UBC Faculty Research and Publications

A combination of p300 and Braf expression in the diagnosis and prognosis of melanoma Bhandaru, Madhuri; Ardekani, Gholamreza S; Zhang, Guohong; Martinka, Magdalena; McElwee, Kevin J; Li, Gang; Rotte, Anand Jun 3, 2014

Your browser doesn't seem to have a PDF viewer, please download the PDF to view this item.

Item Metadata


52383-12885_2013_Article_4572.pdf [ 713.92kB ]
JSON: 52383-1.0223338.json
JSON-LD: 52383-1.0223338-ld.json
RDF/XML (Pretty): 52383-1.0223338-rdf.xml
RDF/JSON: 52383-1.0223338-rdf.json
Turtle: 52383-1.0223338-turtle.txt
N-Triples: 52383-1.0223338-rdf-ntriples.txt
Original Record: 52383-1.0223338-source.json
Full Text

Full Text

RESEARCH ARTICLE Open AccessA combination of p300 and Braf expression in thediagnosis and prognosis of melanomaMadhuri Bhandaru1, Gholamreza Safaee Ardekani1, Guohong Zhang1, Magdalena Martinka2, Kevin J McElwee1,Gang Li1 and Anand Rotte1*AbstractBackground: To date only a handful of drugs are available for the treatment of melanoma. Among themvemurafenib, a BrafV600E specific inhibitor, showed promising results in terms of response rate and increase inmedian survival time. However, its effectiveness is limited by development of resistance and the search foradditional drugs for melanoma treatment is ongoing. The present study was performed to analyze the correlationbetween Braf expression and the expression of p300, a known down stream target of the mitogen activatedprotein kinase (MAPK) pathway, which was recently shown by us to be a prognostic marker for melanomaprogression and patient survival.Methods: The expression of Braf and p300 expression were correlated and analyzed by Chi-square test. A total of327 melanoma patient cases (193 primary melanoma and 134 metastatic melanoma) were used for the study.Classification & regression tree (CRT), Kaplan-Meier, and multivariate Cox regression analysis were used to elucidatethe significance of the combination of Braf and p300 expression in the diagnosis and prognosis of melanoma.Results: Our results demonstrate that Braf expression is inversely correlated with nuclear p300 and positivelycorrelated with cytoplasmic p300 expression. Braf and cytoplasmic p300 were found to be associated withmelanoma progression, tumor size and ulceration status. CRT analysis revealed that a combination of Braf and p300expression (nuclear and cytoplasmic), could be used to distinguish between nevi and melanoma, and primary frommetastatic melanoma lesions. The combination of Braf and nuclear p300 was significantly associated with patientsurvival and nuclear p300 was found to be an independent predictor of patient survival.Conclusion: Our results indicate a cross-talk between Braf and p300 in melanoma and demonstrate the importanceBraf and p300 expression in the diagnosis and prognosis of melanoma.Keywords: p300, Braf, Melanoma, Prognosis, AJCC, Patient survivalBackgroundMelanoma, a type of cancer caused due to uncontrolledproliferation of melanocytes in epidermis of skin, is oneof the most frequent cancers in fair skinned populations[1,2]. According to recently published statistics based ondata from United States of America, it is the fifth mostcommon cancer in men and seventh most common can-cer in women [3]. Melanoma is known for its rapidprogression, metastasis, and poor prognosis, and is re-sponsible for over 80% of deaths from skin cancer [1].Early diagnosis allows for surgical excision of the tumorsand the patients can be managed with a relapse freeinterval of up to 10 years [4,5]. But, approximately 1 in35 patients develop metastatic tumors, and metastaticmelanoma has a very poor prognosis with an overall sur-vival between 8 to 18 months. Only 15% of patients withmetastatic melanoma survive for 5 years [3,6].There has been limited progress in the treatment ofmelanoma; metastatic melanoma is notorious for its re-sistance to conventional radiotherapy and chemotherapy.Until recently, dacarbazine, a DNA alkylating agent, wasthe only FDA approved drug available for the treatmentof melanoma [6]. In 2011, vemurafenib, a specific inhibi-tor of BrafV600E (BRAF harbouring a point mutation* Correspondence: anand.rotte@gmail.com1Skin Cancer Biology Laboratory, Department of Dermatology and SkinScience, University of British Columbia, Research Pavilion, 828 West, 10thAvenue, Vancouver, BC V5Z 1 L8, CanadaFull list of author information is available at the end of the article© 2014 Bhandaru et al.; licensee BioMed Central Ltd. This is an Open Access article distributed under the terms of the CreativeCommons Attribution License (http://creativecommons.org/licenses/by/2.0), which permits unrestricted use, distribution, andreproduction in any medium, provided the original work is properly credited. The Creative Commons Public DomainDedication waiver (http://creativecommons.org/publicdomain/zero/1.0/) applies to the data made available in this article,unless otherwise stated.Bhandaru et al. BMC Cancer 2014, 14:398http://www.biomedcentral.com/1471-2407/14/398resulting from a substitution of valine at amino-acid 600with glutamine), and ipilimumab, a monoclonal antibodyagainst cytotoxic T-lymphocyte associated antigen 4(CTLA-4), have been approved for the treatment of mel-anoma [6]. However, the success of their use is limitedby effectiveness only in a restricted population, potentialdevelopment of lethal resistance with vemurafenib treat-ment, and only a small increase in median survival timein the case of ipilimumab [6]. Our lab previouslyreported a significant association between increased Brafexpression and melanoma progression, and an inverserelationship between Braf expression and patientprognosis [7,8]. Considering the significance of Brafinhibitors in melanoma treatment, several studies haveattempted to decipher the mechanisms for resistanceand suggested both mitogen activated protein kinase(MAP kinase) dependent and independent pathways asreasons for vemurafenib resistance [6]. A number ofstrategies to overcome the resistance, including a com-bination therapy of Braf and MEK1/2 inhibitors, havebeen proposed and are in various stages of clinical stud-ies [6]. However, there are no results on the efficiency ofthe combination therapies in clinical settings and thesearch for alternative and additional drugs for the treat-ment of melanoma is ongoing.We analyzed the expression of p300, a well studiedhistone acetyl transferase (HAT) [9], in melanoma pa-tient samples and found that loss of p300 expression inthe nucleus was correlated with disease progression andworse survival in melanoma patients [10]. Furthermore,we also found that nuclear p300 expression was an inde-pendent prognostic factor, suggesting the importance oftargeting the functions of histone acetyltransferases(HAT) in melanoma therapy [10]. Stability and activityof p300 protein have been shown to be regulated byphosphorylation, and phosphorylation of p300 by mito-gen activated protein kinase (MAPK) and extracellularsignal-regulated kinase (ERK1/2) has been reported topromote the degradation of p300 protein [11,12]. Sinceour previous studies in melanoma patients showed anincrease in Braf expression, which is known to be up-stream of MAPK in the signaling cascade, we hypothe-sized a potential for correlation between p300 and Braf[8]. To test our hypothesis, and to explore the possibleopportunity of targeting histone acetylation and Braf inmelanoma treatment, we studied the association be-tween p300 and Braf expression in patient samples.MethodsPatient specimens and tissue microarray constructionThe collection of patient specimens and the constructionof the tissue microarray (TMA) have been previously de-scribed [13]. Briefly, we used patient data collected from1990 to 2009. Of 748 patients specimens collected, 369biopsies including 327 melanoma cases (193 primary mel-anoma and 134 metastatic melanoma) and 42 cases of nevi(21 normal nevi and 21 dysplastic nevi) could be evaluatedfor comparing p300 and Braf staining in this study, due toloss of biopsy cores or insufficient tumor cells present inthe cores. The demographic characteristics of melanomapatients are detailed in Table 1. All specimens were ob-tained from the archives of the Department of Pathology,Vancouver General Hospital. The use of human skin tissuesand the waiver of patient consent in this study were ap-proved by the Clinical Research Ethics Board of the Univer-sity of British Columbia [14]. The study was conductedTable 1 Demographics and clinical characteristics of 327melanoma patientsVariables Total PercentageAll melanomaAge≤ 62 166 50.8%> 62 161 49.2%GenderMale 196 59.9%Female 131 40.1%AJCCI 80 24.5%II 113 34.6%III 55 16.8%IV 79 24.2%Primary melanoma (n = 193)Age≤ 62 89 46.1%> 62 104 53.9%GenderMale 109 56.5%Female 84 43.5%Thickness≤ 2.0 mm 91 47.2%> 2.0 mm 102 52.8%UlcerationAbsent 144 74.6%Present 49 25.4%Metastatic melanoma (n = 134)Age≤ 62 77 57.5%> 62 57 42.5%GenderMale 87 64.9%Female 47 35.1%Bhandaru et al. BMC Cancer 2014, 14:398 Page 2 of 11http://www.biomedcentral.com/1471-2407/14/398according to the principles expressed in the Declaration ofHelsinki.From the original tissue biopsies, the most representa-tive tumor area was carefully selected and marked onhematoxylin and eosin stained slides. Tissue cores of 0.6-mm thickness were taken in duplicate from each biopsyand the TMAs were assembled using a tissue-array instru-ment (Beecher Instruments, Silver Spring, MD). Using aLeica microtome, multiple 4 μM sections were cut andtransferred to adhesive-coated slides using regular histo-logical procedures. One section from each TMA was rou-tinely stained with hematoxylin and eosin while theremaining sections were stored at room temperature forimmunohistochemical staining.ImmunohistochemistryTissue microarray (TMA) slides were dewaxed at 55°C for20 min followed by three 5 min washes with xylene. Thetissues were then rehydrated by washing the slides for5 min each with 100%, 95%, 80% ethanol and finally withdistilled water. The slides were then heated to 95°C for30 min in 10 mmol/L sodium citrate (pH 6.0) for antigenretrieval and then treated with 3% hydrogen peroxide for1 hour to block the endogenous peroxidase activity. Afterblocking the slides with the universal blocking serum (DakoDiagnostics, Carpinteria, CA, USA), the sections were incu-bated overnight with monoclonal mouse anti-p300 anti-body (1:50 dilution; Millipore, USA) or with mousepolyclonal anti-Braf antibody (1:100 dilution; Sigma, USA)at 4°C. The sections were then incubated for 30 minwith a biotin-labeled secondary antibody and then withstreptavidin-peroxidase (Dako Diagnostics). The sampleswere developed by treatment with 3,3′-diamino-benzidinesubstrate (Vector Laboratories, Burlington, Ontario,Canada) and with hematoxylin to counter-stain the nuclei.Negative controls were done by omitting the p300/Brafantibody during the primary antibody incubation.Evaluation of immunostainingThe evaluation of p300 and Braf staining was done blindlyby microscopic examination of the tissue sections by onedermatopathologist and two other observers simultan-eously, using a multiple viewing microscope and a consen-sus was reached for the score of each core. p300/Brafstaining intensity was scored as 0+, 1+, 2+, 3+ whereas thepercentage of p300/Braf positive cells was scored as 1 (1-25%), 2 (26-50%), 3 (51-75%) and 4 (76-100%). In cases ofdiscrepancy between duplicated cores, the higher scorefrom the two tissue cores was taken as the final score. Theproduct of intensity and percentage was taken as the im-munoreactive score (IRS) [15]. Based on IRS, p300 & Brafstaining in the tissue sections was categorized as negative(IRS 0), weak (IRS 1–4), moderate (IRS 6–8), or strong (IRS9–12). Since p300 was found to be expressed in bothnucleus and cytoplasm [10], the nuclear and cytoplasmicstaining was evaluated in parallel at the same time. Thechoice of the optimum cut-off values for the IRS were de-rived based on the IRS pattern in nevi and melanoma casesand are described previously [7,10].Statistical analysisCorrelation between p300 and Braf, and clinicopathologicparameters was evaluated by Chi-square test among the pa-tient subgroups. Survival time was calculated from the dateof melanoma diagnosis to the date of death or last follow-up. The effect of p300 and Braf on the overall and disease-specific survival was evaluated by Kaplan-Meier analysisand log-rank test. Additionally, multivariate Cox propor-tional hazards regression models were preformed to esti-mate the hazard ratios (HRs) and their 95% confidentialintervals (CIs). Classification tree was constructed by theclassification and regression tree (CRT) model as describedpreviously to examine possibility of using a Braf and p300combination to identify different stages of melanoma [16].The decision trees depicting the classification rules weregenerated through recursive partitioning. When growingeach tree, equal prior probabilities to the normal and can-cer cohorts, and equal misclassification costs were assigned.To assess the amount of over-fitting, 10-fold cross-validation experiments was performed using the SE rule asdescribed previously [16]. P-value <0.05 was considered asstatistically significant. All the statistical analyses were per-formed using SPSS version 16.0 (SPSS Inc, Chicago, IL)software.ResultsBraf expression correlates inversely with nuclear p300and directly with cytoplasmic p300 expressionPrevious studies showed that phosphorylation by MAP kin-ase resulted in accelerated degradation of p300 in cardiaccells [11]. Since Braf is known to be an up stream kinase inthe MAP kinase pathway, we asked if its expression couldbe inversely associated with p300 expression in the tumorsamples from melanoma patients. Based on the previouslyreported cut-off values for immunoreactive scores (IRS), wedivided the staining into low (IRS 0 to 4) and high (IRS6 to 12), and matched the expression of Braf and p300in the melanoma patients [7,10]. Chi-square analysis ofthe matched data revealed that Braf expression inverselycorrelated with nuclear p300 and directly correlatedwith cytoplasmic p300 expression suggesting Braf nega-tively regulates the nuclear accumulation of p300(Figure 1A & B).Braf and cytoplasmic p300 expression are associated withdisease progressionWe next asked if the association between Braf and p300expression was particularly correlated with diseaseBhandaru et al. BMC Cancer 2014, 14:398 Page 3 of 11http://www.biomedcentral.com/1471-2407/14/398progression or tumor size or ulceration status. We firstdivided the data based on American Joint Committee forCancer (AJCC) staging and performed Chi-square testanalysis. As shown in Table 2, the percentage of patientswith high Braf expression or high cytoplasmic expressionwas significantly increased as melanoma progressedfrom AJCC stage I to stage III and then slightly de-creased from stage III to stage IV. Accordingly, the per-centage of patients with high Braf and high cytoplasmicp300 expression was significantly increased from AJCCstage I through stage III and slightly decreased fromstage III to stage IV (Figure 1C). Interestingly, the differ-ence in percentage of patients with high Braf and highcytoplasmic p300 expression was highest between stage Iand II, which differ mainly based on the tumor size(Figure 1C) [17]. On the other hand, increase in the per-centage of cases with high Braf and low nuclear p300 ex-pression was more apparent between stages II and III,which differ based on the presence of tumor cells in thelymph nodes, an indicator of migration and metastasis(Table 2) [17].Next we separated the cases based on tumor size (≤2 mmversus >2 mm) and then based on ulceration status (no ul-ceration versus ulceration). Braf expression was found to besignificantly associated with tumor size and ulceration sta-tus, whereas cytoplasmic p300 expression was associatedwith tumor size but not with ulceration status (Table 3).Nuclear p300 expression was not associated with tumorsize or ulceration status (Table 3). As seen with melanomaprogression, the incidence of larger tumors was significantlyFigure 1 Braf expression correlates with p300 expression in melanoma patients. (A) Negative correlation between Braf and nuclear p300expression in melanoma patient biopsies. Melanomas which have high Braf expression have a significantly higher percentage of low nuclearp300 staining (p = 0.006, χ2 test). (B) Positive correlation between Braf and cytoplasmic p300 expression in melanoma patient biopsies.Melanomas which have high Braf expression also have a significantly higher percentage of high cytoplasmic p300 staining (p = 0.001, χ2 test).High Braf and high cytoplasmic p300 expression is significantly associated with AJCC progression (C) and tumor size (D), but not with ulcerationstatus (E). p-values, 8.7×10−5, 0.001 & 0.119 respectively (χ2 test).Bhandaru et al. BMC Cancer 2014, 14:398 Page 4 of 11http://www.biomedcentral.com/1471-2407/14/398higher (Figure 1C), and presence of ulcerated tumorstended to be higher (Figure 1D), in patients with high Brafand high cytoplasmic p300 expression. Though patientswith low nuclear p300 tended to be associated with ad-vanced stages of melanoma, larger tumor size and presenceof ulcerated tumors, the difference did not reach statisticalsignificance (Table 3).Combination of Braf and p300 in the diagnosis ofmelanomaSince we found Braf and p300 to be significantly associ-ated with markers of advanced melanoma stages, weasked if a combination of Braf and p300 expressioncould be used to separate nevi from melanoma in skinbiopsies. Classification and regression tree (CRT) ana-lysis of the patient expression data was previously shownto be useful in differentiating nevi and melanoma [16].We categorized the nevi and melanoma values asdependent variables and Braf, nuclear p300 and cyto-plasmic p300 expression as independent variables, andperformed CRT analysis on the data. As seen in Figure 2,Braf expression was the best marker to predict melan-oma cases, followed by cytoplasmic p300 expression andnuclear p300 expression. We then used CRT analysis totest if the combination of Braf and p300 could be usedto classify the primary melanoma cases and metastaticmelanoma cases. As seen in Figure 3, cytoplasmic p300expression was the best marker to separate the primarymelanoma from metastatic melanoma cases, whichcould be further classified, using Braf and nuclear p300expression.Combination of Braf and p300 in patient prognosisIn order to test the significance of Braf and p300 in pa-tient prognosis, we analyzed the correlation betweenBraf and p300 expression and patient survival usingKaplan-Meier analysis. We first confirmed the previouslyreported association between nuclear p300 and patientsurvival, and then tested a combination of Braf and nu-clear p300 and studied the 5-year patient survival. Asseen in Figure 4A & B, patients with low nuclear p300expression had significantly worse 5-year survival. Intri-guingly, patients with high Braf and low nuclear p300had significantly worse 5-year survival, and patients withlow Braf and high nuclear p300 had better 5-year sur-vival, indicating the opposing effects of Braf and nuclearp300 on patient survival (Figure 5A & B). On the otherhand, a combination of cytoplasmic p300 and Brafexpression tended to be associated with worse prognosisand the patients with high Braf and high cytoplasmicp300 had the worst 5-year overall and disease-specificsurvival compared to the other categories (Figure 5CTable 2 Correlation between Braf/p300 staining and AJCC stage in 327 melanoma patientsStage I Stage II Stage III Stage IV p-value*BrafLow 37 (46.3%) 26 (23.0%) 11 (20.0%) 20 (25.3%) 9.8 × 10−4High 43 (53.8%) 87 (77.0%) 44 (80.0%) 59 (74.7%)Nuclear p300Low 29 (36.3%) 47 (41.6%) 30 (54.5%) 33 (41.8%) 0.204High 51 (63.7%) 66 (58.4%) 25 (45.5%) 46 (59.2%)Cytoplasmic p300Low 48 (60.0%) 53 (46.9%) 17 (30.9%) 36 (45.6%) 0.011High 32 (40.0%) 60 (53.1%) 38 (69.1%) 43 (54.4%)Braf and nuclear p300Low braf low p300 11 (13.8%) 13 (11.5%) 6 (10.9%) 6 (7.6%) 0.010Low braf high p300 26 (32.5%) 14 (12.4%) 6 (10.9%) 16 (20.3%)High braf low p300 18 (22.5%) 34 (30.1%) 24 (43.6%) 27 (34.2%)High braf high p300 25 (31.3%) 52 (46.0%) 19 (34.6%) 30 (38.0%)Braf and cytoplasmic p300Low braf low p300 21 (26.3%) 19 (16.8%) 5 (9.1%) 7 (8.8%) 8.7 × 10−5Low braf high p300 16 (20.0%) 7 (6.2%) 6 (10.9%) 13 (16.5%)High braf low p300 27 (33.7%) 34 (30.1%) 12 (21.8%) 29 (36.7%)High braf high p300 16 (20.0%) 53 (46.9%) 32 (58.2%) 30 (38.0%)*- χ2 test.Bhandaru et al. BMC Cancer 2014, 14:398 Page 5 of 11http://www.biomedcentral.com/1471-2407/14/398& D). However, the differences were not strong enoughand failed to reach statistical significance.Nuclear p300 expression independently regulates patientsurvivalWe then performed multivariate Cox regression analysis totest if Braf and/or p300 expression could independentlyregulate the patient survival. We used AJCC staging, nu-clear p300, cytoplasmic p300, and Braf expression as vari-ables in the model. As shown in Table 4, multivariate Coxregression analysis revealed that AJCC staging and nuclearp300 were significantly associated with patient survival,whereas the association between Braf and cytoplasmicp300, and patient survival did not reach statistical signifi-cance. Our results are in line with the previously publisheddata showing that Braf expression was not an independentprognostic factor. It was suggested that due to the close as-sociation with the AJCC stages, tumor size and ulcerationstatus, Braf expression could not independently predict pa-tient survival [7].DiscussionThe key to successful management of melanoma includesboth early and accurate diagnosis, followed by medicalintervention in the form of surgery and chemotherapy. Ac-curacy of the diagnosis is particularly important as misdiag-nosis of the melanoma patients might lead to inadequatetreatment and allow spread of the disease. Melanoma is dis-tinguished from dysplastic nevi with a fair degree of successusing routine pathological examination, but ambiguous le-sions could still pose problems due to the wide variation inmorphologic features and because of the overlap in theclinical and histologic features between dysplastic neviand melanoma [16,18-21]. Our results suggest that acombination of Braf and p300 expression can be used fordifferentiating melanoma from nevi. The protocol for im-munohistochemical staining of the tissue samples is a sim-ple technique to perform and can give results relatively fast[22]. Since the expression of only two markers is needed tocompletely separate nevi from melanoma, the experimentalcosts are also relatively small. Our study could thus be usedto develop a practical protocol, which would complementroutine pathological examination and provide a clarificationwhen tissue sections show overlapping morphologic andhistologic features.Despite significant progress in the identification of mo-lecular pathways that drive tumorigenesis, melanoma stillposes a challenge to the scientific community. Owing to itsnotorious resistance to chemotherapy, patients with malig-nant melanoma have limited treatment options and have apoor prognosis. Although, vemurafenib, a BrafV600E specificinhibitor, showed impressive results in terms of responserate and progression free survival, the responses are mostlyshort-lived as seen by development of resistance in nearlyTable 3 Correlation between Braf/p300 staining andtumor size, and ulceration status in 327 melanomapatientsTumor size≤ 2 mm > 2 mm p-value*BrafLow 38 (41.8%) 25 (24.5%) 0.011High 53 (58.2%) 77 (75.5%)Nuclear p300Low 33 (36.3%) 43 (42.2%) 0.403High 58 (63.7%) 59 (57.8%)Cytoplasmic p300Low 55 (60.4%) 46 (45.1%) 0.033High 36 (39.6%) 56 (54.9%)Braf and nuclear p300Low Braf and low p300 11 (12.1%) 13 (12.7%) 0.035Low Braf and high p300 27 (29.7%) 13 (12.7%)High Braf and low p300 22 (24.2%) 30 (29.4%)High Braf and high p300 31 (34.1%) 46 (45.1%)Braf and cytoplasmic p300Low Braf and low p300 22 (24.2%) 18 (17.6%) 0.001Low Braf and high p300 16 (17.6%) 7 (6.9%)High Braf and low p300 33 (36.2%) 28 (27.5%)High Braf and high p300 20 (22.0%) 49 (48.0%)Ulceration statusAbsent PresentBrafLow 53 (36.8%) 10 (20.4%) 0.034High 91 (63.2%) 39 (79.6%)Nuclear p300Low 55 (38.2%) 21 (42.9%) 0.564High 89 (61.8%) 28 (57.1%)Cytoplasmic p300Low 79 (54.9%) 22 (44.9%) 0.223High 65 (45.1%) 27 (55.1%)Braf and nuclear p300Low Braf and low p300 18 (12.5%) 6 (12.2%) 0.199Low Braf and high p300 35 (24.3%) 5 (10.2%)High Braf and low p300 37 (25.7%) 15 (30.6%)High Braf and high p300 54 (37.5%) 23 (46.9%)Braf and cytoplasmic p300Low Braf and low p300 32 (20.8%) 8 (16.3%) 0.119Low Braf and high p300 21 (13.6%) 2 (4.1%)High Braf and low p300 47 (30.5%) 14 (28.6%)High Braf and high p300 54 (35.1%) 25 (51.1%)*- χ2 test.Bhandaru et al. BMC Cancer 2014, 14:398 Page 6 of 11http://www.biomedcentral.com/1471-2407/14/398every case [23-25]. Several strategies to increase the effect-iveness, like combining Braf inhibitors with MEK1/2 inhibi-tors or small molecule inhibitors of the PI-3 kinasepathway, are in various stages of clinical studies, but it istoo early to predict their clinical efficacy [6,25].Our results from patient survival show that patientswith low Braf and high nuclear p300 expression havebetter survival, hinting at the benefits of simultaneouslytargeting Braf and nuclear p300 in treatment of melan-oma. Data from our previous study showed that thoughcytoplasmic p300 expression was significantly associatedwith clinico-pathologic characteristics of melanoma, onlynuclear p300 had prognostic significance [10]. Even inthe present study, cytoplasmic p300 expression was onlyinformative during the diagnosis part of the analysis butwas not a significant prognostic factor (Table 4). Besides,the major site of activity of p300 is in the nucleus whereit regulates critically important processes like transcrip-tion and DNA repair [26-28]. Interestingly, loss ofanother well known histone acetyltransferase, TIP60,was reported to be associated with worse prognosis inmelanoma patients [29]. We therefore think thatcombining Braf inhibitors with HDAC inhibitors mightbe beneficial in the chemotherapy of melanoma. Strik-ingly, two HDAC inhibitors, vorinostat (Merck) andromidepsin (Gloucester Pharmaceuticals), which report-edly showed inhibitory effects on melanoma growth,were approved by the US FDA for the treatment of cuta-neous T-cell lymphoma [30-34]. A combination of tyro-sine kinase & C-Raf inhibitor, Sorafenib and vorinostat iscurrently being studied in the treatment of advancedcancers [35], but we could not find any studies per-formed using a combination of B-raf inhibitors and vori-nostat or romidepsin. Our findings encourage furtherresearch on the potential improved efficacy of coadmin-istration of Braf and HDAC inhibitors.Another finding of our study is the inverse correlationbetween Braf and nuclear p300 and direct correlationFigure 2 Classification and Regression tree for differentiating nevi from melanoma using Braf, nuclear p300 and cytoplasmic p300expression. Nevi samples include both normal and dysplastic nevi cases. Melanoma samples include both primary and metastatic melanomacases. ‘n’ indicates the number of samples and ‘%’ indicates the percentage of samples available at the respective node. Improvement is anindicator of separation achieved by the application of the respective marker to classify the parent node.Bhandaru et al. BMC Cancer 2014, 14:398 Page 7 of 11http://www.biomedcentral.com/1471-2407/14/398Figure 3 Classification and Regression tree for differentiating primary melanoma from metastatic melanoma using Braf, nuclear p300and cytoplasmic p300 expression. PM, primary melanoma, includes AJCC stages I and II cases. MM, metastatic melanoma, includes stages IIIand IV. ‘n’ indicates the number of samples and ‘%’ indicates the percentage of samples available at the respective node. Improvement is anindicator of separation achieved by the application of the respective marker to classify the parent node.Figure 4 Nuclear p300 expression and 5-year patient survival. Kaplan-Meier survival analyses of correlation between nuclear p300 expressionand 5-year overall (A) and disease-specific (B) survival of melanoma patients. The cases with low nuclear p300 expression are represented by‘blue’ line and the cases with high expression are represented by ‘pink’ line.Bhandaru et al. BMC Cancer 2014, 14:398 Page 8 of 11http://www.biomedcentral.com/1471-2407/14/398between Braf and cytoplasmic p300 expression whichsuggests possible cross-talk between Braf and p300. Pre-vious studies showed that phosphorylation of p300 coulddifferentially regulate its activity and protein stability[36,37]. For example, while protein kinase C (PKC) andsalt inducible kinase 2 mediated phosphorylation atserine-89 was reported to inhibit the HAT activity[38,39], Akt mediated phosphorylation at serine-1834,serine-2279, serine-2315, and serine-2366 was shown toenhance the HAT activity of p300 [40-42]. Along thoselines, Akt and ERK2 mediated phosphorylationwas shown to stabilize p300 protein levels, but phos-phorylation by mitogen activated protein kinase(MAPK) resulted in degradation of the p300 protein[11,12,36,40,43]. However, none of the studies have sofar focused on the effect of phosphorylation on intracel-lular distribution of p300. Our findings point to thepossible phosphorylation and altered localization ofp300 by Braf/MAPK signaling, which needs furtherinvestigation.While our database was relatively large with details ofseveral clinical characteristics, further studies are war-ranted before drawing firm conclusions on the benefitsof combined Braf and HDAC inhibitors. Though the sig-nificance of finding a correlation in patient biopsies can-not be underestimated, evidence from studies at theFigure 5 Braf and p300 expression and 5-year patient survival. Kaplan-Meier survival analyses of correlation between Braf & p300 expressionand 5-year overall (left panels) and disease-specific (right panels) survival of melanoma patients. (A and B) correlation between Braf and nuclearp300 expression, and patient survival. (C and D) correlation between Braf and cytoplasmic p300 expression, and patient survival. Blue linerepresents the cases with low Braf and low p300; pink line, cases with low Braf and high p300; yellow line, high Braf and low p300; sky blue line,high Braf and high p300.Table 4 Multivariate Cox regression analysis on overall and disease-specific survival of primary melanoma patientsVariablesOverall survival Disease-specific survival߆ SE HR (95% CI) p-value ß SE HR (95% CI) p-valueAJCC 1.302 0.170 3.68 (2.63-5.13) 1.98 × 10−14 1.457 0.182 4.29 (3.01-6.13) 1.13 × 10−14Braf 0.178 0.190 1.20 (0.82-1.73) 0.348 0.110 0.196 1.12 (0.76-1.64) 0.575Nuclear p300 −0.508 0.161 0.60 (0.44-0.83) 0.002 −0.525 0.169 0.59 (0.42-0.82) 0.002Cytoplasmic p300 0.049 0.163 1.05 (0.76-1.45) 0.764 0.091 0.171 1.10 (0.78-1.53) 0.595Coding of variables: AJCC was coded as 1 (stages I & II) and 2 (stages (III & IV). Braf and p300 expression was coded as 1 (low staining) and 2 (high staining).†β: regression coefficient.Abbreviations: SE standard error of β, HR hazard ratio, CI confidence interval.Bhandaru et al. BMC Cancer 2014, 14:398 Page 9 of 11http://www.biomedcentral.com/1471-2407/14/398cellular level is needed to convincingly establish the rela-tionship between Braf and p300. Furthermore, we didnot have enough cases with information on the status ofBraf mutations, so we were unable to analyze the poten-tial correlation between BrafV600E and p300.ConclusionsOur study elucidates the cross talk between Braf andp300 in melanoma and suggests that Braf might nega-tively regulate the accumulation of p300 in the nucleusand promote the cytoplasmic localization of p300. Wealso show that using a combination of Braf and p300 ex-pression, it is possible to separate nevi and melanomasamples, and primary and metastatic melanoma samples.We show that patients with low Braf and high p300 ex-pression have better prognosis, suggesting the possibilityof combining Braf and HDAC inhibitors in melanomatreatment.Competing interestsThe authors declare that they have no competing interests.Authors’ contributionsConceived and designed the project: AR, analyzed the data: MB, MM, GA, GL,GZ, AR, and KM, wrote the manuscript: AR, KM and MB. All authors read andapproved the final manuscript.AcknowledgementsMB, GA, GZ, GL were supported by funds from Canadian Institute of HealthResearch (CCI-117958, MOP-110974, MOP-93810), KM was supported byfunds from Canadian Dermatology Foundation. The funding organizationshad no role in study design; in the collection, analysis, and interpretation ofdata; in the writing of the manuscript; and in the decision to submit themanuscript for publication.Author details1Skin Cancer Biology Laboratory, Department of Dermatology and SkinScience, University of British Columbia, Research Pavilion, 828 West, 10thAvenue, Vancouver, BC V5Z 1 L8, Canada. 2Department of Pathology &Laboratory Medicine, University of British Columbia, Vancouver, BritishColumbia, Canada.Received: 10 November 2013 Accepted: 29 May 2014Published: 3 June 2014References1. Miller AJ, Mihm MC Jr: Melanoma. N Engl J Med 2006, 355(1):51–65.2. Rastrelli M, Alaibac M, Stramare R, Chiarion Sileni V, Montesco MC, VecchiatoA, Campana LG, Rossi CR: Melanoma m (zero): diagnosis and therapy.ISRN Dermatol 2013, 2013:616170.3. Siegel R, Naishadham D, Jemal A: Cancer statistics, 2013. CA Cancer J Clin2013, 63(1):11–30.4. Turner RM, Bell KJ, Morton RL, Hayen A, Francken AB, Howard K, ArmstrongB, Thompson JF, Irwig L: Optimizing the frequency of follow-up visits forpatients treated for localized primary cutaneous melanoma. J Clin Oncol2011, 29(35):4641–4646.5. Bilimoria KY, Raval MV, Bentrem DJ, Wayne JD, Balch CM, Ko CY: Nationalassessment of melanoma care using formally developed qualityindicators. J Clin Oncol 2009, 27(32):5445–5451.6. Finn L, Markovic SN, Joseph RW: Therapy for metastatic melanoma: thepast, present, and future. BMC Med 2012, 10:23.7. Safaee Ardekani G, Jafarnejad SM, Khosravi S, Martinka M, Ho V, Li G:Disease progression and patient survival are significantly influenced byBRAF protein expression in primary melanoma. Br J Dermatol 2013,169(2):320–328.8. Safaee Ardekani G, Jafarnejad SM, Tan L, Saeedi A, Li G: The prognosticvalue of BRAF mutation in colorectal cancer and melanoma: asystematic review and meta-analysis. PLoS One 2012, 7(10):e47054.9. Roth SY, Denu JM, Allis CD: Histone acetyltransferases. Annu Rev Biochem2001, 70:81–120.10. Rotte A, Bhandaru M, Cheng Y, Sjoestroem C, Martinka M, Li G: Decreasedexpression of nuclear p300 is associated with disease progression andworse prognosis of melanoma patients. PLoS One 2013, 8(9):e75405.11. Poizat C, Puri PL, Bai Y, Kedes L: Phosphorylation-dependent degradationof p300 by doxorubicin-activated p38 mitogen-activated protein kinasein cardiac cells. Mol Cell Biol 2005, 25(7):2673–2687.12. Wang SA, Hung CY, Chuang JY, Chang WC, Hsu TI, Hung JJ: Phosphorylation ofp300 increases its protein degradation to enhance the lung cancerprogression. Biochim Biophys Acta 2014, 1843(6):1135–1149.13. Zhang Z, Chen G, Cheng Y, Martinka M, Li G: Prognostic significance ofRUNX3 expression in human melanoma. Cancer 2011, 117(12):2719–2727.14. Chen G, Cheng Y, Zhang Z, Martinka M, Li G: Cytoplasmic Skp2 expressionis increased in human melanoma and correlated with patient survival.PLoS One 2011, 6(2):e17578.15. Remmele W, Stegner HE: [Recommendation for uniform definition of animmunoreactive score (IRS) for immunohistochemical estrogen receptordetection (ER-ICA) in breast cancer tissue]. Pathologe 1987, 8(3):138–140.16. Zhang G, Li G: Novel multiple markers to distinguish melanoma fromdysplastic nevi. PLoS One 2012, 7(9):e45037.17. Balch CM, Gershenwald JE, Soong SJ, Thompson JF, Atkins MB, Byrd DR,Buzaid AC, Cochran AJ, Coit DG, Eggermont AM DS, Flaherty KT, GimottyPA, Kirkwood JM, McMasters KM, Mihm MC Jr, Morton DL, Ross MI, SoberAJ, Sondak VK: Final version of 2009 AJCC melanoma staging andclassification. J Clin Oncol 2009, 27(36):6199–6206.18. Pellacani G, Longo C, Ferrara G, Cesinaro AM, Bassoli S, Guitera P, MenziesSW, Seidenari S: Spitz nevi: In vivo confocal microscopic features,dermatoscopic aspects, histopathologic correlates, and diagnosticsignificance. J Am Acad Dermatol 2009, 60(2):236–247.19. Duffy K, Grossman D: The dysplastic nevus: from historical perspective tomanagement in the modern era: part II. Molecular aspects and clinicalmanagement. J Am Acad Dermatol 2012, 67(1):19 e11–12. quiz 31–12.20. Duffy K, Grossman D: The dysplastic nevus: from historical perspective tomanagement in the modern era: part I. Historical, histologic, and clinicalaspects. J Am Acad Dermatol 2012, 67(1):1 e1–16. quiz 17–18.21. Rabkin MS: The limited specificity of histological examination in thediagnosis of dysplastic nevi. J Cutan Pathol 2008, 35(Suppl 2):20–23.22. Camp RL, Neumeister V, Rimm DL: A decade of tissue microarrays:progress in the discovery and validation of cancer biomarkers. J ClinOncol 2008, 26(34):5630–5637.23. Chapman PB, Hauschild A, Robert C, Haanen JB, Ascierto P, Larkin J,Dummer R, Garbe C, Testori A, Maio M, Hogg D, Lorigan P, Lebbe C, JouaryT, Schadendorf D, Ribas A, O'Day SJ, Sosman JA, Kirkwood JM, EggermontAMM, Dreno B, Nolop K, Li J, Nelson B, Hou J, Lee RJ, Flaherty KT, McArthurGA, for the BRIM-3 Study Group: Improved survival with vemurafenib inmelanoma with BRAF V600E mutation. N Engl J Med 2011,364(26):2507–2516.24. Flaherty KT, Puzanov I, Kim KB, Ribas A, McArthur GA, Sosman JA, O’Dwyer PJ, LeeRJ, Grippo JF, Nolop K, Chapman PB: Inhibition of mutated, activated BRAF inmetastatic melanoma. N Engl J Med 2010, 363(9):809–819.25. Kudchadkar R, Paraiso KH, Smalley KS: Targeting mutant BRAF inmelanoma: current status and future development of combinationtherapy strategies. Cancer J 2012, 18(2):124–131.26. Iyer NG, Ozdag H, Caldas C: p300/CBP and cancer. Oncogene 2004,23(24):4225–4231.27. Reed SH: Nucleotide excision repair in chromatin: damage removal atthe drop of a HAT. DNA Repair (Amst) 2011, 10(7):734–742.28. Tillhon M, Cazzalini O, Nardo T, Necchi D, Sommatis S, Stivala LA, ScovassiAI, Prosperi E: p300/CBP acetyl transferases interact with and acetylatethe nucleotide excision repair factor XPG. DNA Repair (Amst) 2012,11(10):844–852.29. Chen G, Cheng Y, Tang Y, Martinka M, Li G: Role of Tip60 in humanmelanoma cell migration, metastasis, and patient survival. J InvestDermatol 2012, 132(11):2632–2641.30. Giannini G, Cabri W, Fattorusso C, Rodriquez M: Histone deacetylaseinhibitors in the treatment of cancer: overview and perspectives.Future Med Chem 2012, 4(11):1439–1460.Bhandaru et al. BMC Cancer 2014, 14:398 Page 10 of 11http://www.biomedcentral.com/1471-2407/14/39831. Haigentz M Jr, Kim M, Sarta C, Lin J, Keresztes RS, Culliney B, Gaba AG,Smith RV, Shapiro GI, Chirieac LR Mariadason JM, Belbin TJ, Greally JM,Wright JJ, Haddadf RI: Phase II trial of the histone deacetylase inhibitorromidepsin in patients with recurrent/metastatic head and neck cancer.Oral Oncol 2012, 48(12):1281–1288.32. Murakami T, Sato A, Chun NA, Hara M, Naito Y, Kobayashi Y, Kano Y, OhtsukiM, Furukawa Y, Kobayashi E: Transcriptional modulation using HDACidepsipeptide promotes immune cell-mediated tumor destruction ofmurine B16 melanoma. J Invest Dermatol 2008, 128(6):1506–1516.33. Landreville S, Agapova OA, Matatall KA, Kneass ZT, Onken MD, Lee RS,Bowcock AM, Harbour JW: Histone deacetylase inhibitors induce growtharrest and differentiation in uveal melanoma. Clin Cancer Res 2012,18(2):408–416.34. Dekker FJ, Haisma HJ: Histone acetyl transferases as emerging drugtargets. Drug Discov Today 2009, 14(19–20):942–948.35. Dasari A, Gore L, Messersmith WA, Diab S, Jimeno A, Weekes CD, Lewis KD,Drabkin HA, Flaig TW, Camidge DR: A phase I study of sorafenib andvorinostat in patients with advanced solid tumors with expandedcohorts in renal cell carcinoma and non-small cell lung cancer.Invest New Drugs 2013, 31(1):115–125.36. Wang QE, Han C, Zhao R, Wani G, Zhu Q, Gong L, Battu A, Racoma I,Sharma N, Wani AA: p38 MAPK- and Akt-mediated p300 phosphorylationregulates its degradation to facilitate nucleotide excision repair.Nucleic Acids Res 2013, 41(3):1722–1733.37. Chen J, Li Q: Life and death of transcriptional co-activator p300.Epigenetics 2011, 6(8):957–961.38. Bricambert J, Miranda J, Benhamed F, Girard J, Postic C, Dentin R: Salt-inducible kinase 2 links transcriptional coactivator p300 phosphorylationto the prevention of ChREBP-dependent hepatic steatosis in mice. J ClinInvest 2010, 120(12):4316–4331.39. Yuan LW, Gambee JE: Phosphorylation of p300 at serine 89 by proteinkinase C. J Biol Chem 2000, 275(52):40946–40951.40. Chen J, Halappanavar SS, St-Germain JR, Tsang BK, Li Q: Role of Akt/proteinkinase B in the activity of transcriptional coactivator p300. Cell Mol LifeSci 2004, 61(13):1675–1683.41. Huang WC, Chen CC: Akt phosphorylation of p300 at Ser-1834 is essentialfor its histone acetyltransferase and transcriptional activity. Mol Cell Biol2005, 25(15):6592–6602.42. Liu Y, Denlinger CE, Rundall BK, Smith PW, Jones DR: Suberoylanilidehydroxamic acid induces Akt-mediated phosphorylation of p300, whichpromotes acetylation and transcriptional activation of RelA/p65. J BiolChem 2006, 281(42):31359–31368.43. Chen YJ, Wang YN, Chang WC: ERK2-mediated C-terminal serinephosphorylation of p300 is vital to the regulation of epidermal growthfactor-induced keratin 16 gene expression. J Biol Chem 2007,282(37):27215–27228.doi:10.1186/1471-2407-14-398Cite this article as: Bhandaru et al.: A combination of p300 and Brafexpression in the diagnosis and prognosis of melanoma. BMC Cancer2014 14:398.Submit your next manuscript to BioMed Centraland take full advantage of: • Convenient online submission• Thorough peer review• No space constraints or color figure charges• Immediate publication on acceptance• Inclusion in PubMed, CAS, Scopus and Google Scholar• Research which is freely available for redistributionSubmit your manuscript at www.biomedcentral.com/submitBhandaru et al. BMC Cancer 2014, 14:398 Page 11 of 11http://www.biomedcentral.com/1471-2407/14/398


Citation Scheme:


Citations by CSL (citeproc-js)

Usage Statistics



Customize your widget with the following options, then copy and paste the code below into the HTML of your page to embed this item in your website.
                            <div id="ubcOpenCollectionsWidgetDisplay">
                            <script id="ubcOpenCollectionsWidget"
                            async >
IIIF logo Our image viewer uses the IIIF 2.0 standard. To load this item in other compatible viewers, use this url:


Related Items