Archives of Medical Research 45 (2014) 223e228

ORIGINAL ARTICLE

YAP Expression in Normal and Neoplastic Breast Tissue:An Immunohistochemical Study

Yolanda Jaramillo-Rodr�ıguez,a Ricardo M. Cerda-Flores,b Ruben Ruiz-Ramos,c

Francisco C. L�opez-M�arquez,d and Ana Laura Calder�on-Garcidue~nasc

aInstituto de Medicina Forense, Juan Pablo II, Esq, Reyes Heroles Fracc. Costa Verde, Boca del R�ıo, Veracruz, MexicobFacultad de Enfermer�ıa, Universidad Aut�onoma de Nuevo Le�on, Monterrey, NL, MexicocInstituto de Medicina Forense, Universidad Veracruzana, Boca del R�ıo, Veracruz, Mexico

dCentro de Investigaci�on Biom�edica de la Facultad de Medicina, Universidad Aut�onoma de Coahuila, Torr�eon, Coahuila, Mexico

Received for publication September 25, 2013; accepted January 29, 2014 (ARCMED-D-13-00525).

Address reprint

Researcher, Instituto d

Heroles Fracc. Costa

Phone: +52 (229) 921

þ52 (229) 775 2000

0188-4409/$ - see frohttp://dx.doi.org/10

Background and Aims. Yes-associated protein (YAP) is a transcriptional factor involvedin normal cell proliferation, apoptosis and carcinogenesis; however, its contribution tobreast cancer (BC) is still controversial. We undertook this study to compare the expres-sion of YAP by immunohistochemistry (IHC) in normal breast tissue of women withoutbreast cancer (BC) (controls), non-neoplastic breast tissue in women with cancer (internalcontrols) and in four different subtypes of invasive ductal carcinoma.

Methods. There were 17 controls and 105 tumor cases (53 luminal A, 15 luminal B, 20overexpression of HER2 and 17 triple negative cases) studied by IHC. Statistical analysisincluded c2 for linear trend (Extended Mantel-Haenszel).

Results. There were 40% of internal controls that showed expression of YAP in myoepi-thelial cells, whereas in controls expression was 100%. In controls, 3/17 (17.6%) showedcytoplasmic staining in luminal cells. There was a significant difference in nuclearexpression between the ductal BC subtypes. Luminal A had 4% of positive cases with!10% of cells affected in each case; in contrast, there were 17e20% of positive casesin the other groups with 50% or more of stained cells. YAP expression in stromal cellswas not observed in controls or in triple-negative cases, and luminal B pattern had thehighest YAP nuclear expression (20%).

Conclusions. YAP showed decreased expression in tumor cells compared with normalbreast tissue. These findings are consistent with a role of YAP as a suppressor gene inBC and show differences in YAP expression in different patterns of ductalBC. � 2014 IMSS. Published by Elsevier Inc.

Key Words: Yes-associated protein, Breast cancer, Immunohistochemistry.

Introduction

Yes-associated protein (YAP) is a transcriptional factorinvolved in normal cell proliferation, apoptosis and carcino-genesis (1,2). In this pathway, mammalian Ste20-like (MST)kinases phosphorylate and activate the large tumor suppres-sor (LATS) kinases. The kinase cascade phosphorylates and

requests to: Ana Laura Calder�on-Garcidue~nas,

e Medicina Forense, S.S. Juan Pablo II Esq, Reyes

Verde C.P. 94294, Boca del Rı́o, Veracruz, Mexico;

8741 and þ52 (229) 775 2000 ext. 22503; FAX:

ext. 22036; E-mail: acald911@hotmail.com

nt matter. Copyright � 2014 IMSS. Published by Elsevier.1016/j.arcmed.2014.01.010

inhibits Yes-associated protein 1 (YAP1), which is retainedin the cytoplasm. Thus, cell cycle-promoting and anti-apoptotic gene transcriptions are turned-off (3).

When the Hippo signaling is absent, YAP1 translocatesto the nucleus, associates with coactivators TEAD1-4, andacts as a transcriptional factor promoting the expressionof genes involved in cell proliferation. YAP in normal con-ditions is a protooncogene downregulated by the Hippopathway (4). In human cancer, amplification and overex-pression of YAP1 gene have been reported in several tumortypes. In meningiomas, YAP1 acts as an oncogene (5). He-patic tumors like cholangiocarcinoma and hepatocellular

Inc.

224 Jaramillo-Rodr�ıguez et al./ Archives of Medical Research 45 (2014) 223e228

carcinoma show increased nuclear expression of YAP1 (2).In hepatocellular carcinoma, increased expression is due togene amplification and transcriptional and post-transcriptional regulation; also, YAP modulates survivinexpression (2). YAP1 overexpression is also described inpancreatic cancer (6).

An increase in YAP nuclear expression has beendescribed in primary human colorectal tumors comparedwith its expression in uninvolved colonic mucosa. In colo-rectal carcinoma, Wnt/b-catenin signaling regulates YAPexpression (7). Overexpression of YAP has also beendescribed in malignant mesothelioma due to the inactiva-tion of NF2, LATS2 and/or SAV1 genes or amplificationof YAP itself (8).

In breast cancer (BC), the role of the YAP gene is stillunder discussion. The location of the gene (11q22.2)frequently shows loss of heterozygosity (LOH) (9). Someauthors, using immunohistochemistry, have described a sig-nificant decrease of YAP protein expression in tumors, andLOH analysis has revealed that protein loss correlates withspecific deletion of the YAP gene locus. These findings sup-port a role as tumor suppressor gene (9). Furthermore, YAPis a co-activator of estrogen (ER) and progesterone (PR) re-ceptors. In invasive carcinoma, reduced expression of YAPis associated with ER negativity (10). The role of YAP inBC requires further studies to define its participation inthe carcinogenic process. The aim of this study was todescribe the expression of YAP by immunohistochemistryin normal breast tissue of women without breast neoplasm,non-neoplastic breast tissue in women with cancer and infour different subtypes of ductal carcinoma.

Materials and Methods

General Procedure

Upon approval of the protocol by the research and ethicscommittees, the pathology files from the hospital werereviewed. Paraffin blocks were obtained from a) breasttissue of autopsy cases (2009e2010) of women between35 and 65 years of age without neoplastic disease, b) breasttissue with and without tumor in patients with invasiveductal carcinoma (surgical cases 2009e2010). All con-secutive patients who had ductal carcinoma and immuno-histochemical study (IHC) for ER, PR and HER2 wereselected. In neoplastic cases, we analyzed the tumors andalso the ‘‘normal’’ tissue present in the biopsy (internalcontrol).

Sample

The sample included 17 controls (breast tissue of womenwithout cancer) and 105 patients with ductal cell carcinoma(53 cases with luminal pattern, 15 with luminal B, 20 withoverexpression of HER2, and 17 triple negative).

Breast Tissue Classification

Samples described as ‘‘controls’’ were obtained from au-topsy cases of women (35e65 years old) with no historyof malignant breast pathology. The tumors were invasiveduct cell carcinomas classified according to the expressionprofile of three markers by IHC: estrogen receptor alpha(ERa), progesterone receptor (PR), and HER2 (11). Fourdifferent profiles were obtained as follows: luminal A(þ,þ,�/þ,�,�/�,þ,�), luminal B (þ,þ,þ/þ,�,þ/�,þ,þ),HER2-overexpressing group (�,�,þ) and triple negativepattern (�,�,�) (11).

Tissue Histology and Immunohistochemistry

Paraffin-embedded sections on polylysine-coated slideswere used for staining. Sections were cut at 4 mm. IHCwas performed as follows: slides were deparaffinized inxylene and rehydrated in a graded alcohol series. Antigenretrieval was achieved by microwaving in 10 mM sodiumcitrate buffer at pH 6 for 10 min. Peroxidase was blockedwith 3% hydrogen peroxide in methanol, and nonspecificprotein binding was blocked with 5% goat serum. The pri-mary antibody for YAP (Cell Signaling Technology, Dan-vers, MA) was diluted with phosphate buffered saline(PBS) containing 5% goat serum (1:100) and incubatedovernight in a humidity chamber at 4�C. After the periodof incubation, samples were washed with PBS and bio-tinylated secondary antibody was then diluted in blockingreagent (1:100) and added. The sample was incubated for30 min at room temperature. Streptavidin was applied for30 min at room temperature. Visualization was performedusing DAB chromogen for 5e10 min.

Control sections were incubated with isotype antibodydiluents instead of primary antibody. Sections were coun-terstained with hematoxylin, dehydrated, and mounted.

Tissue Analysis

All cases were reviewed by two pathologists who wereblinded. At least 500 cells per sample were analyzed in atleast five different areas. Each 4 mm-processed sectionwas digitally captured with a 40x magnification using a6-megapixel Olympus SP-320 camera mounted on anOlympus CX31microscope connected to a PC. Oncecaptured, images were processed using the software ImageJ v1.44 (U.S. National Institutes of Health). We analyzedthe basal and luminal duct cells in non-malignant tissue(normal and internal controls), stromal cells around and be-tween the neoplastic cells, and neoplastic tissue. Positivecells to YAP were counted in a blinded way without knowl-edge of the group they belonged (in the case of neoplasticsamples) by two pathologists. Intra- and inter-observeragreement of the researchers was evaluated at study onsetusing positive slides to YAP. Intra-class correlation coeffi-cient indicated a value of 0.96 and 0.89 for an intra- and

225YAP Expression in Breast Tissue

inter-observer reliability, respectively. We evaluated 1)number of cases with positive staining expressed as a per-centage, 2) in each case number of cells with positiveYAP staining in nucleus and cytoplasm in 500 cells wereanalyzed, 3) cases with positive staining were classified ac-cording to the percentage of positive cells as low positivity(!10% of cells in a sample with positive YAP staining),mild (10e50%) and high (51e100%) (11,12).

Statistical Methods

Statistical data processing was done in Epi-info v. 7.1.0.6program using c2 for linear trend (Extended Mantel-Haenszel). Only those results whose value were p !0.05were considered statistically significant.

Results

General results are shown in Table 1.

Controls

All cases showed positive nuclear staining in 100% of basalcells at normal ducts (Figure 1). Only 3/17 (17.6%) showedcytoplasm staining in the luminal cells and those cases haddiffuse staining in 100% of the cells observed (highstaining).

Patterns in Tumor Cells

Nuclear staining. There was a significant difference be-tween the groups in relation to positive tumor staining.Luminal A cases had barely 4% of cases with positive stain-ing and with !10% of cells affected in each case; thisfinding contrasted with 17e20% of positive cases in the

Table 1. Yes-associated protein (YAP) expression in breast tissue

Control Luminal A

N YAP þ 17/17 2/53

OR 1.00 867.00*

Woolf 95% CI Reference 50.71; 14824.29

c2 (Global test) 5 63.97, df 5 4, p 5 0.0000001 c2 for linear tr

C YAP þ 3/17* 3/53

OR 1.00 3.57

Woolf 95% CI Reference 0.97; 13.20

c2 (Global test)5 6.29, df 5 4, p 5 0.18 c2 for linear trend 5 1

SN YAP þ 0/17 1/53

OR 1.00 1.53

Woolf 95% CI Reference 0.09; 26.14

c2 (Global test) 5 8.65, df 5 4, p 5 0.07 c2 for linear trend 5

NPIC YAP þ 17/17 10/29

OR 1.00 64.60*

Woolf 95% CI Reference 3.61; 1154.63

c2 (Global test) 5 26.42, df 5 4, p 5 0.00003 c2 for linear tren

N, nuclei; C, cytoplasm; SN, stromal nuclei; NPIC, nuclear positivity of interna

other groups, with a high percent of stained cells. In situcomponent, comedo type, showed nuclear staining in my-oepithelial cells; this staining delimited the periphery ofthe structure. Other tumor nuclei scattered in the rest ofthe structure showed also positive staining (Figure 2).

Cytoplasm staining. Significant difference between tumorgroups and controls was not observed. Cytoplasm stainingwas positive in 3/17 controls, 3/53 luminal A, 0/15 luminalB, 2/20 HER2 overexpression and 4/17 triple negativecases. The positive cases showed diffuse cytoplasm stainaffecting O51% of the cells (high) (Figure 3).

Stromal cells. Positive staining was not observed in con-trols or in triple negative pattern. Although a significant dif-ference was not observed between groups, luminal B(Figure 4) was the pattern with the highest percentage ofpositive nuclear stromal cells staining (20%). This positiv-ity included stromal cells surrounding the tumor groups.

Internal controls. In those cases with internal controls, thelocation and extension of the positive YAP staining wasevaluated. Only 28/69 (40%) showed positive expressionin the nuclei of ductal basal cells in contrast to 100% ofnon-neoplastic controls. Therefore, a significant differencewas observed between tumor groups and controls(Table 1). However, although there was a decrease in thepercentage of cases with positive expression, in those withpositive stain, O51% of cells were affected (Figure 5).

Statistical Analysis

Table 1 shows a significant difference between the nuclearexpression of YAP in controls and different types of ductalcarcinoma. There was no significant difference in

Patterns

Luminal B HER2 Triple negative

3/15 4/20 3/17

136.00* 136.00* 158.67*

7.31; 2531.58 7.51; 2462.91 8.66; 2907.55

end 5 8.29, df 5 1, p 5 0.004

0/15 2/20 4/17

6.43 1.93 0.70

1.53; 27.07 0.47; 7.88 0.16; 3.01

.28, df 5 1, p 5 0.258

3/15 1/20 0/17

0.12 0.56 1.00

0.01; 2.19 0.03; 9.99 0.06; 18.02

0.47, df 5 1, p 5 0.493

9/12 6/13 3/15

11.33 39.67* 136.00*

0.45; 288.16 1.97; 797.63 7.31; 2531.58

d 5 10.04, df 5 1, p 5 0.001

l controls; Control*, cytoplasm of luminal ductal cells.

Figure 3. Luminal B pattern-tumor that shows both positive YAP staining

Figure 1. Breast tissue in a woman without carcinoma (control), showing

strong nuclear positivity in basal ductal cells. IHC, YAP staining (x100).

(A color figure can be found in the online version of this article.)

226 Jaramillo-Rodr�ıguez et al./ Archives of Medical Research 45 (2014) 223e228

cytoplasmic expression. The nuclear positivity of internalcontrols was significantly less than in controls withoutcancer.

in nuclei and cytoplasm of tumor cells. IHC, YAP staining (400X). (A co-

lor figure can be found in the online version of this article.)

Discussion

YAP was discovered in 1994. Since then, knowledge of itsfunction has been increased. YAP has a regulatory role insignal transduction and gene transcription in normal cells(13). Hippo signaling pathway is an important molecularmechanism for organ size regulation. Its kinase cascadenegatively regulates the downstream component YAP,through phosphorylation. In normal conditions, phosphory-lation (Ser127) of YAP/TAZ/Yorkie determines its cyto-plasmic localization, induces cell apoptosis and restrictsorgan size overgrowth. When the Hippo pathway is inacti-vated, YAP/TAZ/Yorkie translocates to the nucleus to bind

Figure 2. Invasive ductal cell carcinoma, with a comedo-type ‘‘in situ’’

carcinoma that shows positive nuclear staining in tumor cells at the base

of the structure and other positive nuclei scattered in the tumor. IHC,

YAP staining (x50). (A color figure can be found in the online version

of this article.)

to the transcription enhancer factor (TEAD/TEF) to pro-mote cell growth and proliferation (13). In this study itwas possible to observe normal breast tissue of womenwithout malignant breast pathology. Mammary ducts inthese women showed positive YAP staining in the nucleusof all basal duct cells. This was expected since basal cellsare in replication to replenish the luminal cells. An impor-tant finding was the decrease of positive staining in ‘‘inter-nal controls’’ in the case of patients with cancer. Thismeans that, at least for YAP expression, even the ‘‘normal’’adjacent breast tissue showed decreased expression. Only

Figure 4. Ductal cell carcinoma (Her2 overexpression) shows positive nu-

clear staining in stromal cells surrounding the negative tumor cells. IHC,

YAP staining (x400). (A color figure can be found in the online version

of this article.)

Figure 5. Duct cell carcinoma with triple negative pattern. Internal control

ducts show positive basal nuclear staining. Tumor is negative. YAP anti-

body (x100). (A color figure can be found in the online version of this

article.)

227YAP Expression in Breast Tissue

20% of ‘‘normal ducts’’ (internal controls) showed nuclearstaining in triple negative cases, 34% of luminal A, 46% ofHER2 and 75% of luminal B cases. Overexpressed YAPproteins have been found in human cancers such as liver,gastric, ovarian and esophageal cancer and multiform glio-blastoma, indicating that YAP may participate in tumordevelopment and progression as a carcinogenic factor (14).

In breast tissue, YAP functions partially binding toKLF5, a transcriptional factor that promotes breast cell pro-liferation and survival (15). YAP is inactivated by thetumor-suppressing, Hippo pathway. However, there arecontrasting reports about its function in breast cancer.Studies have demonstrated both oncogenic or tumor sup-pressor functions (16). In the present study, breast tumorswere classified according the expression of ER, PR andHER2 because YAP is a coactivator of ER and PR (10).

A significant difference between cancer groups and con-trols in relation to positive tumor staining was observed.There was no overexpression of YAP, instead, in relationto normal tissue, a remarkable decrease of the expressionin tumor tissue was observed. This finding is consistentwith that found by Yuan (9). LOH analysis has revealed thatprotein loss correlates with specific deletion of the YAPgene locus and support that YAP behaves as a suppressorgene (9). Even more, the decreased expression involvedeven the ‘‘normal’’ breast ducts adjacent to the tumor.The pathologic process affects not only neoplastic tissuebut also histologically ‘‘normal’’ adjacent breast tissue.Cytoplasmic staining did not differ between patients withcancer and controls. Another interesting finding was toobserve positive staining in stromal cells in some types ofcancer patterns. All controls were negative. Luminal Bpattern cases showed positive staining in stromal cells in20% of cases; in those cases with positive nuclei staining

of the stromal cells, the tumors were negative for YAP. Itseems that stromal cells behave differently in differenttypes of carcinoma.

In conclusion, in this work, YAP showed decreasedexpression in tumor cells compared with normal breast tis-sue. ‘‘Normal ducts’’ in tumor cases also showed a decreasein positive staining. Positive staining in the nuclei but not inthe cytoplasm showed significant difference betweendifferent types of carcinoma and controls. Nuclear stainingin the stromal cells adjacent to the tumor is a rare event butwas mainly observed in the luminal B pattern. These find-ings are consistent with a role of YAP as a suppressor genein mammary neoplasms (9).

AcknowledgmentsThe authors acknowledge the financial support provided by theHealth Research Fund of the Mexican Social Security Institute(FIS/IMSS/PROT/G09/743). We also appreciate the valuable com-ments of Dr. Jesus Nares Cisneros.

Conflict of Interest

The authors declare no conflict of interest.

References1. Fujii M. Exploration of a new drug that targets YAP. J Biochem 2012;

152:20911.

2. Bai H, Gayyed MF, Lam-Himlin DM, et al. Expression of Yes-

associated protein modulates Survivin expression in primary liver ma-

lignancies. Hum Pathol 2012;43:1376e1385.

3. Bao Y, Nakagawa K, Yang Z, et al. A cell-based assay to screen stim-

ulators of the Hippo pathway reveals the inhibitory effect of dobut-

amine on the YAP-dependent gene transcription. J Biochem 2011;

150:199e208.

4. Huang JM, Nagatomo I, Suzuki E, et al. YAP modifies cancer cell

sensitivity to EGFR and survivin inhibitors and is negatively regulated

by the non-receptor type protein tyrosine phosphatase 14. Oncogene

2013;32:2220e2229.

5. Baia GS, Caballero OL, Orr BA, et al. Yes-associated protein 1 is acti-

vated and functions as an oncogene in meningiomas. Mol Cancer Res

2012;10:904e913.

6. Diep CH, Zucker KM, Hostetter G, et al. Down-regulation of Yes

Associated Protein 1 expression reduces cell proliferation and clono-

genicity of pancreatic cancer cells. PLoS One 2012;7:e32783.

7. Konsavage WM Jr, Kyler SL, Rennoll SA, et al. Wnt/b-catenin

signaling regulates Yes-associated protein (YAP) gene expression in

colorectal carcinoma cells. J Biol Chem 2012;287:11730e11739.

8. Mizuno T, Murakami H, Fujii M, et al. YAP induces malignant meso-

thelioma cell proliferation by upregulating transcription of cell cycle-

promoting genes. Oncogene 2012;31:5117e5122.

9. Yuan M, Tomlinson V, Lara R, et al. Yes-associated protein (YAP)

functions as a tumor suppressor in breast. Cell Death Differ 2008;

15:1752e1759.

10. Tufail R, Jorda M, Zhao W, et al. Loss of Yes-associated protein

(YAP) expression is associated with estrogen and progesterone recep-

tors negativity in invasive breast carcinomas. Breast Cancer Res Treat

2012;131:743e750.

228 Jaramillo-Rodr�ıguez et al./ Archives of Medical Research 45 (2014) 223e228

11. Calder�on-Garcidue~nas AL, Mart�ınez-Reyes G, Gallardo-G�omez AI,

et al. Triple marker immunohistochemestry analysis in breast cancer

Mexican patients. Patol Rev Latinoam 2012;50:72e79.

12. Rizzardi AE, Johnson AT, Vogel RI, et al. Quantitative comparison of

immunohistochemical staining measured by digital image analysis

versus pathologist visual scoring. Diagn Pathol 2012;20(7):42.

13. Chen L, Loh PG, Song H. Structural and functional insights into the

TEAD-YAP complex in the Hippo signaling pathway. Protein Cell

2010;1:1073e1083.

14. Su LL, Ma WX, Yuan JF, et al. Expression of Yes-associated protein in

non-small cell lung cancer and its relationship with clinical patholog-

ical factors. Chin Med J (Engl) 2012;125:4003e4008.

15. Zhi X, Zhao D, Zhou Z, et al. YAP promotes breast cell proliferation

and survival partially through stabilizing the KLF5 transcription fac-

tor. Am J Pathol 2012;180:2452e2461.16. Wang X, Su L, Ou Q. Yes-associated protein promotes tumor develop-

ment in luminal epithelial derived breast cancer. Eur J Cancer 2012;

48:1227e1234.