A Favorable Role of Prolactin in Human Breast Cancer Reveals Novel
Pathway Based Gene Signatures Indicative of Tumor Differentiation and
Favorable Patient Outcome
Ibrahim Y. Hachim MBChB, MSc, Anwar Shams BMBS, MSc, JeanJacques Lebrun PhD, Suhad Ali PhD
PII:
DOI:
Reference:

S0046-8177(16)00075-7
doi: 10.1016/j.humpath.2016.02.010
YHUPA 3829

To appear in:

Human Pathology

Received date:
Revised date:
Accepted date:

10 December 2015
3 February 2016
12 February 2016

Please cite this article as: Hachim Ibrahim Y., Shams Anwar, Lebrun Jean-Jacques,
Ali Suhad, A Favorable Role of Prolactin in Human Breast Cancer Reveals Novel Pathway Based Gene Signatures Indicative of Tumor Dierentiation and Favorable Patient
Outcome, Human Pathology (2016), doi: 10.1016/j.humpath.2016.02.010

This is a PDF le of an unedited manuscript that has been accepted for publication.
As a service to our customers we are providing this early version of the manuscript.
The manuscript will undergo copyediting, typesetting, and review of the resulting proof
before it is published in its nal form. Please note that during the production process
errors may be discovered which could aect the content, and all legal disclaimers that
apply to the journal pertain.

ACCEPTED MANUSCRIPT
A Favorable Role of Prolactin in Human Breast Cancer Reveals Novel Pathway Based
Gene Signatures Indicative of Tumor Differentiation and Favorable Patient Outcome

RI
P

Prolactin-Induced Mammary Differentiation Program in Breast Cancer Prognosis

Equal Contribution (alphabetic order)

Corresponding Author

MA

NU

SC

Ibrahim Y. Hachim (MBChB, MSc) 1, Anwar Shams (BMBS, MSc)1, Jean-Jacques Lebrun
(PhD)2 and Suhad Ali (PhD)1#
1
Division of Hematology, 2Division of Medical Oncology, Department of Medicine, Cancer
Research Program, Research Institute of the McGill University Health Centre, McGill University

PT

ED

Ibrahim.Hachim@mail.mcgill.ca, Anwar.Shams@mail.mcgill.ca, JJ.Lebrun@mcgill.ca,

Suhad.Ali@mcgill.ca,

AC

Dr. Suhad Ali, PhD

RI-MUHC
Office E026232
1001 Decarie Blvd.
Montreal, Quebec
H4A 3J1
Canada

CE

Contact Information for corresponding author

Tel: 514-934-1934 ex 34-863

E-mail: suhad.ali@mcgill.ca
Key words: Differentiation, prognosis, Jak/Stat, gene signature, mammary
Funding Source: Canadian Institutes of Health Research

ACCEPTED MANUSCRIPT

Abstract

Prolactin (PRL) hormone is known to play a key role in mammary gland development allowing

RI
P

for successful lactation. The role of this hormone in breast tumorigenesis is still controversial.
Here, we evaluated PRL protein and gene expression levels in human breast cancer using tissue

SC

microarray of 100 breast cancer cases, as well as different publically available human breast
cancer gene profiling databases. Interestingly, our results showed a significant down regulation

NU

of PRL expression in breast cancer compared to normal adjacent tissue. Moreover, expression of
PRL was associated with more differentiated tumors, early stage, smaller tumor size and absence

MA

of distant metastasis. Importantly, our results indicate that higher PRL mRNA levels are
significantly associated with prolonged relapse free survival (RFS) in breast cancer patients
(P=3.7E-9). Additionally, examining expression of PRL pathway based gene signature composed

ED

of PRL, PRLR, Jak2 and Stat5a showed a significant association with more differentiated tumors
(P<0.00001), prolonged RFS (P=1.8e-06) as well as overall survival (OS) (P=0.0026). As well,

PT

our results indicate that PRL-directed differentiation program in mammary epithelial cells offer
good prognosis in human breast cancer. Indeed, expression of a gene signature composed of PRL

CE

upregulated genes showed a significant association with well differentiated tumors (P<0.00001).
Whereas expression of a gene signature composed of PRL down regulated genes showed a

AC

significant association with shortened distant metastasis free survival (DMFS) (P=0.0086).
Altogether our results highlight that PRL hormone and its signaling pathway may play an
important role in maintaining tumor differentiation state and in turn better patient outcome.

ACCEPTED MANUSCRIPT

1. Introduction

Mounting evidence suggest that the process of cellular differentiation confers the cells with

RI
P

properties making them resistant to transformation and hence tumorigenesis. Clinically it is

observed that higher tumor grade is usually associated with lower long-term patient survival (1,

SC

2). Indeed, the degree of tumor cell differentiation is an important parameter for prognosis in
breast cancer patients. These observations implicate cellular differentiation pathways as pivotal

NU

in regulating the biological behavior of tumor cells and thereby patient outcome. In spite of this
important role, most of tumor biology studies have focused on molecular pathways involved in

MA

cellular growth, survival, migration and invasion. So far few studies have focused on cellular
differentiation processes and their association with tumorigenesis (3).
In breast tissue, PRL hormone through activation of the PRLR/Jak2/Stat5 pathway is known to

ED

play a key role in mammary gland development and terminal differentiation of the mammary
epithelial cells during lactation (4, 5). While previous studies suggested that PRL plays a role in

PT

the growth of breast cancer cells through PRL/PRLR autocrine function (6, 7) leading to
promoting breast cancer tumorigenesis. However recent study have shown that PRL expression

CE

in breast cancer cell lines and tissues to be low/undetectable (8). We have previously shown that
PRL via its main downstream tyrosine kinase Jak2 regulates epithelial plasticity of breast cancer

AC

cells leading to suppression of their mesenchymal and invasive phenotype (9). Furthermore, our
recent study showed PRLR expression in human breast cancer to be associated with wellestablished good prognostic clinicopathological parameters suggesting that PRL may play a
tumor suppressor role in breast cancer (10). In accordance with these findings
expression/activation of the PRL effector molecule Stat5a was shown to associate positively with
increased levels of histologic differentiation of breast cancer tissues and distinguishes breast
cancer patients with favorable prognosis and response to endocrine therapy (11, 12).
Here, we aimed to evaluate the local expression of PRL in human breast cancer cases in
comparison to normal tissue. Moreover, we investigated the association between PRL expression
levels and well established clinicopathological parameters. Finally, we evaluated the association
between PRL signaling pathway as well as PRL modulated target genes in relation to patient
outcome. Together these studies should help clarify the role of PRL in breast cancer.
3

ACCEPTED MANUSCRIPT
2. Materials and Methods
2.1 Cell culture
HC11 mouse mammary epithelial cells (obtained from N. Hynes (Friedrich Miescher Institute,

RI
P

Basel, Switzerland, 1996) were cultured in RPMI 1640 supplemented with 10% fetal bovine
serum (FBS), mEGF 10ng/ml, and insulin 5g/ml. The human breast cancer cell lines T47D

SC

(obtained from Dr. Morag Park) were cultured in DMEM supplemented with 10% FBS.

2.2 Immunohistochemistry

NU

Tissue microarrays (TMA) of 110 cores were obtained from US BIOMAX Inc. (100 cases of
IDC with various grades and stages plus 10 cases of normal adjacent tissue (NAT). BIOMAX

MA

assured that all patients gave a written informed consent and all the tissues were collected with
high ethical standards under the HIPPA approved protocols. For that reason IRB approval was
not necessary. All the cases were reviewed by anatomic pathologist to ensure that the cores are

ED

pathologically representative of the original tumors and the clinciopathological information were
accurate. In brief, slides were deparaffinized and then rehydrated. After antigen retrieval,

PT

staining was performed using Ultravision LP Detection System (Lab Vision Thermo, USA).
Antibodies used in this study: PRL (Santa Cruz #sc-7805), PRLR-L (Santa Cruz#sc-20992).

CE

Positive controls included are the human breast cancer cells lines T47D and MCF7. All TMA
slides were scanned using image digital capture system, the Aperio XT slide scanner (Leica

AC

Biosystems) at the core facility, Breast Cancer Functional Genomics Group, Goodman Cancer
Research Centre, McGill University. Two investigators evaluated immunostaining independently
and in a blind manner from the clinicopathological data. When there is difference in evaluation,
simultaneous examination by both investigators where done to resolve the discrepancies.
For PRL and PRLR analyses, we used a semiquantitative scoring system as previously described
(13, 14). A score of 0 indicates undetectable expression. Expression <10% of tumor cells, is
considered as (+1). Expression in 10-50% of tumor cells is considered as (+2) and if expression
detected in > 50% of cells it is considered as (+3). Both (0) and (+1) score were considered
negative. The staining was considered positive only if there is membranous and/or granular
cytoplasmic staining in malignant cells.

ACCEPTED MANUSCRIPT
2.3 Kaplan Meier survival analysis
Breast Cancer Kaplan-Meier plotter (2014) which includes 4,142 breast cancer patients was
used to plot patient outcome represented as relapse free survival (RFS) and distant metastasis

free survival (DMFS) and overall survival (OS) (15). Cohorts of patients were separated by

RI
P

median expression using auto-selected best cut off. The PRL pathway gene signature was
obtained using the mean expression levels of PRL , PRLR , Jak2 and Stat5a using the multigene

SC

classifier tool of KM plotter , while the PRL modulated gene signatures was obtained using the
mean expression levels of PRL upregulated genes and PRL downregulated genes respectively

NU

using the same multigene classifier tool in the KM plotter.

MA

2.4 Meta-analysis of PRL pathway genes expression in breast cancer

ONCOMINE database, which consist of large publically available microarray data, was used to
investigate PRL mRNA levels in normal as well as malignant breast cancer cases. For that reason

ED

we used Curtis dataset, which includes 2136 normal and invasive breast cancer cases. Moreover,
we used Curtis dataset as well as Sorlie dataset to investigate the correlation between PRL

PT

mRNA levels and different clinicopathological parameters including grade, stage, tumor size, LN
status and distance metastasis.

CE

Gene expression-based Outcome for Breast cancer Online (GOBO) database , which is another
online analysis tool of 1881 human breast cancer patient , was used to evaluate the association

AC

between PRL pathway gene signature as well as PRL upregulated genes in association with
tumor grade and molecular subtypes.

2.5 Microarray analysis

Gene expression profile of HC11 cells treated with PRL for 24hrs compared to untreated cells
was performed at Genome Quebec, McGill University, Canada.

2.6 Statistical analysis

The analysis of variance (ANOVA) test was used to evaluate the correlation between PRL, PRL
pathway based gene signature, and PRL modulated genes signatures in relation to tumor grade,
ER status and molecular subtypes. Moreover, the log-rank test was performed to predict the 10
year censored overall, distance metastasis and relapse fee survival.
5

ACCEPTED MANUSCRIPT
3. Results
3.1 Association of prolactin protein level with clinicopathological features of breast cancer
patients

For better understanding of the role of PRL hormone in breast tumorigenesis, we examined PRL

RI
P

protein expression level using immunohistochemistry in TMA of 100 human invasive ductal
carcinomas of various grades and stages as well as 10 normal adjacent tissue samples.

SC

Interestingly, as can be seen in (Table 1, Fig. 1a & Supplementary Fig. S1a) the overall PRL
immuno-reactivity was lower in invasive breast cancer 35/97 (36%) compared to normal

NU

adjacent tissue 5/10 (50%).

Moreover, using the same TMA, we further evaluated the association between PRL

MA

immunoreactivity and classical clinicopathological parameters important predictors of patient

survival and outcome (16-18). While, no differences in PRL protein expression level was
observed in relation to tumor stage and size (Table 1), unexpectedly, we observed a minor

ED

difference in PRL protein expression in node negative tumors 26/77 (33.3%) compared to node
positive tumors 9/20 (45%) (Table 1). Moreover, we observed higher PRL protein levels in well-

PT

differentiated (grade I) 9/20 (45%) and moderately differentiated (grade II) 25/66 (37.8%) breast
cancer tumors compared to poorly differentiated tumors (grade III) 1/7 (14.2%) (Fig. 1a & 2a,

CE

Table 1). Next, we investigated the expression of the PRLR using the same TMA mentioned
above. PRLR expression was low/undetectable in all the breast cancer cores examined in

AC

comparison to normal adjacent tissue (Supplementary Fig. S1b and Supplementary Fig. S2).
Taken together the above data indicate that PRL and its receptor are down regulated during the
process of tumorigenesis, further supporting the notion of the tumor suppressor role of PRL
signaling pathway in breast cancer.

ACCEPTED MANUSCRIPT
3.2 Association of prolactin gene expression level with clinicopathological features of breast
cancer patients
To further investigate the role of PRL hormone in breast cancer, we next analysed PRL gene

expression level in a large cohort using Curtis dataset of ONCOMINE database, that includes

RI
P

gene profiling data of 2136 breast cancer cases. Data for PRL gene expression levels were
available for 144 normal tissues and 1556 invasive breast cancer cases. Interestingly, our results

SC

showed a significant downregulation in PRL gene expression level in invasive breast cancer (0.153) compared to normal breast tissue (-0.258) (P= 1.34E-19) (Fig. 1b).

NU

Next, we analysed the association between PRL gene expression level and different
clinicopathological parameters using different datasets of ONCOMINE database. According to

MA

Sorlie dataset containing information for 167 breast cancer cases, the highest PRL mRNA
expression was observed in small tumor T1 (0.332) and least expression was observed in larger
tumor T3 (0.057) and T4 (-0.005) (Fig. 2a). Moreover, using the same dataset higher PRL

ED

mRNA levels was observed in cases with no distant metastasis (0.064) in comparison to cases
with distant metastasis (-0.19) (Fig. 2b). In contrast, less PRL mRNA expression levels were

PT

observed in N0 compared to N1 and N2 (Fig. 2c). Next, we analyzed the association between
PRL mRNA levels and tumor stage, encompassing tumor size (T), lymph node status (N) and

CE

metastases (M) using Curtis dataset which includes 2136 breast cancer cases. Importantly, PRL
mRNA levels were highest in stage 0 (-0.247) and least in advanced stage 4 tumors (-0.329) (Fig.

AC

2d). Finally, a significant association was observed between higher PRL mRNA levels and more
differentiated tumors (P= 0.00168) in a cohort of 1411 breast cancer cases (Fig. 2e). Overall,
these findings together highlight the significant association of PRL gene expression with wellestablished good prognostic clinic-pathological parameters in human breast cancer.

ACCEPTED MANUSCRIPT
3.3 Higher prolactin gene expression levels are significantly associated with better clinical
outcome
To further elaborate the prognostic value of PRL hormone in breast cancer, here we investigated

the role of PRL in association with patient outcome using gene profiling data base of breast

RI
P

cancer, Kaplan-Meier (KM) plotter containing information for 4,142 breast cancer cases (15).
Interestingly, we found that patients with high PRL gene expression levels show significantly

SC

better outcome represented as RFS compared with patients who have low levels (Fig. 2f).

NU

3.4 Prolactin pathway based gene signature and its association with patient outcome
The above findings as well as our previous finding demonstrating PRLR as an independent good

MA

prognostic marker in breast cancer highlight the need to further evaluate the PRL pathway as a
single pathway based gene signature. For that reason we used Gene Set Analysis tool of GOBO
database containing information for 1881 human breast cancer cases to generate a gene signature

ED

representing PRL signaling pathway composed of PRL, PRLR, Jak2 and Stat5a. Next, we
evaluated the prognostic power of this gene signature in relation to tumor grade/differentiation,

PT

ER status and molecular subtypes. Indeed our results showed a significant association between
higher PRL pathway based gene signature and well differentiated tumors (P<0.00001) (Fig. 3a).

CE

Moreover, using the same GOBO dataset as above PRL pathway gene signature showed a
significant association with ER positive tumors (P= 1e05) and luminal A subtype, least

AC

aggressive breast cancer molecular subtype (Fig. 3b & 3c). For more evaluation of the prognostic
ability of PRL pathway gene signature in breast cancer, we next evaluated the association
between this gene signature and patient outcome represented as RFS as well as OS. Interestingly,
patients with higher PRL pathway gene signature showed a significant association with better
patient outcome represented as prolonged RFS (P= 1.8e-0.6) (patient number 1660) and OS
(P=0.026) (patient number 520) (Fig. 3d & 3e).

ACCEPTED MANUSCRIPT
3.5 Prolactin-induced mammary differentiation program in relation to clinicopathological
parameters
The above results prompted us to further characterize the biological effect of PRL pathway in

mammary epithelial cells and its contribution to human breast tumorigenesis. For that reason,

RI
P

we performed exploratory gene profiling microarray analysis of mammary epithelial cells

(HC11) following PRL treatment for 24 hours. These cells represent a well-established model for

SC

examining PRL-mediated differentiation program (19). Interestingly, we found a total of 49

genes modulated (up/down) by PRL 2 folds (Table 2). As expected within the up-regulated

NU

genes we found milk protein genes -casein, WAP and -casein. These findings are consistent
with the well-established effects of PRL in inducing milk protein genes and provide a proof of

MA

principle for validation of the microarray results. As well, we found other Jak/Stat target genes to
be up-regulated such as suppressor of cytokine signaling-2, SOCS2 and cytokine inducible SH2containing protein CISH. The other 44 genes constitute novel PRL target genes. Gene ontology

ED

classification analysis of these PRL modulated genes using the program of data base annotation,
visualization, and integrated discovery (DAVID) demonstrated that these genes belong to various

PT

categories including phosphoproteins, alternative splicing, Zinc finger, milk proteins, repressor,
nucleus, SH-2 domain, Zinc, coiled-coil, signal transduction inhibitor, cell adhesion and DNA

CE

binding (Supplementary Fig. S2, Supplementary Table S1) (20, 21).

For better understanding of the role of PRL modulated genes in mammary epithelial cells in

AC

breast cancer we generated PRL-upregulated gene signature comprising the mean expression
level of all PRL up-regulated genes and PRL down-regulated gene signature comprising the
mean expression level of all PRL down regulated genes (Table 2). Next, we evaluated the
prognostic power of the two gene signatures in association with clinicopathological parameters,
as well as patient outcome using GOBO and KM plotter databases. As can be seen in Figure 4a,
PRL-upregulated gene signature showed significant association with well-moderately
differentiated tumors (P<0.00001) (patient number 1411). Similarly, PRL-upregulated gene
signature showed higher expression in luminal A breast cancer subtype and least expression in
the highly aggressive basal breast cancer subtype (Fig. 4b) (patient number 1881). Furthermore,
PRL-upregulated gene signature showed a weak positive trend with prolonged RFS (P=0.09)
(Fig. 4c) (patient number 1660). On the other hand, PRL-downregulated gene signature, while
still associated with luminal A subtype (data not shown), showed a significant association with
9

ACCEPTED MANUSCRIPT
shortened distant metastasis free survival (DMFS) (P=0.0086) (patient number 664) (Fig. 4d).
Altogether, this data highlight that PRL-mediated mammary differentiation axis provide a
powerful prognostic tool in human breast cancer and indicate the crucial role of PRL modulating

AC

CE

PT

ED

MA

NU

SC

RI
P

tumor phenotype/behavior.

10

ACCEPTED MANUSCRIPT
4. Discussion
The level of differentiation of breast cancer cells are believed to play an important role in

regulating the growth capacity and metastatic potential. Therefore, analysis of cellular

RI
P

differentiation pathways should provide novel approaches for the management of breast cancer.
The hormone PRL is an indispensable lactogenic hormone that is crucial in inducing/maintaining
differentiation of mammary epithelial cells during lactation. Characterizing the role of PRL in

SC

human breast cancer is vitally important and is still a subject of debate. Indeed, it is proposed
that PRL working through its receptor in breast cancer cells in an autocrine/paracrine manner

NU

contribute as a growth factor and thereby blocking this pathway would be of therapeutic benefit
(22, 23).

MA

To further characterize the role of PRL in breast carcinogenesis, we examined the local
expression of PRL and its prognostic role in human breast cancer using commercially available
human TMA of 100 cases of invasive ductal carcinoma as well as publically available gene

ED

profiling databases of breast cancer human samples. Our results revealed that both PRL protein
as well as mRNA levels are down regulated in breast cancer cases in comparison to normal
This finding is similar to recent report showing PRL expression levels to be

PT

tissue.

low/undetectable in breast cancer cell lines and patient samples (8). Together, these results

CE

suggest that PRL signaling is downregulated in breast cancer development.

Further analysis of PRL protein expression and its association with classical clinicopathological

AC

parameters revealed no significant association with tumor size and stage. Interestingly, however,
our in silica analysis using large publically available ONCOMINE database revealed high
expression of PRL mRNA to be associated with early tumor stage, smaller tumor size and
absence of distant metastasis.

Moreover, higher PRL mRNA levels showed a significant

association with favorable patient outcome represented as prolonged relapse free survival. Those
findings elaborate the significant association between high PRL expression levels and well
established favorable prognostic features. These results also imply that PRL may be utilized as a
single biomarker predicting favorable patient outcome.
Due to the important regulatory role of PRL pathway in maintaining and regulating the terminal
differentiation of mammary epithelial cells, we studied the association between PRL protein and
mRNA levels and tumor differentiation state. Indeed PRL protein and mRNA levels showed
higher expression in well differentiated tumors and least expression in poorly differentiated
11

ACCEPTED MANUSCRIPT
tumors. Furthermore, PRL pathway based gene signatures representing PRL signaling
components or PRL-up regulated target genes identified in mammary epithelial cells showed the
same significant association with well differentiated tumors and least-aggressive luminal A
Together, these results elucidate the central role of PRL signaling

breast cancer subtype.

RI
P

pathway in maintaining the luminal-epithelial-differentiated state of tumors and in turn

determining the tumor phenotype and behavior.

SC

Our results suggest that patient stratification according to PRL pathway based gene signatures
and PRL modulated gene signatures might provide more precise tool for prognosis, prediction

NU

and treatment development. Indeed, our results revealed a significant association between high
PRL pathway based gene signatures with better patient outcome represented as prolonged RFS

MA

and/or OS. In contrast, gene signature representing PRL downregulated genes showed significant
association with poor patient outcome. Altogether our results highlight the favorable prognostic
role of PRL and its signaling pathway in breast cancer and the possibility of using PRL or PRL

ED

pathway based gene signature as a tool for patient outcome assessment.

Moreover, recently developed diagnostic tools such as MammaPrint and Oncotype DX might

PT

benefit from incorporating PRL pathway based gene signatures described here. Indeed, both of
those diagnostic tools are based on genes associated with proliferation, ER, HER2, metastasis,

CE

invasion and angiogenesis but mainly lack differentiation associated genes. For that reason,
further studies that includes larger patients cohorts are needed to study the role of incorporating

AC

differentiation pathways such as PRL pathway into those diagnostic tools. Moreover, one of the
limitations of our study that needs to be addressed in future studies is to evaluate the association
between PRL expression and classical biomarkers of breast cancer including ER/PR/Her2 and
Ki67. This information would help in characterizing the role of PRL in different breast cancer
molecular subtypes.

In summary, this data suggests that PRL and its signaling pathway through their prodifferentiation function in normal mammary gland might provide a significant protective role in
normal mammary epithelium and down regulation/loss of this mechanism might play a role in
the process of breast carcinogenesis.

12

ACCEPTED MANUSCRIPT

Acknowledgements
This work was supported by The Canadian Institutes of Health Research (CIHR-MOP115105)

AC

CE

PT

ED

MA

NU

SC

RI
P

granted to Suhad Ali.

13

ACCEPTED MANUSCRIPT
Authors Contributions

IH: Performed analyses of IHC as well as bioinformatical analysis, statistical analysis and

contributed to drafting the article, AS: Performed experiments for microarray stud, performed

RI
P

analyses of IHC, and contributed to drafting the article, JJL: Contributed design and revising the

AC

CE

PT

ED

MA

NU

SC

article, SA: Principal research design, supervision and writing the article

14

ACCEPTED MANUSCRIPT

The authors declare no conflict of interest

All authors have read and approved the manuscript

AC

CE

PT

ED

MA

NU

SC

RI
P

Conflict of Interest

15

ACCEPTED MANUSCRIPT
References

AC

CE

PT

ED

MA

NU

SC

RI
P

1.
Soerjomataram I, Louwman MW, Ribot JG, Roukema JA, Coebergh JW. An overview of
prognostic factors for long-term survivors of breast cancer. Breast cancer research and treatment
2008; 107, 309-330.
2.
Arriagada R, Le MG, Dunant A, Tubiana M, Contesso G. Twenty-five years of follow-up
in patients with operable breast carcinoma: correlation between clinicopathologic factors and the
risk of death in each 5-year period. Cancer 2006; 106, 743-750.
3.
Jogi A, Vaapil M, Johansson M, Pahlman S. Cancer cell differentiation heterogeneity and
aggressive behavior in solid tumors. Upsala journal of medical sciences 2012; 117, 217-224.
4.
Bole-Feysot C, Goffin V, Edery M, Binart N, Kelly PA. Prolactin (PRL) and its receptor:
actions, signal transduction pathways and phenotypes observed in PRL receptor knockout mice.
Endocr Rev 1998; 19, 225-268.
5.
Oakes SR, Rogers RL, Naylor MJ, Ormandy CJ. Prolactin regulation of mammary gland
development. Journal of mammary gland biology and neoplasia 2008; 13, 13-28.
6.
Fernandez I, Touraine P, Goffin V. Prolactin and human tumourogenesis. Journal of
neuroendocrinology 2010; 22, 771-777.
7.
Arendt LM, Rose-Hellekant TA, Sandgren EP, Schuler LA. Prolactin potentiates
transforming growth factor alpha induction of mammary neoplasia in transgenic mice. The
American journal of pathology 2006; 168, 1365-1374.
8.
Nitze LM, Galsgaard ED, Din N, Lund VL, Rasmussen BB, Berchtold MW, Christensen
L, Panina S. Reevaluation of the proposed autocrine proliferative function of prolactin in breast
cancer. Breast Cancer Res Treat 2013; 142, 31-44.
9.
Nouhi Z, Chughtai N, Hartley S, Cocolakis E, Lebrun JJ, Ali S. Defining the role of
prolactin as an invasion suppressor hormone in breast cancer cells. Cancer Res 2006; 66, 18241832.
10.
Hachim IY, Hachim MY, Lopez VM, Lebrun J-J, Ali S. Prolactin Receptor Expression is
an Independent Favorable Prognostic Marker in Human Breast Cancer. Applied
Immunohistochemistry & Molecular Morphology 2014; Accepted for publication.
11.
Sultan AS, Brim H, Sherif ZA. Co-overexpression of Janus kinase 2 and signal
transducer and activator of transcription 5a promotes differentiation of mammary cancer cells
through reversal of epithelial-mesenchymal transition. Cancer Sci 2008; 99, 272-279.
12.
Yamashita H, Nishio M, Ando Y, Zhang Z, Hamaguchi M, Mita K, Kobayashi S, Fujii Y,
Iwase H. Stat5 expression predicts response to endocrine therapy and improves survival in
estrogen receptor-positive breast cancer. Endocr Relat Cancer 2006; 13, 885-893.
13.
Bhatavdekar JM, Patel DD, Shah NG, Vora HH, Suthar TP, Ghosh N, Chikhlikar PR,
Trivedi TI. Prolactin as a local growth promoter in patients with breast cancer: GCRI experience.
European journal of surgical oncology : the journal of the European Society of Surgical
Oncology and the British Association of Surgical Oncology 2000; 26, 540-547.
14.
Pedrini JL, Francalacci Savaris R, Casales Schorr M, Cambruzi E, Grudzinski M, Zettler
CG. The effect of neoadjuvant chemotherapy on hormone receptor status, HER2/neu and
prolactin in breast cancer. Tumori 2011; 97, 704-710.

16

ACCEPTED MANUSCRIPT

AC

CE

PT

ED

MA

NU

SC

RI
P

15.
Gyorffy B, Lanczky A, Eklund AC, Denkert C, Budczies J, Li Q, Szallasi Z. An online
survival analysis tool to rapidly assess the effect of 22,277 genes on breast cancer prognosis
using microarray data of 1,809 patients. Breast cancer research and treatment 2010; 123, 725731.
16.
Carter CL, Allen C, Henson DE. Relation of tumor size, lymph node status, and survival
in 24,740 breast cancer cases. Cancer 1989; 63, 181-187.
17.
Rosenberg J, Chia YL, Plevritis S. The effect of age, race, tumor size, tumor grade, and
disease stage on invasive ductal breast cancer survival in the U.S. SEER database. Breast cancer
research and treatment 2005; 89, 47-54.
18.
Wu SG, He ZY, Li Q, Sun JY, Li FY, Lin Q, Lin HX, Guan XX. Prognostic value of
metastatic axillary lymph node ratio for Chinese breast cancer patients. PLoS One 2013; 8,
e61410.
19.
Ball RK, Friis RR, Schoenenberger CA, Doppler W, Groner B. Prolactin regulation of
beta-casein gene expression and of a cytosolic 120-kd protein in a cloned mouse mammary
epithelial cell line. Embo J 1988; 7, 2089-2095.
20.
Huang da W, Sherman BT, Lempicki RA. Systematic and integrative analysis of large
gene lists using DAVID bioinformatics resources. Nature protocols 2009; 4, 44-57.
21.
Huang da W, Sherman BT, Lempicki RA. Bioinformatics enrichment tools: paths toward
the comprehensive functional analysis of large gene lists. Nucleic acids research 2009; 37, 1-13.
22.
Jacobson EM, Hugo ER, Tuttle TR, Papoian R, Ben-Jonathan N. Unexploited therapies
in breast and prostate cancer: blockade of the prolactin receptor. Trends in endocrinology and
metabolism: TEM 2010; 21, 691-698.
23.
Damiano JS, Wasserman E. Molecular pathways: blockade of the PRLR signaling
pathway as a novel antihormonal approach for the treatment of breast and prostate cancer.
Clinical cancer research : an official journal of the American Association for Cancer Research
2013; 19, 1644-1650.

17

ACCEPTED MANUSCRIPT
Figure Legends
Figure 1: Prolactin expression in invasive breast cancer cases
a) H & E stain with examples of positive and negative IHC stain in NAT and different tumor

grades for PRL.

RI
P

b) Gene expression levels of PRL in normal and human breast cancer samples using Curtis

SC

dataset of ONCOMINE database

Figure 2: Prolactin expression and its association with different clinicopathological

NU

parameters and patient outcome in breast cancer

a) PRL mRNA level and its association with tumor size using Sorlie dataset of ONCOMINE

MA

database.

b) PRL mRNA level and its association with distant metastasis status using Sorlie dataset of
ONCOMINE database.

ED

c) PRL mRNA level and its association with LN status using Curtis dataset of ONCOMINE
database.

PT

d) PRL mRNA level and its association with tumor stage using Curtis dataset of ONCOMINE
database.

CE

e) IHC expression of PRL protein in different tumor grades (left panel), PRL mRNA levels and
its association with tumor grade using GOBO database (right panel).

AC

f) PRL mRNA level and it association with relapse free survival using Kaplan-Meier survival
curves plotting tool

Figure 3: Prolactin pathway based gene signature and its association with patient outcome
a) PRL pathway gene signature and its association with tumor grade using GOBO database.
b) PRL pathway gene signature and its association with ER status using GOBO database.
c) PRL pathway gene signature and its association with molecular subtypes using PAM50
subclassification method in GOBO database.
d) PRL pathway gene signature and it association with RFS using Kaplan-Meier survival
curves plotting tool.
e) PRL pathway gene signature and it association with OS using Kaplan-Meier survival curves
plotting tool.
18

ACCEPTED MANUSCRIPT
Figure 4: Prolactin modulated gene signatures in mammary epithelial cells and their
association with clinicopathological parameters and patient outcome.
a) PRL upregulated gene signature and its association with tumor grade using GOBO

database.

RI
P

b) PRL upregulated gene signature and its association with molecular subtypes using
PAM50 subclassification method in GOBO database.

SC

c) PRL upregulated gene signature and it association with RFS using Kaplan-Meier survival
curves plotting tool.

NU

d) PRL downregulated gene signature and its association with OS using Kaplan-Meier

MA

survival curves plotting tool.

Table legends

ED

Table 1: IHC analysis of PRL and its association with different tumor characteristics

Table 2: PRL Modulated Genes in HC11 Mammary Epithelial Cells

PT

Microarray analysis of HC11 cells treated or not with PRL for a period of 24hrs was carried out.
Data presented are of PRL-target genes modulated (up/down) by 2 folds.

CE

Confirmation by RT-qPCR was done for selected genes as shown (ND, Not determined).

AC

Legends to Supplemental Figures and Tables

Supplementary Fig. S1: PRL and PRLR expression in breast cancer cell lines as well as
human breast cancer tissue microarray
a) PRL expression in MCF7 as a positive control as well as TMA of 110 breast cancer cases
and normal adjacent tissue.
b) PRLR expression in T47D as a positive control as well as TMA of 110 breast cancer
cases and normal adjacent tissue
Supplementary Fig. S2: A representative of IHC staining of PRLR in NAT and grade II
tumor
Supplementary Table S1: Gene ontology classification analysis of PRL modulated genes
Gene ontology classification of PRL-modulated genes identified in Table 2.

19

ACCEPTED MANUSCRIPT
Prolactin Expression
Positive
Negative
5 (50%)
5 (50%)
35(36.1%) 62(63.9%)

11(55%)
41(62.1)
6 (85.7%)

SC

9(45%)
25(37.8%)
1(14.29%)

Total
10
97

RI
P

Parameter
Lesion
Normal adjacent tissue
Invasive breast cancer
Grade (a)
I
II
III
Stage
I
II
III, IV
Tumor size (b)
2
>2
LN
Negative
Positive

20
66
7
14
65
18

5(33.3%)
24(34.7%)

10(66.7%)
45(65.2%)

15
69

26 (33.7%)
9 (45%)

51(66.3%)
11(55%)

77
20

ED

MA

NU

5(35.7%)
9(64.3%)
23(35.38%) 42(64.6%)
7(38.89%) 11(61.1%)

AC

CE

PT

(a) Four cases had missing grade data

(b) T4 cases were not involved as they represent tumors of any size growing
into the chest wall or skin.

Table 1

20

ACCEPTED MANUSCRIPT
Table 2: PRL-modulated genes in HC11 cells
Modulation

Tnc
Socs2
Btn1a1
Csn2
Csn1s1
Col9a1
Cish
Wfdc12
Anp32a
Amigo2
Sh3d19
Bcl6
Zbtb20
Dock7
Brcc3
Luc7l2
Akap9
Foxp1
Hmox1
Fut9
Fus
Prmt1
Ykt6
Cdk6
Ssbp2
Eps15
Fam188a
Ccny
Ctnna1
Nudt21
Exoc5
Zcchc7
Tcf12
Erbb2ip
Gsk3b
Dusp16
Mbtd1
Phf21a
Sorbs2
Cux1
Nbeal1
Zfr
Rabgap1
Cstf3
Kif5b
Chka
Nufip2
Cblb
Fat4

tenascin C
suppressor of cytokine signaling 2
butyrophilin, subfamily 1, member A1
casein beta
casein alpha s1
collagen, type IX, alpha 1
cytokine inducible SH2-containing protein
WAP four-disulfide core domain 12
acidic (leucine-rich) nuclear phosphoprotein 32 family, member A
adhesion molecule with Ig like domain 2
SH3 domain protein D19
B-cell leukemia/lymphoma 6
zinc finger and BTB domain containing 20
dedicator of cytokinesis 7
BRCA1/BRCA2-containing complex, subunit 3
LUC7-like 2 (S. cerevisiae)
kinase (PRKA) anchor protein (yotiao) 9A
Forkhead box P1
heme oxygenase (decycling) 1
fucosyltransferase 9
fusion, derived from t(12;16) malignant liposarcoma (human)
protein arginine N-methyltransferase 1
YKT6 homolog (S. Cerevisiae)
cyclin-dependent kinase 6
single-stranded DNA binding protein 2
epidermal growth factor receptor pathway substrate 15
family with sequence similarity 188, member A
cyclin Y
catenin (cadherin associated protein), alpha 1
nudix (nucleoside diphosphate linked moiety X)-type motif 21
exocyst complex component 5
zinc finger, CCHC domain containing 7
Transcription factor 12
Erbb2 interacting protein
glycogen synthase kinase 3 beta
dual specificity phosphatase 16
mbt domain containing 1
PHD finger protein 21A
sorbin and SH3 domain containing 2
Cut-like homeobox 1
neurobeachin like 1
Zinc finger RNA binding protein
RAB GTPase activating protein 1
cleavage stimulation factor, 3' pre-RNA, subunit 3
Kinesin family member 5B
choline kinase alpha
nuclear fragile X mental retardation protein interacting protein 2
Casitas B-lineage lymphoma b
FAT tumor suppressor homolog 4 (Drosophila)

SC

NU

MA

ED

PT

CE

AC

Gene descriptor

RI
P

Gene code

Confirmation
by qPCR
1.12
4
2.09
103.96
4.05
ND
2.15
ND
ND
ND
ND
ND
0.4
ND
0.95
ND
ND
ND
ND
ND
ND
0.77
0.73
ND
ND
ND
ND
ND
ND
0.26
0.67
ND
ND
ND
ND
ND
ND
ND
0.45
ND
ND
0.37
ND
ND
0.78
ND
ND
ND
ND

21

ACCEPTED MANUSCRIPT
List of abbreviations

CE

PT

ED

MA

NU

SC

RI
P

Distant metastasis free survival

Fetal bovine serum
Health Insurance Portability and Accountability Act
Institutional review board
Normal adjacent tissue
Overall survival
Prolactin
Prolactin receptor
Relapse free survival
Tissue microarray

AC

DMFS
FBS
HIPPA
IRB
NAT
OS
PRL
PRLR
RFS
TMA

22

AC

CE

PT

ED

MA

NU

SC

RI
P

ACCEPTED MANUSCRIPT

23

AC

CE

PT

ED

MA

NU

SC

RI
P

ACCEPTED MANUSCRIPT

24

AC

CE

PT

ED

MA

NU

SC

RI
P

ACCEPTED MANUSCRIPT

25

Figure 4

AC

CE

PT

ED

MA

NU

SC

RI
P

ACCEPTED MANUSCRIPT

26