromes arediseases presenting chromosomal breakpoints and typicallyhave been related to human malignancies; hence, FXSpatients constitute an interesting focus of attention forcancer research. A number of tumors occurring in patientswith this condition have been sporadically reported world-wide (10e22); nevertheless, extensive population studiescarried out in Denmark (23) and Finland (24) reported thatthe incidence of cancer in these patients is lower than in thegeneral population.
It is well known that the product of the FMR1 gene(FMRP), absent in FXS patients, is an RNA binding protein
ARCMED1450_proof �
and an essential component of messenger ribonucleopro-teins (mRNPs) regulating RNA translation and stability.Furthermore, evidence that FMRP exhibits an inhibitoryactivity within the translation complex has been reported(43). FMRP appears to be also linked to micro-RNAs, add-ing more complexity to the role that it plays in regulatingthe RNA transport and translation (44e47). Abnormaltranslation processing of a number of genes in FXS patientsmay explain not only its phenotype and clinical manifesta-tions but the malfunction of metabolic regulatory networks.Some of the pathways in which FMRP may be involvedhave been envisaged; however, knowledge about the disrup-tion of the transcriptome and proteome that produce FXSremains to be elucidated (48).
With these antecedents, it is reasonable to assume thatFXS patients may exhibit a differential expression of some
20-3-2010 2-44-28
TFigure 3. Graphics display the median (dark lines), 25the75th percentile
(boxes) and interquartile ranges (whiskers) from our data. (a) Box plot
graphic shows the normalized ratio of WNT7A expression of each patient
with the control group setting as 1 (DD-CP). (b) Comparison between
FXS patients and healthy controls considering all data (D-CP).
ARTICLE IN PRESS
7Reduced expression of Wnt7a in Fragile X syndrome patients
proteins, particularly those related to regulation of tumorsuppressor genes and/or oncogenes. This would explainthe decreased cancer frequency in these patients. Few geneexpression profiling studies in FXS patients have been pub-lished (49,50) but none has explored the apparent relation-ship of these patients with cancer.
Results of gene expression achieved in our laboratory bymicroarray analysis in FXS patients showed significantchanges in the expression of a number of genes whencompared with normal males (Tables 1 and 2). Some ofthese up- or downregulated genes have not demonstrateda clear biological function or are quite unknown; severalother genes show well-established roles in different tissuesbut are unrelated to the cellular cycle or oncogenic or tumorsuppressor activities. However, significant changes inexpression of some genes involved in tumorigenicprocesses were also detected. In particular, the microarrayassay showed upregulation of the PPP2R5C gene anddiminished expression of the WNT7A gene, which areknown members of oncogene and tumor suppressor fami-lies, respectively. Interestingly, both genes are involved inthe canonical b-catenin pathway. On the other hand, as ex-pected, expression of the FMR1 gene was significantlyreduced or absent in individuals from the FXS group, whichis congruent with the overmethylation observed in thepromoter regions of the FMR1 genes from all these patients.
Wnt molecules (name derived from two genes:drosophila wingless and mouse Int1) are 19 cysteine-richsecreted glycoproteins serving as extracellular signalingmolecules and play significant roles in normal and malig-nant developmental processes (51e53). Wnt proteins acton target cells via frizzled receptors (FzeR). The Wntfamily is biologically categorized into two classes: classicaland non-classical Wnts, which usually activate the canon-
ARCMED1450_proof �
EDPROOF
ical and noncanonical intracellular signaling pathways,respectively (53). Canonical signaling pathway leads toincrease in cytoplasmic level of b-catenin and activationof downstream signaling molecules (54). Noncanonicalsignaling pathway leads to intracellular Ca2þ flux (it acti-vates Ca2þ dependent effecter enzymes such as PKC andCamkII) or JNKs activation (jun amino-terminal kinase),which can suppress canonical signaling cascade (55,56).Recently, Wnts have drawn attention as a set of factorsoperating in embryonic development, growth, regulationof adult tissues and cancer formation (57).
The Wnt7a molecule can bind to different transmembranereceptors Frizzled. When it binds to Fzd 9 or 10, it is associ-ated with noncanonical signaling (52,58); when it binds toFzd 2, 5 or 7, it is associated with canonical b-cateninsignaling (51,52). In the fascinating canonical b-cateninsignaling pathway (Figure 1), Wnt7a, through the differenttransmembrane receptors Frizzled (2, 5 or 7), give rise to acti-vation of the cytoplasmic disheveled (Dsh) protein, whichparticipates in the b-catenin degradation complex (adenoma-tous polyposis coli gene product, glycogen synthase kinase-3b and axin). In the absence of Wnt signaling, b-catenin isphosphorylated by GSK-3b, which leads to its rapiddegradation by the ubiquitin pathway. In response to Wntsignals, b-catenin is no longer targeted for degradation, andthis protein is accumulated in the cytoplasm. Later, stabilizedb-catenin molecules translocate into the nucleus where theybind with members of the transcription factors family asT-cell factor (Tcf) and lymphoid enhancer factor (Lef) andactivate the transcription of target genes of the Wnt signalingpathway (c-Myc, c-Jun, cyclin-D, PPARd, fra-1).
The precise biochemical mechanism by which theb-catenin degradation complex is regulated is not yet known.Alterations in protein phosphorylation states are likely centralto this regulation because all elements of the b-catenindegradation complex are phosphoproteins. A number ofprotein kinases have been shown to influence Wnt/b-cateninsignaling including GSK3b, protein kinase C and caseinkinase I and II. The protein phosphatase involved in thissignaling pathway is the catalytic subunit B56 (PPP2R5C)of the protein phosphatase serine/threonine 2A (PP2A), whichexerts a positive role on Wnt/b-catenin pathway activity inmammalian cells and Xenopus embryo explants (41,59).
Quantitative RT-PCR achieved to validate the downregu-lation of the Wnt7a gene as a candidate to explain thedecreased cancer risk in patients with FXS showed a realreduction of the signal intensity in FXS males in compar-ison with healthy males. However, quantitative RT-PCRresults for the PPP2R5C gene were not able to validatethe upregulation observed with the microarray assays inmost FXS patients.
Diminished expression of the WNT7A gene, a mainregulator of the canonical b-catenin pathway, certainlyconstitutes a profound change for both the activity of thispathway as well as for the accomplishment of other
20-3-2010 2-44-29
T
ARTICLE IN PRESS
8 Rosales-Reynoso et al./ Archives of Medical Research - (2010) -
multiple interconnected biochemical pathways. Given thecrucial importance of the WNT7A/b-catenin pathway intothe general cellular function, downregulation of WNT7Ashould be totally or partially counteracted by the actionof other involved regulators. Therefore, it is explicable thattarget genes of the WNT7A/b-catenin pathway (c-Myc,c-Jun, cyclin-D and PPARd) show in this study onlya moderate downregulation in most patients in comparisonto controls. b-catenin target genes can be not wholly absentin any cell; hence, downregulation of this pathway causedby the poor stimulation of a scant Wnt7a protein couldnot completely block the expression of the c-Myc, c-Junor cyclin-D genes but limits a possible overproduction ofthese genes. Such a moderate downregulation would evadepotential abnormalities in the cellular cycle and, conse-quently, oncogenic process.
Although the overwhelming majority of publishedreports invoke a transforming role for Wnt proteins,a limited number of studies have also invoked a tumorsuppressor role for specific Wnt proteins (58,59). Recently,several authors have remarked on this duality about thefunction of Wnt7a. Upregulation of WNT7A in endometrialtissue induced by progestogens is a finding that mayaccount for the antineoplastic effect of these hormones onthe endometrium. Also, analysis of uterine leiomyomasdemonstrated a frequent reduction in WNT7A expressionrelative to the adjacent myometria (52,60). Downregulationof the WNT7A gene has been related to the activation of thecanonical signaling pathway (52). Memarian et al. (2007)found that downexpression of WNT7A is not associatedwith suppressive effects, but probably the noncanonicalsignaling pathway can be activated later and the effectsare highly associated with the cellular receptor context.Additionally, different expression patterns of molecules ofWnt in various types of cancer may be due to the heteroge-neity of cancers and also to different mechanisms involvedin tumorigenesis (53).
Among the upregulated transcripts, the ARMCX2 genewas also present. This gene has mainly been related tothe family of proteins named ALEX and involved in devel-opment, maintenance and tissue integrity. Tumorsuppressor activity has been also proposed for this gene(38,39). Quantitative confirmation of its increased expres-sion and which molecular pathway is involved constitutean interesting task in this group of patients.
Patients with FXS represent a very interesting naturalbiological model to study a large group of molecular andmetabolic pathways that may be affected by the absenceof FMRP. Although FXS patients present with no excessivenumber of phenotypic abnormalities, a number of biochem-ical alterations affecting mainly the central nervous systemprobably are occurring. These features are caused by theabsence of FMRP that has a central role for nuclear andcytoplasmic activities, especially those related with thetranscription processes.
ARCMED1450_proof �
In conclusion, these results suggest that the diminishedexpression of the WNT7A gene and its consequent downre-gulation of the b-catenin pathway may be related toa supposed protection of FXS patients to present cancer.
F
AcknowledgmentsThe authors thank the technical assistance of Dr. Jorge Ramırez,Lorena Chavez Gonzalez, Simon Guzman Leon and Jose SantillanTorres with the microarray preparation and the GenArise software.This study was supported by the Health Research Coordination ofthe Mexican Institute of Social Security.
EDPROO
References1. Howard-Peebles P, Maddalena A, Black S, et al. Population screening
for fragile X syndrome. Lancet 1993;341:770.
2. Warren S, Sherman S. The fragile X syndrome. In: Scriver CR,
Beaudet AL, Sly WS, Valle D, eds. The Metabolic Basis of Inherited
Disease. 7th edn. New York: McGraw-Hill;2001. pp. 1257e1289.
3. Lisik M, Sieron A. X-linked mental retardation. Med Sci Monit 2008;
14:RA221e229.
4. Bell M, Hirts M, Nakahori Y, et al. Physical mapping across the fragile
X: hypermethylation and clinical expression of the fragile X
syndrome. Cell 1991;64:861e866.
5. Nolin S, Lewis F, Ye L, et al. Familial transmission of the FMR1 CGG
repeat. Am J Hum Genet 1996;59:1252e1261.
6. Bourgeois J, Coffey S, Rivera S, et al. Fragile X permutation
disorders-expanding the psychiatric perspective. J Clin Psychiatry
2009;70:852e862.
7. Durkin S, Glover T. Chromosome fragile sites. Annu Rev Genet 2007;
41:169e192.
8. Le Beau M. Chromosomal fragile sites and cancer-specific rearrange-
ments. Blood 1986;67:849e858.
9. Panzer S, Kuhl D, Caskey C. Unstable triplet repeat sequences:
a source of cancer mutations. Stem Cells 1995;13:146e157.
10. Del Pozo B, Millard P. Demonstration of the fra(X) in lymphocytes,
fibroblasts, and bone marrow in a patient with a testicular tumor.
J Med Genet 1982;20:225e227.
11. Phelan M, Stevenson R, Collins J, et al. Fragile X syndrome and
neoplasia. Am J Med Genet 1988;30:77e82.
12. Rodewald L, Miller D, Sciorra L, et al. Brief clinical report: central
nervous system neoplasm in a young man with Martin Bell
syndrome-fra (X)eXLMR. Am J Med Genet 1987;26:7e12.
13. Cunningham B, Dickerman J. Fragile X syndrome and acute lympho-
blastic leukaemia. Cancer 1988;62:2383e2386.
14. Au W, Man C, Pang A, et al. Acute lymphoblastic leukemia in a patient
with fragile X syndrome. Haematologica 2003;88. ECR13.
15. Shabtai F, Hart J, Klar D, et al. Fragile X expression in Martin-Bell
syndrome, intellectually normal individuals, and neoplasia, interpreted
by a viral hypothesis. Am J Med Genet 1988;30:697e702.
16. Drouin V, Vannier J, Moirot H, et al. Nephroblastoma and fragile X
syndrome. Arch Fr Pediatr 1992;49:477.
17. Vorts E, Levene N, Nisani R, et al. Fragile X syndrome and myelodys-
plasia discovered during pregnancy. Br J Haematol 1993;85:415e416.
18. de Graaff E, Willensem R, Zhong N, et al. Instability of the CGG
repeat and expression of the FMR1 protein in a male fragile X patient
with a lung tumor. Am J Hum Genet 1995;57:609e618.
19. Fulchinogni-Lataud M, Olchwang S, Serre J. The fragile X CGG
repeats show a marked level of instability in hereditary non-
polyposis colorectal cancer patients. Eur J Hum Genet 1997;5:89e93.
20-3-2010 2-44-29
T
ARTICLE IN PRESS
9Reduced expression of Wnt7a in Fragile X syndrome patients